1 // -*- mode:C++; tab-width:8; c-basic-offset:2; indent-tabs-mode:t -*-
2 // vim: ts=8 sw=2 smarttab
3
4 #include "acconfig.h"
5 #include "options.h"
6 #include "common/Formatter.h"
7
8 // Helpers for validators
9 #include "include/stringify.h"
10 #include <boost/algorithm/string.hpp>
11 #include <boost/lexical_cast.hpp>
12 #include <regex>
13
14 // Definitions for enums
15 #include "common/perf_counters.h"
16
17 // rbd feature validation
18 #include "librbd/Features.h"
19
20 namespace {
21 class printer : public boost::static_visitor<> {
22 ostream& out;
23 public:
24 explicit printer(ostream& os)
25 : out(os) {}
26 template<typename T>
27 void operator()(const T& v) const {
28 out << v;
29 }
30 void operator()(boost::blank blank) const {
31 return;
32 }
33 void operator()(bool v) const {
34 out << (v ? "true" : "false");
35 }
36 void operator()(double v) const {
37 out << std::fixed << v << std::defaultfloat;
38 }
39 void operator()(const Option::size_t& v) const {
40 out << v.value;
41 }
42 void operator()(const std::chrono::seconds v) const {
43 out << v.count();
44 }
45 };
46 }
47
48 ostream& operator<<(ostream& os, const Option::value_t& v) {
49 printer p{os};
50 v.apply_visitor(p);
51 return os;
52 }
53
54 void Option::dump_value(const char *field_name,
55 const Option::value_t &v, Formatter *f) const
56 {
57 if (boost::get<boost::blank>(&v)) {
58 // This should be nil but Formatter doesn't allow it.
59 f->dump_string(field_name, "");
60 return;
61 }
62 switch (type) {
63 case TYPE_INT:
64 f->dump_int(field_name, boost::get<int64_t>(v)); break;
65 case TYPE_UINT:
66 f->dump_unsigned(field_name, boost::get<uint64_t>(v)); break;
67 case TYPE_STR:
68 f->dump_string(field_name, boost::get<std::string>(v)); break;
69 case TYPE_FLOAT:
70 f->dump_float(field_name, boost::get<double>(v)); break;
71 case TYPE_BOOL:
72 f->dump_bool(field_name, boost::get<bool>(v)); break;
73 default:
74 f->dump_stream(field_name) << v; break;
75 }
76 }
77
78 int Option::pre_validate(std::string *new_value, std::string *err) const
79 {
80 if (validator) {
81 return validator(new_value, err);
82 } else {
83 return 0;
84 }
85 }
86
87 int Option::validate(const Option::value_t &new_value, std::string *err) const
88 {
89 // Generic validation: min
90 if (!boost::get<boost::blank>(&(min))) {
91 if (new_value < min) {
92 std::ostringstream oss;
93 oss << "Value '" << new_value << "' is below minimum " << min;
94 *err = oss.str();
95 return -EINVAL;
96 }
97 }
98
99 // Generic validation: max
100 if (!boost::get<boost::blank>(&(max))) {
101 if (new_value > max) {
102 std::ostringstream oss;
103 oss << "Value '" << new_value << "' exceeds maximum " << max;
104 *err = oss.str();
105 return -EINVAL;
106 }
107 }
108
109 // Generic validation: enum
110 if (!enum_allowed.empty() && type == Option::TYPE_STR) {
|
(1) Event fun_call_w_exception: |
Called function throws an exception of type "boost::exception_detail::clone_impl<boost::exception_detail::error_info_injector<boost::bad_get> >". [details] |
111 auto found = std::find(enum_allowed.begin(), enum_allowed.end(),
112 boost::get<std::string>(new_value));
113 if (found == enum_allowed.end()) {
114 std::ostringstream oss;
115 oss << "'" << new_value << "' is not one of the permitted "
116 "values: " << joinify(enum_allowed.begin(),
117 enum_allowed.end(),
118 std::string(", "));
119 *err = oss.str();
120 return -EINVAL;
121 }
122 }
123
124 return 0;
125 }
126
127 int Option::parse_value(
128 const std::string& raw_val,
129 value_t *out,
130 std::string *error_message,
131 std::string *normalized_value) const
132 {
133 std::string val = raw_val;
134
135 int r = pre_validate(&val, error_message);
136 if (r != 0) {
137 return r;
138 }
139
140 if (type == Option::TYPE_INT) {
141 int64_t f = strict_si_cast<int64_t>(val.c_str(), error_message);
142 if (!error_message->empty()) {
143 return -EINVAL;
144 }
145 *out = f;
146 } else if (type == Option::TYPE_UINT) {
147 uint64_t f = strict_si_cast<uint64_t>(val.c_str(), error_message);
148 if (!error_message->empty()) {
149 return -EINVAL;
150 }
151 *out = f;
152 } else if (type == Option::TYPE_STR) {
153 *out = val;
154 } else if (type == Option::TYPE_FLOAT) {
155 double f = strict_strtod(val.c_str(), error_message);
156 if (!error_message->empty()) {
157 return -EINVAL;
158 } else {
159 *out = f;
160 }
161 } else if (type == Option::TYPE_BOOL) {
162 if (strcasecmp(val.c_str(), "false") == 0) {
163 *out = false;
164 } else if (strcasecmp(val.c_str(), "true") == 0) {
165 *out = true;
166 } else {
167 int b = strict_strtol(val.c_str(), 10, error_message);
168 if (!error_message->empty()) {
169 return -EINVAL;
170 }
171 *out = (bool)!!b;
172 }
173 } else if (type == Option::TYPE_ADDR) {
174 entity_addr_t addr;
175 if (!addr.parse(val.c_str())){
176 return -EINVAL;
177 }
178 *out = addr;
179 } else if (type == Option::TYPE_ADDR) {
180 entity_addrvec_t addr;
181 if (!addr.parse(val.c_str())){
182 return -EINVAL;
183 }
184 *out = addr;
185 } else if (type == Option::TYPE_UUID) {
186 uuid_d uuid;
187 if (!uuid.parse(val.c_str())) {
188 return -EINVAL;
189 }
190 *out = uuid;
191 } else if (type == Option::TYPE_SIZE) {
192 Option::size_t sz{strict_iecstrtoll(val.c_str(), error_message)};
193 if (!error_message->empty()) {
194 return -EINVAL;
195 }
196 *out = sz;
197 } else if (type == Option::TYPE_SECS) {
198 try {
199 *out = parse_timespan(val);
200 } catch (const invalid_argument& e) {
201 *error_message = e.what();
202 return -EINVAL;
203 }
204 } else {
205 ceph_abort();
206 }
207
|
(1) Event fun_call_w_exception: |
Called function throws an exception of type "boost::exception_detail::clone_impl<boost::exception_detail::error_info_injector<boost::bad_get> >". [details] |
208 r = validate(*out, error_message);
209 if (r != 0) {
210 return r;
211 }
212
213 if (normalized_value) {
214 *normalized_value = to_str(*out);
215 }
216 return 0;
217 }
218
219 void Option::dump(Formatter *f) const
220 {
221 f->dump_string("name", name);
222
223 f->dump_string("type", type_to_str(type));
224
225 f->dump_string("level", level_to_str(level));
226
227 f->dump_string("desc", desc);
228 f->dump_string("long_desc", long_desc);
229
230 dump_value("default", value, f);
231 dump_value("daemon_default", daemon_value, f);
232
233 f->open_array_section("tags");
234 for (const auto t : tags) {
235 f->dump_string("tag", t);
236 }
237 f->close_section();
238
239 f->open_array_section("services");
240 for (const auto s : services) {
241 f->dump_string("service", s);
242 }
243 f->close_section();
244
245 f->open_array_section("see_also");
246 for (const auto sa : see_also) {
247 f->dump_string("see_also", sa);
248 }
249 f->close_section();
250
251 if (type == TYPE_STR) {
252 f->open_array_section("enum_values");
253 for (const auto &ea : enum_allowed) {
254 f->dump_string("enum_value", ea);
255 }
256 f->close_section();
257 }
258
259 dump_value("min", min, f);
260 dump_value("max", max, f);
261
262 f->dump_bool("can_update_at_runtime", can_update_at_runtime());
263
264 f->open_array_section("flags");
265 if (has_flag(FLAG_RUNTIME)) {
266 f->dump_string("option", "runtime");
267 }
268 if (has_flag(FLAG_NO_MON_UPDATE)) {
269 f->dump_string("option", "no_mon_update");
270 }
271 if (has_flag(FLAG_STARTUP)) {
272 f->dump_string("option", "startup");
273 }
274 if (has_flag(FLAG_CLUSTER_CREATE)) {
275 f->dump_string("option", "cluster_create");
276 }
277 if (has_flag(FLAG_CREATE)) {
278 f->dump_string("option", "create");
279 }
280 f->close_section();
281 }
282
283 std::string Option::to_str(const Option::value_t& v)
284 {
285 return stringify(v);
286 }
287
288 void Option::print(ostream *out) const
289 {
290 *out << name << " - " << desc << "\n";
291 *out << " (" << type_to_str(type) << ", " << level_to_str(level) << ")\n";
292 if (!boost::get<boost::blank>(&daemon_value)) {
293 *out << " Default (non-daemon): " << stringify(value) << "\n";
294 *out << " Default (daemon): " << stringify(daemon_value) << "\n";
295 } else {
296 *out << " Default: " << stringify(value) << "\n";
297 }
298 if (!enum_allowed.empty()) {
299 *out << " Possible values: ";
300 for (auto& i : enum_allowed) {
301 *out << " " << stringify(i);
302 }
303 *out << "\n";
304 }
305 if (!boost::get<boost::blank>(&min)) {
306 *out << " Minimum: " << stringify(min) << "\n"
307 << " Maximum: " << stringify(max) << "\n";
308 }
309 *out << " Can update at runtime: "
310 << (can_update_at_runtime() ? "true" : "false") << "\n";
311 if (!services.empty()) {
312 *out << " Services: " << services << "\n";
313 }
314 if (!tags.empty()) {
315 *out << " Tags: " << tags << "\n";
316 }
317 if (!see_also.empty()) {
318 *out << " See also: " << see_also << "\n";
319 }
320
321 if (long_desc.size()) {
322 *out << "\n" << long_desc << "\n";
323 }
324 }
325
326 constexpr unsigned long long operator"" _min (unsigned long long min) {
327 return min * 60;
328 }
329 constexpr unsigned long long operator"" _hr (unsigned long long hr) {
330 return hr * 60 * 60;
331 }
332 constexpr unsigned long long operator"" _day (unsigned long long day) {
333 return day * 60 * 60 * 24;
334 }
335 constexpr unsigned long long operator"" _K (unsigned long long n) {
336 return n << 10;
337 }
338 constexpr unsigned long long operator"" _M (unsigned long long n) {
339 return n << 20;
340 }
341 constexpr unsigned long long operator"" _G (unsigned long long n) {
342 return n << 30;
343 }
344 constexpr unsigned long long operator"" _T (unsigned long long n) {
345 return n << 40;
346 }
347
348 std::vector<Option> get_global_options() {
349 return std::vector<Option>({
350 Option("host", Option::TYPE_STR, Option::LEVEL_BASIC)
351 .set_description("local hostname")
352 .set_long_description("if blank, ceph assumes the short hostname (hostname -s)")
353 .set_flag(Option::FLAG_NO_MON_UPDATE)
354 .add_service("common")
355 .add_tag("network"),
356
357 Option("fsid", Option::TYPE_UUID, Option::LEVEL_BASIC)
358 .set_description("cluster fsid (uuid)")
359 .set_flag(Option::FLAG_NO_MON_UPDATE)
360 .set_flag(Option::FLAG_STARTUP)
361 .add_service("common")
362 .add_tag("service"),
363
364 Option("public_addr", Option::TYPE_ADDR, Option::LEVEL_BASIC)
365 .set_description("public-facing address to bind to")
366 .set_flag(Option::FLAG_STARTUP)
367 .add_service({"mon", "mds", "osd", "mgr"}),
368
369 Option("public_bind_addr", Option::TYPE_ADDR, Option::LEVEL_ADVANCED)
370 .set_default(entity_addr_t())
371 .set_flag(Option::FLAG_STARTUP)
372 .add_service("mon")
373 .set_description(""),
374
375 Option("cluster_addr", Option::TYPE_ADDR, Option::LEVEL_BASIC)
376 .set_description("cluster-facing address to bind to")
377 .add_service("osd")
378 .set_flag(Option::FLAG_STARTUP)
379 .add_tag("network"),
380
381 Option("public_network", Option::TYPE_STR, Option::LEVEL_ADVANCED)
382 .add_service({"mon", "mds", "osd", "mgr"})
383 .set_flag(Option::FLAG_STARTUP)
384 .add_tag("network")
385 .set_description("Network(s) from which to choose a public address to bind to"),
386
387 Option("public_network_interface", Option::TYPE_STR, Option::LEVEL_ADVANCED)
388 .add_service({"mon", "mds", "osd", "mgr"})
389 .add_tag("network")
390 .set_flag(Option::FLAG_STARTUP)
391 .set_description("Interface name(s) from which to choose an address from a public_network to bind to; public_network must also be specified.")
392 .add_see_also("public_network"),
393
394 Option("cluster_network", Option::TYPE_STR, Option::LEVEL_ADVANCED)
395 .add_service("osd")
396 .set_flag(Option::FLAG_STARTUP)
397 .add_tag("network")
398 .set_description("Network(s) from which to choose a cluster address to bind to"),
399
400 Option("cluster_network_interface", Option::TYPE_STR, Option::LEVEL_ADVANCED)
401 .add_service({"mon", "mds", "osd", "mgr"})
402 .set_flag(Option::FLAG_STARTUP)
403 .add_tag("network")
404 .set_description("Interface name(s) from which to choose an address from a cluster_network to bind to; cluster_network must also be specified.")
405 .add_see_also("cluster_network"),
406
407 Option("monmap", Option::TYPE_STR, Option::LEVEL_ADVANCED)
408 .set_description("path to MonMap file")
409 .set_long_description("This option is normally used during mkfs, but can also "
410 "be used to identify which monitors to connect to.")
411 .set_flag(Option::FLAG_NO_MON_UPDATE)
412 .add_service("mon")
413 .set_flag(Option::FLAG_CREATE),
414
415 Option("mon_host", Option::TYPE_STR, Option::LEVEL_BASIC)
416 .set_description("list of hosts or addresses to search for a monitor")
417 .set_long_description("This is a comma, whitespace, or semicolon separated "
418 "list of IP addresses or hostnames. Hostnames are "
419 "resolved via DNS and all A or AAAA records are "
420 "included in the search list.")
421 .set_flag(Option::FLAG_NO_MON_UPDATE)
422 .set_flag(Option::FLAG_STARTUP)
423 .add_service("common"),
424
425 Option("mon_dns_srv_name", Option::TYPE_STR, Option::LEVEL_ADVANCED)
426 .set_default("ceph-mon")
427 .set_description("name of DNS SRV record to check for monitor addresses")
428 .set_flag(Option::FLAG_NO_MON_UPDATE)
429 .set_flag(Option::FLAG_STARTUP)
430 .add_service("common")
431 .add_tag("network")
432 .add_see_also("mon_host"),
433
434 Option("image", Option::TYPE_STR, Option::LEVEL_BASIC)
435 .set_description("container image (used by ssh orchestrator)")
436 .set_flag(Option::FLAG_STARTUP)
437 .set_default("ceph/daemon-base"),
438
439 // lockdep
440 Option("lockdep", Option::TYPE_BOOL, Option::LEVEL_DEV)
441 .set_description("enable lockdep lock dependency analyzer")
442 .set_flag(Option::FLAG_NO_MON_UPDATE)
443 .set_flag(Option::FLAG_STARTUP)
444 .add_service("common"),
445
446 Option("lockdep_force_backtrace", Option::TYPE_BOOL, Option::LEVEL_DEV)
447 .set_description("always gather current backtrace at every lock")
448 .set_flag(Option::FLAG_STARTUP)
449 .add_service("common")
450 .add_see_also("lockdep"),
451
452 Option("run_dir", Option::TYPE_STR, Option::LEVEL_ADVANCED)
453 .set_default("/var/run/ceph")
454 .set_flag(Option::FLAG_STARTUP)
455 .set_description("path for the 'run' directory for storing pid and socket files")
456 .add_service("common")
457 .add_see_also("admin_socket"),
458
459 Option("admin_socket", Option::TYPE_STR, Option::LEVEL_ADVANCED)
460 .set_default("")
461 .set_daemon_default("$run_dir/$cluster-$name.asok")
462 .set_flag(Option::FLAG_STARTUP)
463 .set_description("path for the runtime control socket file, used by the 'ceph daemon' command")
464 .add_service("common"),
465
466 Option("admin_socket_mode", Option::TYPE_STR, Option::LEVEL_ADVANCED)
467 .set_description("file mode to set for the admin socket file, e.g, '0755'")
468 .set_flag(Option::FLAG_STARTUP)
469 .add_service("common")
470 .add_see_also("admin_socket"),
471
472 // daemon
473 Option("daemonize", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
474 .set_default(false)
475 .set_daemon_default(true)
476 .set_description("whether to daemonize (background) after startup")
477 .set_flag(Option::FLAG_STARTUP)
478 .set_flag(Option::FLAG_NO_MON_UPDATE)
479 .add_service({"mon", "mgr", "osd", "mds"})
480 .add_tag("service")
481 .add_see_also({"pid_file", "chdir"}),
482
483 Option("setuser", Option::TYPE_STR, Option::LEVEL_ADVANCED)
484 .set_flag(Option::FLAG_STARTUP)
485 .set_description("uid or user name to switch to on startup")
486 .set_long_description("This is normally specified by the systemd unit file.")
487 .add_service({"mon", "mgr", "osd", "mds"})
488 .add_tag("service")
489 .add_see_also("setgroup"),
490
491 Option("setgroup", Option::TYPE_STR, Option::LEVEL_ADVANCED)
492 .set_flag(Option::FLAG_STARTUP)
493 .set_description("gid or group name to switch to on startup")
494 .set_long_description("This is normally specified by the systemd unit file.")
495 .add_service({"mon", "mgr", "osd", "mds"})
496 .add_tag("service")
497 .add_see_also("setuser"),
498
499 Option("setuser_match_path", Option::TYPE_STR, Option::LEVEL_ADVANCED)
500 .set_flag(Option::FLAG_STARTUP)
501 .set_description("if set, setuser/setgroup is condition on this path matching ownership")
502 .set_long_description("If setuser or setgroup are specified, and this option is non-empty, then the uid/gid of the daemon will only be changed if the file or directory specified by this option has a matching uid and/or gid. This exists primarily to allow switching to user ceph for OSDs to be conditional on whether the osd data contents have also been chowned after an upgrade. This is normally specified by the systemd unit file.")
503 .add_service({"mon", "mgr", "osd", "mds"})
504 .add_tag("service")
505 .add_see_also({"setuser", "setgroup"}),
506
507 Option("pid_file", Option::TYPE_STR, Option::LEVEL_ADVANCED)
508 .set_flag(Option::FLAG_STARTUP)
509 .set_description("path to write a pid file (if any)")
510 .add_service({"mon", "mgr", "osd", "mds"})
511 .add_tag("service"),
512
513 Option("chdir", Option::TYPE_STR, Option::LEVEL_ADVANCED)
514 .set_description("path to chdir(2) to after daemonizing")
515 .set_flag(Option::FLAG_STARTUP)
516 .set_flag(Option::FLAG_NO_MON_UPDATE)
517 .add_service({"mon", "mgr", "osd", "mds"})
518 .add_tag("service")
519 .add_see_also("daemonize"),
520
521 Option("fatal_signal_handlers", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
522 .set_default(true)
523 .set_flag(Option::FLAG_STARTUP)
524 .set_description("whether to register signal handlers for SIGABRT etc that dump a stack trace")
525 .set_long_description("This is normally true for daemons and values for libraries.")
526 .add_service({"mon", "mgr", "osd", "mds"})
527 .add_tag("service"),
528
529 Option("crash_dir", Option::TYPE_STR, Option::LEVEL_ADVANCED)
530 .set_flag(Option::FLAG_STARTUP)
531 .set_default("/var/lib/ceph/crash")
532 .set_description("Directory where crash reports are archived"),
533
534 // restapi
535 Option("restapi_log_level", Option::TYPE_STR, Option::LEVEL_ADVANCED)
536 .set_description("default set by python code"),
537
538 Option("restapi_base_url", Option::TYPE_STR, Option::LEVEL_ADVANCED)
539 .set_description("default set by python code"),
540
541 Option("erasure_code_dir", Option::TYPE_STR, Option::LEVEL_ADVANCED)
542 .set_default(CEPH_PKGLIBDIR"/erasure-code")
543 .set_flag(Option::FLAG_STARTUP)
544 .set_description("directory where erasure-code plugins can be found")
545 .add_service({"mon", "osd"}),
546
547 // logging
548 Option("log_file", Option::TYPE_STR, Option::LEVEL_BASIC)
549 .set_default("")
550 .set_daemon_default("/var/log/ceph/$cluster-$name.log")
551 .set_description("path to log file")
552 .add_see_also({"log_to_file",
553 "log_to_stderr",
554 "err_to_stderr",
555 "log_to_syslog",
556 "err_to_syslog"}),
557
558 Option("log_max_new", Option::TYPE_INT, Option::LEVEL_ADVANCED)
559 .set_default(1000)
560 .set_description("max unwritten log entries to allow before waiting to flush to the log")
561 .add_see_also("log_max_recent"),
562
563 Option("log_max_recent", Option::TYPE_INT, Option::LEVEL_ADVANCED)
564 .set_default(500)
565 .set_daemon_default(10000)
566 .set_description("recent log entries to keep in memory to dump in the event of a crash")
567 .set_long_description("The purpose of this option is to log at a higher debug level only to the in-memory buffer, and write out the detailed log messages only if there is a crash. Only log entries below the lower log level will be written unconditionally to the log. For example, debug_osd=1/5 will write everything <= 1 to the log unconditionally but keep entries at levels 2-5 in memory. If there is a seg fault or assertion failure, all entries will be dumped to the log."),
568
569 Option("log_to_file", Option::TYPE_BOOL, Option::LEVEL_BASIC)
570 .set_default(true)
571 .set_description("send log lines to a file")
572 .add_see_also("log_file"),
573
574 Option("log_to_stderr", Option::TYPE_BOOL, Option::LEVEL_BASIC)
575 .set_default(true)
576 .set_daemon_default(false)
577 .set_description("send log lines to stderr"),
578
579 Option("err_to_stderr", Option::TYPE_BOOL, Option::LEVEL_BASIC)
580 .set_default(false)
581 .set_daemon_default(true)
582 .set_description("send critical error log lines to stderr"),
583
584 Option("log_stderr_prefix", Option::TYPE_STR, Option::LEVEL_ADVANCED)
585 .set_description("String to prefix log messages with when sent to stderr")
586 .set_long_description("This is useful in container environments when combined with mon_cluster_log_to_stderr. The mon log prefixes each line with the channel name (e.g., 'default', 'audit'), while log_stderr_prefix can be set to 'debug '.")
587 .add_see_also("mon_cluster_log_to_stderr"),
588
589 Option("log_to_syslog", Option::TYPE_BOOL, Option::LEVEL_BASIC)
590 .set_default(false)
591 .set_description("send log lines to syslog facility"),
592
593 Option("err_to_syslog", Option::TYPE_BOOL, Option::LEVEL_BASIC)
594 .set_default(false)
595 .set_description("send critical error log lines to syslog facility"),
596
597 Option("log_flush_on_exit", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
598 .set_default(false)
599 .set_description("set a process exit handler to ensure the log is flushed on exit"),
600
601 Option("log_stop_at_utilization", Option::TYPE_FLOAT, Option::LEVEL_BASIC)
602 .set_default(.97)
603 .set_min_max(0.0, 1.0)
604 .set_description("stop writing to the log file when device utilization reaches this ratio")
605 .add_see_also("log_file"),
606
607 Option("log_to_graylog", Option::TYPE_BOOL, Option::LEVEL_BASIC)
608 .set_default(false)
609 .set_description("send log lines to remote graylog server")
610 .add_see_also({"err_to_graylog",
611 "log_graylog_host",
612 "log_graylog_port"}),
613
614 Option("err_to_graylog", Option::TYPE_BOOL, Option::LEVEL_BASIC)
615 .set_default(false)
616 .set_description("send critical error log lines to remote graylog server")
617 .add_see_also({"log_to_graylog",
618 "log_graylog_host",
619 "log_graylog_port"}),
620
621 Option("log_graylog_host", Option::TYPE_STR, Option::LEVEL_BASIC)
622 .set_default("127.0.0.1")
623 .set_description("address or hostname of graylog server to log to")
624 .add_see_also({"log_to_graylog",
625 "err_to_graylog",
626 "log_graylog_port"}),
627
628 Option("log_graylog_port", Option::TYPE_INT, Option::LEVEL_BASIC)
629 .set_default(12201)
630 .set_description("port number for the remote graylog server")
631 .add_see_also("log_graylog_host"),
632
633 Option("log_coarse_timestamps", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
634 .set_default(true)
635 .set_description("timestamp log entries from coarse system clock "
636 "to improve performance")
637 .add_service("common")
638 .add_tag("performance")
639 .add_tag("service"),
640
641
642 // unmodified
643 Option("clog_to_monitors", Option::TYPE_STR, Option::LEVEL_ADVANCED)
644 .set_default("default=true")
645 .set_flag(Option::FLAG_RUNTIME)
646 .set_description("Make daemons send cluster log messages to monitors"),
647
648 Option("clog_to_syslog", Option::TYPE_STR, Option::LEVEL_ADVANCED)
649 .set_default("false")
650 .set_flag(Option::FLAG_RUNTIME)
651 .set_description("Make daemons send cluster log messages to syslog"),
652
653 Option("clog_to_syslog_level", Option::TYPE_STR, Option::LEVEL_ADVANCED)
654 .set_default("info")
655 .set_flag(Option::FLAG_RUNTIME)
656 .set_description("Syslog level for cluster log messages")
657 .add_see_also("clog_to_syslog"),
658
659 Option("clog_to_syslog_facility", Option::TYPE_STR, Option::LEVEL_ADVANCED)
660 .set_default("default=daemon audit=local0")
661 .set_flag(Option::FLAG_RUNTIME)
662 .set_description("Syslog facility for cluster log messages")
663 .add_see_also("clog_to_syslog"),
664
665 Option("clog_to_graylog", Option::TYPE_STR, Option::LEVEL_ADVANCED)
666 .set_default("false")
667 .set_flag(Option::FLAG_RUNTIME)
668 .set_description("Make daemons send cluster log to graylog"),
669
670 Option("clog_to_graylog_host", Option::TYPE_STR, Option::LEVEL_ADVANCED)
671 .set_default("127.0.0.1")
672 .set_flag(Option::FLAG_RUNTIME)
673 .set_description("Graylog host to cluster log messages")
674 .add_see_also("clog_to_graylog"),
675
676 Option("clog_to_graylog_port", Option::TYPE_STR, Option::LEVEL_ADVANCED)
677 .set_default("12201")
678 .set_flag(Option::FLAG_RUNTIME)
679 .set_description("Graylog port number for cluster log messages")
680 .add_see_also("clog_to_graylog"),
681
682 Option("mon_cluster_log_to_stderr", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
683 .set_default(false)
684 .add_service("mon")
685 .set_flag(Option::FLAG_RUNTIME)
686 .set_description("Make monitor send cluster log messages to stderr (prefixed by channel)")
687 .add_see_also("log_stderr_prefix"),
688
689 Option("mon_cluster_log_to_syslog", Option::TYPE_STR, Option::LEVEL_ADVANCED)
690 .set_default("default=false")
691 .set_flag(Option::FLAG_RUNTIME)
692 .add_service("mon")
693 .set_description("Make monitor send cluster log messages to syslog"),
694
695 Option("mon_cluster_log_to_syslog_level", Option::TYPE_STR, Option::LEVEL_ADVANCED)
696 .set_default("info")
697 .add_service("mon")
698 .set_flag(Option::FLAG_RUNTIME)
699 .set_description("Syslog level for cluster log messages")
700 .add_see_also("mon_cluster_log_to_syslog"),
701
702 Option("mon_cluster_log_to_syslog_facility", Option::TYPE_STR, Option::LEVEL_ADVANCED)
703 .set_default("daemon")
704 .add_service("mon")
705 .set_flag(Option::FLAG_RUNTIME)
706 .set_description("Syslog facility for cluster log messages")
707 .add_see_also("mon_cluster_log_to_syslog"),
708
709 Option("mon_cluster_log_to_file", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
710 .set_default(true)
711 .set_flag(Option::FLAG_RUNTIME)
712 .add_service("mon")
713 .set_description("Make monitor send cluster log messages to file")
714 .add_see_also("mon_cluster_log_file"),
715
716 Option("mon_cluster_log_file", Option::TYPE_STR, Option::LEVEL_ADVANCED)
717 .set_default("default=/var/log/ceph/$cluster.$channel.log cluster=/var/log/ceph/$cluster.log")
718 .set_flag(Option::FLAG_RUNTIME)
719 .add_service("mon")
720 .set_description("File(s) to write cluster log to")
721 .set_long_description("This can either be a simple file name to receive all messages, or a list of key/value pairs where the key is the log channel and the value is the filename, which may include $cluster and $channel metavariables")
722 .add_see_also("mon_cluster_log_to_file"),
723
724 Option("mon_cluster_log_file_level", Option::TYPE_STR, Option::LEVEL_ADVANCED)
725 .set_default("debug")
726 .set_flag(Option::FLAG_RUNTIME)
727 .add_service("mon")
728 .set_description("Lowest level to include is cluster log file")
729 .add_see_also("mon_cluster_log_file"),
730
731 Option("mon_cluster_log_to_graylog", Option::TYPE_STR, Option::LEVEL_ADVANCED)
732 .set_default("false")
733 .set_flag(Option::FLAG_RUNTIME)
734 .add_service("mon")
735 .set_description("Make monitor send cluster log to graylog"),
736
737 Option("mon_cluster_log_to_graylog_host", Option::TYPE_STR, Option::LEVEL_ADVANCED)
738 .set_default("127.0.0.1")
739 .set_flag(Option::FLAG_RUNTIME)
740 .add_service("mon")
741 .set_description("Graylog host for cluster log messages")
742 .add_see_also("mon_cluster_log_to_graylog"),
743
744 Option("mon_cluster_log_to_graylog_port", Option::TYPE_STR, Option::LEVEL_ADVANCED)
745 .set_default("12201")
746 .set_flag(Option::FLAG_RUNTIME)
747 .add_service("mon")
748 .set_description("Graylog port for cluster log messages")
749 .add_see_also("mon_cluster_log_to_graylog"),
750
751 Option("enable_experimental_unrecoverable_data_corrupting_features", Option::TYPE_STR, Option::LEVEL_ADVANCED)
752 .set_flag(Option::FLAG_RUNTIME)
753 .set_default("")
754 .set_description("Enable named (or all with '*') experimental features that may be untested, dangerous, and/or cause permanent data loss"),
755
756 Option("plugin_dir", Option::TYPE_STR, Option::LEVEL_ADVANCED)
757 .set_default(CEPH_PKGLIBDIR)
758 .set_flag(Option::FLAG_STARTUP)
759 .add_service({"mon", "osd"})
760 .set_description("Base directory for dynamically loaded plugins"),
761
762 // Compressor
763 Option("compressor_zlib_isal", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
764 .set_default(false)
765 .set_description("Use Intel ISA-L accelerated zlib implementation if available"),
766
767 Option("compressor_zlib_level", Option::TYPE_INT, Option::LEVEL_ADVANCED)
768 .set_default(5)
769 .set_description("Zlib compression level to use"),
770
771 Option("qat_compressor_enabled", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
772 .set_default(false)
773 .set_description("Enable Intel QAT acceleration support for compression if available"),
774
775 Option("plugin_crypto_accelerator", Option::TYPE_STR, Option::LEVEL_ADVANCED)
776 .set_default("crypto_isal")
777 .set_description("Crypto accelerator library to use"),
778
779 Option("mempool_debug", Option::TYPE_BOOL, Option::LEVEL_DEV)
780 .set_default(false)
781 .set_flag(Option::FLAG_NO_MON_UPDATE)
782 .set_description(""),
783
784 Option("key", Option::TYPE_STR, Option::LEVEL_ADVANCED)
785 .set_default("")
786 .set_description("Authentication key")
787 .set_long_description("A CephX authentication key, base64 encoded. It normally looks something like 'AQAtut9ZdMbNJBAAHz6yBAWyJyz2yYRyeMWDag=='.")
788 .set_flag(Option::FLAG_STARTUP)
789 .set_flag(Option::FLAG_NO_MON_UPDATE)
790 .add_see_also("keyfile")
791 .add_see_also("keyring"),
792
793 Option("keyfile", Option::TYPE_STR, Option::LEVEL_ADVANCED)
794 .set_default("")
795 .set_description("Path to a file containing a key")
796 .set_long_description("The file should contain a CephX authentication key and optionally a trailing newline, but nothing else.")
797 .set_flag(Option::FLAG_STARTUP)
798 .set_flag(Option::FLAG_NO_MON_UPDATE)
799 .add_see_also("key"),
800
801 Option("keyring", Option::TYPE_STR, Option::LEVEL_ADVANCED)
802 .set_default(
803 "/etc/ceph/$cluster.$name.keyring,/etc/ceph/$cluster.keyring,"
804 "/etc/ceph/keyring,/etc/ceph/keyring.bin,"
805 #if defined(__FreeBSD)
806 "/usr/local/etc/ceph/$cluster.$name.keyring,"
807 "/usr/local/etc/ceph/$cluster.keyring,"
808 "/usr/local/etc/ceph/keyring,/usr/local/etc/ceph/keyring.bin,"
809 #endif
810 )
811 .set_description("Path to a keyring file.")
812 .set_long_description("A keyring file is an INI-style formatted file where the section names are client or daemon names (e.g., 'osd.0') and each section contains a 'key' property with CephX authentication key as the value.")
813 .set_flag(Option::FLAG_STARTUP)
814 .set_flag(Option::FLAG_NO_MON_UPDATE)
815 .add_see_also("key")
816 .add_see_also("keyfile"),
817
818 Option("heartbeat_interval", Option::TYPE_INT, Option::LEVEL_ADVANCED)
819 .set_default(5)
820 .set_flag(Option::FLAG_STARTUP)
821 .set_description("Frequency of internal heartbeat checks (seconds)"),
822
823 Option("heartbeat_file", Option::TYPE_STR, Option::LEVEL_ADVANCED)
824 .set_default("")
825 .set_flag(Option::FLAG_STARTUP)
826 .set_description("File to touch on successful internal heartbeat")
827 .set_long_description("If set, this file will be touched every time an internal heartbeat check succeeds.")
828 .add_see_also("heartbeat_interval"),
829
830 Option("heartbeat_inject_failure", Option::TYPE_INT, Option::LEVEL_DEV)
831 .set_default(0)
832 .set_description(""),
833
834 Option("perf", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
835 .set_default(true)
836 .set_description("Enable internal performance metrics")
837 .set_long_description("If enabled, collect and expose internal health metrics"),
838
839 Option("ms_type", Option::TYPE_STR, Option::LEVEL_ADVANCED)
840 .set_flag(Option::FLAG_STARTUP)
841 .set_default("async+posix")
842 .set_description("Messenger implementation to use for network communication"),
843
844 Option("ms_public_type", Option::TYPE_STR, Option::LEVEL_ADVANCED)
845 .set_default("")
846 .set_flag(Option::FLAG_STARTUP)
847 .set_description("Messenger implementation to use for the public network")
848 .set_long_description("If not specified, use ms_type")
849 .add_see_also("ms_type"),
850
851 Option("ms_cluster_type", Option::TYPE_STR, Option::LEVEL_ADVANCED)
852 .set_default("")
853 .set_flag(Option::FLAG_STARTUP)
854 .set_description("Messenger implementation to use for the internal cluster network")
855 .set_long_description("If not specified, use ms_type")
856 .add_see_also("ms_type"),
857
858 Option("ms_mon_cluster_mode", Option::TYPE_STR, Option::LEVEL_BASIC)
859 .set_default("secure crc")
860 .set_flag(Option::FLAG_STARTUP)
861 .set_description("Connection modes (crc, secure) for intra-mon connections in order of preference")
862 .add_see_also("ms_mon_service_mode")
863 .add_see_also("ms_mon_client_mode")
864 .add_see_also("ms_service_mode")
865 .add_see_also("ms_cluster_mode")
866 .add_see_also("ms_client_mode"),
867
868 Option("ms_mon_service_mode", Option::TYPE_STR, Option::LEVEL_BASIC)
869 .set_default("secure crc")
870 .set_flag(Option::FLAG_STARTUP)
871 .set_description("Allowed connection modes (crc, secure) for connections to mons")
872 .add_see_also("ms_service_mode")
873 .add_see_also("ms_mon_cluster_mode")
874 .add_see_also("ms_mon_client_mode")
875 .add_see_also("ms_cluster_mode")
876 .add_see_also("ms_client_mode"),
877
878 Option("ms_mon_client_mode", Option::TYPE_STR, Option::LEVEL_BASIC)
879 .set_default("secure crc")
880 .set_flag(Option::FLAG_STARTUP)
881 .set_description("Connection modes (crc, secure) for connections from clients to monitors in order of preference")
882 .add_see_also("ms_mon_service_mode")
883 .add_see_also("ms_mon_cluster_mode")
884 .add_see_also("ms_service_mode")
885 .add_see_also("ms_cluster_mode")
886 .add_see_also("ms_client_mode"),
887
888 Option("ms_cluster_mode", Option::TYPE_STR, Option::LEVEL_BASIC)
889 .set_default("crc secure")
890 .set_flag(Option::FLAG_STARTUP)
891 .set_description("Connection modes (crc, secure) for intra-cluster connections in order of preference")
892 .add_see_also("ms_service_mode")
893 .add_see_also("ms_client_mode"),
894
895 Option("ms_service_mode", Option::TYPE_STR, Option::LEVEL_BASIC)
896 .set_default("crc secure")
897 .set_flag(Option::FLAG_STARTUP)
898 .set_description("Allowed connection modes (crc, secure) for connections to daemons")
899 .add_see_also("ms_cluster_mode")
900 .add_see_also("ms_client_mode"),
901
902 Option("ms_client_mode", Option::TYPE_STR, Option::LEVEL_BASIC)
903 .set_default("crc secure")
904 .set_flag(Option::FLAG_STARTUP)
905 .set_description("Connection modes (crc, secure) for connections from clients in order of preference")
906 .add_see_also("ms_cluster_mode")
907 .add_see_also("ms_service_mode"),
908
909 Option("ms_learn_addr_from_peer", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
910 .set_default(true)
911 .set_description("Learn address from what IP our first peer thinks we connect from")
912 .set_long_description("Use the IP address our first peer (usually a monitor) sees that we are connecting from. This is useful if a client is behind some sort of NAT and we want to see it identified by its local (not NATed) address."),
913
914 Option("ms_tcp_nodelay", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
915 .set_default(true)
916 .set_description("Disable Nagle's algorithm and send queued network traffic immediately"),
917
918 Option("ms_tcp_rcvbuf", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
919 .set_default(0)
920 .set_description("Size of TCP socket receive buffer"),
921
922 Option("ms_tcp_prefetch_max_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
923 .set_default(4_K)
924 .set_description("Maximum amount of data to prefetch out of the socket receive buffer"),
925
926 Option("ms_initial_backoff", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
927 .set_default(.2)
928 .set_description("Initial backoff after a network error is detected (seconds)"),
929
930 Option("ms_max_backoff", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
931 .set_default(15.0)
932 .set_description("Maximum backoff after a network error before retrying (seconds)")
933 .add_see_also("ms_initial_backoff"),
934
935 Option("ms_crc_data", Option::TYPE_BOOL, Option::LEVEL_DEV)
936 .set_default(true)
937 .set_description("Set and/or verify crc32c checksum on data payload sent over network"),
938
939 Option("ms_crc_header", Option::TYPE_BOOL, Option::LEVEL_DEV)
940 .set_default(true)
941 .set_description("Set and/or verify crc32c checksum on header payload sent over network"),
942
943 Option("ms_die_on_bad_msg", Option::TYPE_BOOL, Option::LEVEL_DEV)
944 .set_default(false)
945 .set_description("Induce a daemon crash/exit when a bad network message is received"),
946
947 Option("ms_die_on_unhandled_msg", Option::TYPE_BOOL, Option::LEVEL_DEV)
948 .set_default(false)
949 .set_description("Induce a daemon crash/exit when an unrecognized message is received"),
950
951 Option("ms_die_on_old_message", Option::TYPE_BOOL, Option::LEVEL_DEV)
952 .set_default(false)
953 .set_description("Induce a daemon crash/exit when a old, undecodable message is received"),
954
955 Option("ms_die_on_skipped_message", Option::TYPE_BOOL, Option::LEVEL_DEV)
956 .set_default(false)
957 .set_description("Induce a daemon crash/exit if sender skips a message sequence number"),
958
959 Option("ms_die_on_bug", Option::TYPE_BOOL, Option::LEVEL_DEV)
960 .set_default(false)
961 .set_description("Induce a crash/exit on various bugs (for testing purposes)"),
962
963 Option("ms_dispatch_throttle_bytes", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
964 .set_default(100_M)
965 .set_description("Limit messages that are read off the network but still being processed"),
966
967 Option("ms_msgr2_sign_messages", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
968 .set_default(false)
969 .set_description("Sign msgr2 frames' payload")
970 .add_see_also("ms_msgr2_encrypt_messages"),
971
972 Option("ms_msgr2_encrypt_messages", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
973 .set_default(false)
974 .set_description("Encrypt msgr2 frames' payload")
975 .add_see_also("ms_msgr2_sign_messages"),
976
977 Option("ms_bind_ipv4", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
978 .set_default(true)
979 .set_description("Bind servers to IPv4 address(es)")
980 .add_see_also("ms_bind_ipv6"),
981
982 Option("ms_bind_ipv6", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
983 .set_default(false)
984 .set_description("Bind servers to IPv6 address(es)")
985 .add_see_also("ms_bind_ipv4"),
986
987 Option("ms_bind_prefer_ipv4", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
988 .set_default(false)
989 .set_description("Prefer IPV4 over IPV6 address(es)"),
990
991 Option("ms_bind_msgr1", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
992 .set_default(true)
993 .set_description("Bind servers to msgr1 (legacy) protocol address(es)")
994 .add_see_also("ms_bind_msgr2"),
995
996 Option("ms_bind_msgr2", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
997 .set_default(true)
998 .set_description("Bind servers to msgr2 (nautilus+) protocol address(es)")
999 .add_see_also("ms_bind_msgr1"),
1000
1001 Option("ms_bind_port_min", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1002 .set_default(6800)
1003 .set_description("Lowest port number to bind daemon(s) to"),
1004
1005 Option("ms_bind_port_max", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1006 .set_default(7300)
1007 .set_description("Highest port number to bind daemon(s) to"),
1008
1009 Option("ms_bind_retry_count", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1010 #if !defined(__FreeBSD__)
1011 .set_default(3)
1012 #else
1013 // FreeBSD does not use SO_REAUSEADDR so allow for a bit more time per default
1014 .set_default(6)
1015 #endif
1016 .set_description("Number of attempts to make while bind(2)ing to a port"),
1017
1018 Option("ms_bind_retry_delay", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1019 #if !defined(__FreeBSD__)
1020 .set_default(5)
1021 #else
1022 // FreeBSD does not use SO_REAUSEADDR so allow for a bit more time per default
1023 .set_default(6)
1024 #endif
1025 .set_description("Delay between bind(2) attempts (seconds)"),
1026
1027 Option("ms_bind_before_connect", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
1028 .set_default(false)
1029 .set_description("Call bind(2) on client sockets"),
1030
1031 Option("ms_tcp_listen_backlog", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1032 .set_default(512)
1033 .set_description("Size of queue of incoming connections for accept(2)"),
1034
1035
1036 Option("ms_connection_ready_timeout", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
1037 .set_default(10)
1038 .set_description("Time before we declare a not yet ready connection as dead (seconds)"),
1039
1040 Option("ms_connection_idle_timeout", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
1041 .set_default(900)
1042 .set_description("Time before an idle connection is closed (seconds)"),
1043
1044 Option("ms_pq_max_tokens_per_priority", Option::TYPE_UINT, Option::LEVEL_DEV)
1045 .set_default(16777216)
1046 .set_description(""),
1047
1048 Option("ms_pq_min_cost", Option::TYPE_SIZE, Option::LEVEL_DEV)
1049 .set_default(65536)
1050 .set_description(""),
1051
1052 Option("ms_inject_socket_failures", Option::TYPE_UINT, Option::LEVEL_DEV)
1053 .set_default(0)
1054 .set_description("Inject a socket failure every Nth socket operation"),
1055
1056 Option("ms_inject_delay_type", Option::TYPE_STR, Option::LEVEL_DEV)
1057 .set_default("")
1058 .set_description("Entity type to inject delays for")
1059 .set_flag(Option::FLAG_RUNTIME),
1060
1061 Option("ms_inject_delay_msg_type", Option::TYPE_STR, Option::LEVEL_DEV)
1062 .set_default("")
1063 .set_description("Message type to inject delays for"),
1064
1065 Option("ms_inject_delay_max", Option::TYPE_FLOAT, Option::LEVEL_DEV)
1066 .set_default(1)
1067 .set_description("Max delay to inject"),
1068
1069 Option("ms_inject_delay_probability", Option::TYPE_FLOAT, Option::LEVEL_DEV)
1070 .set_default(0)
1071 .set_description(""),
1072
1073 Option("ms_inject_internal_delays", Option::TYPE_FLOAT, Option::LEVEL_DEV)
1074 .set_default(0)
1075 .set_description("Inject various internal delays to induce races (seconds)"),
1076
1077 Option("ms_blackhole_osd", Option::TYPE_BOOL, Option::LEVEL_DEV)
1078 .set_default(false)
1079 .set_description(""),
1080
1081 Option("ms_blackhole_mon", Option::TYPE_BOOL, Option::LEVEL_DEV)
1082 .set_default(false)
1083 .set_description(""),
1084
1085 Option("ms_blackhole_mds", Option::TYPE_BOOL, Option::LEVEL_DEV)
1086 .set_default(false)
1087 .set_description(""),
1088
1089 Option("ms_blackhole_mgr", Option::TYPE_BOOL, Option::LEVEL_DEV)
1090 .set_default(false)
1091 .set_description(""),
1092
1093 Option("ms_blackhole_client", Option::TYPE_BOOL, Option::LEVEL_DEV)
1094 .set_default(false)
1095 .set_description(""),
1096
1097 Option("ms_dump_on_send", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
1098 .set_default(false)
1099 .set_description("Hexdump message to debug log on message send"),
1100
1101 Option("ms_dump_corrupt_message_level", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1102 .set_default(1)
1103 .set_description("Log level at which to hexdump corrupt messages we receive"),
1104
1105 Option("ms_async_op_threads", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
1106 .set_default(3)
1107 .set_min_max(1, 24)
1108 .set_description("Threadpool size for AsyncMessenger (ms_type=async)"),
1109
1110 Option("ms_async_max_op_threads", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
1111 .set_default(5)
1112 .set_description("Maximum threadpool size of AsyncMessenger")
1113 .add_see_also("ms_async_op_threads"),
1114
1115 Option("ms_async_rdma_device_name", Option::TYPE_STR, Option::LEVEL_ADVANCED)
1116 .set_default("")
1117 .set_description(""),
1118
1119 Option("ms_async_rdma_enable_hugepage", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
1120 .set_default(false)
1121 .set_description(""),
1122
1123 Option("ms_async_rdma_buffer_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
1124 .set_default(128_K)
1125 .set_description(""),
1126
1127 Option("ms_async_rdma_send_buffers", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
1128 .set_default(1_K)
1129 .set_description(""),
1130
1131 Option("ms_async_rdma_receive_buffers", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
1132 .set_default(32768)
1133 .set_description(""),
1134
1135 Option("ms_async_rdma_receive_queue_len", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
1136 .set_default(4096)
1137 .set_description(""),
1138
1139 Option("ms_async_rdma_support_srq", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
1140 .set_default(true)
1141 .set_description(""),
1142
1143 Option("ms_async_rdma_port_num", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
1144 .set_default(1)
1145 .set_description(""),
1146
1147 Option("ms_async_rdma_polling_us", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
1148 .set_default(1000)
1149 .set_description(""),
1150
1151 Option("ms_async_rdma_local_gid", Option::TYPE_STR, Option::LEVEL_ADVANCED)
1152 .set_default("")
1153 .set_description(""),
1154
1155 Option("ms_async_rdma_roce_ver", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1156 .set_default(1)
1157 .set_description(""),
1158
1159 Option("ms_async_rdma_sl", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1160 .set_default(3)
1161 .set_description(""),
1162
1163 Option("ms_async_rdma_dscp", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1164 .set_default(96)
1165 .set_description(""),
1166
1167 Option("ms_max_accept_failures", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1168 .set_default(4)
1169 .set_description("The maximum number of consecutive failed accept() calls before "
1170 "considering the daemon is misconfigured and abort it."),
1171
1172 Option("ms_async_rdma_cm", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
1173 .set_default(false)
1174 .set_description(""),
1175
1176 Option("ms_async_rdma_type", Option::TYPE_STR, Option::LEVEL_ADVANCED)
1177 .set_default("ib")
1178 .set_description(""),
1179
1180 Option("ms_dpdk_port_id", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1181 .set_default(0)
1182 .set_description(""),
1183
1184 Option("ms_dpdk_coremask", Option::TYPE_STR, Option::LEVEL_ADVANCED)
1185 .set_default("0xF") //begin with 0x for the string
1186 .set_description("")
1187 .add_see_also("ms_async_op_threads"),
1188
1189 Option("ms_dpdk_memory_channel", Option::TYPE_STR, Option::LEVEL_ADVANCED)
1190 .set_default("4")
1191 .set_description(""),
1192
1193 Option("ms_dpdk_hugepages", Option::TYPE_STR, Option::LEVEL_ADVANCED)
1194 .set_default("")
1195 .set_description(""),
1196
1197 Option("ms_dpdk_pmd", Option::TYPE_STR, Option::LEVEL_ADVANCED)
1198 .set_default("")
1199 .set_description(""),
1200
1201 Option("ms_dpdk_host_ipv4_addr", Option::TYPE_STR, Option::LEVEL_ADVANCED)
1202 .set_default("")
1203 .set_description(""),
1204
1205 Option("ms_dpdk_gateway_ipv4_addr", Option::TYPE_STR, Option::LEVEL_ADVANCED)
1206 .set_default("")
1207 .set_description(""),
1208
1209 Option("ms_dpdk_netmask_ipv4_addr", Option::TYPE_STR, Option::LEVEL_ADVANCED)
1210 .set_default("")
1211 .set_description(""),
1212
1213 Option("ms_dpdk_lro", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
1214 .set_default(true)
1215 .set_description(""),
1216
1217 Option("ms_dpdk_hw_flow_control", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
1218 .set_default(true)
1219 .set_description(""),
1220
1221 Option("ms_dpdk_hw_queue_weight", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
1222 .set_default(1)
1223 .set_description(""),
1224
1225 Option("ms_dpdk_debug_allow_loopback", Option::TYPE_BOOL, Option::LEVEL_DEV)
1226 .set_default(false)
1227 .set_description(""),
1228
1229 Option("ms_dpdk_rx_buffer_count_per_core", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1230 .set_default(8192)
1231 .set_description(""),
1232
1233 Option("inject_early_sigterm", Option::TYPE_BOOL, Option::LEVEL_DEV)
1234 .set_default(false)
1235 .set_description("send ourselves a SIGTERM early during startup"),
1236
1237 // MON
1238 Option("mon_enable_op_tracker", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
1239 .set_default(true)
1240 .add_service("mon")
1241 .set_description("enable/disable MON op tracking"),
1242
1243 Option("mon_op_complaint_time", Option::TYPE_SECS, Option::LEVEL_ADVANCED)
1244 .set_default(30)
1245 .add_service("mon")
1246 .set_description("time after which to consider a monitor operation blocked "
1247 "after no updates"),
1248
1249 Option("mon_op_log_threshold", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1250 .set_default(5)
1251 .add_service("mon")
1252 .set_description("max number of slow ops to display"),
1253
1254 Option("mon_op_history_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
1255 .set_default(20)
1256 .add_service("mon")
1257 .set_description("max number of completed ops to track"),
1258
1259 Option("mon_op_history_duration", Option::TYPE_SECS, Option::LEVEL_ADVANCED)
1260 .set_default(600)
1261 .add_service("mon")
1262 .set_description("expiration time in seconds of historical MON OPS"),
1263
1264 Option("mon_op_history_slow_op_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
1265 .set_default(20)
1266 .add_service("mon")
1267 .set_description("max number of slow historical MON OPS to keep"),
1268
1269 Option("mon_op_history_slow_op_threshold", Option::TYPE_SECS, Option::LEVEL_ADVANCED)
1270 .set_default(10)
1271 .add_service("mon")
1272 .set_description("duration of an op to be considered as a historical slow op"),
1273
1274 Option("mon_data", Option::TYPE_STR, Option::LEVEL_ADVANCED)
1275 .set_flag(Option::FLAG_NO_MON_UPDATE)
1276 .set_default("/var/lib/ceph/mon/$cluster-$id")
1277 .add_service("mon")
1278 .set_description("path to mon database"),
1279
1280 Option("mon_initial_members", Option::TYPE_STR, Option::LEVEL_ADVANCED)
1281 .set_default("")
1282 .add_service("mon")
1283 .set_flag(Option::FLAG_NO_MON_UPDATE)
1284 .set_flag(Option::FLAG_CLUSTER_CREATE)
1285 .set_description(""),
1286
1287 Option("mon_compact_on_start", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
1288 .set_default(false)
1289 .add_service("mon")
1290 .set_description(""),
1291
1292 Option("mon_compact_on_bootstrap", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
1293 .set_default(false)
1294 .add_service("mon")
1295 .set_description(""),
1296
1297 Option("mon_compact_on_trim", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
1298 .set_default(true)
1299 .add_service("mon")
1300 .set_description(""),
1301
1302 /* -- mon: osdmap prune (begin) -- */
1303 Option("mon_osdmap_full_prune_enabled", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
1304 .set_default(true)
1305 .add_service("mon")
1306 .set_description("enables pruning full osdmap versions when we go over a given number of maps")
1307 .add_see_also("mon_osdmap_full_prune_min")
1308 .add_see_also("mon_osdmap_full_prune_interval")
1309 .add_see_also("mon_osdmap_full_prune_txsize"),
1310
1311 Option("mon_osdmap_full_prune_min", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
1312 .set_default(10000)
1313 .add_service("mon")
1314 .set_description("minimum number of versions in the store to trigger full map pruning")
1315 .add_see_also("mon_osdmap_full_prune_enabled")
1316 .add_see_also("mon_osdmap_full_prune_interval")
1317 .add_see_also("mon_osdmap_full_prune_txsize"),
1318
1319 Option("mon_osdmap_full_prune_interval", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
1320 .set_default(10)
1321 .add_service("mon")
1322 .set_description("interval between maps that will not be pruned; maps in the middle will be pruned.")
1323 .add_see_also("mon_osdmap_full_prune_enabled")
1324 .add_see_also("mon_osdmap_full_prune_interval")
1325 .add_see_also("mon_osdmap_full_prune_txsize"),
1326
1327 Option("mon_osdmap_full_prune_txsize", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
1328 .set_default(100)
1329 .add_service("mon")
1330 .set_description("number of maps we will prune per iteration")
1331 .add_see_also("mon_osdmap_full_prune_enabled")
1332 .add_see_also("mon_osdmap_full_prune_interval")
1333 .add_see_also("mon_osdmap_full_prune_txsize"),
1334 /* -- mon: osdmap prune (end) -- */
1335
1336 Option("mon_osd_cache_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1337 .set_default(500)
1338 .add_service("mon")
1339 .set_description("maximum number of OSDMaps to cache in memory"),
1340
1341 Option("mon_osd_cache_size_min", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
1342 .set_default(128_M)
1343 .add_service("mon")
1344 .set_description("The minimum amount of bytes to be kept mapped in memory for osd monitor caches."),
1345
1346 Option("mon_memory_target", Option::TYPE_SIZE, Option::LEVEL_BASIC)
1347 .set_default(2_G)
1348 .set_flag(Option::FLAG_RUNTIME)
1349 .add_service("mon")
1350 .set_description("The amount of bytes pertaining to osd monitor caches and kv cache to be kept mapped in memory with cache auto-tuning enabled"),
1351
1352 Option("mon_memory_autotune", Option::TYPE_BOOL, Option::LEVEL_BASIC)
1353 .set_default(true)
1354 .set_flag(Option::FLAG_RUNTIME)
1355 .add_service("mon")
1356 .set_description("Autotune the cache memory being used for osd monitors and kv database"),
1357
1358 Option("mon_cpu_threads", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1359 .set_default(4)
1360 .add_service("mon")
1361 .set_description("worker threads for CPU intensive background work"),
1362
1363 Option("mon_osd_mapping_pgs_per_chunk", Option::TYPE_INT, Option::LEVEL_DEV)
1364 .set_default(4096)
1365 .add_service("mon")
1366 .set_description("granularity of PG placement calculation background work"),
1367
1368 Option("mon_clean_pg_upmaps_per_chunk", Option::TYPE_UINT, Option::LEVEL_DEV)
1369 .set_default(256)
1370 .add_service("mon")
1371 .set_description("granularity of PG upmap validation background work"),
1372
1373 Option("mon_osd_max_creating_pgs", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1374 .set_default(1024)
1375 .add_service("mon")
1376 .set_description("maximum number of PGs the mon will create at once"),
1377
1378 Option("mon_osd_max_initial_pgs", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1379 .set_default(1024)
1380 .add_service("mon")
1381 .set_description("maximum number of PGs a pool will created with")
1382 .set_long_description("If the user specifies more PGs than this, the cluster will subsequently split PGs after the pool is created in order to reach the target."),
1383
1384 Option("mon_tick_interval", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1385 .set_default(5)
1386 .add_service("mon")
1387 .set_description("interval for internal mon background checks"),
1388
1389 Option("mon_session_timeout", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1390 .set_default(300)
1391 .add_service("mon")
1392 .set_description("close inactive mon client connections after this many seconds"),
1393
1394 Option("mon_subscribe_interval", Option::TYPE_FLOAT, Option::LEVEL_DEV)
1395 .set_default(1_day)
1396 .add_service("mon")
1397 .set_description("subscribe interval for pre-jewel clients"),
1398
1399 Option("mon_delta_reset_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
1400 .set_default(10)
1401 .add_service("mon")
1402 .add_service("mon")
1403 .set_description("window duration for rate calculations in 'ceph status'"),
1404
1405 Option("mon_osd_laggy_halflife", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1406 .set_default(1_hr)
1407 .add_service("mon")
1408 .set_description("halflife of OSD 'lagginess' factor"),
1409
1410 Option("mon_osd_laggy_weight", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
1411 .set_default(.3)
1412 .set_min_max(0.0, 1.0)
1413 .add_service("mon")
1414 .set_description("how heavily to weight OSD marking itself back up in overall laggy_probability")
1415 .set_long_description("1.0 means that an OSD marking itself back up (because it was marked down but not actually dead) means a 100% laggy_probability; 0.0 effectively disables tracking of laggy_probability."),
1416
1417 Option("mon_osd_laggy_max_interval", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1418 .set_default(300)
1419 .add_service("mon")
1420 .set_description("cap value for period for OSD to be marked for laggy_interval calculation"),
1421
1422 Option("mon_osd_adjust_heartbeat_grace", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
1423 .set_default(true)
1424 .add_service("mon")
1425 .set_description("increase OSD heartbeat grace if peers appear to be laggy")
1426 .set_long_description("If an OSD is marked down but then marks itself back up, it implies it wasn't actually down but was unable to respond to heartbeats. If this option is true, we can use the laggy_probability and laggy_interval values calculated to model this situation to increase the heartbeat grace period for this OSD so that it isn't marked down again. laggy_probability is an estimated probability that the given OSD is down because it is laggy (not actually down), and laggy_interval is an estiate on how long it stays down when it is laggy.")
1427 .add_see_also("mon_osd_laggy_halflife")
1428 .add_see_also("mon_osd_laggy_weight")
1429 .add_see_also("mon_osd_laggy_max_interval"),
1430
1431 Option("mon_osd_adjust_down_out_interval", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
1432 .set_default(true)
1433 .add_service("mon")
1434 .set_description("increase the mon_osd_down_out_interval if an OSD appears to be laggy")
1435 .add_see_also("mon_osd_adjust_heartbeat_grace"),
1436
1437 Option("mon_osd_auto_mark_in", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
1438 .set_default(false)
1439 .add_service("mon")
1440 .set_description("mark any OSD that comes up 'in'"),
1441
1442 Option("mon_osd_auto_mark_auto_out_in", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
1443 .set_default(true)
1444 .add_service("mon")
1445 .set_description("mark any OSD that comes up that was automatically marked 'out' back 'in'")
1446 .add_see_also("mon_osd_down_out_interval"),
1447
1448 Option("mon_osd_auto_mark_new_in", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
1449 .set_default(true)
1450 .add_service("mon")
1451 .set_description("mark any new OSD that comes up 'in'"),
1452
1453 Option("mon_osd_destroyed_out_interval", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1454 .set_default(600)
1455 .add_service("mon")
1456 .set_description("mark any OSD 'out' that has been 'destroy'ed for this long (seconds)"),
1457
1458 Option("mon_osd_down_out_interval", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1459 .set_default(600)
1460 .add_service("mon")
1461 .set_description("mark any OSD 'out' that has been 'down' for this long (seconds)"),
1462
1463 Option("mon_osd_down_out_subtree_limit", Option::TYPE_STR, Option::LEVEL_ADVANCED)
1464 .set_default("rack")
1465 .set_flag(Option::FLAG_RUNTIME)
1466 .add_service("mon")
1467 .set_description("do not automatically mark OSDs 'out' if an entire subtree of this size is down")
1468 .add_see_also("mon_osd_down_out_interval"),
1469
1470 Option("mon_osd_min_up_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
1471 .set_default(.3)
1472 .add_service("mon")
1473 .set_description("do not automatically mark OSDs 'out' if fewer than this many OSDs are 'up'")
1474 .add_see_also("mon_osd_down_out_interval"),
1475
1476 Option("mon_osd_min_in_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
1477 .set_default(.75)
1478 .add_service("mon")
1479 .set_description("do not automatically mark OSDs 'out' if fewer than this many OSDs are 'in'")
1480 .add_see_also("mon_osd_down_out_interval"),
1481
1482 Option("mon_osd_warn_op_age", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
1483 .set_default(32)
1484 .add_service("mgr")
1485 .set_description("issue REQUEST_SLOW health warning if OSD ops are slower than this age (seconds)"),
1486
1487 Option("mon_osd_err_op_age_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
1488 .set_default(128)
1489 .add_service("mgr")
1490 .set_description("issue REQUEST_STUCK health error if OSD ops are slower than is age (seconds)"),
1491
1492 Option("mon_osd_prime_pg_temp", Option::TYPE_BOOL, Option::LEVEL_DEV)
1493 .set_default(true)
1494 .add_service("mon")
1495 .set_description("minimize peering work by priming pg_temp values after a map change"),
1496
1497 Option("mon_osd_prime_pg_temp_max_time", Option::TYPE_FLOAT, Option::LEVEL_DEV)
1498 .set_default(.5)
1499 .add_service("mon")
1500 .set_description("maximum time to spend precalculating PG mappings on map change (seconds)"),
1501
1502 Option("mon_osd_prime_pg_temp_max_estimate", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
1503 .set_default(.25)
1504 .add_service("mon")
1505 .set_description("calculate all PG mappings if estimated fraction of PGs that change is above this amount"),
1506
1507 Option("mon_stat_smooth_intervals", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
1508 .set_default(6)
1509 .set_min(1)
1510 .add_service("mgr")
1511 .set_description("number of PGMaps stats over which we calc the average read/write throughput of the whole cluster"),
1512
1513 Option("mon_election_timeout", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
1514 .set_default(5)
1515 .add_service("mon")
1516 .set_description("maximum time for a mon election (seconds)"),
1517
1518 Option("mon_lease", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
1519 .set_default(5)
1520 .add_service("mon")
1521 .set_description("lease interval between quorum monitors (seconds)")
1522 .set_long_description("This setting controls how sensitive your mon quorum is to intermittent network issues or other failures."),
1523
1524 Option("mon_lease_renew_interval_factor", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
1525 .set_default(.6)
1526 .set_min_max((double)0.0, (double).9999999)
1527 .add_service("mon")
1528 .set_description("multiple of mon_lease for the lease renewal interval")
1529 .set_long_description("Leases must be renewed before they time out. A smaller value means frequent renewals, while a value close to 1 makes a lease expiration more likely.")
1530 .add_see_also("mon_lease"),
1531
1532 Option("mon_lease_ack_timeout_factor", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
1533 .set_default(2.0)
1534 .set_min_max(1.0001, 100.0)
1535 .add_service("mon")
1536 .set_description("multiple of mon_lease for the lease ack interval before calling new election")
1537 .add_see_also("mon_lease"),
1538
1539 Option("mon_accept_timeout_factor", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
1540 .set_default(2.0)
1541 .add_service("mon")
1542 .set_description("multiple of mon_lease for follower mons to accept proposed state changes before calling a new election")
1543 .add_see_also("mon_lease"),
1544
1545 Option("mon_clock_drift_allowed", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
1546 .set_default(.050)
1547 .add_service("mon")
1548 .set_description("allowed clock drift (in seconds) between mons before issuing a health warning"),
1549
1550 Option("mon_clock_drift_warn_backoff", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
1551 .set_default(5)
1552 .add_service("mon")
1553 .set_description("exponential backoff factor for logging clock drift warnings in the cluster log"),
1554
1555 Option("mon_timecheck_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
1556 .set_default(300.0)
1557 .add_service("mon")
1558 .set_description("frequency of clock synchronization checks between monitors (seconds)"),
1559
1560 Option("mon_timecheck_skew_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
1561 .set_default(30.0)
1562 .add_service("mon")
1563 .set_description("frequency of clock synchronization (re)checks between monitors while clocks are believed to be skewed (seconds)")
1564 .add_see_also("mon_timecheck_interval"),
1565
1566 Option("mon_pg_stuck_threshold", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1567 .set_default(60)
1568 .set_description("number of seconds after which pgs can be considered stuck inactive, unclean, etc")
1569 .set_long_description("see doc/control.rst under dump_stuck for more info")
1570 .add_service("mgr"),
1571
1572 Option("mon_pg_warn_min_per_osd", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
1573 .set_default(0)
1574 .add_service("mgr")
1575 .set_description("minimal number PGs per (in) osd before we warn the admin"),
1576
1577 Option("mon_max_pg_per_osd", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
1578 .set_min(1)
1579 .set_default(250)
1580 .add_service("mgr")
1581 .set_description("Max number of PGs per OSD the cluster will allow")
1582 .set_long_description("If the number of PGs per OSD exceeds this, a "
1583 "health warning will be visible in `ceph status`. This is also used "
1584 "in automated PG management, as the threshold at which some pools' "
1585 "pg_num may be shrunk in order to enable increasing the pg_num of "
1586 "others."),
1587
1588 Option("mon_target_pg_per_osd", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
1589 .set_min(1)
1590 .set_default(100)
1591 .set_description("Automated PG management creates this many PGs per OSD")
1592 .set_long_description("When creating pools, the automated PG management "
1593 "logic will attempt to reach this target. In some circumstances, it "
1594 "may exceed this target, up to the ``mon_max_pg_per_osd`` limit. "
1595 "Conversely, a lower number of PGs per OSD may be created if the "
1596 "cluster is not yet fully utilised"),
1597
1598 Option("mon_pg_warn_max_object_skew", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
1599 .set_default(10.0)
1600 .set_description("max skew few average in objects per pg")
1601 .add_service("mgr"),
1602
1603 Option("mon_pg_warn_min_objects", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1604 .set_default(10000)
1605 .set_description("do not warn below this object #")
1606 .add_service("mgr"),
1607
1608 Option("mon_pg_warn_min_pool_objects", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1609 .set_default(1000)
1610 .set_description("do not warn on pools below this object #")
1611 .add_service("mgr"),
1612
1613 Option("mon_pg_check_down_all_threshold", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
1614 .set_default(.5)
1615 .set_description("threshold of down osds after which we check all pgs")
1616 .add_service("mgr"),
1617
1618 Option("mon_cache_target_full_warn_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
1619 .set_default(.66)
1620 .add_service("mgr")
1621 .set_flag(Option::FLAG_NO_MON_UPDATE)
1622 .set_flag(Option::FLAG_CLUSTER_CREATE)
1623 .set_description("issue CACHE_POOL_NEAR_FULL health warning when cache pool utilization exceeds this ratio of usable space"),
1624
1625 Option("mon_osd_full_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
1626 .set_default(.95)
1627 .set_flag(Option::FLAG_NO_MON_UPDATE)
1628 .set_flag(Option::FLAG_CLUSTER_CREATE)
1629 .set_description("full ratio of OSDs to be set during initial creation of the cluster"),
1630
1631 Option("mon_osd_backfillfull_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
1632 .set_default(.90)
1633 .set_flag(Option::FLAG_NO_MON_UPDATE)
1634 .set_flag(Option::FLAG_CLUSTER_CREATE)
1635 .set_description(""),
1636
1637 Option("mon_osd_nearfull_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
1638 .set_default(.85)
1639 .set_flag(Option::FLAG_NO_MON_UPDATE)
1640 .set_flag(Option::FLAG_CLUSTER_CREATE)
1641 .set_description("nearfull ratio for OSDs to be set during initial creation of cluster"),
1642
1643 Option("mon_osd_initial_require_min_compat_client", Option::TYPE_STR, Option::LEVEL_ADVANCED)
1644 .set_default("jewel")
1645 .set_flag(Option::FLAG_NO_MON_UPDATE)
1646 .set_flag(Option::FLAG_CLUSTER_CREATE)
1647 .set_description(""),
1648
1649 Option("mon_allow_pool_delete", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
1650 .set_default(false)
1651 .add_service("mon")
1652 .set_description("allow pool deletions"),
1653
1654 Option("mon_fake_pool_delete", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
1655 .set_default(false)
1656 .add_service("mon")
1657 .set_description("fake pool deletions by renaming the rados pool"),
1658
1659 Option("mon_globalid_prealloc", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
1660 .set_default(10000)
1661 .add_service("mon")
1662 .set_description("number of globalid values to preallocate")
1663 .set_long_description("This setting caps how many new clients can authenticate with the cluster before the monitors have to perform a write to preallocate more. Large values burn through the 64-bit ID space more quickly."),
1664
1665 Option("mon_osd_report_timeout", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1666 .set_default(900)
1667 .add_service("mon")
1668 .set_description("time before OSDs who do not report to the mons are marked down (seconds)"),
1669
1670 Option("mon_warn_on_msgr2_not_enabled", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
1671 .set_default(true)
1672 .add_service("mon")
1673 .set_description("issue MON_MSGR2_NOT_ENABLED health warning if monitors are all running Nautilus but not all binding to a msgr2 port")
1674 .add_see_also("ms_bind_msgr2"),
1675
1676 Option("mon_warn_on_legacy_crush_tunables", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
1677 .set_default(true)
1678 .add_service("mgr")
1679 .set_description("issue OLD_CRUSH_TUNABLES health warning if CRUSH tunables are older than mon_crush_min_required_version")
1680 .add_see_also("mon_crush_min_required_version"),
1681
1682 Option("mon_crush_min_required_version", Option::TYPE_STR, Option::LEVEL_ADVANCED)
1683 .set_default("hammer")
1684 .add_service("mgr")
1685 .set_description("minimum ceph release to use for mon_warn_on_legacy_crush_tunables")
1686 .add_see_also("mon_warn_on_legacy_crush_tunables"),
1687
1688 Option("mon_warn_on_crush_straw_calc_version_zero", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
1689 .set_default(true)
1690 .add_service("mgr")
1691 .set_description("issue OLD_CRUSH_STRAW_CALC_VERSION health warning if the CRUSH map's straw_calc_version is zero"),
1692
1693 Option("mon_warn_on_osd_down_out_interval_zero", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
1694 .set_default(true)
1695 .add_service("mgr")
1696 .set_description("issue OSD_NO_DOWN_OUT_INTERVAL health warning if mon_osd_down_out_interval is zero")
1697 .set_long_description("Having mon_osd_down_out_interval set to 0 means that down OSDs are not marked out automatically and the cluster does not heal itself without administrator intervention.")
1698 .add_see_also("mon_osd_down_out_interval"),
1699
1700 Option("mon_warn_on_cache_pools_without_hit_sets", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
1701 .set_default(true)
1702 .add_service("mgr")
1703 .set_description("issue CACHE_POOL_NO_HIT_SET health warning for cache pools that do not have hit sets configured"),
1704
1705 Option("mon_warn_on_pool_no_app", Option::TYPE_BOOL, Option::LEVEL_DEV)
1706 .set_default(true)
1707 .add_service("mgr")
1708 .set_description("issue POOL_APP_NOT_ENABLED health warning if pool has not application enabled"),
1709
1710 Option("mon_warn_on_pool_pg_num_not_power_of_two", Option::TYPE_BOOL, Option::LEVEL_DEV)
1711 .set_default(true)
1712 .add_service("mon")
1713 .set_description("issue POOL_PG_NUM_NOT_POWER_OF_TWO warning if pool has a non-power-of-two pg_num value"),
1714
1715 Option("mon_warn_on_misplaced", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
1716 .set_default(false)
1717 .add_service("mgr")
1718 .set_description("Issue a health warning if there are misplaced objects"),
1719
1720 Option("mon_warn_on_too_few_osds", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
1721 .set_default(true)
1722 .add_service("mgr")
1723 .set_description("Issue a health warning if there are fewer OSDs than osd_pool_default_size"),
1724
1725 Option("mon_warn_on_slow_ping_time", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
1726 .set_default(0)
1727 .add_service("mgr")
1728 .set_description("Override mon_warn_on_slow_ping_ratio with specified threshold in milliseconds")
1729 .add_see_also("mon_warn_on_slow_ping_ratio"),
1730
1731 Option("mon_warn_on_slow_ping_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
1732 .set_default(.05)
1733 .add_service("mgr")
1734 .set_description("Issue a health warning if heartbeat ping longer than percentage of osd_heartbeat_grace")
1735 .add_see_also("osd_heartbeat_grace")
1736 .add_see_also("mon_warn_on_slow_ping_time"),
1737
1738 Option("mon_max_snap_prune_per_epoch", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
1739 .set_default(100)
1740 .add_service("mon")
1741 .set_description("max number of pruned snaps we will process in a single OSDMap epoch"),
1742
1743 Option("mon_min_osdmap_epochs", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1744 .set_default(500)
1745 .add_service("mon")
1746 .set_description("min number of OSDMaps to store"),
1747
1748 Option("mon_max_log_epochs", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1749 .set_default(500)
1750 .add_service("mon")
1751 .set_description("max number of past cluster log epochs to store"),
1752
1753 Option("mon_max_mdsmap_epochs", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1754 .set_default(500)
1755 .add_service("mon")
1756 .set_description("max number of FSMaps/MDSMaps to store"),
1757
1758 Option("mon_max_mgrmap_epochs", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1759 .set_default(500)
1760 .add_service("mon")
1761 .set_description("max number of MgrMaps to store"),
1762
1763 Option("mon_max_osd", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1764 .set_default(10000)
1765 .add_service("mon")
1766 .set_description("max number of OSDs in a cluster"),
1767
1768 Option("mon_probe_timeout", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
1769 .set_default(2.0)
1770 .add_service("mon")
1771 .set_description("timeout for querying other mons during bootstrap pre-election phase (seconds)"),
1772
1773 Option("mon_client_bytes", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
1774 .set_default(100ul << 20)
1775 .add_service("mon")
1776 .set_description("max bytes of outstanding client messages mon will read off the network"),
1777
1778 Option("mon_daemon_bytes", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
1779 .set_default(400ul << 20)
1780 .add_service("mon")
1781 .set_description("max bytes of outstanding mon messages mon will read off the network"),
1782
1783 Option("mon_mgr_proxy_client_bytes_ratio", Option::TYPE_FLOAT, Option::LEVEL_DEV)
1784 .set_default(.3)
1785 .add_service("mon")
1786 .set_description("ratio of mon_client_bytes that can be consumed by "
1787 "proxied mgr commands before we error out to client"),
1788
1789 Option("mon_log_max_summary", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
1790 .set_default(50)
1791 .add_service("mon")
1792 .set_description("number of recent cluster log messages to retain"),
1793
1794 Option("mon_max_log_entries_per_event", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1795 .set_default(4096)
1796 .add_service("mon")
1797 .set_description("max cluster log entries per paxos event"),
1798
1799 Option("mon_reweight_min_pgs_per_osd", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
1800 .set_default(10)
1801 .add_service("mgr")
1802 .set_description(""),
1803
1804 Option("mon_reweight_min_bytes_per_osd", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
1805 .set_default(100_M)
1806 .add_service("mgr")
1807 .set_description(""),
1808
1809 Option("mon_reweight_max_osds", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1810 .set_default(4)
1811 .add_service("mgr")
1812 .set_description(""),
1813
1814 Option("mon_reweight_max_change", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
1815 .set_default(0.05)
1816 .add_service("mgr")
1817 .set_description(""),
1818
1819 Option("mon_health_to_clog", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
1820 .set_default(true)
1821 .add_service("mon")
1822 .set_description("log monitor health to cluster log"),
1823
1824 Option("mon_health_to_clog_interval", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1825 .set_default(1_hr)
1826 .add_service("mon")
1827 .set_description("frequency to log monitor health to cluster log")
1828 .add_see_also("mon_health_to_clog"),
1829
1830 Option("mon_health_to_clog_tick_interval", Option::TYPE_FLOAT, Option::LEVEL_DEV)
1831 .set_default(60.0)
1832 .add_service("mon")
1833 .set_description(""),
1834
1835 Option("mon_health_max_detail", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
1836 .set_default(50)
1837 .add_service("mon")
1838 .set_description("max detailed pgs to report in health detail"),
1839
1840 Option("mon_health_log_update_period", Option::TYPE_INT, Option::LEVEL_DEV)
1841 .set_default(5)
1842 .add_service("mon")
1843 .set_description("minimum time in seconds between log messages about "
1844 "each health check")
1845 .set_min(0),
1846
1847 Option("mon_data_avail_crit", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1848 .set_default(5)
1849 .add_service("mon")
1850 .set_description("issue MON_DISK_CRIT health error when mon available space below this percentage"),
1851
1852 Option("mon_data_avail_warn", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1853 .set_default(30)
1854 .add_service("mon")
1855 .set_description("issue MON_DISK_LOW health warning when mon available space below this percentage"),
1856
1857 Option("mon_data_size_warn", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
1858 .set_default(15_G)
1859 .add_service("mon")
1860 .set_description("issue MON_DISK_BIG health warning when mon database is above this size"),
1861
1862 Option("mon_warn_pg_not_scrubbed_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
1863 .set_default(0.5)
1864 .set_min(0)
1865 .set_description("Percentage of the scrub max interval past the scrub max interval to warn")
1866 .set_long_description("")
1867 .add_see_also("osd_scrub_max_interval"),
1868
1869 Option("mon_warn_pg_not_deep_scrubbed_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
1870 .set_default(0.75)
1871 .set_min(0)
1872 .set_description("Percentage of the deep scrub interval past the deep scrub interval to warn")
1873 .set_long_description("")
1874 .add_see_also("osd_deep_scrub_interval"),
1875
1876 Option("mon_scrub_interval", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1877 .set_default(1_day)
1878 .add_service("mon")
1879 .set_description("frequency for scrubbing mon database"),
1880
1881 Option("mon_scrub_timeout", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1882 .set_default(5_min)
1883 .add_service("mon")
1884 .set_description("timeout to restart scrub of mon quorum participant does not respond for the latest chunk"),
1885
1886 Option("mon_scrub_max_keys", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1887 .set_default(100)
1888 .add_service("mon")
1889 .set_description("max keys per on scrub chunk/step"),
1890
1891 Option("mon_scrub_inject_crc_mismatch", Option::TYPE_FLOAT, Option::LEVEL_DEV)
1892 .set_default(0.0)
1893 .add_service("mon")
1894 .set_description("probability for injecting crc mismatches into mon scrub"),
1895
1896 Option("mon_scrub_inject_missing_keys", Option::TYPE_FLOAT, Option::LEVEL_DEV)
1897 .set_default(0.0)
1898 .add_service("mon")
1899 .set_description("probability for injecting missing keys into mon scrub"),
1900
1901 Option("mon_config_key_max_entry_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
1902 .set_default(64_K)
1903 .add_service("mon")
1904 .set_description("Defines the number of bytes allowed to be held in a "
1905 "single config-key entry"),
1906
1907 Option("mon_sync_timeout", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
1908 .set_default(60.0)
1909 .add_service("mon")
1910 .set_description("timeout before canceling sync if syncing mon does not respond"),
1911
1912 Option("mon_sync_max_payload_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
1913 .set_default(1_M)
1914 .add_service("mon")
1915 .set_description("target max message payload for mon sync"),
1916
1917 Option("mon_sync_debug", Option::TYPE_BOOL, Option::LEVEL_DEV)
1918 .set_default(false)
1919 .add_service("mon")
1920 .set_description("enable extra debugging during mon sync"),
1921
1922 Option("mon_inject_sync_get_chunk_delay", Option::TYPE_FLOAT, Option::LEVEL_DEV)
1923 .set_default(0)
1924 .add_service("mon")
1925 .set_description("inject delay during sync (seconds)"),
1926
1927 Option("mon_osd_min_down_reporters", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
1928 .set_default(2)
1929 .add_service("mon")
1930 .set_description("number of OSDs from different subtrees who need to report a down OSD for it to count")
1931 .add_see_also("mon_osd_reporter_subtree_level"),
1932
1933 Option("mon_osd_reporter_subtree_level", Option::TYPE_STR, Option::LEVEL_ADVANCED)
1934 .set_default("host")
1935 .add_service("mon")
1936 .set_flag(Option::FLAG_RUNTIME)
1937 .set_description("in which level of parent bucket the reporters are counted"),
1938
1939 Option("mon_osd_snap_trim_queue_warn_on", Option::TYPE_INT, Option::LEVEL_ADVANCED)
1940 .set_default(32768)
1941 .add_service("mon")
1942 .set_description("Warn when snap trim queue is that large (or larger).")
1943 .set_long_description("Warn when snap trim queue length for at least one PG crosses this value, as this is indicator of snap trimmer not keeping up, wasting disk space"),
1944
1945 Option("mon_osd_force_trim_to", Option::TYPE_INT, Option::LEVEL_DEV)
1946 .set_default(0)
1947 .add_service("mon")
1948 .set_description("force mons to trim osdmaps through this epoch"),
1949
1950 Option("mon_mds_force_trim_to", Option::TYPE_INT, Option::LEVEL_DEV)
1951 .set_default(0)
1952 .add_service("mon")
1953 .set_description("force mons to trim mdsmaps/fsmaps through this epoch"),
1954
1955 Option("mon_mds_skip_sanity", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
1956 .set_default(false)
1957 .add_service("mon")
1958 .set_description("skip sanity checks on fsmap/mdsmap"),
1959
1960 Option("mon_debug_extra_checks", Option::TYPE_BOOL, Option::LEVEL_DEV)
1961 .set_default(false)
1962 .add_service("mon")
1963 .set_description("Enable some additional monitor checks")
1964 .set_long_description(
1965 "Enable some additional monitor checks that would be too expensive "
1966 "to run on production systems, or would only be relevant while "
1967 "testing or debugging."),
1968
1969 Option("mon_debug_block_osdmap_trim", Option::TYPE_BOOL, Option::LEVEL_DEV)
1970 .set_default(false)
1971 .add_service("mon")
1972 .set_description("Block OSDMap trimming while the option is enabled.")
1973 .set_long_description(
1974 "Blocking OSDMap trimming may be quite helpful to easily reproduce "
1975 "states in which the monitor keeps (hundreds of) thousands of "
1976 "osdmaps."),
1977
1978 Option("mon_debug_deprecated_as_obsolete", Option::TYPE_BOOL, Option::LEVEL_DEV)
1979 .set_default(false)
1980 .add_service("mon")
1981 .set_description("treat deprecated mon commands as obsolete"),
1982
1983 Option("mon_debug_dump_transactions", Option::TYPE_BOOL, Option::LEVEL_DEV)
1984 .set_default(false)
1985 .add_service("mon")
1986 .set_description("dump paxos transactions to log")
1987 .add_see_also("mon_debug_dump_location"),
1988
1989 Option("mon_debug_dump_json", Option::TYPE_BOOL, Option::LEVEL_DEV)
1990 .set_default(false)
1991 .add_service("mon")
1992 .set_description("dump paxos transasctions to log as json")
1993 .add_see_also("mon_debug_dump_transactions"),
1994
1995 Option("mon_debug_dump_location", Option::TYPE_STR, Option::LEVEL_DEV)
1996 .set_default("/var/log/ceph/$cluster-$name.tdump")
1997 .add_service("mon")
1998 .set_description("file to dump paxos transactions to")
1999 .add_see_also("mon_debug_dump_transactions"),
2000
2001 Option("mon_debug_no_require_nautilus", Option::TYPE_BOOL, Option::LEVEL_DEV)
2002 .set_default(false)
2003 .add_service("mon")
2004 .set_flag(Option::FLAG_CLUSTER_CREATE)
2005 .set_description("do not set nautilus feature for new mon clusters"),
2006
2007 Option("mon_debug_no_require_octopus", Option::TYPE_BOOL, Option::LEVEL_DEV)
2008 .set_default(false)
2009 .add_service("mon")
2010 .set_flag(Option::FLAG_CLUSTER_CREATE)
2011 .set_description("do not set octopus feature for new mon clusters"),
2012
2013 Option("mon_debug_no_require_bluestore_for_ec_overwrites", Option::TYPE_BOOL, Option::LEVEL_DEV)
2014 .set_default(false)
2015 .add_service("mon")
2016 .set_description("do not require bluestore OSDs to enable EC overwrites on a rados pool"),
2017
2018 Option("mon_debug_no_initial_persistent_features", Option::TYPE_BOOL, Option::LEVEL_DEV)
2019 .set_default(false)
2020 .add_service("mon")
2021 .set_flag(Option::FLAG_CLUSTER_CREATE)
2022 .set_description("do not set any monmap features for new mon clusters"),
2023
2024 Option("mon_inject_transaction_delay_max", Option::TYPE_FLOAT, Option::LEVEL_DEV)
2025 .set_default(10.0)
2026 .add_service("mon")
2027 .set_description("max duration of injected delay in paxos"),
2028
2029 Option("mon_inject_transaction_delay_probability", Option::TYPE_FLOAT, Option::LEVEL_DEV)
2030 .set_default(0)
2031 .add_service("mon")
2032 .set_description("probability of injecting a delay in paxos"),
2033
2034 Option("mon_inject_pg_merge_bounce_probability", Option::TYPE_FLOAT, Option::LEVEL_DEV)
2035 .set_default(0)
2036 .add_service("mon")
2037 .set_description("probability of failing and reverting a pg_num decrement"),
2038
2039 Option("mon_sync_provider_kill_at", Option::TYPE_INT, Option::LEVEL_DEV)
2040 .set_default(0)
2041 .add_service("mon")
2042 .set_description("kill mon sync requester at specific point"),
2043
2044 Option("mon_sync_requester_kill_at", Option::TYPE_INT, Option::LEVEL_DEV)
2045 .set_default(0)
2046 .add_service("mon")
2047 .set_description("kill mon sync requestor at specific point"),
2048
2049 Option("mon_force_quorum_join", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
2050 .set_default(false)
2051 .add_service("mon")
2052 .set_description("force mon to rejoin quorum even though it was just removed"),
2053
2054 Option("mon_keyvaluedb", Option::TYPE_STR, Option::LEVEL_ADVANCED)
2055 .set_default("rocksdb")
2056 .set_enum_allowed({"leveldb", "rocksdb"})
2057 .set_flag(Option::FLAG_CREATE)
2058 .add_service("mon")
2059 .set_description("database backend to use for the mon database"),
2060
2061 Option("mon_debug_unsafe_allow_tier_with_nonempty_snaps", Option::TYPE_BOOL, Option::LEVEL_DEV)
2062 .set_default(false)
2063 .add_service("mon")
2064 .set_description(""),
2065
2066 Option("mon_osd_blacklist_default_expire", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
2067 .set_default(1_hr)
2068 .add_service("mon")
2069 .set_description("Duration in seconds that blacklist entries for clients "
2070 "remain in the OSD map"),
2071
2072 Option("mon_mds_blacklist_interval", Option::TYPE_FLOAT, Option::LEVEL_DEV)
2073 .set_default(1_day)
2074 .set_min(1_hr)
2075 .add_service("mon")
2076 .set_description("Duration in seconds that blacklist entries for MDS "
2077 "daemons remain in the OSD map"),
2078
2079 Option("mon_osd_crush_smoke_test", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
2080 .set_default(true)
2081 .add_service("mon")
2082 .set_description("perform a smoke test on any new CRUSH map before accepting changes"),
2083
2084 Option("mon_smart_report_timeout", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
2085 .set_default(5)
2086 .add_service("mon")
2087 .set_description("Timeout (in seconds) for smarctl to run, default is set to 5"),
2088
2089
2090 // PAXOS
2091
2092 Option("paxos_stash_full_interval", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2093 .set_default(25)
2094 .add_service("mon")
2095 .set_description(""),
2096
2097 Option("paxos_max_join_drift", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2098 .set_default(10)
2099 .add_service("mon")
2100 .set_description(""),
2101
2102 Option("paxos_propose_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
2103 .set_default(1.0)
2104 .add_service("mon")
2105 .set_description(""),
2106
2107 Option("paxos_min_wait", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
2108 .set_default(0.05)
2109 .add_service("mon")
2110 .set_description(""),
2111
2112 Option("paxos_min", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2113 .set_default(500)
2114 .add_service("mon")
2115 .set_description(""),
2116
2117 Option("paxos_trim_min", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2118 .set_default(250)
2119 .add_service("mon")
2120 .set_description(""),
2121
2122 Option("paxos_trim_max", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2123 .set_default(500)
2124 .add_service("mon")
2125 .set_description(""),
2126
2127 Option("paxos_service_trim_min", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2128 .set_default(250)
2129 .add_service("mon")
2130 .set_description(""),
2131
2132 Option("paxos_service_trim_max", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2133 .set_default(500)
2134 .add_service("mon")
2135 .set_description(""),
2136
2137 Option("paxos_kill_at", Option::TYPE_INT, Option::LEVEL_DEV)
2138 .set_default(0)
2139 .add_service("mon")
2140 .set_description(""),
2141
2142
2143 // AUTH
2144
2145 Option("auth_cluster_required", Option::TYPE_STR, Option::LEVEL_ADVANCED)
2146 .set_default("cephx")
2147 .set_description("authentication methods required by the cluster"),
2148
2149 Option("auth_service_required", Option::TYPE_STR, Option::LEVEL_ADVANCED)
2150 .set_default("cephx")
2151 .set_description("authentication methods required by service daemons"),
2152
2153 Option("auth_client_required", Option::TYPE_STR, Option::LEVEL_ADVANCED)
2154 .set_default("cephx, none")
2155 .set_flag(Option::FLAG_MINIMAL_CONF)
2156 .set_description("authentication methods allowed by clients"),
2157
2158 Option("auth_supported", Option::TYPE_STR, Option::LEVEL_ADVANCED)
2159 .set_default("")
2160 .set_description("authentication methods required (deprecated)"),
2161
2162 Option("max_rotating_auth_attempts", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2163 .set_default(10)
2164 .set_description("number of attempts to initialize rotating keys before giving up"),
2165
2166 Option("rotating_keys_bootstrap_timeout", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2167 .set_default(30)
2168 .set_description("timeout for obtaining rotating keys during bootstrap phase (seconds)"),
2169
2170 Option("rotating_keys_renewal_timeout", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2171 .set_default(10)
2172 .set_description("timeout for updating rotating keys (seconds)"),
2173
2174 Option("cephx_require_signatures", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
2175 .set_default(false)
2176 .set_description(""),
2177
2178 Option("cephx_require_version", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2179 .set_default(1)
2180 .set_description("Cephx version required (1 = pre-mimic, 2 = mimic+)"),
2181
2182 Option("cephx_cluster_require_signatures", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
2183 .set_default(false)
2184 .set_description(""),
2185
2186 Option("cephx_cluster_require_version", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2187 .set_default(1)
2188 .set_description("Cephx version required by the cluster from clients (1 = pre-mimic, 2 = mimic+)"),
2189
2190 Option("cephx_service_require_signatures", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
2191 .set_default(false)
2192 .set_description(""),
2193
2194 Option("cephx_service_require_version", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2195 .set_default(1)
2196 .set_description("Cephx version required from ceph services (1 = pre-mimic, 2 = mimic+)"),
2197
2198 Option("cephx_sign_messages", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
2199 .set_default(true)
2200 .set_description(""),
2201
2202 Option("auth_mon_ticket_ttl", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
2203 .set_default(12_hr)
2204 .set_description(""),
2205
2206 Option("auth_service_ticket_ttl", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
2207 .set_default(1_hr)
2208 .set_description(""),
2209
2210 Option("auth_debug", Option::TYPE_BOOL, Option::LEVEL_DEV)
2211 .set_default(false)
2212 .set_description(""),
2213
2214 Option("mon_client_hunt_parallel", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
2215 .set_default(3)
2216 .set_description(""),
2217
2218 Option("mon_client_hunt_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
2219 .set_default(3.0)
2220 .set_description(""),
2221
2222 Option("mon_client_ping_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
2223 .set_default(10.0)
2224 .set_description(""),
2225
2226 Option("mon_client_ping_timeout", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
2227 .set_default(30.0)
2228 .set_description(""),
2229
2230 Option("mon_client_hunt_interval_backoff", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
2231 .set_default(1.5)
2232 .set_description(""),
2233
2234 Option("mon_client_hunt_interval_min_multiple", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
2235 .set_default(1.0)
2236 .set_description(""),
2237
2238 Option("mon_client_hunt_interval_max_multiple", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
2239 .set_default(10.0)
2240 .set_description(""),
2241
2242 Option("mon_client_max_log_entries_per_message", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2243 .set_default(1000)
2244 .set_description(""),
2245
2246 Option("mon_client_directed_command_retry", Option::TYPE_INT, Option::LEVEL_DEV)
2247 .set_default(2)
2248 .set_description("Number of times to try sending a comamnd directed at a specific monitor"),
2249
2250 Option("mon_max_pool_pg_num", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
2251 .set_default(65536)
2252 .set_description(""),
2253
2254 Option("mon_pool_quota_warn_threshold", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2255 .set_default(0)
2256 .set_description("percent of quota at which to issue warnings")
2257 .add_service("mgr"),
2258
2259 Option("mon_pool_quota_crit_threshold", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2260 .set_default(0)
2261 .set_description("percent of quota at which to issue errors")
2262 .add_service("mgr"),
2263
2264 Option("crush_location", Option::TYPE_STR, Option::LEVEL_ADVANCED)
2265 .set_default("")
2266 .set_description(""),
2267
2268 Option("crush_location_hook", Option::TYPE_STR, Option::LEVEL_ADVANCED)
2269 .set_default("")
2270 .set_description(""),
2271
2272 Option("crush_location_hook_timeout", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2273 .set_default(10)
2274 .set_description(""),
2275
2276 Option("objecter_tick_interval", Option::TYPE_FLOAT, Option::LEVEL_DEV)
2277 .set_default(5.0)
2278 .set_description(""),
2279
2280 Option("objecter_timeout", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
2281 .set_default(10.0)
2282 .set_description("Seconds before in-flight op is considered 'laggy' and we query mon for the latest OSDMap"),
2283
2284 Option("objecter_inflight_op_bytes", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
2285 .set_default(100_M)
2286 .set_description("Max in-flight data in bytes (both directions)"),
2287
2288 Option("objecter_inflight_ops", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
2289 .set_default(1024)
2290 .set_description("Max in-flight operations"),
2291
2292 Option("objecter_completion_locks_per_session", Option::TYPE_UINT, Option::LEVEL_DEV)
2293 .set_default(32)
2294 .set_description(""),
2295
2296 Option("objecter_inject_no_watch_ping", Option::TYPE_BOOL, Option::LEVEL_DEV)
2297 .set_default(false)
2298 .set_description(""),
2299
2300 Option("objecter_retry_writes_after_first_reply", Option::TYPE_BOOL, Option::LEVEL_DEV)
2301 .set_default(false)
2302 .set_description(""),
2303
2304 Option("objecter_debug_inject_relock_delay", Option::TYPE_BOOL, Option::LEVEL_DEV)
2305 .set_default(false)
2306 .set_description(""),
2307
2308 Option("filer_max_purge_ops", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
2309 .set_default(10)
2310 .set_description("Max in-flight operations for purging a striped range (e.g., MDS journal)"),
2311
2312 Option("filer_max_truncate_ops", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
2313 .set_default(128)
2314 .set_description("Max in-flight operations for truncating/deleting a striped sequence (e.g., MDS journal)"),
2315
2316 Option("journaler_write_head_interval", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2317 .set_default(15)
2318 .set_description("Interval in seconds between journal header updates (to help bound replay time)"),
2319
2320 // * journal object size
2321 Option("journaler_prefetch_periods", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
2322 .set_default(10)
2323 .set_min(2) // we need at least 2 periods to make progress.
2324 .set_description("Number of striping periods to prefetch while reading MDS journal"),
2325
2326 // * journal object size
2327 Option("journaler_prezero_periods", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
2328 .set_default(5)
2329 // we need to zero at least two periods, minimum, to ensure that we
2330 // have a full empty object/period in front of us.
2331 .set_min(2)
2332 .set_description("Number of striping periods to zero head of MDS journal write position"),
2333
2334 // -- OSD --
2335 Option("osd_calc_pg_upmaps_aggressively", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
2336 .set_default(true)
2337 .set_flag(Option::FLAG_RUNTIME)
2338 .set_description("try to calculate PG upmaps more aggressively, e.g., "
2339 "by doing a fairly exhaustive search of existing PGs "
2340 "that can be unmapped or upmapped"),
2341
2342 Option("osd_calc_pg_upmaps_max_stddev", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
2343 .set_default(1.0)
2344 .set_flag(Option::FLAG_RUNTIME)
2345 .set_description("standard deviation below which there is no attempt made "
2346 "while trying to calculate PG upmaps"),
2347
2348 Option("osd_calc_pg_upmaps_local_fallback_retries", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
2349 .set_default(100)
2350 .set_flag(Option::FLAG_RUNTIME)
2351 .set_description("Maximum number of PGs we can attempt to unmap or upmap "
2352 "for a specific overfull or underfull osd per iteration "),
2353
2354 Option("osd_numa_prefer_iface", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
2355 .set_default(true)
2356 .set_flag(Option::FLAG_STARTUP)
2357 .set_description("prefer IP on network interface on same numa node as storage")
2358 .add_see_also("osd_numa_auto_affinity"),
2359
2360 Option("osd_numa_auto_affinity", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
2361 .set_default(true)
2362 .set_flag(Option::FLAG_STARTUP)
2363 .set_description("automatically set affinity to numa node when storage and network match"),
2364
2365 Option("osd_numa_node", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2366 .set_default(-1)
2367 .set_flag(Option::FLAG_STARTUP)
2368 .set_description("set affinity to a numa node (-1 for none)")
2369 .add_see_also("osd_numa_auto_affinity"),
2370
2371 Option("osd_smart_report_timeout", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
2372 .set_default(5)
2373 .set_description("Timeout (in seconds) for smarctl to run, default is set to 5"),
2374
2375 Option("osd_check_max_object_name_len_on_startup", Option::TYPE_BOOL, Option::LEVEL_DEV)
2376 .set_default(true)
2377 .set_description(""),
2378
2379 Option("osd_max_backfills", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
2380 .set_default(1)
2381 .set_description("Maximum number of concurrent local and remote backfills or recoveries per OSD ")
2382 .set_long_description("There can be osd_max_backfills local reservations AND the same remote reservations per OSD. So a value of 1 lets this OSD participate as 1 PG primary in recovery and 1 shard of another recovering PG."),
2383
2384 Option("osd_min_recovery_priority", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2385 .set_default(0)
2386 .set_description("Minimum priority below which recovery is not performed")
2387 .set_long_description("The purpose here is to prevent the cluster from doing *any* lower priority work (e.g., rebalancing) below this threshold and focus solely on higher priority work (e.g., replicating degraded objects)."),
2388
2389 Option("osd_backfill_retry_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
2390 .set_default(30.0)
2391 .set_description("how frequently to retry backfill reservations after being denied (e.g., due to a full OSD)"),
2392
2393 Option("osd_recovery_retry_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
2394 .set_default(30.0)
2395 .set_description("how frequently to retry recovery reservations after being denied (e.g., due to a full OSD)"),
2396
2397 Option("osd_agent_max_ops", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2398 .set_default(4)
2399 .set_description("maximum concurrent tiering operations for tiering agent"),
2400
2401 Option("osd_agent_max_low_ops", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2402 .set_default(2)
2403 .set_description("maximum concurrent low-priority tiering operations for tiering agent"),
2404
2405 Option("osd_agent_min_evict_effort", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
2406 .set_default(.1)
2407 .set_min_max(0.0, .99)
2408 .set_description("minimum effort to expend evicting clean objects"),
2409
2410 Option("osd_agent_quantize_effort", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
2411 .set_default(.1)
2412 .set_description("size of quantize unit for eviction effort"),
2413
2414 Option("osd_agent_delay_time", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
2415 .set_default(5.0)
2416 .set_description("how long agent should sleep if it has no work to do"),
2417
2418 Option("osd_find_best_info_ignore_history_les", Option::TYPE_BOOL, Option::LEVEL_DEV)
2419 .set_default(false)
2420 .set_description("ignore last_epoch_started value when peering AND PROBABLY LOSE DATA")
2421 .set_long_description("THIS IS AN EXTREMELY DANGEROUS OPTION THAT SHOULD ONLY BE USED AT THE DIRECTION OF A DEVELOPER. It makes peering ignore the last_epoch_started value when peering, which can allow the OSD to believe an OSD has an authoritative view of a PG's contents even when it is in fact old and stale, typically leading to data loss (by believing a stale PG is up to date)."),
2422
2423 Option("osd_agent_hist_halflife", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2424 .set_default(1000)
2425 .set_description("halflife of agent atime and temp histograms"),
2426
2427 Option("osd_agent_slop", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
2428 .set_default(.02)
2429 .set_description("slop factor to avoid switching tiering flush and eviction mode"),
2430
2431 Option("osd_uuid", Option::TYPE_UUID, Option::LEVEL_ADVANCED)
2432 .set_default(uuid_d())
2433 .set_flag(Option::FLAG_CREATE)
2434 .set_description("uuid label for a new OSD"),
2435
2436 Option("osd_data", Option::TYPE_STR, Option::LEVEL_ADVANCED)
2437 .set_default("/var/lib/ceph/osd/$cluster-$id")
2438 .set_flag(Option::FLAG_NO_MON_UPDATE)
2439 .set_description("path to OSD data"),
2440
2441 Option("osd_journal", Option::TYPE_STR, Option::LEVEL_ADVANCED)
2442 .set_default("/var/lib/ceph/osd/$cluster-$id/journal")
2443 .set_flag(Option::FLAG_NO_MON_UPDATE)
2444 .set_description("path to OSD journal (when FileStore backend is in use)"),
2445
2446 Option("osd_journal_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
2447 .set_default(5120)
2448 .set_flag(Option::FLAG_CREATE)
2449 .set_description("size of FileStore journal (in MiB)"),
2450
2451 Option("osd_journal_flush_on_shutdown", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
2452 .set_default(true)
2453 .set_description("flush FileStore journal contents during clean OSD shutdown"),
2454
2455 Option("osd_os_flags", Option::TYPE_UINT, Option::LEVEL_DEV)
2456 .set_default(0)
2457 .set_description("flags to skip filestore omap or journal initialization"),
2458
2459 Option("osd_max_write_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
2460 .set_min(4)
2461 .set_default(90)
2462 .set_description("Maximum size of a RADOS write operation in megabytes")
2463 .set_long_description("This setting prevents clients from doing "
2464 "very large writes to RADOS. If you set this to a value "
2465 "below what clients expect, they will receive an error "
2466 "when attempting to write to the cluster."),
2467
2468 Option("osd_max_pgls", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
2469 .set_default(1024)
2470 .set_description("maximum number of results when listing objects in a pool"),
2471
2472 Option("osd_client_message_size_cap", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
2473 .set_default(500_M)
2474 .set_description("maximum memory to devote to in-flight client requests")
2475 .set_long_description("If this value is exceeded, the OSD will not read any new client data off of the network until memory is freed."),
2476
2477 Option("osd_client_message_cap", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
2478 .set_default(100)
2479 .set_description("maximum number of in-flight client requests"),
2480
2481 Option("osd_crush_update_weight_set", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
2482 .set_default(true)
2483 .set_description("update CRUSH weight-set weights when updating weights")
2484 .set_long_description("If this setting is true, we will update the weight-set weights when adjusting an item's weight, effectively making changes take effect immediately, and discarding any previous optimization in the weight-set value. Setting this value to false will leave it to the balancer to (slowly, presumably) adjust weights to approach the new target value."),
2485
2486 Option("osd_crush_chooseleaf_type", Option::TYPE_INT, Option::LEVEL_DEV)
2487 .set_default(1)
2488 .set_flag(Option::FLAG_CREATE)
2489 .set_description("default chooseleaf type for osdmaptool --create"),
2490
2491 Option("osd_pool_use_gmt_hitset", Option::TYPE_BOOL, Option::LEVEL_DEV)
2492 .set_default(true)
2493 .set_description("use UTC for hitset timestamps")
2494 .set_long_description("This setting only exists for compatibility with hammer (and older) clusters."),
2495
2496 Option("osd_crush_update_on_start", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
2497 .set_default(true)
2498 .set_description("update OSD CRUSH location on startup"),
2499
2500 Option("osd_class_update_on_start", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
2501 .set_default(true)
2502 .set_description("set OSD device class on startup"),
2503
2504 Option("osd_crush_initial_weight", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
2505 .set_default(-1)
2506 .set_description("if >= 0, initial CRUSH weight for newly created OSDs")
2507 .set_long_description("If this value is negative, the size of the OSD in TiB is used."),
2508
2509 Option("osd_pool_default_ec_fast_read", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
2510 .set_default(false)
2511 .set_description("set ec_fast_read for new erasure-coded pools")
2512 .add_service("mon"),
2513
2514 Option("osd_pool_default_crush_rule", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2515 .set_default(-1)
2516 .set_description("CRUSH rule for newly created pools")
2517 .add_service("mon"),
2518
2519 Option("osd_pool_erasure_code_stripe_unit", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
2520 .set_default(4_K)
2521 .set_description("the amount of data (in bytes) in a data chunk, per stripe")
2522 .add_service("mon"),
2523
2524 Option("osd_pool_default_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
2525 .set_default(3)
2526 .set_min_max(0, 10)
2527 .set_flag(Option::FLAG_RUNTIME)
2528 .set_description("the number of copies of an object for new replicated pools")
2529 .add_service("mon"),
2530
2531 Option("osd_pool_default_min_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
2532 .set_default(0)
2533 .set_min_max(0, 255)
2534 .set_flag(Option::FLAG_RUNTIME)
2535 .set_description("the minimal number of copies allowed to write to a degraded pool for new replicated pools")
2536 .set_long_description("0 means no specific default; ceph will use size-size/2")
2537 .add_see_also("osd_pool_default_size")
2538 .add_service("mon"),
2539
2540 Option("osd_pool_default_pg_num", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
2541 .set_default(8)
2542 .set_description("number of PGs for new pools")
2543 .set_flag(Option::FLAG_RUNTIME)
2544 .add_service("mon"),
2545
2546 Option("osd_pool_default_pgp_num", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
2547 .set_default(0)
2548 .set_description("number of PGs for placement purposes (0 to match pg_num)")
2549 .add_see_also("osd_pool_default_pg_num")
2550 .set_flag(Option::FLAG_RUNTIME)
2551 .add_service("mon"),
2552
2553 Option("osd_pool_default_type", Option::TYPE_STR, Option::LEVEL_ADVANCED)
2554 .set_default("replicated")
2555 .set_enum_allowed({"replicated", "erasure"})
2556 .set_flag(Option::FLAG_RUNTIME)
2557 .set_description("default type of pool to create")
2558 .add_service("mon"),
2559
2560 Option("osd_pool_default_erasure_code_profile", Option::TYPE_STR, Option::LEVEL_ADVANCED)
2561 .set_default("plugin=jerasure technique=reed_sol_van k=2 m=2")
2562 .set_flag(Option::FLAG_RUNTIME)
2563 .set_description("default erasure code profile for new erasure-coded pools")
2564 .add_service("mon"),
2565
2566 Option("osd_erasure_code_plugins", Option::TYPE_STR, Option::LEVEL_ADVANCED)
2567 .set_default("jerasure lrc"
2568 #ifdef HAVE_BETTER_YASM_ELF64
2569 " isa"
2570 #endif
2571 )
2572 .set_flag(Option::FLAG_STARTUP)
2573 .set_description("erasure code plugins to load")
2574 .add_service("mon")
2575 .add_service("osd"),
2576
2577 Option("osd_allow_recovery_below_min_size", Option::TYPE_BOOL, Option::LEVEL_DEV)
2578 .set_default(true)
2579 .set_description("allow replicated pools to recover with < min_size active members")
2580 .add_service("osd"),
2581
2582 Option("osd_pool_default_flags", Option::TYPE_INT, Option::LEVEL_DEV)
2583 .set_default(0)
2584 .set_description("(integer) flags to set on new pools")
2585 .add_service("mon"),
2586
2587 Option("osd_pool_default_flag_hashpspool", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
2588 .set_default(true)
2589 .set_description("set hashpspool (better hashing scheme) flag on new pools")
2590 .add_service("mon"),
2591
2592 Option("osd_pool_default_flag_nodelete", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
2593 .set_default(false)
2594 .set_description("set nodelete flag on new pools")
2595 .add_service("mon"),
2596
2597 Option("osd_pool_default_flag_nopgchange", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
2598 .set_default(false)
2599 .set_description("set nopgchange flag on new pools")
2600 .add_service("mon"),
2601
2602 Option("osd_pool_default_flag_nosizechange", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
2603 .set_default(false)
2604 .set_description("set nosizechange flag on new pools")
2605 .add_service("mon"),
2606
2607 Option("osd_pool_default_hit_set_bloom_fpp", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
2608 .set_default(.05)
2609 .set_description("")
2610 .add_see_also("osd_tier_default_cache_hit_set_type")
2611 .add_service("mon"),
2612
2613 Option("osd_pool_default_cache_target_dirty_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
2614 .set_default(.4)
2615 .set_description(""),
2616
2617 Option("osd_pool_default_cache_target_dirty_high_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
2618 .set_default(.6)
2619 .set_description(""),
2620
2621 Option("osd_pool_default_cache_target_full_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
2622 .set_default(.8)
2623 .set_description(""),
2624
2625 Option("osd_pool_default_cache_min_flush_age", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2626 .set_default(0)
2627 .set_description(""),
2628
2629 Option("osd_pool_default_cache_min_evict_age", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2630 .set_default(0)
2631 .set_description(""),
2632
2633 Option("osd_pool_default_cache_max_evict_check_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2634 .set_default(10)
2635 .set_description(""),
2636
2637 Option("osd_pool_default_pg_autoscale_mode", Option::TYPE_STR, Option::LEVEL_ADVANCED)
2638 .set_default("on")
2639 .set_flag(Option::FLAG_RUNTIME)
2640 .set_enum_allowed({"off", "warn", "on"})
2641 .set_description("Default PG autoscaling behavior for new pools"),
2642
2643 Option("osd_pool_default_read_lease_ratio", Option::TYPE_FLOAT, Option::LEVEL_DEV)
2644 .set_default(.8)
2645 .set_flag(Option::FLAG_RUNTIME)
2646 .set_description("Default read_lease_ratio for a pool, as a multiple of osd_heartbeat_grace")
2647 .set_long_description("This should be <= 1.0 so that the read lease will have expired by the time we decide to mark a peer OSD down.")
2648 .add_see_also("osd_heartbeat_grace"),
2649
2650 Option("osd_hit_set_min_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2651 .set_default(1000)
2652 .set_description(""),
2653
2654 Option("osd_hit_set_max_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2655 .set_default(100000)
2656 .set_description(""),
2657
2658 Option("osd_hit_set_namespace", Option::TYPE_STR, Option::LEVEL_ADVANCED)
2659 .set_default(".ceph-internal")
2660 .set_description(""),
2661
2662 Option("osd_tier_promote_max_objects_sec", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
2663 .set_default(25)
2664 .set_description(""),
2665
2666 Option("osd_tier_promote_max_bytes_sec", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
2667 .set_default(5_M)
2668 .set_description(""),
2669
2670 Option("osd_tier_default_cache_mode", Option::TYPE_STR, Option::LEVEL_ADVANCED)
2671 .set_default("writeback")
2672 .set_enum_allowed({"none", "writeback", "forward",
2673 "readonly", "readforward", "readproxy", "proxy"})
2674 .set_flag(Option::FLAG_RUNTIME)
2675 .set_description(""),
2676
2677 Option("osd_tier_default_cache_hit_set_count", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
2678 .set_default(4)
2679 .set_description(""),
2680
2681 Option("osd_tier_default_cache_hit_set_period", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
2682 .set_default(1200)
2683 .set_description(""),
2684
2685 Option("osd_tier_default_cache_hit_set_type", Option::TYPE_STR, Option::LEVEL_ADVANCED)
2686 .set_default("bloom")
2687 .set_enum_allowed({"bloom", "explicit_hash", "explicit_object"})
2688 .set_flag(Option::FLAG_RUNTIME)
2689 .set_description(""),
2690
2691 Option("osd_tier_default_cache_min_read_recency_for_promote", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
2692 .set_default(1)
2693 .set_description("number of recent HitSets the object must appear in to be promoted (on read)"),
2694
2695 Option("osd_tier_default_cache_min_write_recency_for_promote", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
2696 .set_default(1)
2697 .set_description("number of recent HitSets the object must appear in to be promoted (on write)"),
2698
2699 Option("osd_tier_default_cache_hit_set_grade_decay_rate", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
2700 .set_default(20)
2701 .set_description(""),
2702
2703 Option("osd_tier_default_cache_hit_set_search_last_n", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
2704 .set_default(1)
2705 .set_description(""),
2706
2707 Option("osd_objecter_finishers", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2708 .set_default(1)
2709 .set_flag(Option::FLAG_STARTUP)
2710 .set_description(""),
2711
2712 Option("osd_map_dedup", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
2713 .set_default(true)
2714 .set_description(""),
2715
2716 Option("osd_map_cache_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2717 .set_default(50)
2718 .set_description(""),
2719
2720 Option("osd_map_message_max", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2721 .set_default(40)
2722 .set_description("maximum number of OSDMaps to include in a single message"),
2723
2724 Option("osd_map_message_max_bytes", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
2725 .set_default(10_M)
2726 .set_description("maximum number of bytes worth of OSDMaps to include in a single message"),
2727
2728 Option("osd_map_share_max_epochs", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2729 .set_default(40)
2730 .set_description(""),
2731
2732 Option("osd_pg_epoch_max_lag_factor", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
2733 .set_default(2.0)
2734 .set_description("Max multiple of the map cache that PGs can lag before we throttle map injest")
2735 .add_see_also("osd_map_cache_size"),
2736
2737 Option("osd_inject_bad_map_crc_probability", Option::TYPE_FLOAT, Option::LEVEL_DEV)
2738 .set_default(0)
2739 .set_description(""),
2740
2741 Option("osd_inject_failure_on_pg_removal", Option::TYPE_BOOL, Option::LEVEL_DEV)
2742 .set_default(false)
2743 .set_description(""),
2744
2745 Option("osd_max_markdown_period", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2746 .set_default(600)
2747 .set_description(""),
2748
2749 Option("osd_max_markdown_count", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2750 .set_default(5)
2751 .set_description(""),
2752
2753 Option("osd_op_pq_max_tokens_per_priority", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
2754 .set_default(4194304)
2755 .set_description(""),
2756
2757 Option("osd_op_pq_min_cost", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
2758 .set_default(65536)
2759 .set_description(""),
2760
2761 Option("osd_recover_clone_overlap", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
2762 .set_default(true)
2763 .set_description(""),
2764
2765 Option("osd_op_num_threads_per_shard", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2766 .set_default(0)
2767 .set_flag(Option::FLAG_STARTUP)
2768 .set_description(""),
2769
2770 Option("osd_op_num_threads_per_shard_hdd", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2771 .set_default(1)
2772 .set_flag(Option::FLAG_STARTUP)
2773 .set_description("")
2774 .add_see_also("osd_op_num_threads_per_shard"),
2775
2776 Option("osd_op_num_threads_per_shard_ssd", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2777 .set_default(2)
2778 .set_flag(Option::FLAG_STARTUP)
2779 .set_description("")
2780 .add_see_also("osd_op_num_threads_per_shard"),
2781
2782 Option("osd_op_num_shards", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2783 .set_default(0)
2784 .set_flag(Option::FLAG_STARTUP)
2785 .set_description(""),
2786
2787 Option("osd_op_num_shards_hdd", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2788 .set_default(5)
2789 .set_flag(Option::FLAG_STARTUP)
2790 .set_description("")
2791 .add_see_also("osd_op_num_shards"),
2792
2793 Option("osd_op_num_shards_ssd", Option::TYPE_INT, Option::LEVEL_ADVANCED)
2794 .set_default(8)
2795 .set_flag(Option::FLAG_STARTUP)
2796 .set_description("")
2797 .add_see_also("osd_op_num_shards"),
2798
2799 Option("osd_skip_data_digest", Option::TYPE_BOOL, Option::LEVEL_DEV)
2800 .set_default(false)
2801 .set_description("Do not store full-object checksums if the backend (bluestore) does its own checksums. Only usable with all BlueStore OSDs."),
2802
2803 Option("osd_op_queue", Option::TYPE_STR, Option::LEVEL_ADVANCED)
2804 .set_default("wpq")
2805 .set_enum_allowed( { "wpq", "prioritized", "mclock_opclass", "mclock_client", "debug_random" } )
2806 .set_description("which operation queue algorithm to use")
2807 .set_long_description("which operation queue algorithm to use; mclock_opclass and mclock_client are currently experimental")
2808 .set_flag(Option::FLAG_STARTUP)
2809 .add_see_also("osd_op_queue_cut_off"),
2810
2811 Option("osd_op_queue_cut_off", Option::TYPE_STR, Option::LEVEL_ADVANCED)
2812 .set_default("high")
2813 .set_enum_allowed( { "low", "high", "debug_random" } )
2814 .set_description("the threshold between high priority ops and low priority ops")
2815 .set_long_description("the threshold between high priority ops that use strict priority ordering and low priority ops that use a fairness algorithm that may or may not incorporate priority")
2816 .add_see_also("osd_op_queue"),
2817
2818 Option("osd_op_queue_mclock_client_op_res", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
2819 .set_default(1000.0)
2820 .set_description("mclock reservation of client operator requests")
2821 .set_long_description("mclock reservation of client operator requests when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the reservation")
2822 .add_see_also("osd_op_queue")
2823 .add_see_also("osd_op_queue_mclock_client_op_wgt")
2824 .add_see_also("osd_op_queue_mclock_client_op_lim")
2825 .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
2826 .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
2827 .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
2828 .add_see_also("osd_op_queue_mclock_snap_res")
2829 .add_see_also("osd_op_queue_mclock_snap_wgt")
2830 .add_see_also("osd_op_queue_mclock_snap_lim")
2831 .add_see_also("osd_op_queue_mclock_recov_res")
2832 .add_see_also("osd_op_queue_mclock_recov_wgt")
2833 .add_see_also("osd_op_queue_mclock_recov_lim")
2834 .add_see_also("osd_op_queue_mclock_scrub_res")
2835 .add_see_also("osd_op_queue_mclock_scrub_wgt")
2836 .add_see_also("osd_op_queue_mclock_scrub_lim")
2837 .add_see_also("osd_op_queue_mclock_anticipation_timeout"),
2838
2839 Option("osd_op_queue_mclock_client_op_wgt", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
2840 .set_default(500.0)
2841 .set_description("mclock weight of client operator requests")
2842 .set_long_description("mclock weight of client operator requests when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the weight")
2843 .add_see_also("osd_op_queue")
2844 .add_see_also("osd_op_queue_mclock_client_op_res")
2845 .add_see_also("osd_op_queue_mclock_client_op_lim")
2846 .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
2847 .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
2848 .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
2849 .add_see_also("osd_op_queue_mclock_snap_res")
2850 .add_see_also("osd_op_queue_mclock_snap_wgt")
2851 .add_see_also("osd_op_queue_mclock_snap_lim")
2852 .add_see_also("osd_op_queue_mclock_recov_res")
2853 .add_see_also("osd_op_queue_mclock_recov_wgt")
2854 .add_see_also("osd_op_queue_mclock_recov_lim")
2855 .add_see_also("osd_op_queue_mclock_scrub_res")
2856 .add_see_also("osd_op_queue_mclock_scrub_wgt")
2857 .add_see_also("osd_op_queue_mclock_scrub_lim")
2858 .add_see_also("osd_op_queue_mclock_anticipation_timeout"),
2859
2860 Option("osd_op_queue_mclock_client_op_lim", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
2861 .set_default(0.0)
2862 .set_description("mclock limit of client operator requests")
2863 .set_long_description("mclock limit of client operator requests when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the limit")
2864 .add_see_also("osd_op_queue")
2865 .add_see_also("osd_op_queue_mclock_client_op_res")
2866 .add_see_also("osd_op_queue_mclock_client_op_wgt")
2867 .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
2868 .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
2869 .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
2870 .add_see_also("osd_op_queue_mclock_snap_res")
2871 .add_see_also("osd_op_queue_mclock_snap_wgt")
2872 .add_see_also("osd_op_queue_mclock_snap_lim")
2873 .add_see_also("osd_op_queue_mclock_recov_res")
2874 .add_see_also("osd_op_queue_mclock_recov_wgt")
2875 .add_see_also("osd_op_queue_mclock_recov_lim")
2876 .add_see_also("osd_op_queue_mclock_scrub_res")
2877 .add_see_also("osd_op_queue_mclock_scrub_wgt")
2878 .add_see_also("osd_op_queue_mclock_scrub_lim")
2879 .add_see_also("osd_op_queue_mclock_anticipation_timeout"),
2880
2881 Option("osd_op_queue_mclock_osd_rep_op_res", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
2882 .set_default(1000.0)
2883 .set_description("mclock reservation of osd replication operation requests and replies")
2884 .set_long_description("mclock reservation of replication operation requests and replies when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the reservation")
2885 .add_see_also("osd_op_queue")
2886 .add_see_also("osd_op_queue_mclock_client_op_res")
2887 .add_see_also("osd_op_queue_mclock_client_op_wgt")
2888 .add_see_also("osd_op_queue_mclock_client_op_lim")
2889 .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
2890 .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
2891 .add_see_also("osd_op_queue_mclock_snap_res")
2892 .add_see_also("osd_op_queue_mclock_snap_wgt")
2893 .add_see_also("osd_op_queue_mclock_snap_lim")
2894 .add_see_also("osd_op_queue_mclock_recov_res")
2895 .add_see_also("osd_op_queue_mclock_recov_wgt")
2896 .add_see_also("osd_op_queue_mclock_recov_lim")
2897 .add_see_also("osd_op_queue_mclock_scrub_res")
2898 .add_see_also("osd_op_queue_mclock_scrub_wgt")
2899 .add_see_also("osd_op_queue_mclock_scrub_lim")
2900 .add_see_also("osd_op_queue_mclock_anticipation_timeout"),
2901
2902 Option("osd_op_queue_mclock_osd_rep_op_wgt", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
2903 .set_default(500.0)
2904 .set_description("mclock weight of osd replication operation requests and replies")
2905 .set_long_description("mclock weight of osd replication operation requests and replies when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the weight")
2906 .add_see_also("osd_op_queue")
2907 .add_see_also("osd_op_queue_mclock_client_op_res")
2908 .add_see_also("osd_op_queue_mclock_client_op_wgt")
2909 .add_see_also("osd_op_queue_mclock_client_op_lim")
2910 .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
2911 .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
2912 .add_see_also("osd_op_queue_mclock_snap_res")
2913 .add_see_also("osd_op_queue_mclock_snap_wgt")
2914 .add_see_also("osd_op_queue_mclock_snap_lim")
2915 .add_see_also("osd_op_queue_mclock_recov_res")
2916 .add_see_also("osd_op_queue_mclock_recov_wgt")
2917 .add_see_also("osd_op_queue_mclock_recov_lim")
2918 .add_see_also("osd_op_queue_mclock_scrub_res")
2919 .add_see_also("osd_op_queue_mclock_scrub_wgt")
2920 .add_see_also("osd_op_queue_mclock_scrub_lim")
2921 .add_see_also("osd_op_queue_mclock_anticipation_timeout"),
2922
2923 Option("osd_op_queue_mclock_osd_rep_op_lim", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
2924 .set_default(0.0)
2925 .set_description("mclock limit of osd replication operation requests and replies")
2926 .set_long_description("mclock limit of osd sub-operation requests when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the limit")
2927 .add_see_also("osd_op_queue")
2928 .add_see_also("osd_op_queue_mclock_client_op_res")
2929 .add_see_also("osd_op_queue_mclock_client_op_wgt")
2930 .add_see_also("osd_op_queue_mclock_client_op_lim")
2931 .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
2932 .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
2933 .add_see_also("osd_op_queue_mclock_snap_res")
2934 .add_see_also("osd_op_queue_mclock_snap_wgt")
2935 .add_see_also("osd_op_queue_mclock_snap_lim")
2936 .add_see_also("osd_op_queue_mclock_recov_res")
2937 .add_see_also("osd_op_queue_mclock_recov_wgt")
2938 .add_see_also("osd_op_queue_mclock_recov_lim")
2939 .add_see_also("osd_op_queue_mclock_scrub_res")
2940 .add_see_also("osd_op_queue_mclock_scrub_wgt")
2941 .add_see_also("osd_op_queue_mclock_scrub_lim")
2942 .add_see_also("osd_op_queue_mclock_anticipation_timeout"),
2943
2944 Option("osd_op_queue_mclock_snap_res", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
2945 .set_default(0.0)
2946 .set_description("mclock reservation of snaptrim requests")
2947 .set_long_description("mclock reservation of snaptrim requests when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the reservation")
2948 .add_see_also("osd_op_queue")
2949 .add_see_also("osd_op_queue_mclock_client_op_res")
2950 .add_see_also("osd_op_queue_mclock_client_op_wgt")
2951 .add_see_also("osd_op_queue_mclock_client_op_lim")
2952 .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
2953 .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
2954 .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
2955 .add_see_also("osd_op_queue_mclock_snap_wgt")
2956 .add_see_also("osd_op_queue_mclock_snap_lim")
2957 .add_see_also("osd_op_queue_mclock_recov_res")
2958 .add_see_also("osd_op_queue_mclock_recov_wgt")
2959 .add_see_also("osd_op_queue_mclock_recov_lim")
2960 .add_see_also("osd_op_queue_mclock_scrub_res")
2961 .add_see_also("osd_op_queue_mclock_scrub_wgt")
2962 .add_see_also("osd_op_queue_mclock_scrub_lim")
2963 .add_see_also("osd_op_queue_mclock_anticipation_timeout"),
2964
2965 Option("osd_op_queue_mclock_snap_wgt", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
2966 .set_default(1.0)
2967 .set_description("mclock weight of snaptrim requests")
2968 .set_long_description("mclock weight of snaptrim requests when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the weight")
2969 .add_see_also("osd_op_queue")
2970 .add_see_also("osd_op_queue_mclock_client_op_res")
2971 .add_see_also("osd_op_queue_mclock_client_op_wgt")
2972 .add_see_also("osd_op_queue_mclock_client_op_lim")
2973 .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
2974 .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
2975 .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
2976 .add_see_also("osd_op_queue_mclock_snap_res")
2977 .add_see_also("osd_op_queue_mclock_snap_lim")
2978 .add_see_also("osd_op_queue_mclock_recov_res")
2979 .add_see_also("osd_op_queue_mclock_recov_wgt")
2980 .add_see_also("osd_op_queue_mclock_recov_lim")
2981 .add_see_also("osd_op_queue_mclock_scrub_res")
2982 .add_see_also("osd_op_queue_mclock_scrub_wgt")
2983 .add_see_also("osd_op_queue_mclock_scrub_lim")
2984 .add_see_also("osd_op_queue_mclock_anticipation_timeout"),
2985
2986 Option("osd_op_queue_mclock_snap_lim", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
2987 .set_default(0.001)
2988 .set_description("")
2989 .set_description("mclock limit of snaptrim requests")
2990 .set_long_description("mclock limit of snaptrim requests when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the limit")
2991 .add_see_also("osd_op_queue_mclock_client_op_res")
2992 .add_see_also("osd_op_queue_mclock_client_op_wgt")
2993 .add_see_also("osd_op_queue_mclock_client_op_lim")
2994 .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
2995 .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
2996 .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
2997 .add_see_also("osd_op_queue_mclock_snap_res")
2998 .add_see_also("osd_op_queue_mclock_snap_wgt")
2999 .add_see_also("osd_op_queue_mclock_recov_res")
3000 .add_see_also("osd_op_queue_mclock_recov_wgt")
3001 .add_see_also("osd_op_queue_mclock_recov_lim")
3002 .add_see_also("osd_op_queue_mclock_scrub_res")
3003 .add_see_also("osd_op_queue_mclock_scrub_wgt")
3004 .add_see_also("osd_op_queue_mclock_scrub_lim")
3005 .add_see_also("osd_op_queue_mclock_anticipation_timeout"),
3006
3007 Option("osd_op_queue_mclock_recov_res", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3008 .set_default(0.0)
3009 .set_description("mclock reservation of recovery requests")
3010 .set_long_description("mclock reservation of recovery requests when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the reservation")
3011 .add_see_also("osd_op_queue")
3012 .add_see_also("osd_op_queue_mclock_client_op_res")
3013 .add_see_also("osd_op_queue_mclock_client_op_wgt")
3014 .add_see_also("osd_op_queue_mclock_client_op_lim")
3015 .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
3016 .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
3017 .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
3018 .add_see_also("osd_op_queue_mclock_snap_res")
3019 .add_see_also("osd_op_queue_mclock_snap_wgt")
3020 .add_see_also("osd_op_queue_mclock_snap_lim")
3021 .add_see_also("osd_op_queue_mclock_recov_wgt")
3022 .add_see_also("osd_op_queue_mclock_recov_lim")
3023 .add_see_also("osd_op_queue_mclock_scrub_res")
3024 .add_see_also("osd_op_queue_mclock_scrub_wgt")
3025 .add_see_also("osd_op_queue_mclock_scrub_lim")
3026 .add_see_also("osd_op_queue_mclock_anticipation_timeout"),
3027
3028 Option("osd_op_queue_mclock_recov_wgt", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3029 .set_default(1.0)
3030 .set_description("mclock weight of recovery requests")
3031 .set_long_description("mclock weight of recovery requests when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the weight")
3032 .add_see_also("osd_op_queue")
3033 .add_see_also("osd_op_queue_mclock_client_op_res")
3034 .add_see_also("osd_op_queue_mclock_client_op_wgt")
3035 .add_see_also("osd_op_queue_mclock_client_op_lim")
3036 .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
3037 .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
3038 .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
3039 .add_see_also("osd_op_queue_mclock_snap_res")
3040 .add_see_also("osd_op_queue_mclock_snap_wgt")
3041 .add_see_also("osd_op_queue_mclock_snap_lim")
3042 .add_see_also("osd_op_queue_mclock_recov_res")
3043 .add_see_also("osd_op_queue_mclock_recov_lim")
3044 .add_see_also("osd_op_queue_mclock_scrub_res")
3045 .add_see_also("osd_op_queue_mclock_scrub_wgt")
3046 .add_see_also("osd_op_queue_mclock_scrub_lim")
3047 .add_see_also("osd_op_queue_mclock_anticipation_timeout"),
3048
3049 Option("osd_op_queue_mclock_recov_lim", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3050 .set_default(0.001)
3051 .set_description("mclock limit of recovery requests")
3052 .set_long_description("mclock limit of recovery requests when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the limit")
3053 .add_see_also("osd_op_queue")
3054 .add_see_also("osd_op_queue_mclock_client_op_res")
3055 .add_see_also("osd_op_queue_mclock_client_op_wgt")
3056 .add_see_also("osd_op_queue_mclock_client_op_lim")
3057 .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
3058 .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
3059 .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
3060 .add_see_also("osd_op_queue_mclock_snap_res")
3061 .add_see_also("osd_op_queue_mclock_snap_wgt")
3062 .add_see_also("osd_op_queue_mclock_snap_lim")
3063 .add_see_also("osd_op_queue_mclock_recov_res")
3064 .add_see_also("osd_op_queue_mclock_recov_wgt")
3065 .add_see_also("osd_op_queue_mclock_scrub_res")
3066 .add_see_also("osd_op_queue_mclock_scrub_wgt")
3067 .add_see_also("osd_op_queue_mclock_scrub_lim")
3068 .add_see_also("osd_op_queue_mclock_anticipation_timeout"),
3069
3070 Option("osd_op_queue_mclock_scrub_res", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3071 .set_default(0.0)
3072 .set_description("mclock reservation of scrub requests")
3073 .set_long_description("mclock reservation of scrub requests when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the reservation")
3074 .add_see_also("osd_op_queue")
3075 .add_see_also("osd_op_queue_mclock_client_op_res")
3076 .add_see_also("osd_op_queue_mclock_client_op_wgt")
3077 .add_see_also("osd_op_queue_mclock_client_op_lim")
3078 .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
3079 .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
3080 .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
3081 .add_see_also("osd_op_queue_mclock_snap_res")
3082 .add_see_also("osd_op_queue_mclock_snap_wgt")
3083 .add_see_also("osd_op_queue_mclock_snap_lim")
3084 .add_see_also("osd_op_queue_mclock_recov_res")
3085 .add_see_also("osd_op_queue_mclock_recov_wgt")
3086 .add_see_also("osd_op_queue_mclock_recov_lim")
3087 .add_see_also("osd_op_queue_mclock_scrub_wgt")
3088 .add_see_also("osd_op_queue_mclock_scrub_lim")
3089 .add_see_also("osd_op_queue_mclock_anticipation_timeout"),
3090
3091 Option("osd_op_queue_mclock_scrub_wgt", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3092 .set_default(1.0)
3093 .set_description("mclock weight of scrub requests")
3094 .set_long_description("mclock weight of scrub requests when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the weight")
3095 .add_see_also("osd_op_queue")
3096 .add_see_also("osd_op_queue_mclock_client_op_res")
3097 .add_see_also("osd_op_queue_mclock_client_op_wgt")
3098 .add_see_also("osd_op_queue_mclock_client_op_lim")
3099 .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
3100 .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
3101 .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
3102 .add_see_also("osd_op_queue_mclock_snap_res")
3103 .add_see_also("osd_op_queue_mclock_snap_wgt")
3104 .add_see_also("osd_op_queue_mclock_snap_lim")
3105 .add_see_also("osd_op_queue_mclock_recov_res")
3106 .add_see_also("osd_op_queue_mclock_recov_wgt")
3107 .add_see_also("osd_op_queue_mclock_recov_lim")
3108 .add_see_also("osd_op_queue_mclock_scrub_res")
3109 .add_see_also("osd_op_queue_mclock_scrub_lim")
3110 .add_see_also("osd_op_queue_mclock_anticipation_timeout"),
3111
3112 Option("osd_op_queue_mclock_scrub_lim", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3113 .set_default(0.001)
3114 .set_description("mclock weight of limit requests")
3115 .set_long_description("mclock weight of limit requests when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the limit")
3116 .add_see_also("osd_op_queue")
3117 .add_see_also("osd_op_queue_mclock_client_op_res")
3118 .add_see_also("osd_op_queue_mclock_client_op_wgt")
3119 .add_see_also("osd_op_queue_mclock_client_op_lim")
3120 .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
3121 .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
3122 .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
3123 .add_see_also("osd_op_queue_mclock_snap_res")
3124 .add_see_also("osd_op_queue_mclock_snap_wgt")
3125 .add_see_also("osd_op_queue_mclock_snap_lim")
3126 .add_see_also("osd_op_queue_mclock_recov_res")
3127 .add_see_also("osd_op_queue_mclock_recov_wgt")
3128 .add_see_also("osd_op_queue_mclock_recov_lim")
3129 .add_see_also("osd_op_queue_mclock_scrub_res")
3130 .add_see_also("osd_op_queue_mclock_scrub_wgt")
3131 .add_see_also("osd_op_queue_mclock_anticipation_timeout"),
3132
3133 Option("osd_op_queue_mclock_anticipation_timeout", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3134 .set_default(0.0)
3135 .set_description("mclock anticipation timeout in seconds")
3136 .set_long_description("the amount of time that mclock waits until the unused resource is forfeited")
3137 .add_see_also("osd_op_queue")
3138 .add_see_also("osd_op_queue_mclock_client_op_res")
3139 .add_see_also("osd_op_queue_mclock_client_op_wgt")
3140 .add_see_also("osd_op_queue_mclock_client_op_lim")
3141 .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
3142 .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
3143 .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
3144 .add_see_also("osd_op_queue_mclock_snap_res")
3145 .add_see_also("osd_op_queue_mclock_snap_wgt")
3146 .add_see_also("osd_op_queue_mclock_snap_lim")
3147 .add_see_also("osd_op_queue_mclock_recov_res")
3148 .add_see_also("osd_op_queue_mclock_recov_wgt")
3149 .add_see_also("osd_op_queue_mclock_recov_lim")
3150 .add_see_also("osd_op_queue_mclock_scrub_res")
3151 .add_see_also("osd_op_queue_mclock_scrub_wgt")
3152 .add_see_also("osd_op_queue_mclock_scrub_lim"),
3153
3154 Option("osd_op_queue_mclock_pg_delete_res", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3155 .set_default(0.0)
3156 .set_description("mclock reservation of pg delete work")
3157 .set_long_description("mclock reservation of pg delete work when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the reservation")
3158 .add_see_also("osd_op_queue")
3159 .add_see_also("osd_op_queue_mclock_client_op_res")
3160 .add_see_also("osd_op_queue_mclock_client_op_wgt")
3161 .add_see_also("osd_op_queue_mclock_client_op_lim")
3162 .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
3163 .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
3164 .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
3165 .add_see_also("osd_op_queue_mclock_snap_res")
3166 .add_see_also("osd_op_queue_mclock_snap_wgt")
3167 .add_see_also("osd_op_queue_mclock_snap_lim")
3168 .add_see_also("osd_op_queue_mclock_recov_res")
3169 .add_see_also("osd_op_queue_mclock_recov_wgt")
3170 .add_see_also("osd_op_queue_mclock_recov_lim")
3171 .add_see_also("osd_op_queue_mclock_scrub_wgt")
3172 .add_see_also("osd_op_queue_mclock_scrub_lim"),
3173
3174 Option("osd_op_queue_mclock_pg_delete_wgt", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3175 .set_default(1.0)
3176 .set_description("mclock weight of pg delete work")
3177 .set_long_description("mclock weight of pg delete work when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the weight")
3178 .add_see_also("osd_op_queue")
3179 .add_see_also("osd_op_queue_mclock_client_op_res")
3180 .add_see_also("osd_op_queue_mclock_client_op_wgt")
3181 .add_see_also("osd_op_queue_mclock_client_op_lim")
3182 .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
3183 .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
3184 .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
3185 .add_see_also("osd_op_queue_mclock_snap_res")
3186 .add_see_also("osd_op_queue_mclock_snap_wgt")
3187 .add_see_also("osd_op_queue_mclock_snap_lim")
3188 .add_see_also("osd_op_queue_mclock_recov_res")
3189 .add_see_also("osd_op_queue_mclock_recov_wgt")
3190 .add_see_also("osd_op_queue_mclock_recov_lim")
3191 .add_see_also("osd_op_queue_mclock_scrub_res")
3192 .add_see_also("osd_op_queue_mclock_scrub_lim"),
3193
3194 Option("osd_op_queue_mclock_pg_delete_lim", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3195 .set_default(0.001)
3196 .set_description("mclock weight of pg delete work limit requests")
3197 .set_long_description("mclock weight of limit pg delete work when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the limit")
3198 .add_see_also("osd_op_queue")
3199 .add_see_also("osd_op_queue_mclock_client_op_res")
3200 .add_see_also("osd_op_queue_mclock_client_op_wgt")
3201 .add_see_also("osd_op_queue_mclock_client_op_lim")
3202 .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
3203 .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
3204 .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
3205 .add_see_also("osd_op_queue_mclock_snap_res")
3206 .add_see_also("osd_op_queue_mclock_snap_wgt")
3207 .add_see_also("osd_op_queue_mclock_snap_lim")
3208 .add_see_also("osd_op_queue_mclock_recov_res")
3209 .add_see_also("osd_op_queue_mclock_recov_wgt")
3210 .add_see_also("osd_op_queue_mclock_recov_lim")
3211 .add_see_also("osd_op_queue_mclock_scrub_res")
3212 .add_see_also("osd_op_queue_mclock_scrub_wgt"),
3213
3214 Option("osd_op_queue_mclock_peering_event_res", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3215 .set_default(0.0)
3216 .set_description("mclock reservation of peering events")
3217 .set_long_description("mclock reservation of scrub requests when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the reservation")
3218 .add_see_also("osd_op_queue")
3219 .add_see_also("osd_op_queue_mclock_client_op_res")
3220 .add_see_also("osd_op_queue_mclock_client_op_wgt")
3221 .add_see_also("osd_op_queue_mclock_client_op_lim")
3222 .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
3223 .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
3224 .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
3225 .add_see_also("osd_op_queue_mclock_snap_res")
3226 .add_see_also("osd_op_queue_mclock_snap_wgt")
3227 .add_see_also("osd_op_queue_mclock_snap_lim")
3228 .add_see_also("osd_op_queue_mclock_recov_res")
3229 .add_see_also("osd_op_queue_mclock_recov_wgt")
3230 .add_see_also("osd_op_queue_mclock_recov_lim")
3231 .add_see_also("osd_op_queue_mclock_scrub_wgt")
3232 .add_see_also("osd_op_queue_mclock_scrub_lim"),
3233
3234 Option("osd_op_queue_mclock_peering_event_wgt", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3235 .set_default(1.0)
3236 .set_description("mclock weight of peering events")
3237 .set_long_description("mclock weight of scrub requests when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the weight")
3238 .add_see_also("osd_op_queue")
3239 .add_see_also("osd_op_queue_mclock_client_op_res")
3240 .add_see_also("osd_op_queue_mclock_client_op_wgt")
3241 .add_see_also("osd_op_queue_mclock_client_op_lim")
3242 .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
3243 .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
3244 .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
3245 .add_see_also("osd_op_queue_mclock_snap_res")
3246 .add_see_also("osd_op_queue_mclock_snap_wgt")
3247 .add_see_also("osd_op_queue_mclock_snap_lim")
3248 .add_see_also("osd_op_queue_mclock_recov_res")
3249 .add_see_also("osd_op_queue_mclock_recov_wgt")
3250 .add_see_also("osd_op_queue_mclock_recov_lim")
3251 .add_see_also("osd_op_queue_mclock_scrub_res")
3252 .add_see_also("osd_op_queue_mclock_scrub_lim"),
3253
3254 Option("osd_op_queue_mclock_peering_event_lim", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3255 .set_default(0.001)
3256 .set_description("mclock weight of limit peering events")
3257 .set_long_description("mclock weight of limit requests when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the limit")
3258 .add_see_also("osd_op_queue")
3259 .add_see_also("osd_op_queue_mclock_client_op_res")
3260 .add_see_also("osd_op_queue_mclock_client_op_wgt")
3261 .add_see_also("osd_op_queue_mclock_client_op_lim")
3262 .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
3263 .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
3264 .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
3265 .add_see_also("osd_op_queue_mclock_snap_res")
3266 .add_see_also("osd_op_queue_mclock_snap_wgt")
3267 .add_see_also("osd_op_queue_mclock_snap_lim")
3268 .add_see_also("osd_op_queue_mclock_recov_res")
3269 .add_see_also("osd_op_queue_mclock_recov_wgt")
3270 .add_see_also("osd_op_queue_mclock_recov_lim")
3271 .add_see_also("osd_op_queue_mclock_scrub_res")
3272 .add_see_also("osd_op_queue_mclock_scrub_wgt"),
3273
3274 Option("osd_ignore_stale_divergent_priors", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
3275 .set_default(false)
3276 .set_description(""),
3277
3278 Option("osd_read_ec_check_for_errors", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
3279 .set_default(false)
3280 .set_description(""),
3281
3282 Option("osd_recover_clone_overlap_limit", Option::TYPE_INT, Option::LEVEL_ADVANCED)
3283 .set_default(10)
3284 .set_description(""),
3285
3286 Option("osd_debug_feed_pullee", Option::TYPE_INT, Option::LEVEL_DEV)
3287 .set_default(-1)
3288 .set_description("Feed a pullee, and force primary to pull "
3289 "a currently missing object from it"),
3290
3291 Option("osd_backfill_scan_min", Option::TYPE_INT, Option::LEVEL_ADVANCED)
3292 .set_default(64)
3293 .set_description(""),
3294
3295 Option("osd_backfill_scan_max", Option::TYPE_INT, Option::LEVEL_ADVANCED)
3296 .set_default(512)
3297 .set_description(""),
3298
3299 Option("osd_op_thread_timeout", Option::TYPE_INT, Option::LEVEL_ADVANCED)
3300 .set_default(15)
3301 .set_description(""),
3302
3303 Option("osd_op_thread_suicide_timeout", Option::TYPE_INT, Option::LEVEL_ADVANCED)
3304 .set_default(150)
3305 .set_description(""),
3306
3307 Option("osd_recovery_sleep", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3308 .set_default(0)
3309 .set_description("Time in seconds to sleep before next recovery or backfill op"),
3310
3311 Option("osd_recovery_sleep_hdd", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3312 .set_default(0.1)
3313 .set_description("Time in seconds to sleep before next recovery or backfill op for HDDs"),
3314
3315 Option("osd_recovery_sleep_ssd", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3316 .set_default(0)
3317 .set_description("Time in seconds to sleep before next recovery or backfill op for SSDs")
3318 .add_see_also("osd_recovery_sleep"),
3319
3320 Option("osd_recovery_sleep_hybrid", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3321 .set_default(0.025)
3322 .set_description("Time in seconds to sleep before next recovery or backfill op when data is on HDD and journal is on SSD")
3323 .add_see_also("osd_recovery_sleep"),
3324
3325 Option("osd_snap_trim_sleep", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3326 .set_default(0)
3327 .set_description("Time in seconds to sleep before next snap trim (overrides values below)"),
3328
3329 Option("osd_snap_trim_sleep_hdd", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3330 .set_default(5)
3331 .set_description("Time in seconds to sleep before next snap trim for HDDs"),
3332
3333 Option("osd_snap_trim_sleep_ssd", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3334 .set_default(0)
3335 .set_description("Time in seconds to sleep before next snap trim for SSDs"),
3336
3337 Option("osd_snap_trim_sleep_hybrid", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3338 .set_default(2)
3339 .set_description("Time in seconds to sleep before next snap trim when data is on HDD and journal is on SSD"),
3340
3341 Option("osd_scrub_invalid_stats", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
3342 .set_default(true)
3343 .set_description(""),
3344
3345 Option("osd_command_thread_timeout", Option::TYPE_INT, Option::LEVEL_ADVANCED)
3346 .set_default(10_min)
3347 .set_description(""),
3348
3349 Option("osd_command_thread_suicide_timeout", Option::TYPE_INT, Option::LEVEL_ADVANCED)
3350 .set_default(15_min)
3351 .set_description(""),
3352
3353 Option("osd_heartbeat_interval", Option::TYPE_INT, Option::LEVEL_DEV)
3354 .set_default(6)
3355 .set_min_max(1, 60)
3356 .set_description("Interval (in seconds) between peer pings"),
3357
3358 Option("osd_heartbeat_grace", Option::TYPE_INT, Option::LEVEL_ADVANCED)
3359 .set_default(20)
3360 .set_description(""),
3361
3362 Option("osd_heartbeat_stale", Option::TYPE_INT, Option::LEVEL_ADVANCED)
3363 .set_default(600)
3364 .set_description("Interval (in seconds) we mark an unresponsive heartbeat peer as stale.")
3365 .set_long_description("Automatically mark unresponsive heartbeat sessions as stale and tear them down. "
3366 "The primary benefit is that OSD doesn't need to keep a flood of "
3367 "blocked heartbeat messages around in memory."),
3368
3369 Option("osd_heartbeat_min_peers", Option::TYPE_INT, Option::LEVEL_ADVANCED)
3370 .set_default(10)
3371 .set_description(""),
3372
3373 Option("osd_heartbeat_use_min_delay_socket", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
3374 .set_default(false)
3375 .set_description(""),
3376
3377 Option("osd_heartbeat_min_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
3378 .set_default(2000)
3379 .set_description("Minimum heartbeat packet size in bytes. Will add dummy payload if heartbeat packet is smaller than this."),
3380
3381 Option("osd_pg_max_concurrent_snap_trims", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
3382 .set_default(2)
3383 .set_description(""),
3384
3385 Option("osd_max_trimming_pgs", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
3386 .set_default(2)
3387 .set_description(""),
3388
3389 Option("osd_heartbeat_min_healthy_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3390 .set_default(.33)
3391 .set_description(""),
3392
3393 Option("osd_mon_heartbeat_interval", Option::TYPE_INT, Option::LEVEL_ADVANCED)
3394 .set_default(30)
3395 .set_description(""),
3396
3397 Option("osd_mon_heartbeat_stat_stale", Option::TYPE_INT, Option::LEVEL_ADVANCED)
3398 .set_default(1_hr)
3399 .set_description("Stop reporting on heartbeat ping times not updated for this many seconds.")
3400 .set_long_description("Stop reporting on old heartbeat information unless this is set to zero"),
3401
3402 Option("osd_mon_report_interval", Option::TYPE_INT, Option::LEVEL_ADVANCED)
3403 .set_default(5)
3404 .set_description("Frequency of OSD reports to mon for peer failures, fullness status changes"),
3405
3406 Option("osd_mon_report_max_in_flight", Option::TYPE_INT, Option::LEVEL_ADVANCED)
3407 .set_default(2)
3408 .set_description(""),
3409
3410 Option("osd_beacon_report_interval", Option::TYPE_INT, Option::LEVEL_ADVANCED)
3411 .set_default(300)
3412 .set_description(""),
3413
3414 Option("osd_pg_stat_report_interval_max", Option::TYPE_INT, Option::LEVEL_ADVANCED)
3415 .set_default(500)
3416 .set_description(""),
3417
3418 Option("osd_mon_ack_timeout", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3419 .set_default(30.0)
3420 .set_description(""),
3421
3422 Option("osd_stats_ack_timeout_factor", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3423 .set_default(2.0)
3424 .set_description(""),
3425
3426 Option("osd_stats_ack_timeout_decay", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3427 .set_default(.9)
3428 .set_description(""),
3429
3430 Option("osd_max_snap_prune_intervals_per_epoch", Option::TYPE_UINT, Option::LEVEL_DEV)
3431 .set_default(512)
3432 .set_description("Max number of snap intervals to report to mgr in pg_stat_t"),
3433
3434 Option("osd_default_data_pool_replay_window", Option::TYPE_INT, Option::LEVEL_ADVANCED)
3435 .set_default(45)
3436 .set_description(""),
3437
3438 Option("osd_auto_mark_unfound_lost", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
3439 .set_default(false)
3440 .set_description(""),
3441
3442 Option("osd_recovery_delay_start", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3443 .set_default(0)
3444 .set_description(""),
3445
3446 Option("osd_recovery_max_active", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
3447 .set_default(0)
3448 .set_description("Number of simultaneous active recovery operations per OSD (overrides _ssd and _hdd if non-zero)")
3449 .add_see_also("osd_recovery_max_active_hdd")
3450 .add_see_also("osd_recovery_max_active_ssd"),
3451
3452 Option("osd_recovery_max_active_hdd", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
3453 .set_default(3)
3454 .set_description("Number of simultaneous active recovery oeprations per OSD (for rotational devices)")
3455 .add_see_also("osd_recovery_max_active")
3456 .add_see_also("osd_recovery_max_active_ssd"),
3457
3458 Option("osd_recovery_max_active_ssd", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
3459 .set_default(10)
3460 .set_description("Number of simultaneous active recovery oeprations per OSD (for non-rotational solid state devices)")
3461 .add_see_also("osd_recovery_max_active")
3462 .add_see_also("osd_recovery_max_active_hdd"),
3463
3464 Option("osd_recovery_max_single_start", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
3465 .set_default(1)
3466 .set_description(""),
3467
3468 Option("osd_recovery_max_chunk", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
3469 .set_default(8_M)
3470 .set_description(""),
3471
3472 Option("osd_recovery_max_omap_entries_per_chunk", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
3473 .set_default(8096)
3474 .set_description(""),
3475
3476 Option("osd_copyfrom_max_chunk", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
3477 .set_default(8_M)
3478 .set_description(""),
3479
3480 Option("osd_push_per_object_cost", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
3481 .set_default(1000)
3482 .set_description(""),
3483
3484 Option("osd_max_push_cost", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
3485 .set_default(8<<20)
3486 .set_description(""),
3487
3488 Option("osd_max_push_objects", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
3489 .set_default(10)
3490 .set_description(""),
3491
3492 Option("osd_max_scrubs", Option::TYPE_INT, Option::LEVEL_ADVANCED)
3493 .set_default(1)
3494 .set_description("Maximum concurrent scrubs on a single OSD"),
3495
3496 Option("osd_scrub_during_recovery", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
3497 .set_default(false)
3498 .set_description("Allow scrubbing when PGs on the OSD are undergoing recovery"),
3499
3500 Option("osd_repair_during_recovery", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
3501 .set_default(false)
3502 .set_description("Allow requested repairing when PGs on the OSD are undergoing recovery"),
3503
3504 Option("osd_scrub_begin_hour", Option::TYPE_INT, Option::LEVEL_ADVANCED)
3505 .set_default(0)
3506 .set_description("Restrict scrubbing to this hour of the day or later")
3507 .add_see_also("osd_scrub_end_hour"),
3508
3509 Option("osd_scrub_end_hour", Option::TYPE_INT, Option::LEVEL_ADVANCED)
3510 .set_default(24)
3511 .set_description("Restrict scrubbing to hours of the day earlier than this")
3512 .add_see_also("osd_scrub_begin_hour"),
3513
3514 Option("osd_scrub_begin_week_day", Option::TYPE_INT, Option::LEVEL_ADVANCED)
3515 .set_default(0)
3516 .set_description("Restrict scrubbing to this day of the week or later")
3517 .set_long_description("0 or 7 = Sunday, 1 = Monday, etc.")
3518 .add_see_also("osd_scrub_end_week_day"),
3519
3520 Option("osd_scrub_end_week_day", Option::TYPE_INT, Option::LEVEL_ADVANCED)
3521 .set_default(7)
3522 .set_description("Restrict scrubbing to days of the week earlier than this")
3523 .set_long_description("0 or 7 = Sunday, 1 = Monday, etc.")
3524 .add_see_also("osd_scrub_begin_week_day"),
3525
3526 Option("osd_scrub_load_threshold", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3527 .set_default(0.5)
3528 .set_description("Allow scrubbing when system load divided by number of CPUs is below this value"),
3529
3530 Option("osd_scrub_min_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3531 .set_default(1_day)
3532 .set_description("Scrub each PG no more often than this interval")
3533 .add_see_also("osd_scrub_max_interval"),
3534
3535 Option("osd_scrub_max_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3536 .set_default(7_day)
3537 .set_description("Scrub each PG no less often than this interval")
3538 .add_see_also("osd_scrub_min_interval"),
3539
3540 Option("osd_scrub_interval_randomize_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3541 .set_default(0.5)
3542 .set_description("Ratio of scrub interval to randomly vary")
3543 .set_long_description("This prevents a scrub 'stampede' by randomly varying the scrub intervals so that they are soon uniformly distributed over the week")
3544 .add_see_also("osd_scrub_min_interval"),
3545
3546 Option("osd_scrub_backoff_ratio", Option::TYPE_FLOAT, Option::LEVEL_DEV)
3547 .set_default(.66)
3548 .set_long_description("This is the precentage of ticks that do NOT schedule scrubs, 66% means that 1 out of 3 ticks will schedule scrubs")
3549 .set_description("Backoff ratio for scheduling scrubs"),
3550
3551 Option("osd_scrub_chunk_min", Option::TYPE_INT, Option::LEVEL_ADVANCED)
3552 .set_default(5)
3553 .set_description("Minimum number of objects to scrub in a single chunk")
3554 .add_see_also("osd_scrub_chunk_max"),
3555
3556 Option("osd_scrub_chunk_max", Option::TYPE_INT, Option::LEVEL_ADVANCED)
3557 .set_default(25)
3558 .set_description("Maximum number of objects to scrub in a single chunk")
3559 .add_see_also("osd_scrub_chunk_min"),
3560
3561 Option("osd_scrub_sleep", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3562 .set_default(0)
3563 .set_description("Duration to inject a delay during scrubbing"),
3564
3565 Option("osd_scrub_extended_sleep", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3566 .set_default(0)
3567 .set_description("Duration to inject a delay during scrubbing out of scrubbing hours")
3568 .add_see_also("osd_scrub_begin_hour")
3569 .add_see_also("osd_scrub_end_hour")
3570 .add_see_also("osd_scrub_begin_week_day")
3571 .add_see_also("osd_scrub_end_week_day"),
3572
3573 Option("osd_scrub_auto_repair", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
3574 .set_default(false)
3575 .set_description("Automatically repair damaged objects detected during scrub"),
3576
3577 Option("osd_scrub_auto_repair_num_errors", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
3578 .set_default(5)
3579 .set_description("Maximum number of detected errors to automatically repair")
3580 .add_see_also("osd_scrub_auto_repair"),
3581
3582 Option("osd_scrub_max_preemptions", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
3583 .set_default(5)
3584 .set_description("Set the maximum number of times we will preempt a deep scrub due to a client operation before blocking client IO to complete the scrub"),
3585
3586 Option("osd_deep_scrub_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3587 .set_default(7_day)
3588 .set_description("Deep scrub each PG (i.e., verify data checksums) at least this often"),
3589
3590 Option("osd_deep_scrub_randomize_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3591 .set_default(0.15)
3592 .set_description("Scrubs will randomly become deep scrubs at this rate (0.15 -> 15% of scrubs are deep)")
3593 .set_long_description("This prevents a deep scrub 'stampede' by spreading deep scrubs so they are uniformly distributed over the week"),
3594
3595 Option("osd_deep_scrub_stride", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
3596 .set_default(512_K)
3597 .set_description("Number of bytes to read from an object at a time during deep scrub"),
3598
3599 Option("osd_deep_scrub_keys", Option::TYPE_INT, Option::LEVEL_ADVANCED)
3600 .set_default(1024)
3601 .set_description("Number of keys to read from an object at a time during deep scrub"),
3602
3603 Option("osd_deep_scrub_update_digest_min_age", Option::TYPE_INT, Option::LEVEL_ADVANCED)
3604 .set_default(2_hr)
3605 .set_description("Update overall object digest only if object was last modified longer ago than this"),
3606
3607 Option("osd_deep_scrub_large_omap_object_key_threshold", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
3608 .set_default(200000)
3609 .set_description("Warn when we encounter an object with more omap keys than this")
3610 .add_service("osd")
3611 .add_see_also("osd_deep_scrub_large_omap_object_value_sum_threshold"),
3612
3613 Option("osd_deep_scrub_large_omap_object_value_sum_threshold", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
3614 .set_default(1_G)
3615 .set_description("Warn when we encounter an object with more omap key bytes than this")
3616 .add_service("osd")
3617 .add_see_also("osd_deep_scrub_large_omap_object_key_threshold"),
3618
3619 Option("osd_class_dir", Option::TYPE_STR, Option::LEVEL_ADVANCED)
3620 .set_default(CEPH_LIBDIR "/rados-classes")
3621 .set_description(""),
3622
3623 Option("osd_open_classes_on_start", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
3624 .set_default(true)
3625 .set_description(""),
3626
3627 Option("osd_class_load_list", Option::TYPE_STR, Option::LEVEL_ADVANCED)
3628 .set_default("cephfs hello journal lock log numops " "otp rbd refcount rgw timeindex user version cas")
3629 .set_description(""),
3630
3631 Option("osd_class_default_list", Option::TYPE_STR, Option::LEVEL_ADVANCED)
3632 .set_default("cephfs hello journal lock log numops " "otp rbd refcount rgw timeindex user version cas")
3633 .set_description(""),
3634
3635 Option("osd_check_for_log_corruption", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
3636 .set_default(false)
3637 .set_description(""),
3638
3639 Option("osd_use_stale_snap", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
3640 .set_default(false)
3641 .set_description(""),
3642
3643 Option("osd_rollback_to_cluster_snap", Option::TYPE_STR, Option::LEVEL_ADVANCED)
3644 .set_default("")
3645 .set_description(""),
3646
3647 Option("osd_default_notify_timeout", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
3648 .set_default(30)
3649 .set_description(""),
3650
3651 Option("osd_kill_backfill_at", Option::TYPE_INT, Option::LEVEL_DEV)
3652 .set_default(0)
3653 .set_description(""),
3654
3655 Option("osd_pg_epoch_persisted_max_stale", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
3656 .set_default(40)
3657 .set_description(""),
3658
3659 Option("osd_min_pg_log_entries", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
3660 .set_default(3000)
3661 .set_description("minimum number of entries to maintain in the PG log")
3662 .add_service("osd")
3663 .add_see_also("osd_max_pg_log_entries")
3664 .add_see_also("osd_pg_log_dups_tracked"),
3665
3666 Option("osd_max_pg_log_entries", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
3667 .set_default(3000)
3668 .set_description("maximum number of entries to maintain in the PG log when degraded before we trim")
3669 .add_service("osd")
3670 .add_see_also("osd_min_pg_log_entries")
3671 .add_see_also("osd_pg_log_dups_tracked"),
3672
3673 Option("osd_pg_log_dups_tracked", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
3674 .set_default(3000)
3675 .set_description("how many versions back to track in order to detect duplicate ops; this is combined with both the regular pg log entries and additional minimal dup detection entries")
3676 .add_service("osd")
3677 .add_see_also("osd_min_pg_log_entries")
3678 .add_see_also("osd_max_pg_log_entries"),
3679
3680 Option("osd_object_clean_region_max_num_intervals", Option::TYPE_INT, Option::LEVEL_ADVANCED)
3681 .set_default(10)
3682 .set_description("number of intervals in clean_offsets")
3683 .set_long_description("partial recovery uses multiple intervals to record the clean part of the object"
3684 "when the number of intervals is greater than osd_object_clean_region_max_num_intervals, minimum interval will be trimmed"
3685 "(0 will recovery the entire object data interval)")
3686 .add_service("osd"),
3687
3688 Option("osd_force_recovery_pg_log_entries_factor", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3689 .set_default(1.3)
3690 .set_description(""),
3691
3692 Option("osd_pg_log_trim_min", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
3693 .set_default(100)
3694 .set_description(""),
3695
3696 Option("osd_force_auth_primary_missing_objects", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
3697 .set_default(100)
3698 .set_description("Approximate missing objects above which to force auth_log_shard to be primary temporarily"),
3699
3700 Option("osd_async_recovery_min_cost", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
3701 .set_default(100)
3702 .set_description("A mixture measure of number of current log entries difference "
3703 "and historical missing objects, above which we switch to use "
3704 "asynchronous recovery when appropriate"),
3705
3706 Option("osd_max_pg_per_osd_hard_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3707 .set_default(3)
3708 .set_min(1)
3709 .set_description("Maximum number of PG per OSD, a factor of 'mon_max_pg_per_osd'")
3710 .set_long_description("OSD will refuse to instantiate PG if the number of PG it serves exceeds this number.")
3711 .add_see_also("mon_max_pg_per_osd"),
3712
3713 Option("osd_pg_log_trim_max", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
3714 .set_default(10000)
3715 .set_description("maximum number of entries to remove at once from the PG log")
3716 .add_service("osd")
3717 .add_see_also("osd_min_pg_log_entries")
3718 .add_see_also("osd_max_pg_log_entries"),
3719
3720 Option("osd_op_complaint_time", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3721 .set_default(30)
3722 .set_description(""),
3723
3724 Option("osd_command_max_records", Option::TYPE_INT, Option::LEVEL_ADVANCED)
3725 .set_default(256)
3726 .set_description(""),
3727
3728 Option("osd_max_pg_blocked_by", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
3729 .set_default(16)
3730 .set_description(""),
3731
3732 Option("osd_op_log_threshold", Option::TYPE_INT, Option::LEVEL_ADVANCED)
3733 .set_default(5)
3734 .set_description(""),
3735
3736 Option("osd_backoff_on_unfound", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
3737 .set_default(true)
3738 .set_description(""),
3739
3740 Option("osd_backoff_on_degraded", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
3741 .set_default(false)
3742 .set_description(""),
3743
3744 Option("osd_backoff_on_peering", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
3745 .set_default(false)
3746 .set_description(""),
3747
3748 Option("osd_debug_shutdown", Option::TYPE_BOOL, Option::LEVEL_DEV)
3749 .set_default(false)
3750 .set_description("Turn up debug levels during shutdown"),
3751
3752 Option("osd_debug_crash_on_ignored_backoff", Option::TYPE_BOOL, Option::LEVEL_DEV)
3753 .set_default(false)
3754 .set_description(""),
3755
3756 Option("osd_debug_inject_dispatch_delay_probability", Option::TYPE_FLOAT, Option::LEVEL_DEV)
3757 .set_default(0)
3758 .set_description(""),
3759
3760 Option("osd_debug_inject_dispatch_delay_duration", Option::TYPE_FLOAT, Option::LEVEL_DEV)
3761 .set_default(.1)
3762 .set_description(""),
3763
3764 Option("osd_debug_drop_ping_probability", Option::TYPE_FLOAT, Option::LEVEL_DEV)
3765 .set_default(0)
3766 .set_description(""),
3767
3768 Option("osd_debug_drop_ping_duration", Option::TYPE_INT, Option::LEVEL_DEV)
3769 .set_default(0)
3770 .set_description(""),
3771
3772 Option("osd_debug_op_order", Option::TYPE_BOOL, Option::LEVEL_DEV)
3773 .set_default(false)
3774 .set_description(""),
3775
3776 Option("osd_debug_verify_missing_on_start", Option::TYPE_BOOL, Option::LEVEL_DEV)
3777 .set_default(false)
3778 .set_description(""),
3779
3780 Option("osd_debug_verify_snaps", Option::TYPE_BOOL, Option::LEVEL_DEV)
3781 .set_default(false)
3782 .set_description(""),
3783
3784 Option("osd_debug_verify_stray_on_activate", Option::TYPE_BOOL, Option::LEVEL_DEV)
3785 .set_default(false)
3786 .set_description(""),
3787
3788 Option("osd_debug_skip_full_check_in_backfill_reservation", Option::TYPE_BOOL, Option::LEVEL_DEV)
3789 .set_default(false)
3790 .set_description(""),
3791
3792 Option("osd_debug_reject_backfill_probability", Option::TYPE_FLOAT, Option::LEVEL_DEV)
3793 .set_default(0)
3794 .set_description(""),
3795
3796 Option("osd_debug_inject_copyfrom_error", Option::TYPE_BOOL, Option::LEVEL_DEV)
3797 .set_default(false)
3798 .set_description(""),
3799
3800 Option("osd_debug_misdirected_ops", Option::TYPE_BOOL, Option::LEVEL_DEV)
3801 .set_default(false)
3802 .set_description(""),
3803
3804 Option("osd_debug_skip_full_check_in_recovery", Option::TYPE_BOOL, Option::LEVEL_DEV)
3805 .set_default(false)
3806 .set_description(""),
3807
3808 Option("osd_debug_random_push_read_error", Option::TYPE_FLOAT, Option::LEVEL_DEV)
3809 .set_default(0)
3810 .set_description(""),
3811
3812 Option("osd_debug_verify_cached_snaps", Option::TYPE_BOOL, Option::LEVEL_DEV)
3813 .set_default(false)
3814 .set_description(""),
3815
3816 Option("osd_debug_deep_scrub_sleep", Option::TYPE_FLOAT, Option::LEVEL_DEV)
3817 .set_default(0)
3818 .set_description("Inject an expensive sleep during deep scrub IO to make it easier to induce preemption"),
3819
3820 Option("osd_debug_no_acting_change", Option::TYPE_BOOL, Option::LEVEL_DEV)
3821 .set_default(false),
3822 Option("osd_debug_no_purge_strays", Option::TYPE_BOOL, Option::LEVEL_DEV)
3823 .set_default(false),
3824
3825 Option("osd_debug_pretend_recovery_active", Option::TYPE_BOOL, Option::LEVEL_DEV)
3826 .set_default(false)
3827 .set_description(""),
3828
3829 Option("osd_enable_op_tracker", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
3830 .set_default(true)
3831 .set_description(""),
3832
3833 Option("osd_num_op_tracker_shard", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
3834 .set_default(32)
3835 .set_description(""),
3836
3837 Option("osd_op_history_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
3838 .set_default(20)
3839 .set_description(""),
3840
3841 Option("osd_op_history_duration", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
3842 .set_default(600)
3843 .set_description(""),
3844
3845 Option("osd_op_history_slow_op_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
3846 .set_default(20)
3847 .set_description(""),
3848
3849 Option("osd_op_history_slow_op_threshold", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3850 .set_default(10.0)
3851 .set_description(""),
3852
3853 Option("osd_target_transaction_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
3854 .set_default(30)
3855 .set_description(""),
3856
3857 Option("osd_delete_sleep", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3858 .set_default(0)
3859 .set_description("Time in seconds to sleep before next removal transaction (overrides values below)"),
3860
3861 Option("osd_delete_sleep_hdd", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3862 .set_default(5)
3863 .set_description("Time in seconds to sleep before next removal transaction for HDDs"),
3864
3865 Option("osd_delete_sleep_ssd", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3866 .set_default(0)
3867 .set_description("Time in seconds to sleep before next removal transaction for SSDs"),
3868
3869 Option("osd_delete_sleep_hybrid", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3870 .set_default(2)
3871 .set_description("Time in seconds to sleep before next removal transaction when data is on HDD and journal is on SSD"),
3872
3873 Option("osd_failsafe_full_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3874 .set_default(.97)
3875 .set_description(""),
3876
3877 Option("osd_fast_fail_on_connection_refused", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
3878 .set_default(true)
3879 .set_description(""),
3880
3881 Option("osd_pg_object_context_cache_count", Option::TYPE_INT, Option::LEVEL_ADVANCED)
3882 .set_default(64)
3883 .set_description(""),
3884
3885 Option("osd_tracing", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
3886 .set_default(false)
3887 .set_description(""),
3888
3889 Option("osd_function_tracing", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
3890 .set_default(false)
3891 .set_description(""),
3892
3893 Option("osd_fast_info", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
3894 .set_default(true)
3895 .set_description(""),
3896
3897 Option("osd_debug_pg_log_writeout", Option::TYPE_BOOL, Option::LEVEL_DEV)
3898 .set_default(false)
3899 .set_description(""),
3900
3901 Option("osd_loop_before_reset_tphandle", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
3902 .set_default(64)
3903 .set_description(""),
3904
3905 Option("threadpool_default_timeout", Option::TYPE_INT, Option::LEVEL_ADVANCED)
3906 .set_default(60)
3907 .set_description(""),
3908
3909 Option("threadpool_empty_queue_max_wait", Option::TYPE_INT, Option::LEVEL_ADVANCED)
3910 .set_default(2)
3911 .set_description(""),
3912
3913 Option("leveldb_log_to_ceph_log", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
3914 .set_default(true)
3915 .set_description(""),
3916
3917 Option("leveldb_write_buffer_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
3918 .set_default(8_M)
3919 .set_description(""),
3920
3921 Option("leveldb_cache_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
3922 .set_default(128_M)
3923 .set_description(""),
3924
3925 Option("leveldb_block_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
3926 .set_default(0)
3927 .set_description(""),
3928
3929 Option("leveldb_bloom_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
3930 .set_default(0)
3931 .set_description(""),
3932
3933 Option("leveldb_max_open_files", Option::TYPE_INT, Option::LEVEL_ADVANCED)
3934 .set_default(0)
3935 .set_description(""),
3936
3937 Option("leveldb_compression", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
3938 .set_default(true)
3939 .set_description(""),
3940
3941 Option("leveldb_paranoid", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
3942 .set_default(false)
3943 .set_description(""),
3944
3945 Option("leveldb_log", Option::TYPE_STR, Option::LEVEL_ADVANCED)
3946 .set_default("/dev/null")
3947 .set_description(""),
3948
3949 Option("leveldb_compact_on_mount", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
3950 .set_default(false)
3951 .set_description(""),
3952
3953 Option("rocksdb_log_to_ceph_log", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
3954 .set_default(true)
3955 .set_description(""),
3956
3957 Option("rocksdb_cache_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
3958 .set_default(512_M)
3959 .set_flag(Option::FLAG_RUNTIME)
3960 .set_description(""),
3961
3962 Option("rocksdb_cache_row_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
3963 .set_default(0)
3964 .set_description(""),
3965
3966 Option("rocksdb_cache_shard_bits", Option::TYPE_INT, Option::LEVEL_ADVANCED)
3967 .set_default(4)
3968 .set_description(""),
3969
3970 Option("rocksdb_cache_type", Option::TYPE_STR, Option::LEVEL_ADVANCED)
3971 .set_default("binned_lru")
3972 .set_description(""),
3973
3974 Option("rocksdb_block_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
3975 .set_default(4_K)
3976 .set_description(""),
3977
3978 Option("rocksdb_perf", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
3979 .set_default(false)
3980 .set_description(""),
3981
3982 Option("rocksdb_collect_compaction_stats", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
3983 .set_default(false)
3984 .set_description(""),
3985
3986 Option("rocksdb_collect_extended_stats", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
3987 .set_default(false)
3988 .set_description(""),
3989
3990 Option("rocksdb_collect_memory_stats", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
3991 .set_default(false)
3992 .set_description(""),
3993
3994 Option("rocksdb_bloom_bits_per_key", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
3995 .set_default(20)
3996 .set_description("Number of bits per key to use for RocksDB's bloom filters.")
3997 .set_long_description("RocksDB bloom filters can be used to quickly answer the question of whether or not a key may exist or definitely does not exist in a given RocksDB SST file without having to read all keys into memory. Using a higher bit value decreases the likelihood of false positives at the expense of additional disk space and memory consumption when the filter is loaded into RAM. The current default value of 20 was found to provide significant performance gains when getattr calls are made (such as during new object creation in bluestore) without significant memory overhead or cache pollution when combined with rocksdb partitioned index filters. See: https://github.com/facebook/rocksdb/wiki/Partitioned-Index-Filters for more information."),
3998
3999 Option("rocksdb_cache_index_and_filter_blocks", Option::TYPE_BOOL, Option::LEVEL_DEV)
4000 .set_default(true)
4001 .set_description("Whether to cache indices and filters in block cache")
4002 .set_long_description("By default RocksDB will load an SST file's index and bloom filters into memory when it is opened and remove them from memory when an SST file is closed. Thus, memory consumption by indices and bloom filters is directly tied to the number of concurrent SST files allowed to be kept open. This option instead stores cached indicies and filters in the block cache where they directly compete with other cached data. By default we set this option to true to better account for and bound rocksdb memory usage and keep filters in memory even when an SST file is closed."),
4003
4004 Option("rocksdb_cache_index_and_filter_blocks_with_high_priority", Option::TYPE_BOOL, Option::LEVEL_DEV)
4005 .set_default(true)
4006 .set_description("Whether to cache indices and filters in the block cache with high priority")
4007 .set_long_description("A downside of setting rocksdb_cache_index_and_filter_blocks to true is that regular data can push indices and filters out of memory. Setting this option to true means they are cached with higher priority than other data and should typically stay in the block cache."),
4008
4009 Option("rocksdb_pin_l0_filter_and_index_blocks_in_cache", Option::TYPE_BOOL, Option::LEVEL_DEV)
4010 .set_default(false)
4011 .set_description("Whether to pin Level 0 indices and bloom filters in the block cache")
4012 .set_long_description("A downside of setting rocksdb_cache_index_and_filter_blocks to true is that regular data can push indices and filters out of memory. Setting this option to true means that level 0 SST files will always have their indices and filters pinned in the block cache."),
4013
4014 Option("rocksdb_index_type", Option::TYPE_STR, Option::LEVEL_DEV)
4015 .set_default("binary_search")
4016 .set_description("Type of index for SST files: binary_search, hash_search, two_level")
4017 .set_long_description("This option controls the table index type. binary_search is a space efficient index block that is optimized for block-search-based index. hash_search may improve prefix lookup performance at the expense of higher disk and memory usage and potentially slower compactions. two_level is an experimental index type that uses two binary search indexes and works in conjunction with partition filters. See: http://rocksdb.org/blog/2017/05/12/partitioned-index-filter.html"),
4018
4019 Option("rocksdb_partition_filters", Option::TYPE_BOOL, Option::LEVEL_DEV)
4020 .set_default(false)
4021 .set_description("(experimental) partition SST index/filters into smaller blocks")
4022 .set_long_description("This is an experimental option for rocksdb that works in conjunction with two_level indices to avoid having to keep the entire filter/index in cache when cache_index_and_filter_blocks is true. The idea is to keep a much smaller top-level index in heap/cache and then opportunistically cache the lower level indices. See: https://github.com/facebook/rocksdb/wiki/Partitioned-Index-Filters"),
4023
4024 Option("rocksdb_metadata_block_size", Option::TYPE_SIZE, Option::LEVEL_DEV)
4025 .set_default(4_K)
4026 .set_description("The block size for index partitions. (0 = rocksdb default)"),
4027
4028 Option("mon_rocksdb_options", Option::TYPE_STR, Option::LEVEL_ADVANCED)
4029 .set_default("write_buffer_size=33554432,"
4030 "compression=kNoCompression,"
4031 "level_compaction_dynamic_level_bytes=true")
4032 .set_description(""),
4033
4034 Option("osd_client_op_priority", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
4035 .set_default(63)
4036 .set_description(""),
4037
4038 Option("osd_recovery_op_priority", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
4039 .set_default(3)
4040 .set_description("Priority to use for recovery operations if not specified for the pool"),
4041
4042 Option("osd_peering_op_priority", Option::TYPE_UINT, Option::LEVEL_DEV)
4043 .set_default(255)
4044 .set_description(""),
4045
4046 Option("osd_snap_trim_priority", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
4047 .set_default(5)
4048 .set_description(""),
4049
4050 Option("osd_snap_trim_cost", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4051 .set_default(1<<20)
4052 .set_description(""),
4053
4054 Option("osd_pg_delete_priority", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
4055 .set_default(5)
4056 .set_description(""),
4057
4058 Option("osd_pg_delete_cost", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4059 .set_default(1<<20)
4060 .set_description(""),
4061
4062 Option("osd_scrub_priority", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
4063 .set_default(5)
4064 .set_description("Priority for scrub operations in work queue"),
4065
4066 Option("osd_scrub_cost", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4067 .set_default(50<<20)
4068 .set_description("Cost for scrub operations in work queue"),
4069
4070 Option("osd_requested_scrub_priority", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
4071 .set_default(120)
4072 .set_description(""),
4073
4074 Option("osd_recovery_priority", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
4075 .set_default(5)
4076 .set_description("Priority of recovery in the work queue")
4077 .set_long_description("Not related to a pool's recovery_priority"),
4078
4079 Option("osd_recovery_cost", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4080 .set_default(20<<20)
4081 .set_description(""),
4082
4083 Option("osd_recovery_op_warn_multiple", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
4084 .set_default(16)
4085 .set_description(""),
4086
4087 Option("osd_mon_shutdown_timeout", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
4088 .set_default(5)
4089 .set_description(""),
4090
4091 Option("osd_shutdown_pgref_assert", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
4092 .set_default(false)
4093 .set_description(""),
4094
4095 Option("osd_max_object_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4096 .set_default(128_M)
4097 .set_description(""),
4098
4099 Option("osd_max_object_name_len", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
4100 .set_default(2048)
4101 .set_description(""),
4102
4103 Option("osd_max_object_namespace_len", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
4104 .set_default(256)
4105 .set_description(""),
4106
4107 Option("osd_max_attr_name_len", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
4108 .set_default(100)
4109 .set_description(""),
4110
4111 Option("osd_max_attr_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
4112 .set_default(0)
4113 .set_description(""),
4114
4115 Option("osd_max_omap_entries_per_request", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
4116 .set_default(131072)
4117 .set_description(""),
4118
4119 Option("osd_max_omap_bytes_per_request", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4120 .set_default(1_G)
4121 .set_description(""),
4122
4123 Option("osd_max_write_op_reply_len", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4124 .set_default(32)
4125 .set_description("Max size of the per-op payload for requests with the RETURNVEC flag set")
4126 .set_long_description("This value caps the amount of data (per op; a request may have many ops) that will be sent back to the client and recorded in the PG log."),
4127
4128 Option("osd_objectstore", Option::TYPE_STR, Option::LEVEL_ADVANCED)
4129 .set_default("bluestore")
4130 .set_enum_allowed({"bluestore", "filestore", "memstore", "kstore"})
4131 .set_flag(Option::FLAG_CREATE)
4132 .set_description("backend type for an OSD (like filestore or bluestore)"),
4133
4134 Option("osd_objectstore_tracing", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
4135 .set_default(false)
4136 .set_description(""),
4137
4138 Option("osd_objectstore_fuse", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
4139 .set_default(false)
4140 .set_description(""),
4141
4142 Option("osd_bench_small_size_max_iops", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
4143 .set_default(100)
4144 .set_description(""),
4145
4146 Option("osd_bench_large_size_max_throughput", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4147 .set_default(100_M)
4148 .set_description(""),
4149
4150 Option("osd_bench_max_block_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4151 .set_default(64_M)
4152 .set_description(""),
4153
4154 Option("osd_bench_duration", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
4155 .set_default(30)
4156 .set_description(""),
4157
4158 Option("osd_blkin_trace_all", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
4159 .set_default(false)
4160 .set_description(""),
4161
4162 Option("osdc_blkin_trace_all", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
4163 .set_default(false)
4164 .set_description(""),
4165
4166 Option("osd_discard_disconnected_ops", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
4167 .set_default(true)
4168 .set_description(""),
4169
4170 Option("osd_memory_target", Option::TYPE_SIZE, Option::LEVEL_BASIC)
4171 .set_default(4_G)
4172 .set_min(896_M)
4173 .set_flag(Option::FLAG_RUNTIME)
4174 .add_see_also("bluestore_cache_autotune")
4175 .add_see_also("osd_memory_cache_min")
4176 .add_see_also("osd_memory_base")
4177 .set_description("When tcmalloc and cache autotuning is enabled, try to keep this many bytes mapped in memory.")
4178 .set_long_description("The minimum value must be at least equal to osd_memory_base + osd_memory_cache_min."),
4179
4180 Option("osd_memory_target_cgroup_limit_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
4181 .set_default(0.8)
4182 .set_min_max(0.0, 1.0)
4183 .add_see_also("osd_memory_target")
4184 .set_description("Set the default value for osd_memory_target to the cgroup memory limit (if set) times this value")
4185 .set_long_description("A value of 0 disables this feature."),
4186
4187 Option("osd_memory_base", Option::TYPE_SIZE, Option::LEVEL_DEV)
4188 .set_default(768_M)
4189 .set_flag(Option::FLAG_RUNTIME)
4190 .add_see_also("bluestore_cache_autotune")
4191 .set_description("When tcmalloc and cache autotuning is enabled, estimate the minimum amount of memory in bytes the OSD will need."),
4192
4193 Option("osd_memory_expected_fragmentation", Option::TYPE_FLOAT, Option::LEVEL_DEV)
4194 .set_default(0.15)
4195 .set_min_max(0.0, 1.0)
4196 .set_flag(Option::FLAG_RUNTIME)
4197 .add_see_also("bluestore_cache_autotune")
4198 .set_description("When tcmalloc and cache autotuning is enabled, estimate the percent of memory fragmentation."),
4199
4200 Option("osd_memory_cache_min", Option::TYPE_SIZE, Option::LEVEL_DEV)
4201 .set_default(128_M)
4202 .set_min(128_M)
4203 .set_flag(Option::FLAG_RUNTIME)
4204 .add_see_also("bluestore_cache_autotune")
4205 .set_description("When tcmalloc and cache autotuning is enabled, set the minimum amount of memory used for caches."),
4206
4207 Option("osd_memory_cache_resize_interval", Option::TYPE_FLOAT, Option::LEVEL_DEV)
4208 .set_default(1)
4209 .add_see_also("bluestore_cache_autotune")
4210 .set_description("When tcmalloc and cache autotuning is enabled, wait this many seconds between resizing caches."),
4211
4212 Option("memstore_device_bytes", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4213 .set_default(1_G)
4214 .set_description(""),
4215
4216 Option("memstore_page_set", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
4217 .set_default(false)
4218 .set_description(""),
4219
4220 Option("memstore_page_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4221 .set_default(64_K)
4222 .set_description(""),
4223
4224 Option("memstore_debug_omit_block_device_write", Option::TYPE_BOOL, Option::LEVEL_DEV)
4225 .set_default(false)
4226 .add_see_also("bluestore_debug_omit_block_device_write")
4227 .set_description("write metadata only"),
4228
4229 Option("objectstore_blackhole", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
4230 .set_default(false)
4231 .set_description(""),
4232
4233 // --------------------------
4234 // bluestore
4235
4236 Option("bdev_debug_inflight_ios", Option::TYPE_BOOL, Option::LEVEL_DEV)
4237 .set_default(false)
4238 .set_description(""),
4239
4240 Option("bdev_inject_crash", Option::TYPE_INT, Option::LEVEL_DEV)
4241 .set_default(0)
4242 .set_description(""),
4243
4244 Option("bdev_inject_crash_flush_delay", Option::TYPE_INT, Option::LEVEL_DEV)
4245 .set_default(2)
4246 .set_description(""),
4247
4248 Option("bdev_aio", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
4249 .set_default(true)
4250 .set_description(""),
4251
4252 Option("bdev_aio_poll_ms", Option::TYPE_INT, Option::LEVEL_ADVANCED)
4253 .set_default(250)
4254 .set_description(""),
4255
4256 Option("bdev_aio_max_queue_depth", Option::TYPE_INT, Option::LEVEL_ADVANCED)
4257 .set_default(1024)
4258 .set_description(""),
4259
4260 Option("bdev_aio_reap_max", Option::TYPE_INT, Option::LEVEL_ADVANCED)
4261 .set_default(16)
4262 .set_description(""),
4263
4264 Option("bdev_block_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4265 .set_default(4_K)
4266 .set_description(""),
4267
4268 Option("bdev_debug_aio", Option::TYPE_BOOL, Option::LEVEL_DEV)
4269 .set_default(false)
4270 .set_description(""),
4271
4272 Option("bdev_debug_aio_suicide_timeout", Option::TYPE_FLOAT, Option::LEVEL_DEV)
4273 .set_default(60.0)
4274 .set_description(""),
4275
4276 Option("bdev_debug_aio_log_age", Option::TYPE_FLOAT, Option::LEVEL_DEV)
4277 .set_default(5.0)
4278 .set_description(""),
4279
4280 Option("bdev_nvme_unbind_from_kernel", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
4281 .set_default(false)
4282 .set_description(""),
4283
4284 Option("bdev_nvme_retry_count", Option::TYPE_INT, Option::LEVEL_ADVANCED)
4285 .set_default(-1)
4286 .set_description(""),
4287
4288 Option("bdev_enable_discard", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
4289 .set_default(false)
4290 .set_description(""),
4291
4292 Option("bdev_async_discard", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
4293 .set_default(false)
4294 .set_description(""),
4295
4296 Option("bluefs_alloc_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4297 .set_default(1_M)
4298 .set_description("Allocation unit size for DB and WAL devices"),
4299
4300 Option("bluefs_shared_alloc_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4301 .set_default(64_K)
4302 .set_description("Allocation unit size for primary/shared device"),
4303
4304 Option("bluefs_max_prefetch", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4305 .set_default(1_M)
4306 .set_description(""),
4307
4308 Option("bluefs_min_log_runway", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4309 .set_default(1_M)
4310 .set_description(""),
4311
4312 Option("bluefs_max_log_runway", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4313 .set_default(4194304)
4314 .set_description(""),
4315
4316 Option("bluefs_log_compact_min_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
4317 .set_default(5.0)
4318 .set_description(""),
4319
4320 Option("bluefs_log_compact_min_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4321 .set_default(16_M)
4322 .set_description(""),
4323
4324 Option("bluefs_min_flush_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4325 .set_default(512_K)
4326 .set_description(""),
4327
4328 Option("bluefs_compact_log_sync", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
4329 .set_default(false)
4330 .set_description(""),
4331
4332 Option("bluefs_buffered_io", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
4333 .set_default(true)
4334 .set_description(""),
4335
4336 Option("bluefs_sync_write", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
4337 .set_default(false)
4338 .set_description(""),
4339
4340 Option("bluefs_allocator", Option::TYPE_STR, Option::LEVEL_DEV)
4341 .set_default("bitmap")
4342 .set_description(""),
4343
4344 Option("bluefs_preextend_wal_files", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
4345 .set_default(true)
4346 .set_description("Preextent rocksdb wal files on mkfs to avoid performance penalty for young stores"),
4347
4348 Option("bluestore_bluefs", Option::TYPE_BOOL, Option::LEVEL_DEV)
4349 .set_default(true)
4350 .set_flag(Option::FLAG_CREATE)
4351 .set_description("Use BlueFS to back rocksdb")
4352 .set_long_description("BlueFS allows rocksdb to share the same physical device(s) as the rest of BlueStore. It should be used in all cases unless testing/developing an alternative metadata database for BlueStore."),
4353
4354 Option("bluestore_bluefs_env_mirror", Option::TYPE_BOOL, Option::LEVEL_DEV)
4355 .set_default(false)
4356 .set_flag(Option::FLAG_CREATE)
4357 .set_description("Mirror bluefs data to file system for testing/validation"),
4358
4359 Option("bluestore_bluefs_min", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4360 .set_default(1_G)
4361 .set_description("minimum disk space allocated to BlueFS (e.g., at mkfs)"),
4362
4363 Option("bluestore_bluefs_min_free", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4364 .set_default(1_G)
4365 .set_description("minimum free space allocated to BlueFS"),
4366
4367 Option("bluestore_bluefs_max_free", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4368 .set_default(10_G)
4369 .set_description("Maximum free space allocated to BlueFS"),
4370
4371 Option("bluestore_bluefs_min_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
4372 .set_default(.02)
4373 .set_description("Minimum fraction of free space devoted to BlueFS"),
4374
4375 Option("bluestore_bluefs_max_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
4376 .set_default(.90)
4377 .set_description("Maximum fraction of free storage devoted to BlueFS"),
4378
4379 Option("bluestore_bluefs_gift_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
4380 .set_default(.02)
4381 .set_description("Maximum fraction of free space to give to BlueFS at once"),
4382
4383 Option("bluestore_bluefs_reclaim_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
4384 .set_default(.20)
4385 .set_description("Maximum fraction of free space to reclaim from BlueFS at once"),
4386
4387 Option("bluestore_bluefs_balance_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
4388 .set_default(1)
4389 .set_description("How frequently (in seconds) to balance free space between BlueFS and BlueStore"),
4390
4391 Option("bluestore_bluefs_alloc_failure_dump_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
4392 .set_default(0)
4393 .set_description("How frequently (in seconds) to dump allocator on"
4394 "BlueFS space allocation failure"),
4395
4396 Option("bluestore_bluefs_db_compatibility", Option::TYPE_BOOL, Option::LEVEL_DEV)
4397 .set_default(true)
4398 .set_description("Sync db with legacy bluefs extents info")
4399 .set_long_description("Enforces db sync with legacy bluefs extents information on close."
4400 " Enables downgrades to pre-nautilus releases"),
4401
4402 Option("bluestore_spdk_mem", Option::TYPE_SIZE, Option::LEVEL_DEV)
4403 .set_default(512)
4404 .set_description("Amount of dpdk memory size in MB")
4405 .set_long_description("If running multiple SPDK instances per node, you must specify the amount of dpdk memory size in MB each instance will use, to make sure each instance uses its own dpdk memory"),
4406
4407 Option("bluestore_spdk_coremask", Option::TYPE_STR, Option::LEVEL_DEV)
4408 .set_default("0x1")
4409 .set_description("A hexadecimal bit mask of the cores to run on. Note the core numbering can change between platforms and should be determined beforehand"),
4410
4411 Option("bluestore_spdk_max_io_completion", Option::TYPE_UINT, Option::LEVEL_DEV)
4412 .set_default(0)
4413 .set_description("Maximal I/Os to be batched completed while checking queue pair completions, 0 means let spdk library determine it"),
4414
4415 Option("bluestore_spdk_io_sleep", Option::TYPE_UINT, Option::LEVEL_DEV)
4416 .set_default(5)
4417 .set_description("Time period to wait if there is no completed I/O from polling"),
4418
4419 Option("bluestore_block_path", Option::TYPE_STR, Option::LEVEL_DEV)
4420 .set_default("")
4421 .set_flag(Option::FLAG_CREATE)
4422 .set_description("Path to block device/file"),
4423
4424 Option("bluestore_block_size", Option::TYPE_SIZE, Option::LEVEL_DEV)
4425 .set_default(1_T)
4426 .set_flag(Option::FLAG_CREATE)
4427 .set_description("Size of file to create for backing bluestore"),
4428
4429 Option("bluestore_block_create", Option::TYPE_BOOL, Option::LEVEL_DEV)
4430 .set_default(true)
4431 .set_flag(Option::FLAG_CREATE)
4432 .set_description("Create bluestore_block_path if it doesn't exist")
4433 .add_see_also("bluestore_block_path").add_see_also("bluestore_block_size"),
4434
4435 Option("bluestore_block_db_path", Option::TYPE_STR, Option::LEVEL_DEV)
4436 .set_default("")
4437 .set_flag(Option::FLAG_CREATE)
4438 .set_description("Path for db block device"),
4439
4440 Option("bluestore_block_db_size", Option::TYPE_UINT, Option::LEVEL_DEV)
4441 .set_default(0)
4442 .set_flag(Option::FLAG_CREATE)
4443 .set_description("Size of file to create for bluestore_block_db_path"),
4444
4445 Option("bluestore_block_db_create", Option::TYPE_BOOL, Option::LEVEL_DEV)
4446 .set_default(false)
4447 .set_flag(Option::FLAG_CREATE)
4448 .set_description("Create bluestore_block_db_path if it doesn't exist")
4449 .add_see_also("bluestore_block_db_path")
4450 .add_see_also("bluestore_block_db_size"),
4451
4452 Option("bluestore_block_wal_path", Option::TYPE_STR, Option::LEVEL_DEV)
4453 .set_default("")
4454 .set_flag(Option::FLAG_CREATE)
4455 .set_description("Path to block device/file backing bluefs wal"),
4456
4457 Option("bluestore_block_wal_size", Option::TYPE_SIZE, Option::LEVEL_DEV)
4458 .set_default(96_M)
4459 .set_flag(Option::FLAG_CREATE)
4460 .set_description("Size of file to create for bluestore_block_wal_path"),
4461
4462 Option("bluestore_block_wal_create", Option::TYPE_BOOL, Option::LEVEL_DEV)
4463 .set_default(false)
4464 .set_flag(Option::FLAG_CREATE)
4465 .set_description("Create bluestore_block_wal_path if it doesn't exist")
4466 .add_see_also("bluestore_block_wal_path")
4467 .add_see_also("bluestore_block_wal_size"),
4468
4469 Option("bluestore_block_preallocate_file", Option::TYPE_BOOL, Option::LEVEL_DEV)
4470 .set_default(false)
4471 .set_flag(Option::FLAG_CREATE)
4472 .set_description("Preallocate file created via bluestore_block*_create"),
4473
4474 Option("bluestore_ignore_data_csum", Option::TYPE_BOOL, Option::LEVEL_DEV)
4475 .set_default(false)
4476 .set_flag(Option::FLAG_RUNTIME)
4477 .set_description("Ignore checksum errors on read and do not generate an EIO error"),
4478
4479 Option("bluestore_csum_type", Option::TYPE_STR, Option::LEVEL_ADVANCED)
4480 .set_default("crc32c")
4481 .set_enum_allowed({"none", "crc32c", "crc32c_16", "crc32c_8", "xxhash32", "xxhash64"})
4482 .set_flag(Option::FLAG_RUNTIME)
4483 .set_description("Default checksum algorithm to use")
4484 .set_long_description("crc32c, xxhash32, and xxhash64 are available. The _16 and _8 variants use only a subset of the bits for more compact (but less reliable) checksumming."),
4485
4486 Option("bluestore_retry_disk_reads", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
4487 .set_default(3)
4488 .set_min_max(0, 255)
4489 .set_flag(Option::FLAG_RUNTIME)
4490 .set_description("Number of read retries on checksum validation error")
4491 .set_long_description("Retries to read data from the disk this many times when checksum validation fails to handle spurious read errors gracefully."),
4492
4493 Option("bluestore_min_alloc_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
4494 .set_default(0)
4495 .set_flag(Option::FLAG_CREATE)
4496 .set_description("Minimum allocation size to allocate for an object")
4497 .set_long_description("A smaller allocation size generally means less data is read and then rewritten when a copy-on-write operation is triggered (e.g., when writing to something that was recently snapshotted). Similarly, less data is journaled before performing an overwrite (writes smaller than min_alloc_size must first pass through the BlueStore journal). Larger values of min_alloc_size reduce the amount of metadata required to describe the on-disk layout and reduce overall fragmentation."),
4498
4499 Option("bluestore_min_alloc_size_hdd", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4500 .set_default(64_K)
4501 .set_flag(Option::FLAG_CREATE)
4502 .set_description("Default min_alloc_size value for rotational media")
4503 .add_see_also("bluestore_min_alloc_size"),
4504
4505 Option("bluestore_min_alloc_size_ssd", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4506 .set_default(16_K)
4507 .set_flag(Option::FLAG_CREATE)
4508 .set_description("Default min_alloc_size value for non-rotational (solid state) media")
4509 .add_see_also("bluestore_min_alloc_size"),
4510
4511 Option("bluestore_max_alloc_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4512 .set_default(0)
4513 .set_flag(Option::FLAG_CREATE)
4514 .set_description("Maximum size of a single allocation (0 for no max)"),
4515
4516 Option("bluestore_prefer_deferred_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4517 .set_default(0)
4518 .set_flag(Option::FLAG_RUNTIME)
4519 .set_description("Writes smaller than this size will be written to the journal and then asynchronously written to the device. This can be beneficial when using rotational media where seeks are expensive, and is helpful both with and without solid state journal/wal devices."),
4520
4521 Option("bluestore_prefer_deferred_size_hdd", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4522 .set_default(32768)
4523 .set_flag(Option::FLAG_RUNTIME)
4524 .set_description("Default bluestore_prefer_deferred_size for rotational media")
4525 .add_see_also("bluestore_prefer_deferred_size"),
4526
4527 Option("bluestore_prefer_deferred_size_ssd", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4528 .set_default(0)
4529 .set_flag(Option::FLAG_RUNTIME)
4530 .set_description("Default bluestore_prefer_deferred_size for non-rotational (solid state) media")
4531 .add_see_also("bluestore_prefer_deferred_size"),
4532
4533 Option("bluestore_compression_mode", Option::TYPE_STR, Option::LEVEL_ADVANCED)
4534 .set_default("none")
4535 .set_enum_allowed({"none", "passive", "aggressive", "force"})
4536 .set_flag(Option::FLAG_RUNTIME)
4537 .set_description("Default policy for using compression when pool does not specify")
4538 .set_long_description("'none' means never use compression. 'passive' means use compression when clients hint that data is compressible. 'aggressive' means use compression unless clients hint that data is not compressible. This option is used when the per-pool property for the compression mode is not present."),
4539
4540 Option("bluestore_compression_algorithm", Option::TYPE_STR, Option::LEVEL_ADVANCED)
4541 .set_default("snappy")
4542 .set_enum_allowed({"", "snappy", "zlib", "zstd", "lz4"})
4543 .set_flag(Option::FLAG_RUNTIME)
4544 .set_description("Default compression algorithm to use when writing object data")
4545 .set_long_description("This controls the default compressor to use (if any) if the per-pool property is not set. Note that zstd is *not* recommended for bluestore due to high CPU overhead when compressing small amounts of data."),
4546
4547 Option("bluestore_compression_min_blob_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4548 .set_default(0)
4549 .set_flag(Option::FLAG_RUNTIME)
4550 .set_description("Maximum chunk size to apply compression to when random access is expected for an object.")
4551 .set_long_description("Chunks larger than this are broken into smaller chunks before being compressed"),
4552
4553 Option("bluestore_compression_min_blob_size_hdd", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4554 .set_default(128_K)
4555 .set_flag(Option::FLAG_RUNTIME)
4556 .set_description("Default value of bluestore_compression_min_blob_size for rotational media")
4557 .add_see_also("bluestore_compression_min_blob_size"),
4558
4559 Option("bluestore_compression_min_blob_size_ssd", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4560 .set_default(8_K)
4561 .set_flag(Option::FLAG_RUNTIME)
4562 .set_description("Default value of bluestore_compression_min_blob_size for non-rotational (solid state) media")
4563 .add_see_also("bluestore_compression_min_blob_size"),
4564
4565 Option("bluestore_compression_max_blob_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4566 .set_default(0)
4567 .set_flag(Option::FLAG_RUNTIME)
4568 .set_description("Maximum chunk size to apply compression to when non-random access is expected for an object.")
4569 .set_long_description("Chunks larger than this are broken into smaller chunks before being compressed"),
4570
4571 Option("bluestore_compression_max_blob_size_hdd", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4572 .set_default(512_K)
4573 .set_flag(Option::FLAG_RUNTIME)
4574 .set_description("Default value of bluestore_compression_max_blob_size for rotational media")
4575 .add_see_also("bluestore_compression_max_blob_size"),
4576
4577 Option("bluestore_compression_max_blob_size_ssd", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4578 .set_default(64_K)
4579 .set_flag(Option::FLAG_RUNTIME)
4580 .set_description("Default value of bluestore_compression_max_blob_size for non-rotational (solid state) media")
4581 .add_see_also("bluestore_compression_max_blob_size"),
4582
4583 Option("bluestore_gc_enable_blob_threshold", Option::TYPE_INT, Option::LEVEL_DEV)
4584 .set_default(0)
4585 .set_flag(Option::FLAG_RUNTIME)
4586 .set_description(""),
4587
4588 Option("bluestore_gc_enable_total_threshold", Option::TYPE_INT, Option::LEVEL_DEV)
4589 .set_default(0)
4590 .set_flag(Option::FLAG_RUNTIME)
4591 .set_description(""),
4592
4593 Option("bluestore_max_blob_size", Option::TYPE_SIZE, Option::LEVEL_DEV)
4594 .set_default(0)
4595 .set_flag(Option::FLAG_RUNTIME)
4596 .set_description("")
4597 .set_long_description("Bluestore blobs are collections of extents (ie on-disk data) originating from one or more objects. Blobs can be compressed, typically have checksum data, may be overwritten, may be shared (with an extent ref map), or split. This setting controls the maximum size a blob is allowed to be."),
4598
4599 Option("bluestore_max_blob_size_hdd", Option::TYPE_SIZE, Option::LEVEL_DEV)
4600 .set_default(512_K)
4601 .set_flag(Option::FLAG_RUNTIME)
4602 .set_description("")
4603 .add_see_also("bluestore_max_blob_size"),
4604
4605 Option("bluestore_max_blob_size_ssd", Option::TYPE_SIZE, Option::LEVEL_DEV)
4606 .set_default(64_K)
4607 .set_flag(Option::FLAG_RUNTIME)
4608 .set_description("")
4609 .add_see_also("bluestore_max_blob_size"),
4610
4611 Option("bluestore_compression_required_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
4612 .set_default(.875)
4613 .set_flag(Option::FLAG_RUNTIME)
4614 .set_description("Compression ratio required to store compressed data")
4615 .set_long_description("If we compress data and get less than this we discard the result and store the original uncompressed data."),
4616
4617 Option("bluestore_extent_map_shard_max_size", Option::TYPE_SIZE, Option::LEVEL_DEV)
4618 .set_default(1200)
4619 .set_description("Max size (bytes) for a single extent map shard before splitting"),
4620
4621 Option("bluestore_extent_map_shard_target_size", Option::TYPE_SIZE, Option::LEVEL_DEV)
4622 .set_default(500)
4623 .set_description("Target size (bytes) for a single extent map shard"),
4624
4625 Option("bluestore_extent_map_shard_min_size", Option::TYPE_SIZE, Option::LEVEL_DEV)
4626 .set_default(150)
4627 .set_description("Min size (bytes) for a single extent map shard before merging"),
4628
4629 Option("bluestore_extent_map_shard_target_size_slop", Option::TYPE_FLOAT, Option::LEVEL_DEV)
4630 .set_default(.2)
4631 .set_description("Ratio above/below target for a shard when trying to align to an existing extent or blob boundary"),
4632
4633 Option("bluestore_extent_map_inline_shard_prealloc_size", Option::TYPE_SIZE, Option::LEVEL_DEV)
4634 .set_default(256)
4635 .set_description("Preallocated buffer for inline shards"),
4636
4637 Option("bluestore_cache_trim_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
4638 .set_default(.05)
4639 .set_description("How frequently we trim the bluestore cache"),
4640
4641 Option("bluestore_cache_trim_max_skip_pinned", Option::TYPE_UINT, Option::LEVEL_DEV)
4642 .set_default(64)
4643 .set_description("Max pinned cache entries we consider before giving up"),
4644
4645 Option("bluestore_cache_type", Option::TYPE_STR, Option::LEVEL_DEV)
4646 .set_default("2q")
4647 .set_enum_allowed({"2q", "lru"})
4648 .set_description("Cache replacement algorithm"),
4649
4650 Option("bluestore_2q_cache_kin_ratio", Option::TYPE_FLOAT, Option::LEVEL_DEV)
4651 .set_default(.5)
4652 .set_description("2Q paper suggests .5"),
4653
4654 Option("bluestore_2q_cache_kout_ratio", Option::TYPE_FLOAT, Option::LEVEL_DEV)
4655 .set_default(.5)
4656 .set_description("2Q paper suggests .5"),
4657
4658 Option("bluestore_cache_size", Option::TYPE_SIZE, Option::LEVEL_DEV)
4659 .set_default(0)
4660 .set_description("Cache size (in bytes) for BlueStore")
4661 .set_long_description("This includes data and metadata cached by BlueStore as well as memory devoted to rocksdb's cache(s)."),
4662
4663 Option("bluestore_cache_size_hdd", Option::TYPE_SIZE, Option::LEVEL_DEV)
4664 .set_default(1_G)
4665 .set_description("Default bluestore_cache_size for rotational media")
4666 .add_see_also("bluestore_cache_size"),
4667
4668 Option("bluestore_cache_size_ssd", Option::TYPE_SIZE, Option::LEVEL_DEV)
4669 .set_default(3_G)
4670 .set_description("Default bluestore_cache_size for non-rotational (solid state) media")
4671 .add_see_also("bluestore_cache_size"),
4672
4673 Option("bluestore_cache_meta_ratio", Option::TYPE_FLOAT, Option::LEVEL_DEV)
4674 .set_default(.4)
4675 .add_see_also("bluestore_cache_size")
4676 .set_description("Ratio of bluestore cache to devote to metadata"),
4677
4678 Option("bluestore_cache_kv_ratio", Option::TYPE_FLOAT, Option::LEVEL_DEV)
4679 .set_default(.4)
4680 .add_see_also("bluestore_cache_size")
4681 .set_description("Ratio of bluestore cache to devote to kv database (rocksdb)"),
4682
4683 Option("bluestore_cache_autotune", Option::TYPE_BOOL, Option::LEVEL_DEV)
4684 .set_default(true)
4685 .add_see_also("bluestore_cache_size")
4686 .add_see_also("bluestore_cache_meta_ratio")
4687 .set_description("Automatically tune the ratio of caches while respecting min values."),
4688
4689 Option("bluestore_cache_autotune_interval", Option::TYPE_FLOAT, Option::LEVEL_DEV)
4690 .set_default(5)
4691 .add_see_also("bluestore_cache_autotune")
4692 .set_description("The number of seconds to wait between rebalances when cache autotune is enabled."),
4693
4694 Option("bluestore_kvbackend", Option::TYPE_STR, Option::LEVEL_DEV)
4695 .set_default("rocksdb")
4696 .set_flag(Option::FLAG_CREATE)
4697 .set_description("Key value database to use for bluestore"),
4698
4699 Option("bluestore_allocator", Option::TYPE_STR, Option::LEVEL_ADVANCED)
4700 .set_default("bitmap")
4701 .set_enum_allowed({"bitmap", "stupid"})
4702 .set_description("Allocator policy")
4703 .set_long_description("Allocator to use for bluestore. Stupid should only be used for testing."),
4704
4705 Option("bluestore_freelist_blocks_per_key", Option::TYPE_SIZE, Option::LEVEL_DEV)
4706 .set_default(128)
4707 .set_description("Block (and bits) per database key"),
4708
4709 Option("bluestore_bitmapallocator_blocks_per_zone", Option::TYPE_SIZE, Option::LEVEL_DEV)
4710 .set_default(1024)
4711 .set_description(""),
4712
4713 Option("bluestore_bitmapallocator_span_size", Option::TYPE_SIZE, Option::LEVEL_DEV)
4714 .set_default(1024)
4715 .set_description(""),
4716
4717 Option("bluestore_max_deferred_txc", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
4718 .set_default(32)
4719 .set_description("Max transactions with deferred writes that can accumulate before we force flush deferred writes"),
4720
4721 Option("bluestore_max_defer_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
4722 .set_default(3)
4723 .set_description("max duration to force deferred submit"),
4724
4725 Option("bluestore_rocksdb_options", Option::TYPE_STR, Option::LEVEL_ADVANCED)
4726 .set_default("compression=kNoCompression,max_write_buffer_number=4,min_write_buffer_number_to_merge=1,recycle_log_file_num=4,write_buffer_size=268435456,writable_file_max_buffer_size=0,compaction_readahead_size=2097152,max_background_compactions=2")
4727 .set_description("Rocksdb options"),
4728
4729 Option("bluestore_rocksdb_cf", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
4730 .set_default(false)
4731 .set_description("Enable use of rocksdb column families for bluestore metadata"),
4732
4733 Option("bluestore_rocksdb_cfs", Option::TYPE_STR, Option::LEVEL_DEV)
4734 .set_default("M= P= L=")
4735 .set_description("List of whitespace-separate key/value pairs where key is CF name and value is CF options"),
4736
4737 Option("bluestore_fsck_on_mount", Option::TYPE_BOOL, Option::LEVEL_DEV)
4738 .set_default(false)
4739 .set_description("Run fsck at mount"),
4740
4741 Option("bluestore_fsck_on_mount_deep", Option::TYPE_BOOL, Option::LEVEL_DEV)
4742 .set_default(false)
4743 .set_description("Run deep fsck at mount when bluestore_fsck_on_mount is set to true"),
4744
4745 Option("bluestore_fsck_quick_fix_on_mount", Option::TYPE_BOOL, Option::LEVEL_DEV)
4746 .set_default(true)
4747 .set_description("Do quick-fix for the store at mount"),
4748
4749 Option("bluestore_fsck_on_umount", Option::TYPE_BOOL, Option::LEVEL_DEV)
4750 .set_default(false)
4751 .set_description("Run fsck at umount"),
4752
4753 Option("bluestore_fsck_on_umount_deep", Option::TYPE_BOOL, Option::LEVEL_DEV)
4754 .set_default(false)
4755 .set_description("Run deep fsck at umount when bluestore_fsck_on_umount is set to true"),
4756
4757 Option("bluestore_fsck_on_mkfs", Option::TYPE_BOOL, Option::LEVEL_DEV)
4758 .set_default(true)
4759 .set_description("Run fsck after mkfs"),
4760
4761 Option("bluestore_fsck_on_mkfs_deep", Option::TYPE_BOOL, Option::LEVEL_DEV)
4762 .set_default(false)
4763 .set_description("Run deep fsck after mkfs"),
4764
4765 Option("bluestore_sync_submit_transaction", Option::TYPE_BOOL, Option::LEVEL_DEV)
4766 .set_default(false)
4767 .set_description("Try to submit metadata transaction to rocksdb in queuing thread context"),
4768
4769 Option("bluestore_fsck_read_bytes_cap", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4770 .set_default(64_M)
4771 .set_flag(Option::FLAG_RUNTIME)
4772 .set_description("Maximum bytes read at once by deep fsck"),
4773
4774 Option("bluestore_fsck_quick_fix_threads", Option::TYPE_INT, Option::LEVEL_ADVANCED)
4775 .set_default(2)
4776 .set_description("Number of additional threads to perform quick-fix (shallow fsck) command"),
4777
4778 Option("bluestore_throttle_bytes", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4779 .set_default(64_M)
4780 .set_flag(Option::FLAG_RUNTIME)
4781 .set_description("Maximum bytes in flight before we throttle IO submission"),
4782
4783 Option("bluestore_throttle_deferred_bytes", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4784 .set_default(128_M)
4785 .set_flag(Option::FLAG_RUNTIME)
4786 .set_description("Maximum bytes for deferred writes before we throttle IO submission"),
4787
4788 Option("bluestore_throttle_cost_per_io", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4789 .set_default(0)
4790 .set_flag(Option::FLAG_RUNTIME)
4791 .set_description("Overhead added to transaction cost (in bytes) for each IO"),
4792
4793 Option("bluestore_throttle_cost_per_io_hdd", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
4794 .set_default(670000)
4795 .set_flag(Option::FLAG_RUNTIME)
4796 .set_description("Default bluestore_throttle_cost_per_io for rotational media")
4797 .add_see_also("bluestore_throttle_cost_per_io"),
4798
4799 Option("bluestore_throttle_cost_per_io_ssd", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
4800 .set_default(4000)
4801 .set_flag(Option::FLAG_RUNTIME)
4802 .set_description("Default bluestore_throttle_cost_per_io for non-rotation (solid state) media")
4803 .add_see_also("bluestore_throttle_cost_per_io"),
4804
4805 Option("bluestore_deferred_batch_ops", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
4806 .set_default(0)
4807 .set_flag(Option::FLAG_RUNTIME)
4808 .set_description("Max number of deferred writes before we flush the deferred write queue"),
4809
4810 Option("bluestore_deferred_batch_ops_hdd", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
4811 .set_default(64)
4812 .set_flag(Option::FLAG_RUNTIME)
4813 .set_description("Default bluestore_deferred_batch_ops for rotational media")
4814 .add_see_also("bluestore_deferred_batch_ops"),
4815
4816 Option("bluestore_deferred_batch_ops_ssd", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
4817 .set_default(16)
4818 .set_flag(Option::FLAG_RUNTIME)
4819 .set_description("Default bluestore_deferred_batch_ops for non-rotational (solid state) media")
4820 .add_see_also("bluestore_deferred_batch_ops"),
4821
4822 Option("bluestore_nid_prealloc", Option::TYPE_INT, Option::LEVEL_DEV)
4823 .set_default(1024)
4824 .set_description("Number of unique object ids to preallocate at a time"),
4825
4826 Option("bluestore_blobid_prealloc", Option::TYPE_UINT, Option::LEVEL_DEV)
4827 .set_default(10240)
4828 .set_description("Number of unique blob ids to preallocate at a time"),
4829
4830 Option("bluestore_clone_cow", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
4831 .set_default(true)
4832 .set_flag(Option::FLAG_RUNTIME)
4833 .set_description("Use copy-on-write when cloning objects (versus reading and rewriting them at clone time)"),
4834
4835 Option("bluestore_default_buffered_read", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
4836 .set_default(true)
4837 .set_flag(Option::FLAG_RUNTIME)
4838 .set_description("Cache read results by default (unless hinted NOCACHE or WONTNEED)"),
4839
4840 Option("bluestore_default_buffered_write", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
4841 .set_default(false)
4842 .set_flag(Option::FLAG_RUNTIME)
4843 .set_description("Cache writes by default (unless hinted NOCACHE or WONTNEED)"),
4844
4845 Option("bluestore_debug_misc", Option::TYPE_BOOL, Option::LEVEL_DEV)
4846 .set_default(false)
4847 .set_description(""),
4848
4849 Option("bluestore_debug_no_reuse_blocks", Option::TYPE_BOOL, Option::LEVEL_DEV)
4850 .set_default(false)
4851 .set_description(""),
4852
4853 Option("bluestore_debug_small_allocations", Option::TYPE_INT, Option::LEVEL_DEV)
4854 .set_default(0)
4855 .set_description(""),
4856
4857 Option("bluestore_debug_too_many_blobs_threshold", Option::TYPE_INT, Option::LEVEL_DEV)
4858 .set_default(24*1024)
4859 .set_description(""),
4860
4861 Option("bluestore_debug_freelist", Option::TYPE_BOOL, Option::LEVEL_DEV)
4862 .set_default(false)
4863 .set_description(""),
4864
4865 Option("bluestore_debug_prefill", Option::TYPE_FLOAT, Option::LEVEL_DEV)
4866 .set_default(0)
4867 .set_description("simulate fragmentation"),
4868
4869 Option("bluestore_debug_prefragment_max", Option::TYPE_SIZE, Option::LEVEL_DEV)
4870 .set_default(1_M)
4871 .set_description(""),
4872
4873 Option("bluestore_debug_inject_read_err", Option::TYPE_BOOL, Option::LEVEL_DEV)
4874 .set_default(false)
4875 .set_description(""),
4876
4877 Option("bluestore_debug_randomize_serial_transaction", Option::TYPE_INT, Option::LEVEL_DEV)
4878 .set_default(0)
4879 .set_description(""),
4880
4881 Option("bluestore_debug_omit_block_device_write", Option::TYPE_BOOL, Option::LEVEL_DEV)
4882 .set_default(false)
4883 .set_description(""),
4884
4885 Option("bluestore_debug_fsck_abort", Option::TYPE_BOOL, Option::LEVEL_DEV)
4886 .set_default(false)
4887 .set_description(""),
4888
4889 Option("bluestore_debug_omit_kv_commit", Option::TYPE_BOOL, Option::LEVEL_DEV)
4890 .set_default(false)
4891 .set_description(""),
4892
4893 Option("bluestore_debug_permit_any_bdev_label", Option::TYPE_BOOL, Option::LEVEL_DEV)
4894 .set_default(false)
4895 .set_description(""),
4896
4897 Option("bluestore_debug_random_read_err", Option::TYPE_FLOAT, Option::LEVEL_DEV)
4898 .set_default(0)
4899 .set_description(""),
4900
4901 Option("bluestore_debug_inject_bug21040", Option::TYPE_BOOL, Option::LEVEL_DEV)
4902 .set_default(false)
4903 .set_description(""),
4904
4905 Option("bluestore_debug_inject_csum_err_probability", Option::TYPE_FLOAT, Option::LEVEL_DEV)
4906 .set_default(0.0)
4907 .set_description("inject crc verification errors into bluestore device reads"),
4908
4909 Option("bluestore_fsck_error_on_no_per_pool_stats", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
4910 .set_default(false)
4911 .set_description("Make fsck error (instead of warn) when bluestore lacks per-pool stats, e.g., after an upgrade"),
4912
4913 Option("bluestore_warn_on_bluefs_spillover", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
4914 .set_default(true)
4915 .set_description("Enable health indication on bluefs slow device usage"),
4916
4917 Option("bluestore_warn_on_legacy_statfs", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
4918 .set_default(true)
4919 .set_description("Enable health indication on lack of per-pool statfs reporting from bluestore"),
4920
4921 Option("bluestore_fsck_error_on_no_per_pool_omap", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
4922 .set_default(false)
4923 .set_description("Make fsck error (instead of warn) when objects without per-pool omap are found"),
4924
4925 Option("bluestore_warn_on_no_per_pool_omap", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
4926 .set_default(true)
4927 .set_description("Enable health indication on lack of per-pool omap"),
4928
4929 Option("bluestore_log_op_age", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
4930 .set_default(5)
4931 .set_description("log operation if it's slower than this age (seconds)"),
4932
4933 Option("bluestore_log_omap_iterator_age", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
4934 .set_default(5)
4935 .set_description("log omap iteration operation if it's slower than this age (seconds)"),
4936
4937 Option("bluestore_log_collection_list_age", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
4938 .set_default(60)
4939 .set_description("log collection list operation if it's slower than this age (seconds)"),
4940
4941 Option("bluestore_debug_enforce_settings", Option::TYPE_STR, Option::LEVEL_DEV)
4942 .set_default("default")
4943 .set_enum_allowed({"default", "hdd", "ssd"})
4944 .set_description("Enforces specific hw profile settings")
4945 .set_long_description("'hdd' enforces settings intended for BlueStore above a rotational drive. 'ssd' enforces settings intended for BlueStore above a solid drive. 'default' - using settings for the actual hardware."),
4946
4947
4948 // -----------------------------------------
4949 // kstore
4950
4951 Option("kstore_max_ops", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
4952 .set_default(512)
4953 .set_description(""),
4954
4955 Option("kstore_max_bytes", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4956 .set_default(64_M)
4957 .set_description(""),
4958
4959 Option("kstore_backend", Option::TYPE_STR, Option::LEVEL_ADVANCED)
4960 .set_default("rocksdb")
4961 .set_description(""),
4962
4963 Option("kstore_rocksdb_options", Option::TYPE_STR, Option::LEVEL_ADVANCED)
4964 .set_default("compression=kNoCompression")
4965 .set_description("Options to pass through when RocksDB is used as the KeyValueDB for kstore."),
4966
4967 Option("kstore_fsck_on_mount", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
4968 .set_default(false)
4969 .set_description("Whether or not to run fsck on mount for kstore."),
4970
4971 Option("kstore_fsck_on_mount_deep", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
4972 .set_default(true)
4973 .set_description("Whether or not to run deep fsck on mount for kstore"),
4974
4975 Option("kstore_nid_prealloc", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
4976 .set_default(1024)
4977 .set_description(""),
4978
4979 Option("kstore_sync_transaction", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
4980 .set_default(false)
4981 .set_description(""),
4982
4983 Option("kstore_sync_submit_transaction", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
4984 .set_default(false)
4985 .set_description(""),
4986
4987 Option("kstore_onode_map_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
4988 .set_default(1024)
4989 .set_description(""),
4990
4991 Option("kstore_default_stripe_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
4992 .set_default(65536)
4993 .set_description(""),
4994
4995 // ---------------------
4996 // filestore
4997
4998 Option("filestore_rocksdb_options", Option::TYPE_STR, Option::LEVEL_DEV)
4999 .set_default("max_background_jobs=10,compaction_readahead_size=2097152,compression=kNoCompression")
5000 .set_description("Options to pass through when RocksDB is used as the KeyValueDB for filestore."),
5001
5002 Option("filestore_omap_backend", Option::TYPE_STR, Option::LEVEL_DEV)
5003 .set_default("rocksdb")
5004 .set_enum_allowed({"leveldb", "rocksdb"})
5005 .set_description("The KeyValueDB to use for filestore metadata (ie omap)."),
5006
5007 Option("filestore_omap_backend_path", Option::TYPE_STR, Option::LEVEL_DEV)
5008 .set_default("")
5009 .set_description("The path where the filestore KeyValueDB should store it's database(s)."),
5010
5011 Option("filestore_wbthrottle_enable", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
5012 .set_default(true)
5013 .set_description("Enabling throttling of operations to backing file system"),
5014
5015 Option("filestore_wbthrottle_btrfs_bytes_start_flusher", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
5016 .set_default(41943040)
5017 .set_description("Start flushing (fsyncing) when this many bytes are written(btrfs)"),
5018
5019 Option("filestore_wbthrottle_btrfs_bytes_hard_limit", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
5020 .set_default(419430400)
5021 .set_description("Block writes when this many bytes haven't been flushed (fsynced) (btrfs)"),
5022
5023 Option("filestore_wbthrottle_btrfs_ios_start_flusher", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
5024 .set_default(500)
5025 .set_description("Start flushing (fsyncing) when this many IOs are written (brtrfs)"),
5026
5027 Option("filestore_wbthrottle_btrfs_ios_hard_limit", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
5028 .set_default(5000)
5029 .set_description("Block writes when this many IOs haven't been flushed (fsynced) (btrfs)"),
5030
5031 Option("filestore_wbthrottle_btrfs_inodes_start_flusher", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
5032 .set_default(500)
5033 .set_description("Start flushing (fsyncing) when this many distinct inodes have been modified (btrfs)"),
5034
5035 Option("filestore_wbthrottle_xfs_bytes_start_flusher", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
5036 .set_default(41943040)
5037 .set_description("Start flushing (fsyncing) when this many bytes are written(xfs)"),
5038
5039 Option("filestore_wbthrottle_xfs_bytes_hard_limit", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
5040 .set_default(419430400)
5041 .set_description("Block writes when this many bytes haven't been flushed (fsynced) (xfs)"),
5042
5043 Option("filestore_wbthrottle_xfs_ios_start_flusher", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
5044 .set_default(500)
5045 .set_description("Start flushing (fsyncing) when this many IOs are written (xfs)"),
5046
5047 Option("filestore_wbthrottle_xfs_ios_hard_limit", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
5048 .set_default(5000)
5049 .set_description("Block writes when this many IOs haven't been flushed (fsynced) (xfs)"),
5050
5051 Option("filestore_wbthrottle_xfs_inodes_start_flusher", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
5052 .set_default(500)
5053 .set_description("Start flushing (fsyncing) when this many distinct inodes have been modified (xfs)"),
5054
5055 Option("filestore_wbthrottle_btrfs_inodes_hard_limit", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
5056 .set_default(5000)
5057 .set_description("Block writing when this many inodes have outstanding writes (btrfs)"),
5058
5059 Option("filestore_wbthrottle_xfs_inodes_hard_limit", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
5060 .set_default(5000)
5061 .set_description("Block writing when this many inodes have outstanding writes (xfs)"),
5062
5063 Option("filestore_odsync_write", Option::TYPE_BOOL, Option::LEVEL_DEV)
5064 .set_default(false)
5065 .set_description("Write with O_DSYNC"),
5066
5067 Option("filestore_index_retry_probability", Option::TYPE_FLOAT, Option::LEVEL_DEV)
5068 .set_default(0)
5069 .set_description(""),
5070
5071 Option("filestore_debug_inject_read_err", Option::TYPE_BOOL, Option::LEVEL_DEV)
5072 .set_default(false)
5073 .set_description(""),
5074
5075 Option("filestore_debug_random_read_err", Option::TYPE_FLOAT, Option::LEVEL_DEV)
5076 .set_default(0)
5077 .set_description(""),
5078
5079 Option("filestore_debug_omap_check", Option::TYPE_BOOL, Option::LEVEL_DEV)
5080 .set_default(false)
5081 .set_description(""),
5082
5083 Option("filestore_omap_header_cache_size", Option::TYPE_SIZE, Option::LEVEL_DEV)
5084 .set_default(1024)
5085 .set_description(""),
5086
5087 Option("filestore_max_inline_xattr_size", Option::TYPE_SIZE, Option::LEVEL_DEV)
5088 .set_default(0)
5089 .set_description(""),
5090
5091 Option("filestore_max_inline_xattr_size_xfs", Option::TYPE_SIZE, Option::LEVEL_DEV)
5092 .set_default(65536)
5093 .set_description(""),
5094
5095 Option("filestore_max_inline_xattr_size_btrfs", Option::TYPE_SIZE, Option::LEVEL_DEV)
5096 .set_default(2048)
5097 .set_description(""),
5098
5099 Option("filestore_max_inline_xattr_size_other", Option::TYPE_SIZE, Option::LEVEL_DEV)
5100 .set_default(512)
5101 .set_description(""),
5102
5103 Option("filestore_max_inline_xattrs", Option::TYPE_UINT, Option::LEVEL_DEV)
5104 .set_default(0)
5105 .set_description(""),
5106
5107 Option("filestore_max_inline_xattrs_xfs", Option::TYPE_UINT, Option::LEVEL_DEV)
5108 .set_default(10)
5109 .set_description(""),
5110
5111 Option("filestore_max_inline_xattrs_btrfs", Option::TYPE_UINT, Option::LEVEL_DEV)
5112 .set_default(10)
5113 .set_description(""),
5114
5115 Option("filestore_max_inline_xattrs_other", Option::TYPE_UINT, Option::LEVEL_DEV)
5116 .set_default(2)
5117 .set_description(""),
5118
5119 Option("filestore_max_xattr_value_size", Option::TYPE_SIZE, Option::LEVEL_DEV)
5120 .set_default(0)
5121 .set_description(""),
5122
5123 Option("filestore_max_xattr_value_size_xfs", Option::TYPE_SIZE, Option::LEVEL_DEV)
5124 .set_default(64_K)
5125 .set_description(""),
5126
5127 Option("filestore_max_xattr_value_size_btrfs", Option::TYPE_SIZE, Option::LEVEL_DEV)
5128 .set_default(64_K)
5129 .set_description(""),
5130
5131 Option("filestore_max_xattr_value_size_other", Option::TYPE_SIZE, Option::LEVEL_DEV)
5132 .set_default(1_K)
5133 .set_description(""),
5134
5135 Option("filestore_sloppy_crc", Option::TYPE_BOOL, Option::LEVEL_DEV)
5136 .set_default(false)
5137 .set_description(""),
5138
5139 Option("filestore_sloppy_crc_block_size", Option::TYPE_SIZE, Option::LEVEL_DEV)
5140 .set_default(65536)
5141 .set_description(""),
5142
5143 Option("filestore_max_alloc_hint_size", Option::TYPE_SIZE, Option::LEVEL_DEV)
5144 .set_default(1ULL << 20)
5145 .set_description(""),
5146
5147 Option("filestore_max_sync_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
5148 .set_default(5)
5149 .set_description("Period between calls to syncfs(2) and journal trims (seconds)"),
5150
5151 Option("filestore_min_sync_interval", Option::TYPE_FLOAT, Option::LEVEL_DEV)
5152 .set_default(.01)
5153 .set_description("Minimum period between calls to syncfs(2)"),
5154
5155 Option("filestore_btrfs_snap", Option::TYPE_BOOL, Option::LEVEL_DEV)
5156 .set_default(true)
5157 .set_description(""),
5158
5159 Option("filestore_btrfs_clone_range", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
5160 .set_default(true)
5161 .set_description("Use btrfs clone_range ioctl to efficiently duplicate objects"),
5162
5163 Option("filestore_zfs_snap", Option::TYPE_BOOL, Option::LEVEL_DEV)
5164 .set_default(false)
5165 .set_description(""),
5166
5167 Option("filestore_fsync_flushes_journal_data", Option::TYPE_BOOL, Option::LEVEL_DEV)
5168 .set_default(false)
5169 .set_description(""),
5170
5171 Option("filestore_fiemap", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
5172 .set_default(false)
5173 .set_description("Use fiemap ioctl(2) to determine which parts of objects are sparse"),
5174
5175 Option("filestore_punch_hole", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
5176 .set_default(false)
5177 .set_description("Use fallocate(2) FALLOC_FL_PUNCH_HOLE to efficiently zero ranges of objects"),
5178
5179 Option("filestore_seek_data_hole", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
5180 .set_default(false)
5181 .set_description("Use lseek(2) SEEK_HOLE and SEEK_DATA to determine which parts of objects are sparse"),
5182
5183 Option("filestore_splice", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
5184 .set_default(false)
5185 .set_description("Use splice(2) to more efficiently copy data between files"),
5186
5187 Option("filestore_fadvise", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
5188 .set_default(true)
5189 .set_description("Use posix_fadvise(2) to pass hints to file system"),
5190
5191 Option("filestore_collect_device_partition_information", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
5192 .set_default(true)
5193 .set_description("Collect metadata about the backing file system on OSD startup"),
5194
5195 Option("filestore_xfs_extsize", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
5196 .set_default(false)
5197 .set_description("Use XFS extsize ioctl(2) to hint allocator about expected write sizes"),
5198
5199 Option("filestore_journal_parallel", Option::TYPE_BOOL, Option::LEVEL_DEV)
5200 .set_default(false)
5201 .set_description(""),
5202
5203 Option("filestore_journal_writeahead", Option::TYPE_BOOL, Option::LEVEL_DEV)
5204 .set_default(false)
5205 .set_description(""),
5206
5207 Option("filestore_journal_trailing", Option::TYPE_BOOL, Option::LEVEL_DEV)
5208 .set_default(false)
5209 .set_description(""),
5210
5211 Option("filestore_queue_max_ops", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
5212 .set_default(50)
5213 .set_description("Max IO operations in flight"),
5214
5215 Option("filestore_queue_max_bytes", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
5216 .set_default(100_M)
5217 .set_description("Max (written) bytes in flight"),
5218
5219 Option("filestore_caller_concurrency", Option::TYPE_INT, Option::LEVEL_DEV)
5220 .set_default(10)
5221 .set_description(""),
5222
5223 Option("filestore_expected_throughput_bytes", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
5224 .set_default(200_M)
5225 .set_description("Expected throughput of backend device (aids throttling calculations)"),
5226
5227 Option("filestore_expected_throughput_ops", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
5228 .set_default(200)
5229 .set_description("Expected through of backend device in IOPS (aids throttling calculations)"),
5230
5231 Option("filestore_queue_max_delay_multiple", Option::TYPE_FLOAT, Option::LEVEL_DEV)
5232 .set_default(0)
5233 .set_description(""),
5234
5235 Option("filestore_queue_high_delay_multiple", Option::TYPE_FLOAT, Option::LEVEL_DEV)
5236 .set_default(0)
5237 .set_description(""),
5238
5239 Option("filestore_queue_max_delay_multiple_bytes", Option::TYPE_FLOAT, Option::LEVEL_DEV)
5240 .set_default(0)
5241 .set_description(""),
5242
5243 Option("filestore_queue_high_delay_multiple_bytes", Option::TYPE_FLOAT, Option::LEVEL_DEV)
5244 .set_default(0)
5245 .set_description(""),
5246
5247 Option("filestore_queue_max_delay_multiple_ops", Option::TYPE_FLOAT, Option::LEVEL_DEV)
5248 .set_default(0)
5249 .set_description(""),
5250
5251 Option("filestore_queue_high_delay_multiple_ops", Option::TYPE_FLOAT, Option::LEVEL_DEV)
5252 .set_default(0)
5253 .set_description(""),
5254
5255 Option("filestore_queue_low_threshhold", Option::TYPE_FLOAT, Option::LEVEL_DEV)
5256 .set_default(0.3)
5257 .set_description(""),
5258
5259 Option("filestore_queue_high_threshhold", Option::TYPE_FLOAT, Option::LEVEL_DEV)
5260 .set_default(0.9)
5261 .set_description(""),
5262
5263 Option("filestore_op_threads", Option::TYPE_INT, Option::LEVEL_ADVANCED)
5264 .set_default(2)
5265 .set_description("Threads used to apply changes to backing file system"),
5266
5267 Option("filestore_op_thread_timeout", Option::TYPE_INT, Option::LEVEL_ADVANCED)
5268 .set_default(60)
5269 .set_description("Seconds before a worker thread is considered stalled"),
5270
5271 Option("filestore_op_thread_suicide_timeout", Option::TYPE_INT, Option::LEVEL_ADVANCED)
5272 .set_default(180)
5273 .set_description("Seconds before a worker thread is considered dead"),
5274
5275 Option("filestore_commit_timeout", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
5276 .set_default(600)
5277 .set_description("Seconds before backing file system is considered hung"),
5278
5279 Option("filestore_fiemap_threshold", Option::TYPE_SIZE, Option::LEVEL_DEV)
5280 .set_default(4_K)
5281 .set_description(""),
5282
5283 Option("filestore_merge_threshold", Option::TYPE_INT, Option::LEVEL_DEV)
5284 .set_default(-10)
5285 .set_description(""),
5286
5287 Option("filestore_split_multiple", Option::TYPE_INT, Option::LEVEL_DEV)
5288 .set_default(2)
5289 .set_description(""),
5290
5291 Option("filestore_split_rand_factor", Option::TYPE_UINT, Option::LEVEL_DEV)
5292 .set_default(20)
5293 .set_description(""),
5294
5295 Option("filestore_update_to", Option::TYPE_INT, Option::LEVEL_DEV)
5296 .set_default(1000)
5297 .set_description(""),
5298
5299 Option("filestore_blackhole", Option::TYPE_BOOL, Option::LEVEL_DEV)
5300 .set_default(false)
5301 .set_description(""),
5302
5303 Option("filestore_fd_cache_size", Option::TYPE_INT, Option::LEVEL_DEV)
5304 .set_default(128)
5305 .set_description(""),
5306
5307 Option("filestore_fd_cache_shards", Option::TYPE_INT, Option::LEVEL_DEV)
5308 .set_default(16)
5309 .set_description(""),
5310
5311 Option("filestore_ondisk_finisher_threads", Option::TYPE_INT, Option::LEVEL_DEV)
5312 .set_default(1)
5313 .set_description(""),
5314
5315 Option("filestore_apply_finisher_threads", Option::TYPE_INT, Option::LEVEL_DEV)
5316 .set_default(1)
5317 .set_description(""),
5318
5319 Option("filestore_dump_file", Option::TYPE_STR, Option::LEVEL_DEV)
5320 .set_default("")
5321 .set_description(""),
5322
5323 Option("filestore_kill_at", Option::TYPE_INT, Option::LEVEL_DEV)
5324 .set_default(0)
5325 .set_description(""),
5326
5327 Option("filestore_inject_stall", Option::TYPE_INT, Option::LEVEL_DEV)
5328 .set_default(0)
5329 .set_description(""),
5330
5331 Option("filestore_fail_eio", Option::TYPE_BOOL, Option::LEVEL_DEV)
5332 .set_default(true)
5333 .set_description(""),
5334
5335 Option("filestore_debug_verify_split", Option::TYPE_BOOL, Option::LEVEL_DEV)
5336 .set_default(false)
5337 .set_description(""),
5338
5339 Option("journal_dio", Option::TYPE_BOOL, Option::LEVEL_DEV)
5340 .set_default(true)
5341 .set_description(""),
5342
5343 Option("journal_aio", Option::TYPE_BOOL, Option::LEVEL_DEV)
5344 .set_default(true)
5345 .set_description(""),
5346
5347 Option("journal_force_aio", Option::TYPE_BOOL, Option::LEVEL_DEV)
5348 .set_default(false)
5349 .set_description(""),
5350
5351 Option("journal_block_size", Option::TYPE_SIZE, Option::LEVEL_DEV)
5352 .set_default(4_K)
5353 .set_description(""),
5354
5355 Option("journal_block_align", Option::TYPE_BOOL, Option::LEVEL_DEV)
5356 .set_default(true)
5357 .set_description(""),
5358
5359 Option("journal_write_header_frequency", Option::TYPE_UINT, Option::LEVEL_DEV)
5360 .set_default(0)
5361 .set_description(""),
5362
5363 Option("journal_max_write_bytes", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
5364 .set_default(10_M)
5365 .set_description("Max bytes in flight to journal"),
5366
5367 Option("journal_max_write_entries", Option::TYPE_INT, Option::LEVEL_ADVANCED)
5368 .set_default(100)
5369 .set_description("Max IOs in flight to journal"),
5370
5371 Option("journal_throttle_low_threshhold", Option::TYPE_FLOAT, Option::LEVEL_DEV)
5372 .set_default(0.6)
5373 .set_description(""),
5374
5375 Option("journal_throttle_high_threshhold", Option::TYPE_FLOAT, Option::LEVEL_DEV)
5376 .set_default(0.9)
5377 .set_description(""),
5378
5379 Option("journal_throttle_high_multiple", Option::TYPE_FLOAT, Option::LEVEL_DEV)
5380 .set_default(0)
5381 .set_description(""),
5382
5383 Option("journal_throttle_max_multiple", Option::TYPE_FLOAT, Option::LEVEL_DEV)
5384 .set_default(0)
5385 .set_description(""),
5386
5387 Option("journal_align_min_size", Option::TYPE_SIZE, Option::LEVEL_DEV)
5388 .set_default(64_K)
5389 .set_description(""),
5390
5391 Option("journal_replay_from", Option::TYPE_INT, Option::LEVEL_DEV)
5392 .set_default(0)
5393 .set_description(""),
5394
5395 Option("mgr_stats_threshold", Option::TYPE_INT, Option::LEVEL_ADVANCED)
5396 .set_default((int64_t)PerfCountersBuilder::PRIO_USEFUL)
5397 .set_description("Lowest perfcounter priority collected by mgr")
5398 .set_long_description("Daemons only set perf counter data to the manager "
5399 "daemon if the counter has a priority higher than this.")
5400 .set_min_max((int64_t)PerfCountersBuilder::PRIO_DEBUGONLY,
5401 (int64_t)PerfCountersBuilder::PRIO_CRITICAL + 1),
5402
5403 Option("journal_zero_on_create", Option::TYPE_BOOL, Option::LEVEL_DEV)
5404 .set_default(false)
5405 .set_description(""),
5406
5407 Option("journal_ignore_corruption", Option::TYPE_BOOL, Option::LEVEL_DEV)
5408 .set_default(false)
5409 .set_description(""),
5410
5411 Option("journal_discard", Option::TYPE_BOOL, Option::LEVEL_DEV)
5412 .set_default(false)
5413 .set_description(""),
5414
5415 Option("fio_dir", Option::TYPE_STR, Option::LEVEL_ADVANCED)
5416 .set_default("/tmp/fio")
5417 .set_description(""),
5418
5419 Option("rados_mon_op_timeout", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
5420 .set_default(0)
5421 .set_description(""),
5422
5423 Option("rados_osd_op_timeout", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
5424 .set_default(0)
5425 .set_description(""),
5426
5427 Option("rados_tracing", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
5428 .set_default(false)
5429 .set_description(""),
5430
5431 Option("ceph_daemon_path", Option::TYPE_STR, Option::LEVEL_ADVANCED)
5432 .set_default("/usr/sbin/ceph-daemon")
5433 .add_service("mgr")
5434 .set_description("Path to ceph-daemon utility"),
5435
5436 Option("mgr_module_path", Option::TYPE_STR, Option::LEVEL_ADVANCED)
5437 .set_default(CEPH_DATADIR "/mgr")
5438 .add_service("mgr")
5439 .set_description("Filesystem path to manager modules."),
5440
5441 Option("mgr_initial_modules", Option::TYPE_STR, Option::LEVEL_BASIC)
5442 .set_default("restful iostat")
5443 .set_flag(Option::FLAG_NO_MON_UPDATE)
5444 .set_flag(Option::FLAG_CLUSTER_CREATE)
5445 .add_service("mon")
5446 .set_description("List of manager modules to enable when the cluster is "
5447 "first started")
5448 .set_long_description("This list of module names is read by the monitor "
5449 "when the cluster is first started after installation, to populate "
5450 "the list of enabled manager modules. Subsequent updates are done using "
5451 "the 'mgr module [enable|disable]' commands. List may be comma "
5452 "or space separated."),
5453
5454 Option("mgr_data", Option::TYPE_STR, Option::LEVEL_ADVANCED)
5455 .set_default("/var/lib/ceph/mgr/$cluster-$id")
5456 .set_flag(Option::FLAG_NO_MON_UPDATE)
5457 .add_service("mgr")
5458 .set_description("Filesystem path to the ceph-mgr data directory, used to "
5459 "contain keyring."),
5460
5461 Option("mgr_tick_period", Option::TYPE_SECS, Option::LEVEL_ADVANCED)
5462 .set_default(2)
5463 .add_service("mgr")
5464 .set_description("Period in seconds of beacon messages to monitor"),
5465
5466 Option("mgr_stats_period", Option::TYPE_INT, Option::LEVEL_BASIC)
5467 .set_default(5)
5468 .add_service("mgr")
5469 .set_description("Period in seconds of OSD/MDS stats reports to manager")
5470 .set_long_description("Use this setting to control the granularity of "
5471 "time series data collection from daemons. Adjust "
5472 "upwards if the manager CPU load is too high, or "
5473 "if you simply do not require the most up to date "
5474 "performance counter data."),
5475
5476 Option("mgr_client_bytes", Option::TYPE_SIZE, Option::LEVEL_DEV)
5477 .set_default(128_M)
5478 .add_service("mgr"),
5479
5480 Option("mgr_client_messages", Option::TYPE_UINT, Option::LEVEL_DEV)
5481 .set_default(512)
5482 .add_service("mgr"),
5483
5484 Option("mgr_osd_bytes", Option::TYPE_SIZE, Option::LEVEL_DEV)
5485 .set_default(512_M)
5486 .add_service("mgr"),
5487
5488 Option("mgr_osd_messages", Option::TYPE_UINT, Option::LEVEL_DEV)
5489 .set_default(8192)
5490 .add_service("mgr"),
5491
5492 Option("mgr_mds_bytes", Option::TYPE_SIZE, Option::LEVEL_DEV)
5493 .set_default(128_M)
5494 .add_service("mgr"),
5495
5496 Option("mgr_mds_messages", Option::TYPE_UINT, Option::LEVEL_DEV)
5497 .set_default(128)
5498 .add_service("mgr"),
5499
5500 Option("mgr_mon_bytes", Option::TYPE_SIZE, Option::LEVEL_DEV)
5501 .set_default(128_M)
5502 .add_service("mgr"),
5503
5504 Option("mgr_mon_messages", Option::TYPE_UINT, Option::LEVEL_DEV)
5505 .set_default(128)
5506 .add_service("mgr"),
5507
5508 Option("mgr_connect_retry_interval", Option::TYPE_FLOAT, Option::LEVEL_DEV)
5509 .set_default(1.0)
5510 .add_service("common"),
5511
5512 Option("mgr_service_beacon_grace", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
5513 .set_default(60.0)
5514 .add_service("mgr")
5515 .set_description("Period in seconds from last beacon to manager dropping "
5516 "state about a monitored service (RGW, rbd-mirror etc)"),
5517
5518 Option("mgr_client_service_daemon_unregister_timeout", Option::TYPE_FLOAT, Option::LEVEL_DEV)
5519 .set_default(1.0)
5520 .set_description("Time to wait during shutdown to deregister service with mgr"),
5521
5522 Option("mgr_debug_aggressive_pg_num_changes", Option::TYPE_BOOL, Option::LEVEL_DEV)
5523 .set_default(false)
5524 .set_description("Bypass most throttling and safety checks in pg[p]_num controller")
5525 .add_service("mgr"),
5526
5527 Option("mon_mgr_digest_period", Option::TYPE_INT, Option::LEVEL_DEV)
5528 .set_default(5)
5529 .add_service("mon")
5530 .set_description("Period in seconds between monitor-to-manager "
5531 "health/status updates"),
5532
5533 Option("mon_mgr_beacon_grace", Option::TYPE_SECS, Option::LEVEL_ADVANCED)
5534 .set_default(30)
5535 .add_service("mon")
5536 .set_description("Period in seconds from last beacon to monitor marking "
5537 "a manager daemon as failed"),
5538
5539 Option("mon_mgr_inactive_grace", Option::TYPE_INT, Option::LEVEL_ADVANCED)
5540 .set_default(60)
5541 .add_service("mon")
5542 .set_description("Period in seconds after cluster creation during which "
5543 "cluster may have no active manager")
5544 .set_long_description("This grace period enables the cluster to come "
5545 "up cleanly without raising spurious health check "
5546 "failures about managers that aren't online yet"),
5547
5548 Option("mon_mgr_mkfs_grace", Option::TYPE_INT, Option::LEVEL_ADVANCED)
5549 .set_default(120)
5550 .add_service("mon")
5551 .set_description("Period in seconds that the cluster may have no active "
5552 "manager before this is reported as an ERR rather than "
5553 "a WARN"),
5554
5555 Option("throttler_perf_counter", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
5556 .set_default(true)
5557 .set_description(""),
5558
5559 Option("event_tracing", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
5560 .set_default(false)
5561 .set_description(""),
5562
5563 Option("bluestore_tracing", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
5564 .set_default(false)
5565 .set_description("Enable bluestore event tracing."),
5566
5567 Option("bluestore_throttle_trace_rate", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
5568 .set_default(0)
5569 .set_description("Rate at which to sample bluestore transactions (per second)"),
5570
5571 Option("debug_deliberately_leak_memory", Option::TYPE_BOOL, Option::LEVEL_DEV)
5572 .set_default(false)
5573 .set_description(""),
5574
5575 Option("debug_asserts_on_shutdown", Option::TYPE_BOOL,Option::LEVEL_DEV)
5576 .set_default(false)
5577 .set_description("Enable certain asserts to check for refcounting bugs on shutdown; see http://tracker.ceph.com/issues/21738"),
5578
5579 Option("debug_asok_assert_abort", Option::TYPE_BOOL, Option::LEVEL_DEV)
5580 .set_default(false)
5581 .set_description("allow commands 'assert' and 'abort' via asok for testing crash dumps etc"),
5582
5583 Option("target_max_misplaced_ratio", Option::TYPE_FLOAT, Option::LEVEL_BASIC)
5584 .set_default(.05)
5585 .set_description("Max ratio of misplaced objects to target when throttling data rebalancing activity"),
5586
5587 Option("device_failure_prediction_mode", Option::TYPE_STR, Option::LEVEL_BASIC)
5588 .set_default("none")
5589 .set_flag(Option::FLAG_RUNTIME)
5590 .set_enum_allowed({"none", "local", "cloud"})
5591 .set_description("Method used to predict device failures")
5592 .set_long_description("To disable prediction, use 'none', 'local' uses a prediction model that runs inside the mgr daemon. 'cloud' will share metrics with a cloud service and query the service for devicelife expectancy."),
5593
5594 /* KRB Authentication. */
5595 Option("gss_ktab_client_file", Option::TYPE_STR, Option::LEVEL_ADVANCED)
5596 .set_default("/var/lib/ceph/$name/gss_client_$name.ktab")
5597 .set_description("GSS/KRB5 Keytab file for client authentication")
5598 .add_service({"mon", "osd"})
5599 .set_long_description("This sets the full path for the GSS/Kerberos client keytab file location."),
5600
5601 Option("gss_target_name", Option::TYPE_STR, Option::LEVEL_ADVANCED)
5602 .set_default("ceph")
5603 .set_description("")
5604 .add_service({"mon", "osd"})
5605 .set_long_description("This sets the gss target service name."),
5606
5607 Option("debug_disable_randomized_ping", Option::TYPE_BOOL, Option::LEVEL_DEV)
5608 .set_default(false)
5609 .set_description("Disable heartbeat ping randomization for testing purposes"),
5610
5611 Option("debug_heartbeat_testing_span", Option::TYPE_INT, Option::LEVEL_DEV)
5612 .set_default(0)
5613 .set_description("Override 60 second periods for testing only"),
5614 });
5615 }
5616
5617 std::vector<Option> get_rgw_options() {
5618 return std::vector<Option>({
5619 Option("rgw_acl_grants_max_num", Option::TYPE_INT, Option::LEVEL_ADVANCED)
5620 .set_default(100)
5621 .set_description("Max number of ACL grants in a single request"),
5622
5623 Option("rgw_cors_rules_max_num", Option::TYPE_INT, Option::LEVEL_ADVANCED)
5624 .set_default(100)
5625 .set_description("Max number of cors rules in a single request"),
5626
5627 Option("rgw_delete_multi_obj_max_num", Option::TYPE_INT, Option::LEVEL_ADVANCED)
5628 .set_default(1000)
5629 .set_description("Max number of objects in a single multi-object delete request"),
5630
5631 Option("rgw_website_routing_rules_max_num", Option::TYPE_INT, Option::LEVEL_ADVANCED)
5632 .set_default(50)
5633 .set_description("Max number of website routing rules in a single request"),
5634
5635 Option("rgw_rados_tracing", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
5636 .set_default(false)
5637 .set_description("true if LTTng-UST tracepoints should be enabled"),
5638
5639 Option("rgw_op_tracing", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
5640 .set_default(false)
5641 .set_description("true if LTTng-UST tracepoints should be enabled"),
5642
5643 Option("rgw_max_chunk_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
5644 .set_default(4_M)
5645 .set_description("Set RGW max chunk size")
5646 .set_long_description(
5647 "The chunk size is the size of RADOS I/O requests that RGW sends when accessing "
5648 "data objects. RGW read and write operation will never request more than this amount "
5649 "in a single request. This also defines the rgw object head size, as head operations "
5650 "need to be atomic, and anything larger than this would require more than a single "
5651 "operation."),
5652
5653 Option("rgw_put_obj_min_window_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
5654 .set_default(16_M)
5655 .set_description("The minimum RADOS write window size (in bytes).")
5656 .set_long_description(
5657 "The window size determines the total concurrent RADOS writes of a single rgw object. "
5658 "When writing an object RGW will send multiple chunks to RADOS. The total size of the "
5659 "writes does not exceed the window size. The window size can be automatically "
5660 "in order to better utilize the pipe.")
5661 .add_see_also({"rgw_put_obj_max_window_size", "rgw_max_chunk_size"}),
5662
5663 Option("rgw_put_obj_max_window_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
5664 .set_default(64_M)
5665 .set_description("The maximum RADOS write window size (in bytes).")
5666 .set_long_description("The window size may be dynamically adjusted, but will not surpass this value.")
5667 .add_see_also({"rgw_put_obj_min_window_size", "rgw_max_chunk_size"}),
5668
5669 Option("rgw_max_put_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
5670 .set_default(5_G)
5671 .set_description("Max size (in bytes) of regular (non multi-part) object upload.")
5672 .set_long_description(
5673 "Plain object upload is capped at this amount of data. In order to upload larger "
5674 "objects, a special upload mechanism is required. The S3 API provides the "
5675 "multi-part upload, and Swift provides DLO and SLO."),
5676
5677 Option("rgw_max_put_param_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
5678 .set_default(1_M)
5679 .set_description("The maximum size (in bytes) of data input of certain RESTful requests."),
5680
5681 Option("rgw_max_attr_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
5682 .set_default(0)
5683 .set_description("The maximum length of metadata value. 0 skips the check"),
5684
5685 Option("rgw_max_attr_name_len", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
5686 .set_default(0)
5687 .set_description("The maximum length of metadata name. 0 skips the check"),
5688
5689 Option("rgw_max_attrs_num_in_req", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
5690 .set_default(0)
5691 .set_description("The maximum number of metadata items that can be put via single request"),
5692
5693 Option("rgw_override_bucket_index_max_shards", Option::TYPE_UINT, Option::LEVEL_DEV)
5694 .set_default(0)
5695 .set_description(""),
5696
5697 Option("rgw_bucket_index_max_aio", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
5698 .set_default(128)
5699 .set_description("Max number of concurrent RADOS requests when handling bucket shards."),
5700
5701 Option("rgw_enable_quota_threads", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
5702 .set_default(true)
5703 .set_description("Enables the quota maintenance thread.")
5704 .set_long_description(
5705 "The quota maintenance thread is responsible for quota related maintenance work. "
5706 "The thread itself can be disabled, but in order for quota to work correctly, at "
5707 "least one RGW in each zone needs to have this thread running. Having the thread "
5708 "enabled on multiple RGW processes within the same zone can spread "
5709 "some of the maintenance work between them.")
5710 .add_see_also({"rgw_enable_gc_threads", "rgw_enable_lc_threads"}),
5711
5712 Option("rgw_enable_gc_threads", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
5713 .set_default(true)
5714 .set_description("Enables the garbage collection maintenance thread.")
5715 .set_long_description(
5716 "The garbage collection maintenance thread is responsible for garbage collector "
5717 "maintenance work. The thread itself can be disabled, but in order for garbage "
5718 "collection to work correctly, at least one RGW in each zone needs to have this "
5719 "thread running. Having the thread enabled on multiple RGW processes within the "
5720 "same zone can spread some of the maintenance work between them.")
5721 .add_see_also({"rgw_enable_quota_threads", "rgw_enable_lc_threads"}),
5722
5723 Option("rgw_enable_lc_threads", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
5724 .set_default(true)
5725 .set_description("Enables the lifecycle maintenance thread. This is required on at least one rgw for each zone.")
5726 .set_long_description(
5727 "The lifecycle maintenance thread is responsible for lifecycle related maintenance "
5728 "work. The thread itself can be disabled, but in order for lifecycle to work "
5729 "correctly, at least one RGW in each zone needs to have this thread running. Having"
5730 "the thread enabled on multiple RGW processes within the same zone can spread "
5731 "some of the maintenance work between them.")
5732 .add_see_also({"rgw_enable_gc_threads", "rgw_enable_quota_threads"}),
5733
5734 Option("rgw_data", Option::TYPE_STR, Option::LEVEL_ADVANCED)
5735 .set_default("/var/lib/ceph/radosgw/$cluster-$id")
5736 .set_flag(Option::FLAG_NO_MON_UPDATE)
5737 .set_description("Alternative location for RGW configuration.")
5738 .set_long_description(
5739 "If this is set, the different Ceph system configurables (such as the keyring file "
5740 "will be located in the path that is specified here. "),
5741
5742 Option("rgw_enable_apis", Option::TYPE_STR, Option::LEVEL_ADVANCED)
5743 .set_default("s3, s3website, swift, swift_auth, admin, sts, iam, pubsub")
5744 .set_description("A list of set of RESTful APIs that rgw handles."),
5745
5746 Option("rgw_cache_enabled", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
5747 .set_default(true)
5748 .set_description("Enable RGW metadata cache.")
5749 .set_long_description(
5750 "The metadata cache holds metadata entries that RGW requires for processing "
5751 "requests. Metadata entries can be user info, bucket info, and bucket instance "
5752 "info. If not found in the cache, entries will be fetched from the backing "
5753 "RADOS store.")
5754 .add_see_also("rgw_cache_lru_size"),
5755
5756 Option("rgw_cache_lru_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
5757 .set_default(10000)
5758 .set_description("Max number of items in RGW metadata cache.")
5759 .set_long_description(
5760 "When full, the RGW metadata cache evicts least recently used entries.")
5761 .add_see_also("rgw_cache_enabled"),
5762
5763 Option("rgw_socket_path", Option::TYPE_STR, Option::LEVEL_ADVANCED)
5764 .set_default("")
5765 .set_description("RGW FastCGI socket path (for FastCGI over Unix domain sockets).")
5766 .add_see_also("rgw_fcgi_socket_backlog"),
5767
5768 Option("rgw_host", Option::TYPE_STR, Option::LEVEL_ADVANCED)
5769 .set_default("")
5770 .set_description("RGW FastCGI host name (for FastCGI over TCP)")
5771 .add_see_also({"rgw_port", "rgw_fcgi_socket_backlog"}),
5772
5773 Option("rgw_port", Option::TYPE_STR, Option::LEVEL_BASIC)
5774 .set_default("")
5775 .set_description("RGW FastCGI port number (for FastCGI over TCP)")
5776 .add_see_also({"rgw_host", "rgw_fcgi_socket_backlog"}),
5777
5778 Option("rgw_dns_name", Option::TYPE_STR, Option::LEVEL_ADVANCED)
5779 .set_default("")
5780 .set_description("The host name that RGW uses.")
5781 .set_long_description(
5782 "This is Needed for virtual hosting of buckets to work properly, unless configured "
5783 "via zonegroup configuration."),
5784
5785 Option("rgw_dns_s3website_name", Option::TYPE_STR, Option::LEVEL_ADVANCED)
5786 .set_default("")
5787 .set_description("The host name that RGW uses for static websites (S3)")
5788 .set_long_description(
5789 "This is needed for virtual hosting of buckets, unless configured via zonegroup "
5790 "configuration."),
5791
5792 Option("rgw_service_provider_name", Option::TYPE_STR, Option::LEVEL_ADVANCED)
5793 .set_default("")
5794 .set_description("Service provider name which is contained in http response headers")
5795 .set_long_description(
5796 "As S3 or other cloud storage providers do, http response headers should contain the name of the provider. "
5797 "This name will be placed in http header 'Server'."),
5798
5799 Option("rgw_content_length_compat", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
5800 .set_default(false)
5801 .set_description("Multiple content length headers compatibility")
5802 .set_long_description(
5803 "Try to handle requests with abiguous multiple content length headers "
5804 "(Content-Length, Http-Content-Length)."),
5805
5806 Option("rgw_relaxed_region_enforcement", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
5807 .set_default(false)
5808 .set_description("Disable region constraint enforcement")
5809 .set_long_description(
5810 "Enable requests such as bucket creation to succeed irrespective of region restrictions (Jewel compat)."),
5811
5812 Option("rgw_lifecycle_work_time", Option::TYPE_STR, Option::LEVEL_ADVANCED)
5813 .set_default("00:00-06:00")
5814 .set_description("Lifecycle allowed work time")
5815 .set_long_description("Local time window in which the lifecycle maintenance thread can work."),
5816
5817 Option("rgw_lc_lock_max_time", Option::TYPE_INT, Option::LEVEL_DEV)
5818 .set_default(60)
5819 .set_description(""),
5820
5821 Option("rgw_lc_thread_delay", Option::TYPE_INT, Option::LEVEL_ADVANCED)
5822 .set_default(0)
5823 .set_description("Delay after processing of bucket listing chunks (i.e., per 1000 entries) in milliseconds"),
5824
5825 Option("rgw_lc_max_objs", Option::TYPE_INT, Option::LEVEL_ADVANCED)
5826 .set_default(32)
5827 .set_description("Number of lifecycle data shards")
5828 .set_long_description(
5829 "Number of RADOS objects to use for storing lifecycle index. This can affect "
5830 "concurrency of lifecycle maintenance, but requires multiple RGW processes "
5831 "running on the zone to be utilized."),
5832
5833 Option("rgw_lc_max_rules", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
5834 .set_default(1000)
5835 .set_description("Max number of lifecycle rules set on one bucket")
5836 .set_long_description("Number of lifecycle rules set on one bucket should be limited."),
5837
5838 Option("rgw_lc_debug_interval", Option::TYPE_INT, Option::LEVEL_DEV)
5839 .set_default(-1)
5840 .set_description(""),
5841
5842 Option("rgw_mp_lock_max_time", Option::TYPE_INT, Option::LEVEL_ADVANCED)
5843 .set_default(600)
5844 .set_description("Multipart upload max completion time")
5845 .set_long_description(
5846 "Time length to allow completion of a multipart upload operation. This is done "
5847 "to prevent concurrent completions on the same object with the same upload id."),
5848
5849 Option("rgw_script_uri", Option::TYPE_STR, Option::LEVEL_DEV)
5850 .set_default("")
5851 .set_description(""),
5852
5853 Option("rgw_request_uri", Option::TYPE_STR, Option::LEVEL_DEV)
5854 .set_default("")
5855 .set_description(""),
5856
5857 Option("rgw_ignore_get_invalid_range", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
5858 .set_default(false)
5859 .set_description("Treat invalid (e.g., negative) range request as full")
5860 .set_long_description("Treat invalid (e.g., negative) range request "
5861 "as request for the full object (AWS compatibility)"),
5862
5863 Option("rgw_swift_url", Option::TYPE_STR, Option::LEVEL_ADVANCED)
5864 .set_default("")
5865 .set_description("Swift-auth storage URL")
5866 .set_long_description(
5867 "Used in conjunction with rgw internal swift authentication. This affects the "
5868 "X-Storage-Url response header value.")
5869 .add_see_also("rgw_swift_auth_entry"),
5870
5871 Option("rgw_swift_url_prefix", Option::TYPE_STR, Option::LEVEL_ADVANCED)
5872 .set_default("swift")
5873 .set_description("Swift URL prefix")
5874 .set_long_description("The URL path prefix for swift requests."),
5875
5876 Option("rgw_swift_auth_url", Option::TYPE_STR, Option::LEVEL_ADVANCED)
5877 .set_default("")
5878 .set_description("Swift auth URL")
5879 .set_long_description(
5880 "Default url to which RGW connects and verifies tokens for v1 auth (if not using "
5881 "internal swift auth)."),
5882
5883 Option("rgw_swift_auth_entry", Option::TYPE_STR, Option::LEVEL_ADVANCED)
5884 .set_default("auth")
5885 .set_description("Swift auth URL prefix")
5886 .set_long_description("URL path prefix for internal swift auth requests.")
5887 .add_see_also("rgw_swift_url"),
5888
5889 Option("rgw_swift_tenant_name", Option::TYPE_STR, Option::LEVEL_ADVANCED)
5890 .set_default("")
5891 .set_description("Swift tenant name")
5892 .set_long_description("Tenant name that is used when constructing the swift path.")
5893 .add_see_also("rgw_swift_account_in_url"),
5894
5895 Option("rgw_swift_account_in_url", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
5896 .set_default(false)
5897 .set_description("Swift account encoded in URL")
5898 .set_long_description("Whether the swift account is encoded in the uri path (AUTH_<account>).")
5899 .add_see_also("rgw_swift_tenant_name"),
5900
5901 Option("rgw_swift_enforce_content_length", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
5902 .set_default(false)
5903 .set_description("Send content length when listing containers (Swift)")
5904 .set_long_description(
5905 "Whether content length header is needed when listing containers. When this is "
5906 "set to false, RGW will send extra info for each entry in the response."),
5907
5908 Option("rgw_keystone_url", Option::TYPE_STR, Option::LEVEL_BASIC)
5909 .set_default("")
5910 .set_description("The URL to the Keystone server."),
5911
5912 Option("rgw_keystone_admin_token", Option::TYPE_STR, Option::LEVEL_ADVANCED)
5913 .set_default("")
5914 .set_description("DEPRECATED: The admin token (shared secret) that is used for the Keystone requests."),
5915
5916 Option("rgw_keystone_admin_token_path", Option::TYPE_STR, Option::LEVEL_ADVANCED)
5917 .set_default("")
5918 .set_description("Path to a file containing the admin token (shared secret) that is used for the Keystone requests."),
5919
5920 Option("rgw_keystone_admin_user", Option::TYPE_STR, Option::LEVEL_ADVANCED)
5921 .set_default("")
5922 .set_description("Keystone admin user."),
5923
5924 Option("rgw_keystone_admin_password", Option::TYPE_STR, Option::LEVEL_ADVANCED)
5925 .set_default("")
5926 .set_description("DEPRECATED: Keystone admin password."),
5927
5928 Option("rgw_keystone_admin_password_path", Option::TYPE_STR, Option::LEVEL_ADVANCED)
5929 .set_default("")
5930 .set_description("Path to a file containing the Keystone admin password."),
5931
5932 Option("rgw_keystone_admin_tenant", Option::TYPE_STR, Option::LEVEL_ADVANCED)
5933 .set_default("")
5934 .set_description("Keystone admin user tenant."),
5935
5936 Option("rgw_keystone_admin_project", Option::TYPE_STR, Option::LEVEL_ADVANCED)
5937 .set_default("")
5938 .set_description("Keystone admin user project (for Keystone v3)."),
5939
5940 Option("rgw_keystone_admin_domain", Option::TYPE_STR, Option::LEVEL_ADVANCED)
5941 .set_default("")
5942 .set_description("Keystone admin user domain (for Keystone v3)."),
5943
5944 Option("rgw_keystone_barbican_user", Option::TYPE_STR, Option::LEVEL_ADVANCED)
5945 .set_default("")
5946 .set_description("Keystone user to access barbican secrets."),
5947
5948 Option("rgw_keystone_barbican_password", Option::TYPE_STR, Option::LEVEL_ADVANCED)
5949 .set_default("")
5950 .set_description("Keystone password for barbican user."),
5951
5952 Option("rgw_keystone_barbican_tenant", Option::TYPE_STR, Option::LEVEL_ADVANCED)
5953 .set_default("")
5954 .set_description("Keystone barbican user tenant (Keystone v2.0)."),
5955
5956 Option("rgw_keystone_barbican_project", Option::TYPE_STR, Option::LEVEL_ADVANCED)
5957 .set_default("")
5958 .set_description("Keystone barbican user project (Keystone v3)."),
5959
5960 Option("rgw_keystone_barbican_domain", Option::TYPE_STR, Option::LEVEL_ADVANCED)
5961 .set_default("")
5962 .set_description("Keystone barbican user domain."),
5963
5964 Option("rgw_keystone_api_version", Option::TYPE_INT, Option::LEVEL_ADVANCED)
5965 .set_default(2)
5966 .set_description("Version of Keystone API to use (2 or 3)."),
5967
5968 Option("rgw_keystone_accepted_roles", Option::TYPE_STR, Option::LEVEL_ADVANCED)
5969 .set_default("Member, admin")
5970 .set_description("Only users with one of these roles will be served when doing Keystone authentication."),
5971
5972 Option("rgw_keystone_accepted_admin_roles", Option::TYPE_STR, Option::LEVEL_ADVANCED)
5973 .set_default("")
5974 .set_description("List of roles allowing user to gain admin privileges (Keystone)."),
5975
5976 Option("rgw_keystone_token_cache_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
5977 .set_default(10000)
5978 .set_description("Keystone token cache size")
5979 .set_long_description(
5980 "Max number of Keystone tokens that will be cached. Token that is not cached "
5981 "requires RGW to access the Keystone server when authenticating."),
5982
5983 Option("rgw_keystone_verify_ssl", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
5984 .set_default(true)
5985 .set_description("Should RGW verify the Keystone server SSL certificate."),
5986
5987 Option("rgw_keystone_implicit_tenants", Option::TYPE_STR, Option::LEVEL_ADVANCED)
5988 .set_default("false")
5989 .set_enum_allowed( { "false", "true", "swift", "s3", "both", "0", "1", "none" } )
5990 .set_description("RGW Keystone implicit tenants creation")
5991 .set_long_description(
5992 "Implicitly create new users in their own tenant with the same name when "
5993 "authenticating via Keystone. Can be limited to s3 or swift only."),
5994
5995 Option("rgw_cross_domain_policy", Option::TYPE_STR, Option::LEVEL_ADVANCED)
5996 .set_default("<allow-access-from domain=\"*\" secure=\"false\" />")
5997 .set_description("RGW handle cross domain policy")
5998 .set_long_description("Returned cross domain policy when accessing the crossdomain.xml "
5999 "resource (Swift compatiility)."),
6000
6001 Option("rgw_healthcheck_disabling_path", Option::TYPE_STR, Option::LEVEL_DEV)
6002 .set_default("")
6003 .set_description("Swift health check api can be disabled if a file can be accessed in this path."),
6004
6005 Option("rgw_s3_auth_use_rados", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
6006 .set_default(true)
6007 .set_description("Should S3 authentication use credentials stored in RADOS backend."),
6008
6009 Option("rgw_s3_auth_use_keystone", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
6010 .set_default(false)
6011 .set_description("Should S3 authentication use Keystone."),
6012
6013 Option("rgw_s3_auth_order", Option::TYPE_STR, Option::LEVEL_ADVANCED)
6014 .set_default("sts, external, local")
6015 .set_description("Authentication strategy order to use for s3 authentication")
6016 .set_long_description(
6017 "Order of authentication strategies to try for s3 authentication, the allowed "
6018 "options are a comma separated list of engines external, local. The "
6019 "default order is to try all the externally configured engines before "
6020 "attempting local rados based authentication"),
6021
6022 Option("rgw_barbican_url", Option::TYPE_STR, Option::LEVEL_ADVANCED)
6023 .set_default("")
6024 .set_description("URL to barbican server."),
6025
6026 Option("rgw_ldap_uri", Option::TYPE_STR, Option::LEVEL_ADVANCED)
6027 .set_default("ldaps://<ldap.your.domain>")
6028 .set_description("Space-separated list of LDAP servers in URI format."),
6029
6030 Option("rgw_ldap_binddn", Option::TYPE_STR, Option::LEVEL_ADVANCED)
6031 .set_default("uid=admin,cn=users,dc=example,dc=com")
6032 .set_description("LDAP entry RGW will bind with (user match)."),
6033
6034 Option("rgw_ldap_searchdn", Option::TYPE_STR, Option::LEVEL_ADVANCED)
6035 .set_default("cn=users,cn=accounts,dc=example,dc=com")
6036 .set_description("LDAP search base (basedn)."),
6037
6038 Option("rgw_ldap_dnattr", Option::TYPE_STR, Option::LEVEL_ADVANCED)
6039 .set_default("uid")
6040 .set_description("LDAP attribute containing RGW user names (to form binddns)."),
6041
6042 Option("rgw_ldap_secret", Option::TYPE_STR, Option::LEVEL_ADVANCED)
6043 .set_default("/etc/openldap/secret")
6044 .set_description("Path to file containing credentials for rgw_ldap_binddn."),
6045
6046 Option("rgw_s3_auth_use_ldap", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
6047 .set_default(false)
6048 .set_description("Should S3 authentication use LDAP."),
6049
6050 Option("rgw_ldap_searchfilter", Option::TYPE_STR, Option::LEVEL_ADVANCED)
6051 .set_default("")
6052 .set_description("LDAP search filter."),
6053
6054 Option("rgw_opa_url", Option::TYPE_STR, Option::LEVEL_ADVANCED)
6055 .set_default("")
6056 .set_description("URL to OPA server."),
6057
6058 Option("rgw_opa_token", Option::TYPE_STR, Option::LEVEL_ADVANCED)
6059 .set_default("")
6060 .set_description("The Bearer token OPA uses to authenticate client requests."),
6061
6062 Option("rgw_opa_verify_ssl", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
6063 .set_default(true)
6064 .set_description("Should RGW verify the OPA server SSL certificate."),
6065
6066 Option("rgw_use_opa_authz", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
6067 .set_default(false)
6068 .set_description("Should OPA be used to authorize client requests."),
6069
6070 Option("rgw_admin_entry", Option::TYPE_STR, Option::LEVEL_ADVANCED)
6071 .set_default("admin")
6072 .set_description("Path prefix to be used for accessing RGW RESTful admin API."),
6073
6074 Option("rgw_enforce_swift_acls", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
6075 .set_default(true)
6076 .set_description("RGW enforce swift acls")
6077 .set_long_description(
6078 "Should RGW enforce special Swift-only ACLs. Swift has a special ACL that gives "
6079 "permission to access all objects in a container."),
6080
6081 Option("rgw_swift_token_expiration", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6082 .set_default(1_day)
6083 .set_description("Expiration time (in seconds) for token generated through RGW Swift auth."),
6084
6085 Option("rgw_print_continue", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
6086 .set_default(true)
6087 .set_description("RGW support of 100-continue")
6088 .set_long_description(
6089 "Should RGW explicitly send 100 (continue) responses. This is mainly relevant when "
6090 "using FastCGI, as some FastCGI modules do not fully support this feature."),
6091
6092 Option("rgw_print_prohibited_content_length", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
6093 .set_default(false)
6094 .set_description("RGW RFC-7230 compatibility")
6095 .set_long_description(
6096 "Specifies whether RGW violates RFC 7230 and sends Content-Length with 204 or 304 "
6097 "statuses."),
6098
6099 Option("rgw_remote_addr_param", Option::TYPE_STR, Option::LEVEL_ADVANCED)
6100 .set_default("REMOTE_ADDR")
6101 .set_description("HTTP header that holds the remote address in incoming requests.")
6102 .set_long_description(
6103 "RGW will use this header to extract requests origin. When RGW runs behind "
6104 "a reverse proxy, the remote address header will point at the proxy's address "
6105 "and not at the originator's address. Therefore it is sometimes possible to "
6106 "have the proxy add the originator's address in a separate HTTP header, which "
6107 "will allow RGW to log it correctly."
6108 )
6109 .add_see_also("rgw_enable_ops_log"),
6110
6111 Option("rgw_op_thread_timeout", Option::TYPE_INT, Option::LEVEL_DEV)
6112 .set_default(10*60)
6113 .set_description("Timeout for async rados coroutine operations."),
6114
6115 Option("rgw_op_thread_suicide_timeout", Option::TYPE_INT, Option::LEVEL_DEV)
6116 .set_default(0)
6117 .set_description(""),
6118
6119 Option("rgw_thread_pool_size", Option::TYPE_INT, Option::LEVEL_BASIC)
6120 .set_default(512)
6121 .set_description("RGW requests handling thread pool size.")
6122 .set_long_description(
6123 "This parameter determines the number of concurrent requests RGW can process "
6124 "when using either the civetweb, or the fastcgi frontends. The higher this "
6125 "number is, RGW will be able to deal with more concurrent requests at the "
6126 "cost of more resource utilization."),
6127
6128 Option("rgw_num_control_oids", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6129 .set_default(8)
6130 .set_description("Number of control objects used for cross-RGW communication.")
6131 .set_long_description(
6132 "RGW uses certain control objects to send messages between different RGW "
6133 "processes running on the same zone. These messages include metadata cache "
6134 "invalidation info that is being sent when metadata is modified (such as "
6135 "user or bucket information). A higher number of control objects allows "
6136 "better concurrency of these messages, at the cost of more resource "
6137 "utilization."),
6138
6139 Option("rgw_num_rados_handles", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
6140 .set_default(1)
6141 .set_description("Number of librados handles that RGW uses.")
6142 .set_long_description(
6143 "This param affects the number of separate librados handles it uses to "
6144 "connect to the RADOS backend, which directly affects the number of connections "
6145 "RGW will have to each OSD. A higher number affects resource utilization."),
6146
6147 Option("rgw_verify_ssl", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
6148 .set_default(true)
6149 .set_description("Should RGW verify SSL when connecing to a remote HTTP server")
6150 .set_long_description(
6151 "RGW can send requests to other RGW servers (e.g., in multi-site sync work). "
6152 "This configurable selects whether RGW should verify the certificate for "
6153 "the remote peer and host.")
6154 .add_see_also("rgw_keystone_verify_ssl"),
6155
6156 Option("rgw_nfs_lru_lanes", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6157 .set_default(5)
6158 .set_description(""),
6159
6160 Option("rgw_nfs_lru_lane_hiwat", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6161 .set_default(911)
6162 .set_description(""),
6163
6164 Option("rgw_nfs_fhcache_partitions", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6165 .set_default(3)
6166 .set_description(""),
6167
6168 Option("rgw_nfs_fhcache_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6169 .set_default(2017)
6170 .set_description(""),
6171
6172 Option("rgw_nfs_namespace_expire_secs", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6173 .set_default(300)
6174 .set_min(1)
6175 .set_description(""),
6176
6177 Option("rgw_nfs_max_gc", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6178 .set_default(300)
6179 .set_min(1)
6180 .set_description(""),
6181
6182 Option("rgw_nfs_write_completion_interval_s", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6183 .set_default(10)
6184 .set_description(""),
6185
6186 Option("rgw_nfs_s3_fast_attrs", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
6187 .set_default(false)
6188 .set_description("use fast S3 attrs from bucket index (immutable only)")
6189 .set_long_description("use fast S3 attrs from bucket index (assumes NFS "
6190 "mounts are immutable)"),
6191
6192 Option("rgw_rados_pool_autoscale_bias", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
6193 .set_default(4.0)
6194 .set_min_max(0.01, 100000.0)
6195 .set_description("pg_autoscale_bias value for RGW metadata (omap-heavy) pools"),
6196
6197 Option("rgw_rados_pool_pg_num_min", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
6198 .set_default(8)
6199 .set_min_max(1, 1024)
6200 .set_description("pg_num_min value for RGW metadata (omap-heavy) pools"),
6201
6202 Option("rgw_rados_pool_recovery_priority", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
6203 .set_default(5)
6204 .set_min_max(-10, 10)
6205 .set_description("recovery_priority value for RGW metadata (omap-heavy) pools"),
6206
6207 Option("rgw_zone", Option::TYPE_STR, Option::LEVEL_ADVANCED)
6208 .set_default("")
6209 .set_description("Zone name")
6210 .add_see_also({"rgw_zonegroup", "rgw_realm"}),
6211
6212 Option("rgw_zone_root_pool", Option::TYPE_STR, Option::LEVEL_ADVANCED)
6213 .set_default(".rgw.root")
6214 .set_description("Zone root pool name")
6215 .set_long_description(
6216 "The zone root pool, is the pool where the RGW zone configuration located."
6217 )
6218 .add_see_also({"rgw_zonegroup_root_pool", "rgw_realm_root_pool", "rgw_period_root_pool"}),
6219
6220 Option("rgw_default_zone_info_oid", Option::TYPE_STR, Option::LEVEL_ADVANCED)
6221 .set_default("default.zone")
6222 .set_description("Default zone info object id")
6223 .set_long_description(
6224 "Name of the RADOS object that holds the default zone information."
6225 ),
6226
6227 Option("rgw_region", Option::TYPE_STR, Option::LEVEL_ADVANCED)
6228 .set_default("")
6229 .set_description("Region name")
6230 .set_long_description(
6231 "Obsolete config option. The rgw_zonegroup option should be used instead.")
6232 .add_see_also("rgw_zonegroup"),
6233
6234 Option("rgw_region_root_pool", Option::TYPE_STR, Option::LEVEL_ADVANCED)
6235 .set_default(".rgw.root")
6236 .set_description("Region root pool")
6237 .set_long_description(
6238 "Obsolete config option. The rgw_zonegroup_root_pool should be used instead.")
6239 .add_see_also("rgw_zonegroup_root_pool"),
6240
6241 Option("rgw_default_region_info_oid", Option::TYPE_STR, Option::LEVEL_ADVANCED)
6242 .set_default("default.region")
6243 .set_description("Default region info object id")
6244 .set_long_description(
6245 "Obsolete config option. The rgw_default_zonegroup_info_oid should be used instead.")
6246 .add_see_also("rgw_default_zonegroup_info_oid"),
6247
6248 Option("rgw_zonegroup", Option::TYPE_STR, Option::LEVEL_ADVANCED)
6249 .set_default("")
6250 .set_description("Zonegroup name")
6251 .add_see_also({"rgw_zone", "rgw_realm"}),
6252
6253 Option("rgw_zonegroup_root_pool", Option::TYPE_STR, Option::LEVEL_ADVANCED)
6254 .set_default(".rgw.root")
6255 .set_description("Zonegroup root pool")
6256 .set_long_description(
6257 "The zonegroup root pool, is the pool where the RGW zonegroup configuration located."
6258 )
6259 .add_see_also({"rgw_zone_root_pool", "rgw_realm_root_pool", "rgw_period_root_pool"}),
6260
6261 Option("rgw_default_zonegroup_info_oid", Option::TYPE_STR, Option::LEVEL_ADVANCED)
6262 .set_default("default.zonegroup")
6263 .set_description(""),
6264
6265 Option("rgw_realm", Option::TYPE_STR, Option::LEVEL_ADVANCED)
6266 .set_default("")
6267 .set_description(""),
6268
6269 Option("rgw_realm_root_pool", Option::TYPE_STR, Option::LEVEL_ADVANCED)
6270 .set_default(".rgw.root")
6271 .set_description("Realm root pool")
6272 .set_long_description(
6273 "The realm root pool, is the pool where the RGW realm configuration located."
6274 )
6275 .add_see_also({"rgw_zonegroup_root_pool", "rgw_zone_root_pool", "rgw_period_root_pool"}),
6276
6277 Option("rgw_default_realm_info_oid", Option::TYPE_STR, Option::LEVEL_ADVANCED)
6278 .set_default("default.realm")
6279 .set_description(""),
6280
6281 Option("rgw_period_root_pool", Option::TYPE_STR, Option::LEVEL_ADVANCED)
6282 .set_default(".rgw.root")
6283 .set_description("Period root pool")
6284 .set_long_description(
6285 "The period root pool, is the pool where the RGW period configuration located."
6286 )
6287 .add_see_also({"rgw_zonegroup_root_pool", "rgw_zone_root_pool", "rgw_realm_root_pool"}),
6288
6289 Option("rgw_period_latest_epoch_info_oid", Option::TYPE_STR, Option::LEVEL_DEV)
6290 .set_default(".latest_epoch")
6291 .set_description(""),
6292
6293 Option("rgw_log_nonexistent_bucket", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
6294 .set_default(false)
6295 .set_description("Should RGW log operations on bucket that does not exist")
6296 .set_long_description(
6297 "This config option applies to the ops log. When this option is set, the ops log "
6298 "will log operations that are sent to non existing buckets. These operations "
6299 "inherently fail, and do not correspond to a specific user.")
6300 .add_see_also("rgw_enable_ops_log"),
6301
6302 Option("rgw_log_object_name", Option::TYPE_STR, Option::LEVEL_ADVANCED)
6303 .set_default("%Y-%m-%d-%H-%i-%n")
6304 .set_description("Ops log object name format")
6305 .set_long_description(
6306 "Defines the format of the RADOS objects names that ops log uses to store ops "
6307 "log data")
6308 .add_see_also("rgw_enable_ops_log"),
6309
6310 Option("rgw_log_object_name_utc", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
6311 .set_default(false)
6312 .set_description("Should ops log object name based on UTC")
6313 .set_long_description(
6314 "If set, the names of the RADOS objects that hold the ops log data will be based "
6315 "on UTC time zone. If not set, it will use the local time zone.")
6316 .add_see_also({"rgw_enable_ops_log", "rgw_log_object_name"}),
6317
6318 Option("rgw_usage_max_shards", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6319 .set_default(32)
6320 .set_description("Number of shards for usage log.")
6321 .set_long_description(
6322 "The number of RADOS objects that RGW will use in order to store the usage log "
6323 "data.")
6324 .add_see_also("rgw_enable_usage_log"),
6325
6326 Option("rgw_usage_max_user_shards", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6327 .set_default(1)
6328 .set_min(1)
6329 .set_description("Number of shards for single user in usage log")
6330 .set_long_description(
6331 "The number of shards that a single user will span over in the usage log.")
6332 .add_see_also("rgw_enable_usage_log"),
6333
6334 Option("rgw_enable_ops_log", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
6335 .set_default(false)
6336 .set_description("Enable ops log")
6337 .add_see_also({"rgw_log_nonexistent_bucket", "rgw_log_object_name", "rgw_ops_log_rados",
6338 "rgw_ops_log_socket_path"}),
6339
6340 Option("rgw_enable_usage_log", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
6341 .set_default(false)
6342 .set_description("Enable usage log")
6343 .add_see_also("rgw_usage_max_shards"),
6344
6345 Option("rgw_ops_log_rados", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
6346 .set_default(true)
6347 .set_description("Use RADOS for ops log")
6348 .set_long_description(
6349 "If set, RGW will store ops log information in RADOS.")
6350 .add_see_also({"rgw_enable_ops_log"}),
6351
6352 Option("rgw_ops_log_socket_path", Option::TYPE_STR, Option::LEVEL_ADVANCED)
6353 .set_default("")
6354 .set_description("Unix domain socket path for ops log.")
6355 .set_long_description(
6356 "Path to unix domain socket that RGW will listen for connection on. When connected, "
6357 "RGW will send ops log data through it.")
6358 .add_see_also({"rgw_enable_ops_log", "rgw_ops_log_data_backlog"}),
6359
6360 Option("rgw_ops_log_data_backlog", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
6361 .set_default(5 << 20)
6362 .set_description("Ops log socket backlog")
6363 .set_long_description(
6364 "Maximum amount of data backlog that RGW can keep when ops log is configured to "
6365 "send info through unix domain socket. When data backlog is higher than this, "
6366 "ops log entries will be lost. In order to avoid ops log information loss, the "
6367 "listener needs to clear data (by reading it) quickly enough.")
6368 .add_see_also({"rgw_enable_ops_log", "rgw_ops_log_socket_path"}),
6369
6370 Option("rgw_fcgi_socket_backlog", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6371 .set_default(1024)
6372 .set_description("FastCGI socket connection backlog")
6373 .set_long_description(
6374 "Size of FastCGI connection backlog. This reflects the maximum number of new "
6375 "connection requests that RGW can handle concurrently without dropping any. ")
6376 .add_see_also({"rgw_host", "rgw_socket_path"}),
6377
6378 Option("rgw_usage_log_flush_threshold", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6379 .set_default(1024)
6380 .set_description("Number of entries in usage log before flushing")
6381 .set_long_description(
6382 "This is the max number of entries that will be held in the usage log, before it "
6383 "will be flushed to the backend. Note that the usage log is periodically flushed, "
6384 "even if number of entries does not reach this threshold. A usage log entry "
6385 "corresponds to one or more operations on a single bucket.i")
6386 .add_see_also({"rgw_enable_usage_log", "rgw_usage_log_tick_interval"}),
6387
6388 Option("rgw_usage_log_tick_interval", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6389 .set_default(30)
6390 .set_description("Number of seconds between usage log flush cycles")
6391 .set_long_description(
6392 "The number of seconds between consecutive usage log flushes. The usage log will "
6393 "also flush itself to the backend if the number of pending entries reaches a "
6394 "certain threshold.")
6395 .add_see_also({"rgw_enable_usage_log", "rgw_usage_log_flush_threshold"}),
6396
6397 Option("rgw_init_timeout", Option::TYPE_INT, Option::LEVEL_BASIC)
6398 .set_default(300)
6399 .set_description("Initialization timeout")
6400 .set_long_description(
6401 "The time length (in seconds) that RGW will allow for its initialization. RGW "
6402 "process will give up and quit if initialization is not complete after this amount "
6403 "of time."),
6404
6405 Option("rgw_mime_types_file", Option::TYPE_STR, Option::LEVEL_BASIC)
6406 .set_default("/etc/mime.types")
6407 .set_description("Path to local mime types file")
6408 .set_long_description(
6409 "The mime types file is needed in Swift when uploading an object. If object's "
6410 "content type is not specified, RGW will use data from this file to assign "
6411 "a content type to the object."),
6412
6413 Option("rgw_gc_max_objs", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6414 .set_default(32)
6415 .set_description("Number of shards for garbage collector data")
6416 .set_long_description(
6417 "The number of garbage collector data shards, is the number of RADOS objects that "
6418 "RGW will use to store the garbage collection information on.")
6419 .add_see_also({"rgw_gc_obj_min_wait", "rgw_gc_processor_max_time", "rgw_gc_processor_period", "rgw_gc_max_concurrent_io"}),
6420
6421 Option("rgw_gc_obj_min_wait", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6422 .set_default(2_hr)
6423 .set_description("Garbage collection object expiration time")
6424 .set_long_description(
6425 "The length of time (in seconds) that the RGW collector will wait before purging "
6426 "a deleted object's data. RGW will not remove object immediately, as object could "
6427 "still have readers. A mechanism exists to increase the object's expiration time "
6428 "when it's being read.")
6429 .add_see_also({"rgw_gc_max_objs", "rgw_gc_processor_max_time", "rgw_gc_processor_period", "rgw_gc_max_concurrent_io"}),
6430
6431 Option("rgw_gc_processor_max_time", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6432 .set_default(1_hr)
6433 .set_description("Length of time GC processor can lease shard")
6434 .set_long_description(
6435 "Garbage collection thread in RGW process holds a lease on its data shards. These "
6436 "objects contain the information about the objects that need to be removed. RGW "
6437 "takes a lease in order to prevent multiple RGW processes from handling the same "
6438 "objects concurrently. This time signifies that maximum amount of time (in seconds) that RGW "
6439 "is allowed to hold that lease. In the case where RGW goes down uncleanly, this "
6440 "is the amount of time where processing of that data shard will be blocked.")
6441 .add_see_also({"rgw_gc_max_objs", "rgw_gc_obj_min_wait", "rgw_gc_processor_period", "rgw_gc_max_concurrent_io"}),
6442
6443 Option("rgw_gc_processor_period", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6444 .set_default(1_hr)
6445 .set_description("Garbage collector cycle run time")
6446 .set_long_description(
6447 "The amount of time between the start of consecutive runs of the garbage collector "
6448 "threads. If garbage collector runs takes more than this period, it will not wait "
6449 "before running again.")
6450 .add_see_also({"rgw_gc_max_objs", "rgw_gc_obj_min_wait", "rgw_gc_processor_max_time", "rgw_gc_max_concurrent_io", "rgw_gc_max_trim_chunk"}),
6451
6452 Option("rgw_gc_max_concurrent_io", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6453 .set_default(10)
6454 .set_description("Max concurrent RADOS IO operations for garbage collection")
6455 .set_long_description(
6456 "The maximum number of concurrent IO operations that the RGW garbage collection "
6457 "thread will use when purging old data.")
6458 .add_see_also({"rgw_gc_max_objs", "rgw_gc_obj_min_wait", "rgw_gc_processor_max_time", "rgw_gc_max_trim_chunk"}),
6459
6460 Option("rgw_gc_max_trim_chunk", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6461 .set_default(16)
6462 .set_description("Max number of keys to remove from garbage collector log in a single operation")
6463 .add_see_also({"rgw_gc_max_objs", "rgw_gc_obj_min_wait", "rgw_gc_processor_max_time", "rgw_gc_max_concurrent_io"}),
6464
6465 Option("rgw_s3_success_create_obj_status", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6466 .set_default(0)
6467 .set_description("HTTP return code override for object creation")
6468 .set_long_description(
6469 "If not zero, this is the HTTP return code that will be returned on a successful S3 "
6470 "object creation."),
6471
6472 Option("rgw_resolve_cname", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
6473 .set_default(false)
6474 .set_description("Support vanity domain names via CNAME")
6475 .set_long_description(
6476 "If true, RGW will query DNS when detecting that it's serving a request that was "
6477 "sent to a host in another domain. If a CNAME record is configured for that domain "
6478 "it will use it instead. This gives user to have the ability of creating a unique "
6479 "domain of their own to point at data in their bucket."),
6480
6481 Option("rgw_obj_stripe_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
6482 .set_default(4_M)
6483 .set_description("RGW object stripe size")
6484 .set_long_description(
6485 "The size of an object stripe for RGW objects. This is the maximum size a backing "
6486 "RADOS object will have. RGW objects that are larger than this will span over "
6487 "multiple objects."),
6488
6489 Option("rgw_extended_http_attrs", Option::TYPE_STR, Option::LEVEL_ADVANCED)
6490 .set_default("")
6491 .set_description("RGW support extended HTTP attrs")
6492 .set_long_description(
6493 "Add new set of attributes that could be set on an object. These extra attributes "
6494 "can be set through HTTP header fields when putting the objects. If set, these "
6495 "attributes will return as HTTP fields when doing GET/HEAD on the object."),
6496
6497 Option("rgw_exit_timeout_secs", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6498 .set_default(120)
6499 .set_description("RGW shutdown timeout")
6500 .set_long_description("Number of seconds to wait for a process before exiting unconditionally."),
6501
6502 Option("rgw_get_obj_window_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
6503 .set_default(16_M)
6504 .set_description("RGW object read window size")
6505 .set_long_description("The window size in bytes for a single object read request"),
6506
6507 Option("rgw_get_obj_max_req_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
6508 .set_default(4_M)
6509 .set_description("RGW object read chunk size")
6510 .set_long_description(
6511 "The maximum request size of a single object read operation sent to RADOS"),
6512
6513 Option("rgw_relaxed_s3_bucket_names", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
6514 .set_default(false)
6515 .set_description("RGW enable relaxed S3 bucket names")
6516 .set_long_description("RGW enable relaxed S3 bucket name rules for US region buckets."),
6517
6518 Option("rgw_defer_to_bucket_acls", Option::TYPE_STR, Option::LEVEL_ADVANCED)
6519 .set_default("")
6520 .set_description("Bucket ACLs override object ACLs")
6521 .set_long_description(
6522 "If not empty, a string that selects that mode of operation. 'recurse' will use "
6523 "bucket's ACL for the authorizaton. 'full-control' will allow users that users "
6524 "that have full control permission on the bucket have access to the object."),
6525
6526 Option("rgw_list_buckets_max_chunk", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6527 .set_default(1000)
6528 .set_description("Max number of buckets to retrieve in a single listing operation")
6529 .set_long_description(
6530 "When RGW fetches lists of user's buckets from the backend, this is the max number "
6531 "of entries it will try to retrieve in a single operation. Note that the backend "
6532 "may choose to return a smaller number of entries."),
6533
6534 Option("rgw_md_log_max_shards", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6535 .set_default(64)
6536 .set_description("RGW number of metadata log shards")
6537 .set_long_description(
6538 "The number of shards the RGW metadata log entries will reside in. This affects "
6539 "the metadata sync parallelism as a shard can only be processed by a single "
6540 "RGW at a time"),
6541
6542 Option("rgw_curl_wait_timeout_ms", Option::TYPE_INT, Option::LEVEL_DEV)
6543 .set_default(1000)
6544 .set_description(""),
6545
6546 Option("rgw_curl_low_speed_limit", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6547 .set_default(1024)
6548 .set_long_description(
6549 "It contains the average transfer speed in bytes per second that the "
6550 "transfer should be below during rgw_curl_low_speed_time seconds for libcurl "
6551 "to consider it to be too slow and abort. Set it zero to disable this."),
6552
6553 Option("rgw_curl_low_speed_time", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6554 .set_default(300)
6555 .set_long_description(
6556 "It contains the time in number seconds that the transfer speed should be below "
6557 "the rgw_curl_low_speed_limit for the library to consider it too slow and abort. "
6558 "Set it zero to disable this."),
6559
6560 Option("rgw_copy_obj_progress", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
6561 .set_default(true)
6562 .set_description("Send progress report through copy operation")
6563 .set_long_description(
6564 "If true, RGW will send progress information when copy operation is executed. "),
6565
6566 Option("rgw_copy_obj_progress_every_bytes", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
6567 .set_default(1_M)
6568 .set_description("Send copy-object progress info after these many bytes"),
6569
6570 Option("rgw_obj_tombstone_cache_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6571 .set_default(1000)
6572 .set_description("Max number of entries to keep in tombstone cache")
6573 .set_long_description(
6574 "The tombstone cache is used when doing a multi-zone data sync. RGW keeps "
6575 "there information about removed objects which is needed in order to prevent "
6576 "re-syncing of objects that were already removed."),
6577
6578 Option("rgw_data_log_window", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6579 .set_default(30)
6580 .set_description("Data log time window")
6581 .set_long_description(
6582 "The data log keeps information about buckets that have objectst that were "
6583 "modified within a specific timeframe. The sync process then knows which buckets "
6584 "are needed to be scanned for data sync."),
6585
6586 Option("rgw_data_log_changes_size", Option::TYPE_INT, Option::LEVEL_DEV)
6587 .set_default(1000)
6588 .set_description("Max size of pending changes in data log")
6589 .set_long_description(
6590 "RGW will trigger update to the data log if the number of pending entries reached "
6591 "this number."),
6592
6593 Option("rgw_data_log_num_shards", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6594 .set_default(128)
6595 .set_description("Number of data log shards")
6596 .set_long_description(
6597 "The number of shards the RGW data log entries will reside in. This affects the "
6598 "data sync parallelism as a shard can only be processed by a single RGW at a time."),
6599
6600 Option("rgw_data_log_obj_prefix", Option::TYPE_STR, Option::LEVEL_DEV)
6601 .set_default("data_log")
6602 .set_description(""),
6603
6604 Option("rgw_bucket_quota_ttl", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6605 .set_default(600)
6606 .set_description("Bucket quota stats cache TTL")
6607 .set_long_description(
6608 "Length of time for bucket stats to be cached within RGW instance."),
6609
6610 Option("rgw_bucket_quota_soft_threshold", Option::TYPE_FLOAT, Option::LEVEL_BASIC)
6611 .set_default(0.95)
6612 .set_description("RGW quota soft threshold")
6613 .set_long_description(
6614 "Threshold from which RGW doesn't rely on cached info for quota "
6615 "decisions. This is done for higher accuracy of the quota mechanism at "
6616 "cost of performance, when getting close to the quota limit. The value "
6617 "configured here is the ratio between the data usage to the max usage "
6618 "as specified by the quota."),
6619
6620 Option("rgw_bucket_quota_cache_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6621 .set_default(10000)
6622 .set_description("RGW quota stats cache size")
6623 .set_long_description(
6624 "Maximum number of entries in the quota stats cache."),
6625
6626 Option("rgw_bucket_default_quota_max_objects", Option::TYPE_INT, Option::LEVEL_BASIC)
6627 .set_default(-1)
6628 .set_description("Default quota for max objects in a bucket")
6629 .set_long_description(
6630 "The default quota configuration for max number of objects in a bucket. A "
6631 "negative number means 'unlimited'."),
6632
6633 Option("rgw_bucket_default_quota_max_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6634 .set_default(-1)
6635 .set_description("Default quota for total size in a bucket")
6636 .set_long_description(
6637 "The default quota configuration for total size of objects in a bucket. A "
6638 "negative number means 'unlimited'."),
6639
6640 Option("rgw_expose_bucket", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
6641 .set_default(false)
6642 .set_description("Send Bucket HTTP header with the response")
6643 .set_long_description(
6644 "If true, RGW will send a Bucket HTTP header with the responses. The header will "
6645 "contain the name of the bucket the operation happened on."),
6646
6647 Option("rgw_frontends", Option::TYPE_STR, Option::LEVEL_BASIC)
6648 .set_default("beast port=7480")
6649 .set_description("RGW frontends configuration")
6650 .set_long_description(
6651 "A comma delimited list of frontends configuration. Each configuration contains "
6652 "the type of the frontend followed by an optional space delimited set of "
6653 "key=value config parameters."),
6654
6655 Option("rgw_user_quota_bucket_sync_interval", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6656 .set_default(180)
6657 .set_description("User quota bucket sync interval")
6658 .set_long_description(
6659 "Time period for accumulating modified buckets before syncing these stats."),
6660
6661 Option("rgw_user_quota_sync_interval", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6662 .set_default(1_day)
6663 .set_description("User quota sync interval")
6664 .set_long_description(
6665 "Time period for accumulating modified buckets before syncing entire user stats."),
6666
6667 Option("rgw_user_quota_sync_idle_users", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
6668 .set_default(false)
6669 .set_description("Should sync idle users quota")
6670 .set_long_description(
6671 "Whether stats for idle users be fully synced."),
6672
6673 Option("rgw_user_quota_sync_wait_time", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6674 .set_default(1_day)
6675 .set_description("User quota full-sync wait time")
6676 .set_long_description(
6677 "Minimum time between two full stats sync for non-idle users."),
6678
6679 Option("rgw_user_default_quota_max_objects", Option::TYPE_INT, Option::LEVEL_BASIC)
6680 .set_default(-1)
6681 .set_description("User quota max objects")
6682 .set_long_description(
6683 "The default quota configuration for total number of objects for a single user. A "
6684 "negative number means 'unlimited'."),
6685
6686 Option("rgw_user_default_quota_max_size", Option::TYPE_INT, Option::LEVEL_BASIC)
6687 .set_default(-1)
6688 .set_description("User quota max size")
6689 .set_long_description(
6690 "The default quota configuration for total size of objects for a single user. A "
6691 "negative number means 'unlimited'."),
6692
6693 Option("rgw_multipart_min_part_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
6694 .set_default(5_M)
6695 .set_description("Minimum S3 multipart-upload part size")
6696 .set_long_description(
6697 "When doing a multipart upload, each part (other than the last part) should be "
6698 "at least this size."),
6699
6700 Option("rgw_multipart_part_upload_limit", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6701 .set_default(10000)
6702 .set_description("Max number of parts in multipart upload"),
6703
6704 Option("rgw_max_slo_entries", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6705 .set_default(1000)
6706 .set_description("Max number of entries in Swift Static Large Object manifest"),
6707
6708 Option("rgw_olh_pending_timeout_sec", Option::TYPE_INT, Option::LEVEL_DEV)
6709 .set_default(1_hr)
6710 .set_description("Max time for pending OLH change to complete")
6711 .set_long_description(
6712 "OLH is a versioned object's logical head. Operations on it are journaled and "
6713 "as pending before completion. If an operation doesn't complete with this amount "
6714 "of seconds, we remove the operation from the journal."),
6715
6716 Option("rgw_user_max_buckets", Option::TYPE_INT, Option::LEVEL_BASIC)
6717 .set_default(1000)
6718 .set_description("Max number of buckets per user")
6719 .set_long_description(
6720 "A user can create this many buckets. Zero means unlimmited, negative number means "
6721 "user cannot create any buckets (although user will retain buckets already created."),
6722
6723 Option("rgw_objexp_gc_interval", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
6724 .set_default(10_min)
6725 .set_description("Swift objects expirer garbage collector interval"),
6726
6727 Option("rgw_objexp_hints_num_shards", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
6728 .set_default(127)
6729 .set_description("Number of object expirer data shards")
6730 .set_long_description(
6731 "The number of shards the (Swift) object expirer will store its data on."),
6732
6733 Option("rgw_objexp_chunk_size", Option::TYPE_UINT, Option::LEVEL_DEV)
6734 .set_default(100)
6735 .set_description(""),
6736
6737 Option("rgw_enable_static_website", Option::TYPE_BOOL, Option::LEVEL_BASIC)
6738 .set_default(false)
6739 .set_description("Enable static website APIs")
6740 .set_long_description(
6741 "This configurable controls whether RGW handles the website control APIs. RGW can "
6742 "server static websites if s3website hostnames are configured, and unrelated to "
6743 "this configurable."),
6744
6745 Option("rgw_user_unique_email", Option::TYPE_BOOL, Option::LEVEL_BASIC)
6746 .set_default(true)
6747 .set_description("Require local RGW users to have unique email addresses")
6748 .set_long_description(
6749 "Enforce builtin user accounts to have unique email addresses. This "
6750 "setting is historical. In future, non-enforcement of email address "
6751 "uniqueness is likely to become the default."),
6752
6753 Option("rgw_log_http_headers", Option::TYPE_STR, Option::LEVEL_BASIC)
6754 .set_default("")
6755 .set_description("List of HTTP headers to log")
6756 .set_long_description(
6757 "A comma delimited list of HTTP headers to log when seen, ignores case (e.g., "
6758 "http_x_forwarded_for)."),
6759
6760 Option("rgw_num_async_rados_threads", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6761 .set_default(32)
6762 .set_description("Number of concurrent RADOS operations in multisite sync")
6763 .set_long_description(
6764 "The number of concurrent RADOS IO operations that will be triggered for handling "
6765 "multisite sync operations. This includes control related work, and not the actual "
6766 "sync operations."),
6767
6768 Option("rgw_md_notify_interval_msec", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6769 .set_default(200)
6770 .set_description("Length of time to aggregate metadata changes")
6771 .set_long_description(
6772 "Length of time (in milliseconds) in which the master zone aggregates all the "
6773 "metadata changes that occurred, before sending notifications to all the other "
6774 "zones."),
6775
6776 Option("rgw_run_sync_thread", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
6777 .set_default(true)
6778 .set_description("Should run sync thread"),
6779
6780 Option("rgw_sync_lease_period", Option::TYPE_INT, Option::LEVEL_DEV)
6781 .set_default(120)
6782 .set_description(""),
6783
6784 Option("rgw_sync_log_trim_interval", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6785 .set_default(1200)
6786 .set_description("Sync log trim interval")
6787 .set_long_description(
6788 "Time in seconds between attempts to trim sync logs."),
6789
6790 Option("rgw_sync_log_trim_max_buckets", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6791 .set_default(16)
6792 .set_description("Maximum number of buckets to trim per interval")
6793 .set_long_description("The maximum number of buckets to consider for bucket index log trimming each trim interval, regardless of the number of bucket index shards. Priority is given to buckets with the most sync activity over the last trim interval.")
6794 .add_see_also("rgw_sync_log_trim_interval")
6795 .add_see_also("rgw_sync_log_trim_min_cold_buckets")
6796 .add_see_also("rgw_sync_log_trim_concurrent_buckets"),
6797
6798 Option("rgw_sync_log_trim_min_cold_buckets", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6799 .set_default(4)
6800 .set_description("Minimum number of cold buckets to trim per interval")
6801 .set_long_description("Of the `rgw_sync_log_trim_max_buckets` selected for bucket index log trimming each trim interval, at least this many of them must be 'cold' buckets. These buckets are selected in order from the list of all bucket instances, to guarantee that all buckets will be visited eventually.")
6802 .add_see_also("rgw_sync_log_trim_interval")
6803 .add_see_also("rgw_sync_log_trim_max_buckets")
6804 .add_see_also("rgw_sync_log_trim_concurrent_buckets"),
6805
6806 Option("rgw_sync_log_trim_concurrent_buckets", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6807 .set_default(4)
6808 .set_description("Maximum number of buckets to trim in parallel")
6809 .add_see_also("rgw_sync_log_trim_interval")
6810 .add_see_also("rgw_sync_log_trim_max_buckets")
6811 .add_see_also("rgw_sync_log_trim_min_cold_buckets"),
6812
6813 Option("rgw_sync_data_inject_err_probability", Option::TYPE_FLOAT, Option::LEVEL_DEV)
6814 .set_default(0)
6815 .set_description(""),
6816
6817 Option("rgw_sync_meta_inject_err_probability", Option::TYPE_FLOAT, Option::LEVEL_DEV)
6818 .set_default(0)
6819 .set_description(""),
6820
6821 Option("rgw_sync_trace_history_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
6822 .set_default(4096)
6823 .set_description("Sync trace history size")
6824 .set_long_description(
6825 "Maximum number of complete sync trace entries to keep."),
6826
6827 Option("rgw_sync_trace_per_node_log_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6828 .set_default(32)
6829 .set_description("Sync trace per-node log size")
6830 .set_long_description(
6831 "The number of log entries to keep per sync-trace node."),
6832
6833 Option("rgw_sync_trace_servicemap_update_interval", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6834 .set_default(10)
6835 .set_description("Sync-trace service-map update interval")
6836 .set_long_description(
6837 "Number of seconds between service-map updates of sync-trace events."),
6838
6839 Option("rgw_period_push_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
6840 .set_default(2)
6841 .set_description("Period push interval")
6842 .set_long_description(
6843 "Number of seconds to wait before retrying 'period push' operation."),
6844
6845 Option("rgw_period_push_interval_max", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
6846 .set_default(30)
6847 .set_description("Period push maximum interval")
6848 .set_long_description(
6849 "The max number of seconds to wait before retrying 'period push' after exponential "
6850 "backoff."),
6851
6852 Option("rgw_safe_max_objects_per_shard", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6853 .set_default(100*1024)
6854 .set_description("Safe number of objects per shard")
6855 .set_long_description(
6856 "This is the max number of objects per bucket index shard that RGW considers "
6857 "safe. RGW will warn if it identifies a bucket where its per-shard count is "
6858 "higher than a percentage of this number.")
6859 .add_see_also("rgw_shard_warning_threshold"),
6860
6861 Option("rgw_shard_warning_threshold", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
6862 .set_default(90)
6863 .set_description("Warn about max objects per shard")
6864 .set_long_description(
6865 "Warn if number of objects per shard in a specific bucket passed this percentage "
6866 "of the safe number.")
6867 .add_see_also("rgw_safe_max_objects_per_shard"),
6868
6869 Option("rgw_swift_versioning_enabled", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
6870 .set_default(false)
6871 .set_description("Enable Swift versioning"),
6872
6873 Option("rgw_swift_custom_header", Option::TYPE_STR, Option::LEVEL_ADVANCED)
6874 .set_default("")
6875 .set_description("Enable swift custom header")
6876 .set_long_description(
6877 "If not empty, specifies a name of HTTP header that can include custom data. When "
6878 "uploading an object, if this header is passed RGW will store this header info "
6879 "and it will be available when listing the bucket."),
6880
6881 Option("rgw_swift_need_stats", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
6882 .set_default(true)
6883 .set_description("Enable stats on bucket listing in Swift"),
6884
6885 Option("rgw_reshard_num_logs", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
6886 .set_default(16)
6887 .set_min(1)
6888 .set_description("")
6889 .add_service("rgw"),
6890
6891 Option("rgw_reshard_bucket_lock_duration", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
6892 .set_default(360)
6893 .set_min(30)
6894 .set_description("Number of seconds the timeout on the reshard locks (bucket reshard lock and reshard log lock) are set to. As a reshard proceeds these locks can be renewed/extended. If too short, reshards cannot complete and will fail, causing a future reshard attempt. If too long a hung or crashed reshard attempt will keep the bucket locked for an extended period, not allowing RGW to detect the failed reshard attempt and recover.")
6895 .add_tag("performance")
6896 .add_service("rgw"),
6897
6898 Option("rgw_reshard_batch_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
6899 .set_default(64)
6900 .set_min(8)
6901 .set_description("Number of reshard entries to batch together before sending the operations to the CLS back-end")
6902 .add_tag("performance")
6903 .add_service("rgw"),
6904
6905 Option("rgw_reshard_max_aio", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
6906 .set_default(128)
6907 .set_min(16)
6908 .set_description("Maximum number of outstanding asynchronous I/O operations to allow at a time during resharding")
6909 .add_tag("performance")
6910 .add_service("rgw"),
6911
6912 Option("rgw_trust_forwarded_https", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
6913 .set_default(false)
6914 .set_description("Trust Forwarded and X-Forwarded-Proto headers")
6915 .set_long_description(
6916 "When a proxy in front of radosgw is used for ssl termination, radosgw "
6917 "does not know whether incoming http connections are secure. Enable "
6918 "this option to trust the Forwarded and X-Forwarded-Proto headers sent "
6919 "by the proxy when determining whether the connection is secure. This "
6920 "is required for some features, such as server side encryption.")
6921 .add_see_also("rgw_crypt_require_ssl"),
6922
6923 Option("rgw_crypt_require_ssl", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
6924 .set_default(true)
6925 .set_description("Requests including encryption key headers must be sent over ssl"),
6926
6927 Option("rgw_crypt_default_encryption_key", Option::TYPE_STR, Option::LEVEL_DEV)
6928 .set_default("")
6929 .set_description(""),
6930
6931 Option("rgw_crypt_s3_kms_backend", Option::TYPE_STR, Option::LEVEL_ADVANCED)
6932 .set_default("barbican")
6933 .set_enum_allowed({"barbican", "vault", "testing"})
6934 .set_description(
6935 "Where the SSE-KMS encryption keys are stored. Supported KMS "
6936 "systems are OpenStack Barbican ('barbican', the default) and HashiCorp "
6937 "Vault ('vault')."),
6938
6939 Option("rgw_crypt_s3_kms_encryption_keys", Option::TYPE_STR, Option::LEVEL_DEV)
6940 .set_default("")
6941 .set_description(""),
6942
6943 Option("rgw_crypt_vault_auth", Option::TYPE_STR, Option::LEVEL_ADVANCED)
6944 .set_default("token")
6945 .set_enum_allowed({"token"})
6946 .set_description(
6947 "Type of authentication method to be used with Vault. "
6948 "The only currently supported method is 'token'.")
6949 .add_see_also({
6950 "rgw_crypt_s3_kms_backend",
6951 "rgw_crypt_vault_addr",
6952 "rgw_crypt_vault_token_file"}),
6953
6954 Option("rgw_crypt_vault_token_file", Option::TYPE_STR, Option::LEVEL_ADVANCED)
6955 .set_default("")
6956 .set_description(
6957 "If authentication method is 'token', provide a path to the token file, "
6958 "which for security reasons should readable only by Rados Gateway.")
6959 .add_see_also({
6960 "rgw_crypt_s3_kms_backend",
6961 "rgw_crypt_vault_auth",
6962 "rgw_crypt_vault_addr"}),
6963
6964 Option("rgw_crypt_vault_addr", Option::TYPE_STR, Option::LEVEL_ADVANCED)
6965 .set_default("")
6966 .set_description("Base URL to the Vault server.")
6967 .add_see_also({
6968 "rgw_crypt_s3_kms_backend",
6969 "rgw_crypt_vault_auth",
6970 "rgw_crypt_vault_token_file"}),
6971
6972 Option("rgw_crypt_suppress_logs", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
6973 .set_default(true)
6974 .set_description("Suppress logs that might print client key"),
6975
6976 Option("rgw_list_bucket_min_readahead", Option::TYPE_INT, Option::LEVEL_ADVANCED)
6977 .set_default(1000)
6978 .set_description("Minimum number of entries to request from rados for bucket listing"),
6979
6980 Option("rgw_rest_getusage_op_compat", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
6981 .set_default(false)
6982 .set_description("REST GetUsage request backward compatibility"),
6983
6984 Option("rgw_torrent_flag", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
6985 .set_default(false)
6986 .set_description("When true, uploaded objects will calculate and store "
6987 "a SHA256 hash of object data so the object can be "
6988 "retrieved as a torrent file"),
6989
6990 Option("rgw_torrent_tracker", Option::TYPE_STR, Option::LEVEL_ADVANCED)
6991 .set_default("")
6992 .set_description("Torrent field announce and announce list"),
6993
6994 Option("rgw_torrent_createby", Option::TYPE_STR, Option::LEVEL_ADVANCED)
6995 .set_default("")
6996 .set_description("torrent field created by"),
6997
6998 Option("rgw_torrent_comment", Option::TYPE_STR, Option::LEVEL_ADVANCED)
6999 .set_default("")
7000 .set_description("Torrent field comment"),
7001
7002 Option("rgw_torrent_encoding", Option::TYPE_STR, Option::LEVEL_ADVANCED)
7003 .set_default("")
7004 .set_description("torrent field encoding"),
7005
7006 Option("rgw_data_notify_interval_msec", Option::TYPE_INT, Option::LEVEL_ADVANCED)
7007 .set_default(200)
7008 .set_description("data changes notification interval to followers"),
7009
7010 Option("rgw_torrent_origin", Option::TYPE_STR, Option::LEVEL_ADVANCED)
7011 .set_default("")
7012 .set_description("Torrent origin"),
7013
7014 Option("rgw_torrent_sha_unit", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
7015 .set_default(512*1024)
7016 .set_description(""),
7017
7018 Option("rgw_dynamic_resharding", Option::TYPE_BOOL, Option::LEVEL_BASIC)
7019 .set_default(true)
7020 .set_description("Enable dynamic resharding")
7021 .set_long_description(
7022 "If true, RGW will dynamicall increase the number of shards in buckets that have "
7023 "a high number of objects per shard.")
7024 .add_see_also("rgw_max_objs_per_shard"),
7025
7026 Option("rgw_max_objs_per_shard", Option::TYPE_UINT, Option::LEVEL_BASIC)
7027 .set_default(100000)
7028 .set_description("Max objects per shard for dynamic resharding")
7029 .set_long_description(
7030 "This is the max number of objects per bucket index shard that RGW will "
7031 "allow with dynamic resharding. RGW will trigger an automatic reshard operation "
7032 "on the bucket if it exceeds this number.")
7033 .add_see_also("rgw_dynamic_resharding"),
7034
7035 Option("rgw_reshard_thread_interval", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7036 .set_default(10_min)
7037 .set_min(10)
7038 .set_description("Number of seconds between processing of reshard log entries"),
7039
7040 Option("rgw_cache_expiry_interval", Option::TYPE_UINT,
7041 Option::LEVEL_ADVANCED)
7042 .set_default(15_min)
7043 .set_description("Number of seconds before entries in the cache are "
7044 "assumed stale and re-fetched. Zero is never.")
7045 .add_tag("performance")
7046 .add_service("rgw")
7047 .set_long_description("The Rados Gateway stores metadata and objects in "
7048 "an internal cache. This should be kept consistent "
7049 "by the OSD's relaying notify events between "
7050 "multiple watching RGW processes. In the event "
7051 "that this notification protocol fails, bounding "
7052 "the length of time that any data in the cache will "
7053 "be assumed valid will ensure that any RGW instance "
7054 "that falls out of sync will eventually recover. "
7055 "This seems to be an issue mostly for large numbers "
7056 "of RGW instances under heavy use. If you would like "
7057 "to turn off cache expiry, set this value to zero."),
7058
7059 Option("rgw_inject_notify_timeout_probability", Option::TYPE_FLOAT,
7060 Option::LEVEL_DEV)
7061 .set_default(0)
7062 .add_tag("fault injection")
7063 .add_tag("testing")
7064 .add_service("rgw")
7065 .set_min_max(0.0, 1.0)
7066 .set_description("Likelihood of ignoring a notify")
7067 .set_long_description("This is the probability that the RGW cache will "
7068 "ignore a cache notify message. It exists to help "
7069 "with the development and testing of cache "
7070 "consistency and recovery improvements. Please "
7071 "do not set it in a production cluster, as it "
7072 "actively causes failures. Set this to a floating "
7073 "point value between 0 and 1."),
7074 Option("rgw_max_notify_retries", Option::TYPE_UINT,
7075 Option::LEVEL_ADVANCED)
7076 .set_default(3)
7077 .add_tag("error recovery")
7078 .add_service("rgw")
7079 .set_description("Number of attempts to notify peers before giving up.")
7080 .set_long_description("The number of times we will attempt to update "
7081 "a peer's cache in the event of error before giving "
7082 "up. This is unlikely to be an issue unless your "
7083 "cluster is very heavily loaded. Beware that "
7084 "increasing this value may cause some operations to "
7085 "take longer in exceptional cases and thus may, "
7086 "rarely, cause clients to time out."),
7087 Option("rgw_sts_entry", Option::TYPE_STR, Option::LEVEL_ADVANCED)
7088 .set_default("sts")
7089 .set_description("STS URL prefix")
7090 .set_long_description("URL path prefix for internal STS requests."),
7091
7092 Option("rgw_sts_key", Option::TYPE_STR, Option::LEVEL_ADVANCED)
7093 .set_default("sts")
7094 .set_description("STS Key")
7095 .set_long_description("Key used for encrypting/ decrypting session token."),
7096
7097 Option("rgw_s3_auth_use_sts", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
7098 .set_default(false)
7099 .set_description("Should S3 authentication use STS."),
7100
7101 Option("rgw_sts_max_session_duration", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7102 .set_default(43200)
7103 .set_description("Session token max duration")
7104 .set_long_description("Max duration in seconds for which the session token is valid."),
7105
7106 Option("rgw_max_listing_results", Option::TYPE_UINT,
7107 Option::LEVEL_ADVANCED)
7108 .set_default(1000)
7109 .set_min_max(1, 100000)
7110 .add_service("rgw")
7111 .set_description("Upper bound on results in listing operations, ListBucket max-keys")
7112 .set_long_description("This caps the maximum permitted value for listing-like operations in RGW S3. "
7113 "Affects ListBucket(max-keys), "
7114 "ListBucketVersions(max-keys), "
7115 "ListBucketMultipartUploads(max-uploads), "
7116 "ListMultipartUploadParts(max-parts)"),
7117
7118 Option("rgw_sts_token_introspection_url", Option::TYPE_STR, Option::LEVEL_ADVANCED)
7119 .set_default("")
7120 .set_description("STS Web Token introspection URL")
7121 .set_long_description("URL for introspecting an STS Web Token."),
7122
7123 Option("rgw_sts_client_id", Option::TYPE_STR, Option::LEVEL_ADVANCED)
7124 .set_default("")
7125 .set_description("Client Id")
7126 .set_long_description("Client Id needed for introspecting a Web Token."),
7127
7128 Option("rgw_sts_client_secret", Option::TYPE_STR, Option::LEVEL_ADVANCED)
7129 .set_default("")
7130 .set_description("Client Secret")
7131 .set_long_description("Client Secret needed for introspecting a Web Token."),
7132
7133 Option("rgw_max_concurrent_requests", Option::TYPE_INT, Option::LEVEL_BASIC)
7134 .set_default(1024)
7135 .set_description("Maximum number of concurrent HTTP requests.")
7136 .set_long_description(
7137 "Maximum number of concurrent HTTP requests that the beast frontend "
7138 "will process. Tuning this can help to limit memory usage under heavy "
7139 "load.")
7140 .add_tag("performance")
7141 .add_see_also("rgw_frontends"),
7142
7143 Option("rgw_scheduler_type", Option::TYPE_STR, Option::LEVEL_ADVANCED)
7144 .set_default("throttler")
7145 .set_description("Set the type of dmclock scheduler, defaults to throttler "
7146 "Other valid values are dmclock which is experimental"),
7147
7148 Option("rgw_dmclock_admin_res", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7149 .set_default(100.0)
7150 .set_description("mclock reservation for admin requests")
7151 .add_see_also("rgw_dmclock_admin_wgt")
7152 .add_see_also("rgw_dmclock_admin_lim"),
7153
7154 Option("rgw_dmclock_admin_wgt", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7155 .set_default(100.0)
7156 .set_description("mclock weight for admin requests")
7157 .add_see_also("rgw_dmclock_admin_res")
7158 .add_see_also("rgw_dmclock_admin_lim"),
7159
7160 Option("rgw_dmclock_admin_lim", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7161 .set_default(0.0)
7162 .set_description("mclock limit for admin requests")
7163 .add_see_also("rgw_dmclock_admin_res")
7164 .add_see_also("rgw_dmclock_admin_wgt"),
7165
7166 Option("rgw_dmclock_auth_res", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7167 .set_default(200.0)
7168 .set_description("mclock reservation for object data requests")
7169 .add_see_also("rgw_dmclock_auth_wgt")
7170 .add_see_also("rgw_dmclock_auth_lim"),
7171
7172 Option("rgw_dmclock_auth_wgt", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7173 .set_default(100.0)
7174 .set_description("mclock weight for object data requests")
7175 .add_see_also("rgw_dmclock_auth_res")
7176 .add_see_also("rgw_dmclock_auth_lim"),
7177
7178 Option("rgw_dmclock_auth_lim", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7179 .set_default(0.0)
7180 .set_description("mclock limit for object data requests")
7181 .add_see_also("rgw_dmclock_auth_res")
7182 .add_see_also("rgw_dmclock_auth_wgt"),
7183
7184 Option("rgw_dmclock_data_res", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7185 .set_default(500.0)
7186 .set_description("mclock reservation for object data requests")
7187 .add_see_also("rgw_dmclock_data_wgt")
7188 .add_see_also("rgw_dmclock_data_lim"),
7189
7190 Option("rgw_dmclock_data_wgt", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7191 .set_default(500.0)
7192 .set_description("mclock weight for object data requests")
7193 .add_see_also("rgw_dmclock_data_res")
7194 .add_see_also("rgw_dmclock_data_lim"),
7195
7196 Option("rgw_dmclock_data_lim", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7197 .set_default(0.0)
7198 .set_description("mclock limit for object data requests")
7199 .add_see_also("rgw_dmclock_data_res")
7200 .add_see_also("rgw_dmclock_data_wgt"),
7201
7202 Option("rgw_dmclock_metadata_res", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7203 .set_default(500.0)
7204 .set_description("mclock reservation for metadata requests")
7205 .add_see_also("rgw_dmclock_metadata_wgt")
7206 .add_see_also("rgw_dmclock_metadata_lim"),
7207
7208 Option("rgw_dmclock_metadata_wgt", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7209 .set_default(500.0)
7210 .set_description("mclock weight for metadata requests")
7211 .add_see_also("rgw_dmclock_metadata_res")
7212 .add_see_also("rgw_dmclock_metadata_lim"),
7213
7214 Option("rgw_dmclock_metadata_lim", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7215 .set_default(0.0)
7216 .set_description("mclock limit for metadata requests")
7217 .add_see_also("rgw_dmclock_metadata_res")
7218 .add_see_also("rgw_dmclock_metadata_wgt"),
7219
7220 });
7221 }
7222
7223 static std::vector<Option> get_rbd_options() {
7224 return std::vector<Option>({
7225 Option("rbd_default_pool", Option::TYPE_STR, Option::LEVEL_ADVANCED)
7226 .set_default("rbd")
7227 .set_description("default pool for storing new images")
7228 .set_validator([](std::string *value, std::string *error_message){
7229 std::regex pattern("^[^@/]+$");
7230 if (!std::regex_match (*value, pattern)) {
7231 *value = "rbd";
7232 *error_message = "invalid RBD default pool, resetting to 'rbd'";
7233 }
7234 return 0;
7235 }),
7236
7237 Option("rbd_default_data_pool", Option::TYPE_STR, Option::LEVEL_ADVANCED)
7238 .set_default("")
7239 .set_description("default pool for storing data blocks for new images")
7240 .set_validator([](std::string *value, std::string *error_message){
7241 std::regex pattern("^[^@/]*$");
7242 if (!std::regex_match (*value, pattern)) {
7243 *value = "";
7244 *error_message = "ignoring invalid RBD data pool";
7245 }
7246 return 0;
7247 }),
7248
7249 Option("rbd_default_features", Option::TYPE_STR, Option::LEVEL_ADVANCED)
7250 .set_default("layering,exclusive-lock,object-map,fast-diff,deep-flatten")
7251 .set_description("default v2 image features for new images")
7252 .set_long_description(
7253 "RBD features are only applicable for v2 images. This setting accepts "
7254 "either an integer bitmask value or comma-delimited string of RBD "
7255 "feature names. This setting is always internally stored as an integer "
7256 "bitmask value. The mapping between feature bitmask value and feature "
7257 "name is as follows: +1 -> layering, +2 -> striping, "
7258 "+4 -> exclusive-lock, +8 -> object-map, +16 -> fast-diff, "
7259 "+32 -> deep-flatten, +64 -> journaling, +128 -> data-pool")
7260 .set_flag(Option::FLAG_RUNTIME)
7261 .set_validator([](std::string *value, std::string *error_message) {
7262 ostringstream ss;
7263 uint64_t features = librbd::rbd_features_from_string(*value, &ss);
7264 // Leave this in integer form to avoid breaking Cinder. Someday
7265 // we would like to present this in string form instead...
7266 *value = stringify(features);
7267 if (ss.str().size()) {
7268 return -EINVAL;
7269 }
7270 return 0;
7271 }),
7272
7273 Option("rbd_op_threads", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7274 .set_default(1)
7275 .set_description("number of threads to utilize for internal processing"),
7276
7277 Option("rbd_op_thread_timeout", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7278 .set_default(60)
7279 .set_description("time in seconds for detecting a hung thread"),
7280
7281 Option("rbd_non_blocking_aio", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
7282 .set_default(true)
7283 .set_description("process AIO ops from a dispatch thread to prevent blocking"),
7284
7285 Option("rbd_cache", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
7286 .set_default(true)
7287 .set_description("whether to enable caching (writeback unless rbd_cache_max_dirty is 0)"),
7288
7289 Option("rbd_cache_policy", Option::TYPE_STR, Option::LEVEL_ADVANCED)
7290 .set_enum_allowed({"writethrough", "writeback", "writearound"})
7291 .set_default("writearound")
7292 .set_description("cache policy for handling writes."),
7293
7294 Option("rbd_cache_writethrough_until_flush", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
7295 .set_default(true)
7296 .set_description("whether to make writeback caching writethrough until "
7297 "flush is called, to be sure the user of librbd will send "
7298 "flushes so that writeback is safe"),
7299
7300 Option("rbd_cache_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
7301 .set_default(32_M)
7302 .set_description("cache size in bytes"),
7303
7304 Option("rbd_cache_max_dirty", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
7305 .set_default(24_M)
7306 .set_description("dirty limit in bytes - set to 0 for write-through caching"),
7307
7308 Option("rbd_cache_target_dirty", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
7309 .set_default(16_M)
7310 .set_description("target dirty limit in bytes"),
7311
7312 Option("rbd_cache_max_dirty_age", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7313 .set_default(1.0)
7314 .set_description("seconds in cache before writeback starts"),
7315
7316 Option("rbd_cache_max_dirty_object", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7317 .set_default(0)
7318 .set_description("dirty limit for objects - set to 0 for auto calculate from rbd_cache_size"),
7319
7320 Option("rbd_cache_block_writes_upfront", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
7321 .set_default(false)
7322 .set_description("whether to block writes to the cache before the aio_write call completes"),
7323
7324 Option("rbd_parent_cache_enabled", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
7325 .set_default(false)
7326 .set_description("whether to enable rbd shared ro cache"),
7327
7328 Option("rbd_concurrent_management_ops", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7329 .set_default(10)
7330 .set_min(1)
7331 .set_description("how many operations can be in flight for a management operation like deleting or resizing an image"),
7332
7333 Option("rbd_balance_snap_reads", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
7334 .set_default(false)
7335 .set_description("distribute snap read requests to random OSD"),
7336
7337 Option("rbd_localize_snap_reads", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
7338 .set_default(false)
7339 .set_description("localize snap read requests to closest OSD"),
7340
7341 Option("rbd_balance_parent_reads", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
7342 .set_default(false)
7343 .set_description("distribute parent read requests to random OSD"),
7344
7345 Option("rbd_localize_parent_reads", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
7346 .set_default(false)
7347 .set_description("localize parent requests to closest OSD"),
7348
7349 Option("rbd_sparse_read_threshold_bytes", Option::TYPE_SIZE,
7350 Option::LEVEL_ADVANCED)
7351 .set_default(64_K)
7352 .set_description("threshold for issuing a sparse-read")
7353 .set_long_description("minimum number of sequential bytes to read against "
7354 "an object before issuing a sparse-read request to "
7355 "the cluster. 0 implies it must be a full object read "
7356 "to issue a sparse-read, 1 implies always use "
7357 "sparse-read, and any value larger than the maximum "
7358 "object size will disable sparse-read for all "
7359 "requests"),
7360
7361 Option("rbd_readahead_trigger_requests", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7362 .set_default(10)
7363 .set_description("number of sequential requests necessary to trigger readahead"),
7364
7365 Option("rbd_readahead_max_bytes", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
7366 .set_default(512_K)
7367 .set_description("set to 0 to disable readahead"),
7368
7369 Option("rbd_readahead_disable_after_bytes", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
7370 .set_default(50_M)
7371 .set_description("how many bytes are read in total before readahead is disabled"),
7372
7373 Option("rbd_clone_copy_on_read", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
7374 .set_default(false)
7375 .set_description("copy-up parent image blocks to clone upon read request"),
7376
7377 Option("rbd_blacklist_on_break_lock", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
7378 .set_default(true)
7379 .set_description("whether to blacklist clients whose lock was broken"),
7380
7381 Option("rbd_blacklist_expire_seconds", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7382 .set_default(0)
7383 .set_description("number of seconds to blacklist - set to 0 for OSD default"),
7384
7385 Option("rbd_request_timed_out_seconds", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7386 .set_default(30)
7387 .set_description("number of seconds before maintenance request times out"),
7388
7389 Option("rbd_skip_partial_discard", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
7390 .set_default(true)
7391 .set_description("skip discard (zero) of unaligned extents within an object"),
7392
7393 Option("rbd_discard_granularity_bytes", Option::TYPE_UINT,
7394 Option::LEVEL_ADVANCED)
7395 .set_default(64_K)
7396 .set_min_max(4_K, 32_M)
7397 .set_validator([](std::string *value, std::string *error_message){
7398 uint64_t f = strict_si_cast<uint64_t>(value->c_str(), error_message);
7399 if (!error_message->empty()) {
7400 return -EINVAL;
7401 } else if (!isp2(f)) {
7402 *error_message = "value must be a power of two";
7403 return -EINVAL;
7404 }
7405 return 0;
7406 })
7407 .set_description("minimum aligned size of discard operations"),
7408
7409 Option("rbd_enable_alloc_hint", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
7410 .set_default(true)
7411 .set_description("when writing a object, it will issue a hint to osd backend to indicate the expected size object need"),
7412
7413 Option("rbd_tracing", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
7414 .set_default(false)
7415 .set_description("true if LTTng-UST tracepoints should be enabled"),
7416
7417 Option("rbd_blkin_trace_all", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
7418 .set_default(false)
7419 .set_description("create a blkin trace for all RBD requests"),
7420
7421 Option("rbd_validate_pool", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
7422 .set_default(true)
7423 .set_description("validate empty pools for RBD compatibility"),
7424
7425 Option("rbd_validate_names", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
7426 .set_default(true)
7427 .set_description("validate new image names for RBD compatibility"),
7428
7429 Option("rbd_auto_exclusive_lock_until_manual_request", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
7430 .set_default(true)
7431 .set_description("automatically acquire/release exclusive lock until it is explicitly requested"),
7432
7433 Option("rbd_move_to_trash_on_remove", Option::TYPE_BOOL, Option::LEVEL_BASIC)
7434 .set_default(false)
7435 .set_description("automatically move images to the trash when deleted"),
7436
7437 Option("rbd_move_to_trash_on_remove_expire_seconds", Option::TYPE_UINT, Option::LEVEL_BASIC)
7438 .set_default(0)
7439 .set_description("default number of seconds to protect deleted images in the trash"),
7440
7441 Option("rbd_move_parent_to_trash_on_remove", Option::TYPE_BOOL, Option::LEVEL_BASIC)
7442 .set_default(false)
7443 .set_description("move parent with clone format v2 children to the trash when deleted"),
7444
7445 Option("rbd_mirroring_resync_after_disconnect", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
7446 .set_default(false)
7447 .set_description("automatically start image resync after mirroring is disconnected due to being laggy"),
7448
7449 Option("rbd_mirroring_delete_delay", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7450 .set_default(0)
7451 .set_description("time-delay in seconds for rbd-mirror delete propagation"),
7452
7453 Option("rbd_mirroring_replay_delay", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7454 .set_default(0)
7455 .set_description("time-delay in seconds for rbd-mirror asynchronous replication"),
7456
7457 Option("rbd_default_format", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7458 .set_default(2)
7459 .set_description("default image format for new images"),
7460
7461 Option("rbd_default_order", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7462 .set_default(22)
7463 .set_description("default order (data block object size) for new images"),
7464
7465 Option("rbd_default_stripe_count", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7466 .set_default(0)
7467 .set_description("default stripe count for new images"),
7468
7469 Option("rbd_default_stripe_unit", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
7470 .set_default(0)
7471 .set_description("default stripe width for new images"),
7472
7473 Option("rbd_default_map_options", Option::TYPE_STR, Option::LEVEL_ADVANCED)
7474 .set_default("")
7475 .set_description("default krbd map options"),
7476
7477 Option("rbd_default_clone_format", Option::TYPE_STR, Option::LEVEL_ADVANCED)
7478 .set_enum_allowed({"1", "2", "auto"})
7479 .set_default("auto")
7480 .set_description("default internal format for handling clones")
7481 .set_long_description("This sets the internal format for tracking cloned "
7482 "images. The setting of '1' requires attaching to "
7483 "protected snapshots that cannot be removed until "
7484 "the clone is removed/flattened. The setting of '2' "
7485 "will allow clones to be attached to any snapshot "
7486 "and permits removing in-use parent snapshots but "
7487 "requires Mimic or later clients. The default "
7488 "setting of 'auto' will use the v2 format if the "
7489 "cluster is configured to require mimic or later "
7490 "clients.")
7491 .set_flag(Option::FLAG_RUNTIME),
7492
7493 Option("rbd_journal_order", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7494 .set_min_max(12, 26)
7495 .set_default(24)
7496 .set_description("default order (object size) for journal data objects"),
7497
7498 Option("rbd_journal_splay_width", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7499 .set_default(4)
7500 .set_description("number of active journal objects"),
7501
7502 Option("rbd_journal_commit_age", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7503 .set_default(5)
7504 .set_description("commit time interval, seconds"),
7505
7506 Option("rbd_journal_object_writethrough_until_flush", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
7507 .set_default(true)
7508 .set_description("when enabled, the rbd_journal_object_flush* configuration "
7509 "options are ignored until the first flush so that batched "
7510 "journal IO is known to be safe for consistency"),
7511
7512 Option("rbd_journal_object_flush_interval", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7513 .set_default(0)
7514 .set_description("maximum number of pending commits per journal object"),
7515
7516 Option("rbd_journal_object_flush_bytes", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
7517 .set_default(1_M)
7518 .set_description("maximum number of pending bytes per journal object"),
7519
7520 Option("rbd_journal_object_flush_age", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7521 .set_default(0)
7522 .set_description("maximum age (in seconds) for pending commits"),
7523
7524 Option("rbd_journal_object_max_in_flight_appends", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7525 .set_default(0)
7526 .set_description("maximum number of in-flight appends per journal object"),
7527
7528 Option("rbd_journal_pool", Option::TYPE_STR, Option::LEVEL_ADVANCED)
7529 .set_default("")
7530 .set_description("pool for journal objects"),
7531
7532 Option("rbd_journal_max_payload_bytes", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
7533 .set_default(16384)
7534 .set_description("maximum journal payload size before splitting"),
7535
7536 Option("rbd_journal_max_concurrent_object_sets", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7537 .set_default(0)
7538 .set_description("maximum number of object sets a journal client can be behind before it is automatically unregistered"),
7539
7540 Option("rbd_qos_iops_limit", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7541 .set_default(0)
7542 .set_description("the desired limit of IO operations per second"),
7543
7544 Option("rbd_qos_bps_limit", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7545 .set_default(0)
7546 .set_description("the desired limit of IO bytes per second"),
7547
7548 Option("rbd_qos_read_iops_limit", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7549 .set_default(0)
7550 .set_description("the desired limit of read operations per second"),
7551
7552 Option("rbd_qos_write_iops_limit", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7553 .set_default(0)
7554 .set_description("the desired limit of write operations per second"),
7555
7556 Option("rbd_qos_read_bps_limit", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7557 .set_default(0)
7558 .set_description("the desired limit of read bytes per second"),
7559
7560 Option("rbd_qos_write_bps_limit", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7561 .set_default(0)
7562 .set_description("the desired limit of write bytes per second"),
7563
7564 Option("rbd_qos_iops_burst", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7565 .set_default(0)
7566 .set_description("the desired burst limit of IO operations"),
7567
7568 Option("rbd_qos_bps_burst", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7569 .set_default(0)
7570 .set_description("the desired burst limit of IO bytes"),
7571
7572 Option("rbd_qos_read_iops_burst", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7573 .set_default(0)
7574 .set_description("the desired burst limit of read operations"),
7575
7576 Option("rbd_qos_write_iops_burst", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7577 .set_default(0)
7578 .set_description("the desired burst limit of write operations"),
7579
7580 Option("rbd_qos_read_bps_burst", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7581 .set_default(0)
7582 .set_description("the desired burst limit of read bytes"),
7583
7584 Option("rbd_qos_write_bps_burst", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7585 .set_default(0)
7586 .set_description("the desired burst limit of write bytes"),
7587
7588 Option("rbd_qos_schedule_tick_min", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7589 .set_default(50)
7590 .set_min(1)
7591 .set_description("minimum schedule tick (in milliseconds) for QoS"),
7592
7593 Option("rbd_discard_on_zeroed_write_same", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
7594 .set_default(true)
7595 .set_description("discard data on zeroed write same instead of writing zero"),
7596
7597 Option("rbd_mtime_update_interval", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7598 .set_default(60)
7599 .set_min(0)
7600 .set_description("RBD Image modify timestamp refresh interval. Set to 0 to disable modify timestamp update."),
7601
7602 Option("rbd_atime_update_interval", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7603 .set_default(60)
7604 .set_min(0)
7605 .set_description("RBD Image access timestamp refresh interval. Set to 0 to disable access timestamp update."),
7606
7607 Option("rbd_io_scheduler", Option::TYPE_STR, Option::LEVEL_ADVANCED)
7608 .set_default("simple")
7609 .set_enum_allowed({"none", "simple"})
7610 .set_description("RBD IO scheduler"),
7611
7612 Option("rbd_io_scheduler_simple_max_delay", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7613 .set_default(0)
7614 .set_min(0)
7615 .set_description("maximum io delay (in milliseconds) for simple io scheduler (if set to 0 dalay is calculated based on latency stats)"),
7616 });
7617 }
7618
7619 static std::vector<Option> get_rbd_mirror_options() {
7620 return std::vector<Option>({
7621 Option("rbd_mirror_journal_commit_age", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7622 .set_default(5)
7623 .set_description("commit time interval, seconds"),
7624
7625 Option("rbd_mirror_journal_poll_age", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7626 .set_default(5)
7627 .set_description("maximum age (in seconds) between successive journal polls"),
7628
7629 Option("rbd_mirror_sync_point_update_age", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7630 .set_default(30)
7631 .set_description("number of seconds between each update of the image sync point object number"),
7632
7633 Option("rbd_mirror_concurrent_image_syncs", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7634 .set_default(5)
7635 .set_description("maximum number of image syncs in parallel"),
7636
7637 Option("rbd_mirror_pool_replayers_refresh_interval", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7638 .set_default(30)
7639 .set_description("interval to refresh peers in rbd-mirror daemon"),
7640
7641 Option("rbd_mirror_concurrent_image_deletions", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7642 .set_default(1)
7643 .set_min(1)
7644 .set_description("maximum number of image deletions in parallel"),
7645
7646 Option("rbd_mirror_delete_retry_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7647 .set_default(30)
7648 .set_description("interval to check and retry the failed deletion requests"),
7649
7650 Option("rbd_mirror_image_state_check_interval", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7651 .set_default(30)
7652 .set_min(1)
7653 .set_description("interval to get images from pool watcher and set sources in replayer"),
7654
7655 Option("rbd_mirror_leader_heartbeat_interval", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7656 .set_default(5)
7657 .set_min(1)
7658 .set_description("interval (in seconds) between mirror leader heartbeats"),
7659
7660 Option("rbd_mirror_leader_max_missed_heartbeats", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7661 .set_default(2)
7662 .set_description("number of missed heartbeats for non-lock owner to attempt to acquire lock"),
7663
7664 Option("rbd_mirror_leader_max_acquire_attempts_before_break", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7665 .set_default(3)
7666 .set_description("number of failed attempts to acquire lock after missing heartbeats before breaking lock"),
7667
7668 Option("rbd_mirror_image_policy_type", Option::TYPE_STR, Option::LEVEL_ADVANCED)
7669 .set_default("simple")
7670 .set_enum_allowed({"none", "simple"})
7671 .set_description("active/active policy type for mapping images to instances"),
7672
7673 Option("rbd_mirror_image_policy_migration_throttle", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7674 .set_default(300)
7675 .set_description("number of seconds after which an image can be reshuffled (migrated) again"),
7676
7677 Option("rbd_mirror_image_policy_update_throttle_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7678 .set_default(1)
7679 .set_min(1)
7680 .set_description("interval (in seconds) to throttle images for mirror daemon peer updates"),
7681
7682 Option("rbd_mirror_image_policy_rebalance_timeout", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7683 .set_default(0)
7684 .set_description("number of seconds policy should be idle before trigerring reshuffle (rebalance) of images"),
7685
7686 Option("rbd_mirror_perf_stats_prio", Option::TYPE_INT, Option::LEVEL_ADVANCED)
7687 .set_default((int64_t)PerfCountersBuilder::PRIO_USEFUL)
7688 .set_description("Priority level for mirror daemon replication perf counters")
7689 .set_long_description("The daemon will send perf counter data to the "
7690 "manager daemon if the priority is not lower than "
7691 "mgr_stats_threshold.")
7692 .set_min_max((int64_t)PerfCountersBuilder::PRIO_DEBUGONLY,
7693 (int64_t)PerfCountersBuilder::PRIO_CRITICAL + 1),
7694
7695 Option("rbd_mirror_memory_autotune", Option::TYPE_BOOL, Option::LEVEL_DEV)
7696 .set_default(true)
7697 .add_see_also("rbd_mirror_memory_target")
7698 .set_description("Automatically tune the ratio of caches while respecting min values."),
7699
7700 Option("rbd_mirror_memory_target", Option::TYPE_SIZE, Option::LEVEL_BASIC)
7701 .set_default(4_G)
7702 .add_see_also("rbd_mirror_memory_autotune")
7703 .set_description("When tcmalloc and cache autotuning is enabled, try to keep this many bytes mapped in memory."),
7704
7705 Option("rbd_mirror_memory_base", Option::TYPE_SIZE, Option::LEVEL_DEV)
7706 .set_default(768_M)
7707 .add_see_also("rbd_mirror_memory_autotune")
7708 .set_description("When tcmalloc and cache autotuning is enabled, estimate the minimum amount of memory in bytes the rbd-mirror daemon will need."),
7709
7710 Option("rbd_mirror_memory_expected_fragmentation", Option::TYPE_FLOAT, Option::LEVEL_DEV)
7711 .set_default(0.15)
7712 .set_min_max(0.0, 1.0)
7713 .add_see_also("rbd_mirror_memory_autotune")
7714 .set_description("When tcmalloc and cache autotuning is enabled, estimate the percent of memory fragmentation."),
7715
7716 Option("rbd_mirror_memory_cache_min", Option::TYPE_SIZE, Option::LEVEL_DEV)
7717 .set_default(128_M)
7718 .add_see_also("rbd_mirror_memory_autotune")
7719 .set_description("When tcmalloc and cache autotuning is enabled, set the minimum amount of memory used for cache."),
7720
7721 Option("rbd_mirror_memory_cache_resize_interval", Option::TYPE_FLOAT, Option::LEVEL_DEV)
7722 .set_default(5)
7723 .add_see_also("rbd_mirror_memory_autotune")
7724 .set_description("When tcmalloc and cache autotuning is enabled, wait this many seconds between resizing caches."),
7725
7726 Option("rbd_mirror_memory_cache_autotune_interval", Option::TYPE_FLOAT, Option::LEVEL_DEV)
7727 .set_default(30)
7728 .add_see_also("rbd_mirror_memory_autotune")
7729 .set_description("The number of seconds to wait between rebalances when cache autotune is enabled."),
7730 });
7731 }
7732
7733 static std::vector<Option> get_immutable_object_cache_options() {
7734 return std::vector<Option>({
7735 Option("immutable_object_cache_path", Option::TYPE_STR, Option::LEVEL_ADVANCED)
7736 .set_default("/tmp")
7737 .set_description("immutable object cache data dir"),
7738
7739 Option("immutable_object_cache_sock", Option::TYPE_STR, Option::LEVEL_ADVANCED)
7740 .set_default("/var/run/ceph/immutable_object_cache_sock")
7741 .set_description("immutable object cache domain socket"),
7742
7743 Option("immutable_object_cache_max_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
7744 .set_default(1_G)
7745 .set_description("max immutable object cache data size"),
7746
7747 Option("immutable_object_cache_max_inflight_ops", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7748 .set_default(128)
7749 .set_description("max inflight promoting requests for immutable object cache daemon"),
7750
7751 Option("immutable_object_cache_client_dedicated_thread_num", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7752 .set_default(2)
7753 .set_description("immutable object cache client dedicated thread number"),
7754
7755 Option("immutable_object_cache_watermark", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7756 .set_default(0.1)
7757 .set_description("immutable object cache water mark"),
7758 });
7759 }
7760
7761 std::vector<Option> get_mds_options() {
7762 return std::vector<Option>({
7763 Option("mds_data", Option::TYPE_STR, Option::LEVEL_ADVANCED)
7764 .set_default("/var/lib/ceph/mds/$cluster-$id")
7765 .set_flag(Option::FLAG_NO_MON_UPDATE)
7766 .set_description("path to MDS data and keyring"),
7767
7768 Option("mds_max_xattr_pairs_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
7769 .set_default(64_K)
7770 .set_description("maximum aggregate size of extended attributes on a file"),
7771
7772 Option("mds_cache_trim_interval", Option::TYPE_SECS, Option::LEVEL_ADVANCED)
7773 .set_default(1)
7774 .set_description("interval in seconds between cache trimming"),
7775
7776 Option("mds_cache_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
7777 .set_default(0)
7778 .set_description("maximum number of inodes in MDS cache (<=0 is unlimited)")
7779 .set_long_description("This tunable is no longer recommended. Use mds_cache_memory_limit."),
7780
7781 Option("mds_cache_memory_limit", Option::TYPE_SIZE, Option::LEVEL_BASIC)
7782 .set_default(1*(1LL<<30))
7783 .set_description("target maximum memory usage of MDS cache")
7784 .set_long_description("This sets a target maximum memory usage of the MDS cache and is the primary tunable to limit the MDS memory usage. The MDS will try to stay under a reservation of this limit (by default 95%; 1 - mds_cache_reservation) by trimming unused metadata in its cache and recalling cached items in the client caches. It is possible for the MDS to exceed this limit due to slow recall from clients. The mds_health_cache_threshold (150%) sets a cache full threshold for when the MDS signals a cluster health warning."),
7785
7786 Option("mds_cache_reservation", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7787 .set_default(.05)
7788 .set_description("amount of memory to reserve for future cached objects"),
7789
7790 Option("mds_health_cache_threshold", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7791 .set_default(1.5)
7792 .set_description("threshold for cache size to generate health warning"),
7793
7794 Option("mds_cache_mid", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7795 .set_default(.7)
7796 .set_description("midpoint for MDS cache LRU"),
7797
7798 Option("mds_cache_trim_decay_rate", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7799 .set_default(1)
7800 .set_description("decay rate for trimming MDS cache throttle"),
7801
7802 Option("mds_cache_trim_threshold", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
7803 .set_default(64_K)
7804 .set_description("threshold for number of dentries that can be trimmed"),
7805
7806 Option("mds_max_file_recover", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7807 .set_default(32)
7808 .set_description("maximum number of files to recover file sizes in parallel"),
7809
7810 Option("mds_dir_max_commit_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
7811 .set_default(10)
7812 .set_description("maximum size in megabytes for a RADOS write to a directory"),
7813
7814 Option("mds_dir_keys_per_op", Option::TYPE_INT, Option::LEVEL_ADVANCED)
7815 .set_default(16384)
7816 .set_description("number of directory entries to read in one RADOS operation"),
7817
7818 Option("mds_decay_halflife", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7819 .set_default(5)
7820 .set_description("rate of decay for temperature counters on each directory for balancing"),
7821
7822 Option("mds_beacon_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7823 .set_default(4)
7824 .set_description("interval in seconds between MDS beacons to monitors"),
7825
7826 Option("mds_beacon_grace", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7827 .set_default(15)
7828 .set_description("tolerance in seconds for missed MDS beacons to monitors"),
7829
7830 Option("mds_heartbeat_grace", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7831 .set_default(15)
7832 .set_description("tolerance in seconds for MDS internal heartbeat"),
7833
7834 Option("mds_enforce_unique_name", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
7835 .set_default(true)
7836 .set_description("require MDS name is unique in the cluster"),
7837
7838 Option("mds_session_blacklist_on_timeout", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
7839 .set_default(true)
7840 .set_description("blacklist clients whose sessions have become stale"),
7841
7842 Option("mds_session_blacklist_on_evict", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
7843 .set_default(true)
7844 .set_description("blacklist clients that have been evicted"),
7845
7846 Option("mds_sessionmap_keys_per_op", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7847 .set_default(1024)
7848 .set_description("number of omap keys to read from the SessionMap in one operation"),
7849
7850 Option("mds_recall_max_caps", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
7851 .set_default(5000)
7852 .set_description("maximum number of caps to recall from client session in single recall"),
7853
7854 Option("mds_recall_max_decay_rate", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7855 .set_default(2.5)
7856 .set_description("decay rate for throttle on recalled caps on a session"),
7857
7858 Option("mds_recall_max_decay_threshold", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
7859 .set_default(16_K)
7860 .set_description("decay threshold for throttle on recalled caps on a session"),
7861
7862 Option("mds_recall_global_max_decay_threshold", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
7863 .set_default(64_K)
7864 .set_description("decay threshold for throttle on recalled caps globally"),
7865
7866 Option("mds_recall_warning_threshold", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
7867 .set_default(32_K)
7868 .set_description("decay threshold for warning on slow session cap recall"),
7869
7870 Option("mds_recall_warning_decay_rate", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7871 .set_default(60.0)
7872 .set_description("decay rate for warning on slow session cap recall"),
7873
7874 Option("mds_session_cache_liveness_decay_rate", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7875 .add_see_also("mds_session_cache_liveness_magnitude")
7876 .set_default(5_min)
7877 .set_description("decay rate for session liveness leading to preemptive cap recall")
7878 .set_long_description("This determines how long a session needs to be quiescent before the MDS begins preemptively recalling capabilities. The default of 5 minutes will cause 10 halvings of the decay counter after 1 hour, or 1/1024. The default magnitude of 10 (1^10 or 1024) is chosen so that the MDS considers a previously chatty session (approximately) to be quiescent after 1 hour."),
7879
7880 Option("mds_session_cache_liveness_magnitude", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
7881 .add_see_also("mds_session_cache_liveness_decay_rate")
7882 .set_default(10)
7883 .set_description("decay magnitude for preemptively recalling caps on quiet client")
7884 .set_long_description("This is the order of magnitude difference (in base 2) of the internal liveness decay counter and the number of capabilities the session holds. When this difference occurs, the MDS treats the session as quiescent and begins recalling capabilities."),
7885
7886 Option("mds_freeze_tree_timeout", Option::TYPE_FLOAT, Option::LEVEL_DEV)
7887 .set_default(30)
7888 .set_description(""),
7889
7890 Option("mds_health_summarize_threshold", Option::TYPE_INT, Option::LEVEL_ADVANCED)
7891 .set_default(10)
7892 .set_description("threshold of number of clients to summarize late client recall"),
7893
7894 Option("mds_reconnect_timeout", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7895 .set_default(45)
7896 .set_description("timeout in seconds to wait for clients to reconnect during MDS reconnect recovery state"),
7897
7898 Option("mds_tick_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7899 .set_default(5)
7900 .set_description("time in seconds between upkeep tasks"),
7901
7902 Option("mds_dirstat_min_interval", Option::TYPE_FLOAT, Option::LEVEL_DEV)
7903 .set_default(1)
7904 .set_description(""),
7905
7906 Option("mds_scatter_nudge_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7907 .set_default(5)
7908 .set_description("minimum interval between scatter lock updates"),
7909
7910 Option("mds_client_prealloc_inos", Option::TYPE_INT, Option::LEVEL_ADVANCED)
7911 .set_default(1000)
7912 .set_description("number of unused inodes to pre-allocate to clients for file creation"),
7913
7914 Option("mds_early_reply", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
7915 .set_default(true)
7916 .set_description("additional reply to clients that metadata requests are complete but not yet durable"),
7917
7918 Option("mds_default_dir_hash", Option::TYPE_INT, Option::LEVEL_ADVANCED)
7919 .set_default(CEPH_STR_HASH_RJENKINS)
7920 .set_description("hash function to select directory fragment for dentry name"),
7921
7922 Option("mds_log_pause", Option::TYPE_BOOL, Option::LEVEL_DEV)
7923 .set_default(false)
7924 .set_description(""),
7925
7926 Option("mds_log_skip_corrupt_events", Option::TYPE_BOOL, Option::LEVEL_DEV)
7927 .set_default(false)
7928 .set_description(""),
7929
7930 Option("mds_log_max_events", Option::TYPE_INT, Option::LEVEL_ADVANCED)
7931 .set_default(-1)
7932 .set_description("maximum number of events in the MDS journal (-1 is unlimited)"),
7933
7934 Option("mds_log_events_per_segment", Option::TYPE_INT, Option::LEVEL_ADVANCED)
7935 .set_default(1024)
7936 .set_description("maximum number of events in an MDS journal segment"),
7937
7938 Option("mds_log_segment_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
7939 .set_default(0)
7940 .set_description("size in bytes of each MDS log segment"),
7941
7942 Option("mds_log_max_segments", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
7943 .set_default(128)
7944 .set_description("maximum number of segments which may be untrimmed"),
7945
7946 Option("mds_bal_export_pin", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
7947 .set_default(true)
7948 .set_description("allow setting directory export pins to particular ranks"),
7949
7950 Option("mds_bal_sample_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7951 .set_default(3.0)
7952 .set_description("interval in seconds between balancer ticks"),
7953
7954 Option("mds_bal_replicate_threshold", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7955 .set_default(8000)
7956 .set_description("hot popularity threshold to replicate a subtree"),
7957
7958 Option("mds_bal_unreplicate_threshold", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7959 .set_default(0)
7960 .set_description("cold popularity threshold to merge subtrees"),
7961
7962 Option("mds_bal_split_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
7963 .set_default(10000)
7964 .set_description("minimum size of directory fragment before splitting"),
7965
7966 Option("mds_bal_split_rd", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7967 .set_default(25000)
7968 .set_description("hot read popularity threshold for splitting a directory fragment"),
7969
7970 Option("mds_bal_split_wr", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7971 .set_default(10000)
7972 .set_description("hot write popularity threshold for splitting a directory fragment"),
7973
7974 Option("mds_bal_split_bits", Option::TYPE_INT, Option::LEVEL_ADVANCED)
7975 .set_default(3)
7976 .set_min_max(1, 24)
7977 .set_description("power of two child fragments for a fragment on split"),
7978
7979 Option("mds_bal_merge_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
7980 .set_default(50)
7981 .set_description("size of fragments where merging should occur"),
7982
7983 Option("mds_bal_interval", Option::TYPE_INT, Option::LEVEL_ADVANCED)
7984 .set_default(10)
7985 .set_description("interval between MDS balancer cycles"),
7986
7987 Option("mds_bal_fragment_interval", Option::TYPE_INT, Option::LEVEL_ADVANCED)
7988 .set_default(5)
7989 .set_description("delay in seconds before interrupting client IO to perform splits"),
7990
7991 Option("mds_bal_fragment_size_max", Option::TYPE_INT, Option::LEVEL_ADVANCED)
7992 .set_default(10000*10)
7993 .set_description("maximum size of a directory fragment before new creat/links fail"),
7994
7995 Option("mds_bal_fragment_fast_factor", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
7996 .set_default(1.5)
7997 .set_description("ratio of mds_bal_split_size at which fast fragment splitting occurs"),
7998
7999 Option("mds_bal_fragment_dirs", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
8000 .set_default(true)
8001 .set_description("enable directory fragmentation")
8002 .set_long_description("Directory fragmentation is a standard feature of CephFS that allows sharding directories across multiple objects for performance and stability. Additionally, this allows fragments to be distributed across multiple active MDSs to increase throughput. Disabling (new) fragmentation should only be done in exceptional circumstances and may lead to performance issues."),
8003
8004 Option("mds_bal_idle_threshold", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
8005 .set_default(0)
8006 .set_description("idle metadata popularity threshold before rebalancing"),
8007
8008 Option("mds_bal_max", Option::TYPE_INT, Option::LEVEL_DEV)
8009 .set_default(-1)
8010 .set_description(""),
8011
8012 Option("mds_bal_max_until", Option::TYPE_INT, Option::LEVEL_DEV)
8013 .set_default(-1)
8014 .set_description(""),
8015
8016 Option("mds_bal_mode", Option::TYPE_INT, Option::LEVEL_DEV)
8017 .set_default(0)
8018 .set_description(""),
8019
8020 Option("mds_bal_min_rebalance", Option::TYPE_FLOAT, Option::LEVEL_DEV)
8021 .set_default(.1)
8022 .set_description("amount overloaded over internal target before balancer begins offloading"),
8023
8024 Option("mds_bal_min_start", Option::TYPE_FLOAT, Option::LEVEL_DEV)
8025 .set_default(.2)
8026 .set_description(""),
8027
8028 Option("mds_bal_need_min", Option::TYPE_FLOAT, Option::LEVEL_DEV)
8029 .set_default(.8)
8030 .set_description(""),
8031
8032 Option("mds_bal_need_max", Option::TYPE_FLOAT, Option::LEVEL_DEV)
8033 .set_default(1.2)
8034 .set_description(""),
8035
8036 Option("mds_bal_midchunk", Option::TYPE_FLOAT, Option::LEVEL_DEV)
8037 .set_default(.3)
8038 .set_description(""),
8039
8040 Option("mds_bal_minchunk", Option::TYPE_FLOAT, Option::LEVEL_DEV)
8041 .set_default(.001)
8042 .set_description(""),
8043
8044 Option("mds_bal_target_decay", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
8045 .set_default(10.0)
8046 .set_description("rate of decay for export targets communicated to clients"),
8047
8048 Option("mds_replay_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
8049 .set_default(1.0)
8050 .set_description("time in seconds between replay of updates to journal by standby replay MDS"),
8051
8052 Option("mds_shutdown_check", Option::TYPE_INT, Option::LEVEL_DEV)
8053 .set_default(0)
8054 .set_description(""),
8055
8056 Option("mds_thrash_exports", Option::TYPE_INT, Option::LEVEL_DEV)
8057 .set_default(0)
8058 .set_description(""),
8059
8060 Option("mds_thrash_fragments", Option::TYPE_INT, Option::LEVEL_DEV)
8061 .set_default(0)
8062 .set_description(""),
8063
8064 Option("mds_dump_cache_on_map", Option::TYPE_BOOL, Option::LEVEL_DEV)
8065 .set_default(false)
8066 .set_description(""),
8067
8068 Option("mds_dump_cache_after_rejoin", Option::TYPE_BOOL, Option::LEVEL_DEV)
8069 .set_default(false)
8070 .set_description(""),
8071
8072 Option("mds_verify_scatter", Option::TYPE_BOOL, Option::LEVEL_DEV)
8073 .set_default(false)
8074 .set_description(""),
8075
8076 Option("mds_debug_scatterstat", Option::TYPE_BOOL, Option::LEVEL_DEV)
8077 .set_default(false)
8078 .set_description(""),
8079
8080 Option("mds_debug_frag", Option::TYPE_BOOL, Option::LEVEL_DEV)
8081 .set_default(false)
8082 .set_description(""),
8083
8084 Option("mds_debug_auth_pins", Option::TYPE_BOOL, Option::LEVEL_DEV)
8085 .set_default(false)
8086 .set_description(""),
8087
8088 Option("mds_debug_subtrees", Option::TYPE_BOOL, Option::LEVEL_DEV)
8089 .set_default(false)
8090 .set_description(""),
8091
8092 Option("mds_kill_mdstable_at", Option::TYPE_INT, Option::LEVEL_DEV)
8093 .set_default(0)
8094 .set_description(""),
8095
8096 Option("mds_max_export_size", Option::TYPE_SIZE, Option::LEVEL_DEV)
8097 .set_default(20_M)
8098 .set_description(""),
8099
8100 Option("mds_kill_export_at", Option::TYPE_INT, Option::LEVEL_DEV)
8101 .set_default(0)
8102 .set_description(""),
8103
8104 Option("mds_kill_import_at", Option::TYPE_INT, Option::LEVEL_DEV)
8105 .set_default(0)
8106 .set_description(""),
8107
8108 Option("mds_kill_link_at", Option::TYPE_INT, Option::LEVEL_DEV)
8109 .set_default(0)
8110 .set_description(""),
8111
8112 Option("mds_kill_rename_at", Option::TYPE_INT, Option::LEVEL_DEV)
8113 .set_default(0)
8114 .set_description(""),
8115
8116 Option("mds_kill_openc_at", Option::TYPE_INT, Option::LEVEL_DEV)
8117 .set_default(0)
8118 .set_description(""),
8119
8120 Option("mds_kill_journal_at", Option::TYPE_INT, Option::LEVEL_DEV)
8121 .set_default(0)
8122 .set_description(""),
8123
8124 Option("mds_kill_journal_expire_at", Option::TYPE_INT, Option::LEVEL_DEV)
8125 .set_default(0)
8126 .set_description(""),
8127
8128 Option("mds_kill_journal_replay_at", Option::TYPE_INT, Option::LEVEL_DEV)
8129 .set_default(0)
8130 .set_description(""),
8131
8132 Option("mds_journal_format", Option::TYPE_UINT, Option::LEVEL_DEV)
8133 .set_default(1)
8134 .set_description(""),
8135
8136 Option("mds_kill_create_at", Option::TYPE_INT, Option::LEVEL_DEV)
8137 .set_default(0)
8138 .set_description(""),
8139
8140 Option("mds_inject_traceless_reply_probability", Option::TYPE_FLOAT, Option::LEVEL_DEV)
8141 .set_default(0)
8142 .set_description(""),
8143
8144 Option("mds_wipe_sessions", Option::TYPE_BOOL, Option::LEVEL_DEV)
8145 .set_default(0)
8146 .set_description(""),
8147
8148 Option("mds_wipe_ino_prealloc", Option::TYPE_BOOL, Option::LEVEL_DEV)
8149 .set_default(0)
8150 .set_description(""),
8151
8152 Option("mds_skip_ino", Option::TYPE_INT, Option::LEVEL_DEV)
8153 .set_default(0)
8154 .set_description(""),
8155
8156 Option("mds_enable_op_tracker", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
8157 .set_default(true)
8158 .set_description("track remote operation progression and statistics"),
8159
8160 Option("mds_op_history_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
8161 .set_default(20)
8162 .set_description("maximum size for list of historical operations"),
8163
8164 Option("mds_op_history_duration", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
8165 .set_default(600)
8166 .set_description("expiration time in seconds of historical operations"),
8167
8168 Option("mds_op_complaint_time", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
8169 .set_default(30)
8170 .set_description("time in seconds to consider an operation blocked after no updates"),
8171
8172 Option("mds_op_log_threshold", Option::TYPE_INT, Option::LEVEL_DEV)
8173 .set_default(5)
8174 .set_description(""),
8175
8176 Option("mds_snap_min_uid", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
8177 .set_default(0)
8178 .set_description("minimum uid of client to perform snapshots"),
8179
8180 Option("mds_snap_max_uid", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
8181 .set_default(4294967294)
8182 .set_description("maximum uid of client to perform snapshots"),
8183
8184 Option("mds_snap_rstat", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
8185 .set_default(false)
8186 .set_description("enabled nested rstat for snapshots"),
8187
8188 Option("mds_verify_backtrace", Option::TYPE_UINT, Option::LEVEL_DEV)
8189 .set_default(1)
8190 .set_description(""),
8191
8192 Option("mds_max_completed_flushes", Option::TYPE_UINT, Option::LEVEL_DEV)
8193 .set_default(100000)
8194 .set_description(""),
8195
8196 Option("mds_max_completed_requests", Option::TYPE_UINT, Option::LEVEL_DEV)
8197 .set_default(100000)
8198 .set_description(""),
8199
8200 Option("mds_action_on_write_error", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
8201 .set_default(1)
8202 .set_description("action to take when MDS cannot write to RADOS (0:ignore, 1:read-only, 2:suicide)"),
8203
8204 Option("mds_mon_shutdown_timeout", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
8205 .set_default(5)
8206 .set_description("time to wait for mon to receive damaged MDS rank notification"),
8207
8208 Option("mds_max_purge_files", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
8209 .set_default(64)
8210 .set_description("maximum number of deleted files to purge in parallel"),
8211
8212 Option("mds_max_purge_ops", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
8213 .set_default(8192)
8214 .set_description("maximum number of purge operations performed in parallel"),
8215
8216 Option("mds_max_purge_ops_per_pg", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
8217 .set_default(0.5)
8218 .set_description("number of parallel purge operations performed per PG"),
8219
8220 Option("mds_purge_queue_busy_flush_period", Option::TYPE_FLOAT, Option::LEVEL_DEV)
8221 .set_default(1.0)
8222 .set_description(""),
8223
8224 Option("mds_root_ino_uid", Option::TYPE_INT, Option::LEVEL_ADVANCED)
8225 .set_default(0)
8226 .set_description("default uid for new root directory"),
8227
8228 Option("mds_root_ino_gid", Option::TYPE_INT, Option::LEVEL_ADVANCED)
8229 .set_default(0)
8230 .set_description("default gid for new root directory"),
8231
8232 Option("mds_max_scrub_ops_in_progress", Option::TYPE_INT, Option::LEVEL_ADVANCED)
8233 .set_default(5)
8234 .set_description("maximum number of scrub operations performed in parallel"),
8235
8236 Option("mds_damage_table_max_entries", Option::TYPE_INT, Option::LEVEL_ADVANCED)
8237 .set_default(10000)
8238 .set_description("maximum number of damage table entries"),
8239
8240 Option("mds_client_writeable_range_max_inc_objs", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
8241 .set_default(1024)
8242 .set_description("maximum number of objects in writeable range of a file for a client"),
8243
8244 Option("mds_min_caps_per_client", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
8245 .set_default(100)
8246 .set_description("minimum number of capabilities a client may hold"),
8247
8248 Option("mds_max_caps_per_client", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
8249 .set_default(1_M)
8250 .set_description("maximum number of capabilities a client may hold"),
8251
8252 Option("mds_hack_allow_loading_invalid_metadata", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
8253 .set_default(0)
8254 .set_description("INTENTIONALLY CAUSE DATA LOSS by bypasing checks for invalid metadata on disk. Allows testing repair tools."),
8255
8256 Option("mds_defer_session_stale", Option::TYPE_BOOL, Option::LEVEL_DEV)
8257 .set_default(true),
8258
8259 Option("mds_inject_migrator_session_race", Option::TYPE_BOOL, Option::LEVEL_DEV)
8260 .set_default(false),
8261
8262 Option("mds_request_load_average_decay_rate", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
8263 .set_default(60)
8264 .set_description("rate of decay in seconds for calculating request load average"),
8265
8266 Option("mds_cap_revoke_eviction_timeout", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
8267 .set_default(0)
8268 .set_description("number of seconds after which clients which have not responded to cap revoke messages by the MDS are evicted."),
8269
8270 Option("mds_max_retries_on_remount_failure", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
8271 .set_default(5)
8272 .set_description("number of consecutive failed remount attempts for invalidating kernel dcache after which client would abort."),
8273
8274 Option("mds_dump_cache_threshold_formatter", Option::TYPE_SIZE, Option::LEVEL_DEV)
8275 .set_default(1_G)
8276 .set_description("threshold for cache usage to disallow \"dump cache\" operation to formatter")
8277 .set_long_description("Disallow MDS from dumping caches to formatter via \"dump cache\" command if cache usage exceeds this threshold."),
8278
8279 Option("mds_dump_cache_threshold_file", Option::TYPE_SIZE, Option::LEVEL_DEV)
8280 .set_default(0)
8281 .set_description("threshold for cache usage to disallow \"dump cache\" operation to file")
8282 .set_long_description("Disallow MDS from dumping caches to file via \"dump cache\" command if cache usage exceeds this threshold."),
8283
8284 Option("mds_task_status_update_interval", Option::TYPE_FLOAT, Option::LEVEL_DEV)
8285 .set_default(2.0)
8286 .set_description("task status update interval to manager")
8287 .set_long_description("interval (in seconds) for sending mds task status to ceph manager"),
8288 });
8289 }
8290
8291 std::vector<Option> get_mds_client_options() {
8292 return std::vector<Option>({
8293 Option("client_cache_size", Option::TYPE_SIZE, Option::LEVEL_BASIC)
8294 .set_default(16384)
8295 .set_description("soft maximum number of directory entries in client cache"),
8296
8297 Option("client_cache_mid", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
8298 .set_default(.75)
8299 .set_description("mid-point of client cache LRU"),
8300
8301 Option("client_use_random_mds", Option::TYPE_BOOL, Option::LEVEL_DEV)
8302 .set_default(false)
8303 .set_description("issue new requests to a random active MDS"),
8304
8305 Option("client_mount_timeout", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
8306 .set_default(300.0)
8307 .set_description("timeout for mounting CephFS (seconds)"),
8308
8309 Option("client_tick_interval", Option::TYPE_FLOAT, Option::LEVEL_DEV)
8310 .set_default(1.0)
8311 .set_description("seconds between client upkeep ticks"),
8312
8313 Option("client_trace", Option::TYPE_STR, Option::LEVEL_DEV)
8314 .set_default("")
8315 .set_description("file containing trace of client operations"),
8316
8317 Option("client_readahead_min", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
8318 .set_default(128*1024)
8319 .set_description("minimum bytes to readahead in a file"),
8320
8321 Option("client_readahead_max_bytes", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
8322 .set_default(0)
8323 .set_description("maximum bytes to readahead in a file (zero is unlimited)"),
8324
8325 Option("client_readahead_max_periods", Option::TYPE_INT, Option::LEVEL_ADVANCED)
8326 .set_default(4)
8327 .set_description("maximum stripe periods to readahead in a file"),
8328
8329 Option("client_reconnect_stale", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
8330 .set_default(false)
8331 .set_description("reconnect when the session becomes stale"),
8332
8333 Option("client_snapdir", Option::TYPE_STR, Option::LEVEL_ADVANCED)
8334 .set_default(".snap")
8335 .set_description("pseudo directory for snapshot access to a directory"),
8336
8337 Option("client_mountpoint", Option::TYPE_STR, Option::LEVEL_ADVANCED)
8338 .set_default("/")
8339 .set_description("default mount-point"),
8340
8341 Option("client_mount_uid", Option::TYPE_INT, Option::LEVEL_ADVANCED)
8342 .set_default(-1)
8343 .set_description("uid to mount as"),
8344
8345 Option("client_mount_gid", Option::TYPE_INT, Option::LEVEL_ADVANCED)
8346 .set_default(-1)
8347 .set_description("gid to mount as"),
8348
8349 /* RADOS client option */
8350 Option("client_notify_timeout", Option::TYPE_INT, Option::LEVEL_DEV)
8351 .set_default(10)
8352 .set_description(""),
8353
8354 /* RADOS client option */
8355 Option("osd_client_watch_timeout", Option::TYPE_INT, Option::LEVEL_DEV)
8356 .set_default(30)
8357 .set_description(""),
8358
8359 Option("client_caps_release_delay", Option::TYPE_INT, Option::LEVEL_DEV)
8360 .set_default(5)
8361 .set_description(""),
8362
8363 Option("client_quota_df", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
8364 .set_default(true)
8365 .set_description("show quota usage for statfs (df)"),
8366
8367 Option("client_oc", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
8368 .set_default(true)
8369 .set_description("enable object caching"),
8370
8371 Option("client_oc_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
8372 .set_default(200_M)
8373 .set_description("maximum size of object cache"),
8374
8375 Option("client_oc_max_dirty", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
8376 .set_default(100_M)
8377 .set_description("maximum size of dirty pages in object cache"),
8378
8379 Option("client_oc_target_dirty", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
8380 .set_default(8_M)
8381 .set_description("target size of dirty pages object cache"),
8382
8383 Option("client_oc_max_dirty_age", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
8384 .set_default(5.0)
8385 .set_description("maximum age of dirty pages in object cache (seconds)"),
8386
8387 Option("client_oc_max_objects", Option::TYPE_INT, Option::LEVEL_ADVANCED)
8388 .set_default(1000)
8389 .set_description("maximum number of objects in cache"),
8390
8391 Option("client_debug_getattr_caps", Option::TYPE_BOOL, Option::LEVEL_DEV)
8392 .set_default(false)
8393 .set_description(""),
8394
8395 Option("client_debug_force_sync_read", Option::TYPE_BOOL, Option::LEVEL_DEV)
8396 .set_default(false)
8397 .set_description(""),
8398
8399 Option("client_debug_inject_tick_delay", Option::TYPE_INT, Option::LEVEL_DEV)
8400 .set_default(0)
8401 .set_description(""),
8402
8403 Option("client_max_inline_size", Option::TYPE_SIZE, Option::LEVEL_DEV)
8404 .set_default(4_K)
8405 .set_description(""),
8406
8407 Option("client_inject_release_failure", Option::TYPE_BOOL, Option::LEVEL_DEV)
8408 .set_default(false)
8409 .set_description(""),
8410
8411 Option("client_inject_fixed_oldest_tid", Option::TYPE_BOOL, Option::LEVEL_DEV)
8412 .set_default(false)
8413 .set_description(""),
8414
8415 Option("client_metadata", Option::TYPE_STR, Option::LEVEL_ADVANCED)
8416 .set_default("")
8417 .set_description("metadata key=value comma-delimited pairs appended to session metadata"),
8418
8419 Option("client_acl_type", Option::TYPE_STR, Option::LEVEL_ADVANCED)
8420 .set_default("")
8421 .set_description("ACL type to enforce (none or \"posix_acl\")"),
8422
8423 Option("client_permissions", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
8424 .set_default(true)
8425 .set_description("client-enforced permission checking"),
8426
8427 Option("client_dirsize_rbytes", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
8428 .set_default(true)
8429 .set_description("set the directory size as the number of file bytes recursively used")
8430 .set_long_description("This option enables a CephFS feature that stores the recursive directory size (the bytes used by files in the directory and its descendents) in the st_size field of the stat structure."),
8431
8432 Option("client_force_lazyio", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
8433 .set_default(false)
8434 .set_description(""),
8435
8436 // note: the max amount of "in flight" dirty data is roughly (max - target)
8437 Option("fuse_use_invalidate_cb", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
8438 .set_default(true)
8439 .set_description("use fuse 2.8+ invalidate callback to keep page cache consistent"),
8440
8441 Option("fuse_disable_pagecache", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
8442 .set_default(false)
8443 .set_description("disable page caching in the kernel for this FUSE mount"),
8444
8445 Option("fuse_allow_other", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
8446 .set_default(true)
8447 .set_description("pass allow_other to FUSE on mount"),
8448
8449 Option("fuse_default_permissions", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
8450 .set_default(false)
8451 .set_description("pass default_permisions to FUSE on mount"),
8452
8453 Option("fuse_big_writes", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
8454 .set_default(false)
8455 .set_description("big_writes is deprecated in libfuse 3.0.0"),
8456
8457 Option("fuse_max_write", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
8458 .set_default(0)
8459 .set_description("set the maximum number of bytes in a single write operation")
8460 .set_long_description("Set the maximum number of bytes in a single write operation that may pass atomically through FUSE. The FUSE default is 128kB and may be indicated by setting this option to 0."),
8461
8462 Option("fuse_atomic_o_trunc", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
8463 .set_default(true)
8464 .set_description("pass atomic_o_trunc flag to FUSE on mount"),
8465
8466 Option("fuse_debug", Option::TYPE_BOOL, Option::LEVEL_DEV)
8467 .set_default(false)
8468 .set_description(""),
8469
8470 Option("fuse_multithreaded", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
8471 .set_default(true)
8472 .set_description("allow parallel processing through FUSE library"),
8473
8474 Option("fuse_require_active_mds", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
8475 .set_default(true)
8476 .set_description("require active MDSs in the file system when mounting"),
8477
8478 Option("fuse_syncfs_on_mksnap", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
8479 .set_default(true)
8480 .set_description("synchronize all local metadata/file changes after snapshot"),
8481
8482 Option("fuse_set_user_groups", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
8483 .set_default(true)
8484 .set_description("check for ceph-fuse to consider supplementary groups for permissions"),
8485
8486 Option("client_try_dentry_invalidate", Option::TYPE_BOOL, Option::LEVEL_DEV)
8487 .set_default(false)
8488 .set_description(""),
8489
8490 Option("client_die_on_failed_remount", Option::TYPE_BOOL, Option::LEVEL_DEV)
8491 .set_default(false)
8492 .set_description(""),
8493
8494 Option("client_die_on_failed_dentry_invalidate", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
8495 .set_default(true)
8496 .set_description("kill the client when no dentry invalidation options are available")
8497 .set_long_description("The CephFS client requires a mechanism to invalidate dentries in the caller (e.g. the kernel for ceph-fuse) when capabilities must be recalled. If the client cannot do this then the MDS cache cannot shrink which can cause the MDS to fail."),
8498
8499 Option("client_check_pool_perm", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
8500 .set_default(true)
8501 .set_description("confirm access to inode's data pool/namespace described in file layout"),
8502
8503 Option("client_use_faked_inos", Option::TYPE_BOOL, Option::LEVEL_DEV)
8504 .set_default(false)
8505 .set_description(""),
8506
8507 Option("client_mds_namespace", Option::TYPE_STR, Option::LEVEL_ADVANCED)
8508 .set_default("")
8509
8510 .set_description("CephFS file system name to mount")
8511 .set_long_description("Use this with ceph-fuse, or with any process "
8512 "that uses libcephfs. Programs using libcephfs may also pass "
8513 "the filesystem name into mount(), which will override this setting. "
8514 "If no filesystem name is given in mount() or this setting, the default "
8515 "filesystem will be mounted (usually the first created)."),
8516
8517 Option("fake_statfs_for_testing", Option::TYPE_INT, Option::LEVEL_DEV)
8518 .set_default(0)
8519 .set_description("Set a value for kb and compute kb_used from total of num_bytes"),
8520
8521 Option("debug_allow_any_pool_priority", Option::TYPE_BOOL, Option::LEVEL_DEV)
8522 .set_default(false)
8523 .set_description("Allow any pool priority to be set to test conversion to new range"),
8524 });
8525 }
8526
8527
8528 static std::vector<Option> build_options()
8529 {
8530 std::vector<Option> result = get_global_options();
8531
8532 auto ingest = [&result](std::vector<Option>&& options, const char* svc) {
8533 for (auto &o : options) {
8534 o.add_service(svc);
8535 result.push_back(std::move(o));
8536 }
8537 };
8538
8539 ingest(get_rgw_options(), "rgw");
8540 ingest(get_rbd_options(), "rbd");
8541 ingest(get_rbd_mirror_options(), "rbd-mirror");
8542 ingest(get_immutable_object_cache_options(), "immutable-objet-cache");
8543 ingest(get_mds_options(), "mds");
8544 ingest(get_mds_client_options(), "mds_client");
8545
8546 return result;
8547 }
8548
8549 const std::vector<Option> ceph_options = build_options();
8550