1CTDB(1) CTDB - clustered TDB database CTDB(1)
2
6 ctdb - CTDB management utility
7
9 ctdb [OPTION...] {COMMAND} [COMMAND-ARGS]
10
12 ctdb is a utility to view and manage a CTDB cluster.
13 The following terms are used when referring to nodes in a cluster:
15 PNN
17 Physical Node Number. The physical node number is an integer that
18 describes the node in the cluster. The first node has physical node
19 number 0. in a cluster.
20 PNN-LIST
22 This is either a single PNN, a comma-separate list of PNNs or
23 "all".
24 Commands that reference a database use the following terms:
26 DB
28 This is either a database name, such as locking.tdb or a database
29 ID such as "0x42fe72c5".
30 DB-LIST
32 A space separated list of at least one DB.
33
35 -n PNN
36 The node specified by PNN should be queried for the requested
37 information. Default is to query the daemon running on the local
38 host.
39 -Y
41 Produce output in machine readable form for easier parsing by
42 scripts. This uses a field delimiter of ':'. Not all commands
43 support this option.
44 -x SEPARATOR
46 Use SEPARATOR to delimit fields in machine readable output. This
47 implies -Y.
48 -X
50 Produce output in machine readable form for easier parsing by
51 scripts. This uses a field delimiter of '|'. Not all commands
52 support this option.
53 This is equivalent to "-x|" and avoids some shell quoting issues.
55 -t TIMEOUT
57 Indicates that ctdb should wait up to TIMEOUT seconds for a
58 response to most commands sent to the CTDB daemon. The default is
59 10 seconds.
60 -T TIMELIMIT
62 Indicates that TIMELIMIT is the maximum run time (in seconds) for
63 the ctdb command. When TIMELIMIT is exceeded the ctdb command will
64 terminate with an error. The default is 120 seconds.
65 -? --help
67 Print some help text to the screen.
68 --usage
70 Print usage information to the screen.
71 -d --debug=DEBUGLEVEL
73 Change the debug level for the command. Default is NOTICE.
74
76 These are commands used to monitor and administer a CTDB cluster.
77 pnn
79 This command displays the PNN of the current node.
80 status
82 This command shows the current status of all CTDB nodes based on
83 information from the queried node.
84 Note: If the queried node is INACTIVE then the status might not be
86 current.
87 Node status
89 This includes the number of physical nodes and the status of each
90 node. See ctdb(7) for information about node states.
91 Generation
93 The generation id is a number that indicates the current generation
94 of a cluster instance. Each time a cluster goes through a
95 reconfiguration or a recovery its generation id will be changed.
96 This number does not have any particular meaning other than to keep
98 track of when a cluster has gone through a recovery. It is a random
99 number that represents the current instance of a ctdb cluster and
100 its databases. The CTDB daemon uses this number internally to be
101 able to tell when commands to operate on the cluster and the
102 databases was issued in a different generation of the cluster, to
103 ensure that commands that operate on the databases will not survive
104 across a cluster database recovery. After a recovery, all old
105 outstanding commands will automatically become invalid.
106 Sometimes this number will be shown as "INVALID". This only means
108 that the ctdbd daemon has started but it has not yet merged with
109 the cluster through a recovery. All nodes start with generation
110 "INVALID" and are not assigned a real generation id until they have
111 successfully been merged with a cluster through a recovery.
112 Virtual Node Number (VNN) map
114 Consists of the number of virtual nodes and mapping from virtual
115 node numbers to physical node numbers. Only nodes that are
116 participating in the VNN map can become lmaster for database
117 records.
118 Recovery mode
120 This is the current recovery mode of the cluster. There are two
121 possible modes:
122 NORMAL - The cluster is fully operational.
124 RECOVERY - The cluster databases have all been frozen, pausing all
126 services while the cluster awaits a recovery process to complete. A
127 recovery process should finish within seconds. If a cluster is
128 stuck in the RECOVERY state this would indicate a cluster
129 malfunction which needs to be investigated.
130 Once the leader detects an inconsistency, for example a node
132 becomes disconnected/connected, the recovery daemon will trigger a
133 cluster recovery process, where all databases are remerged across
134 the cluster. When this process starts, the leader will first
135 "freeze" all databases to prevent applications such as samba from
136 accessing the databases and it will also mark the recovery mode as
137 RECOVERY.
138 When the CTDB daemon starts up, it will start in RECOVERY mode.
140 Once the node has been merged into a cluster and all databases have
141 been recovered, the node mode will change into NORMAL mode and the
142 databases will be "thawed", allowing samba to access the databases
143 again.
144 Leader
146 This is the cluster node that is currently designated as the
147 leader. This node is responsible of monitoring the consistency of
148 the cluster and to perform the actual recovery process when
149 reqired.
150 Only one node at a time can be the designated leader. Which node is
152 designated the leader is decided by an election process in the
153 recovery daemons running on each node.
154 Example
156 # ctdb status
157 Number of nodes:4
158 pnn:0 192.168.2.200 OK (THIS NODE)
159 pnn:1 192.168.2.201 OK
160 pnn:2 192.168.2.202 OK
161 pnn:3 192.168.2.203 OK
162 Generation:1362079228
163 Size:4
164 hash:0 lmaster:0
165 hash:1 lmaster:1
166 hash:2 lmaster:2
167 hash:3 lmaster:3
168 Recovery mode:NORMAL (0)
169 Leader:0
170 nodestatus [PNN-LIST]
173 This command is similar to the status command. It displays the "node
174 status" subset of output. The main differences are:
175 • The exit code is the bitwise-OR of the flags for each specified
177 node, while ctdb status exits with 0 if it was able to retrieve
178 status for all nodes.
179 • ctdb status provides status information for all nodes. ctdb
181 nodestatus defaults to providing status for only the current node.
182 If PNN-LIST is provided then status is given for the indicated
183 node(s).
184 A common invocation in scripts is ctdb nodestatus all to check whether
186 all nodes in a cluster are healthy.
187 Example
189 # ctdb nodestatus
190 pnn:0 10.0.0.30 OK (THIS NODE)
191 # ctdb nodestatus all
193 Number of nodes:2
194 pnn:0 10.0.0.30 OK (THIS NODE)
195 pnn:1 10.0.0.31 OK
196 leader
199 This command shows the pnn of the node which is currently the leader.
200 Note: If the queried node is INACTIVE then the status might not be
202 current.
203 uptime
205 This command shows the uptime for the ctdb daemon. When the last
206 recovery or ip-failover completed and how long it took. If the
207 "duration" is shown as a negative number, this indicates that there is
208 a recovery/failover in progress and it started that many seconds ago.
209 Example
211 # ctdb uptime
212 Current time of node : Thu Oct 29 10:38:54 2009
213 Ctdbd start time : (000 16:54:28) Wed Oct 28 17:44:26 2009
214 Time of last recovery/failover: (000 16:53:31) Wed Oct 28 17:45:23 2009
215 Duration of last recovery/failover: 2.248552 seconds
216 listnodes
219 This command shows lists the ip addresses of all the nodes in the
220 cluster.
221 Example
223 # ctdb listnodes
224 192.168.2.200
225 192.168.2.201
226 192.168.2.202
227 192.168.2.203
228 natgw {leader|list|status}
231 This command shows different aspects of NAT gateway status. For an
232 overview of CTDB's NAT gateway functionality please see the NAT GATEWAY
233 section in ctdb(7).
234 leader
236 Show the PNN and private IP address of the current NAT gateway
237 leader node.
238 Example output:
240 1 192.168.2.201
242 list
245 List the private IP addresses of nodes in the current NAT gateway
246 group, annotating the leader node.
247 Example output:
249 192.168.2.200
251 192.168.2.201 LEADER
252 192.168.2.202
253 192.168.2.203
254 status
257 List the nodes in the current NAT gateway group and their status.
258 Example output:
260 pnn:0 192.168.2.200 UNHEALTHY (THIS NODE)
262 pnn:1 192.168.2.201 OK
263 pnn:2 192.168.2.202 OK
264 pnn:3 192.168.2.203 OK
265 ping
268 This command will "ping" specified CTDB nodes in the cluster to verify
269 that they are running.
270 Example
272 # ctdb ping
273 response from 0 time=0.000054 sec (3 clients)
274 ifaces
277 This command will display the list of network interfaces, which could
278 host public addresses, along with their status.
279 Example
281 # ctdb ifaces
282 Interfaces on node 0
283 name:eth5 link:up references:2
284 name:eth4 link:down references:0
285 name:eth3 link:up references:1
286 name:eth2 link:up references:1
287 # ctdb -X ifaces
289 |Name|LinkStatus|References|
290 |eth5|1|2|
291 |eth4|0|0|
292 |eth3|1|1|
293 |eth2|1|1|
294 ip
297 This command will display the list of public addresses that are
298 provided by the cluster and which physical node is currently serving
299 this ip. By default this command will ONLY show those public addresses
300 that are known to the node itself. To see the full list of all public
301 ips across the cluster you must use "ctdb ip all".
302 Example
304 # ctdb ip -v
305 Public IPs on node 0
306 172.31.91.82 node[1] active[] available[eth2,eth3] configured[eth2,eth3]
307 172.31.91.83 node[0] active[eth3] available[eth2,eth3] configured[eth2,eth3]
308 172.31.91.84 node[1] active[] available[eth2,eth3] configured[eth2,eth3]
309 172.31.91.85 node[0] active[eth2] available[eth2,eth3] configured[eth2,eth3]
310 172.31.92.82 node[1] active[] available[eth5] configured[eth4,eth5]
311 172.31.92.83 node[0] active[eth5] available[eth5] configured[eth4,eth5]
312 172.31.92.84 node[1] active[] available[eth5] configured[eth4,eth5]
313 172.31.92.85 node[0] active[eth5] available[eth5] configured[eth4,eth5]
314 # ctdb -X ip -v
316 |Public IP|Node|ActiveInterface|AvailableInterfaces|ConfiguredInterfaces|
317 |172.31.91.82|1||eth2,eth3|eth2,eth3|
318 |172.31.91.83|0|eth3|eth2,eth3|eth2,eth3|
319 |172.31.91.84|1||eth2,eth3|eth2,eth3|
320 |172.31.91.85|0|eth2|eth2,eth3|eth2,eth3|
321 |172.31.92.82|1||eth5|eth4,eth5|
322 |172.31.92.83|0|eth5|eth5|eth4,eth5|
323 |172.31.92.84|1||eth5|eth4,eth5|
324 |172.31.92.85|0|eth5|eth5|eth4,eth5|
325 ipinfo IP
328 This command will display details about the specified public addresses.
329 Example
331 # ctdb ipinfo 172.31.92.85
332 Public IP[172.31.92.85] info on node 0
333 IP:172.31.92.85
334 CurrentNode:0
335 NumInterfaces:2
336 Interface[1]: Name:eth4 Link:down References:0
337 Interface[2]: Name:eth5 Link:up References:2 (active)
338 event run|status|script list|script enable|script disable
341 This command is used to control event daemon and to inspect status of
342 various events.
343 The commands below require a component to be specified. In the current
345 version the only valid component is legacy.
346 run TIMEOUT COMPONENT EVENT [ARGUMENTS]
348 This command can be used to manually run specified EVENT in
349 COMPONENT with optional ARGUMENTS. The event will be allowed to run
350 a maximum of TIMEOUT seconds. If TIMEOUT is 0, then there is no
351 time limit for running the event.
352 status COMPONENT EVENT
354 This command displays the last execution status of the specified
355 EVENT in COMPONENT.
356 The command will terminate with the exit status corresponding to
358 the overall status of event that is displayed.
359 The output is the list of event scripts executed. Each line shows
361 the name, status, duration and start time for each script.
362 Example
364 # ctdb event status legacy monitor
366 00.ctdb OK 0.014 Sat Dec 17 19:39:11 2016
367 01.reclock OK 0.013 Sat Dec 17 19:39:11 2016
368 05.system OK 0.029 Sat Dec 17 19:39:11 2016
369 06.nfs OK 0.014 Sat Dec 17 19:39:11 2016
370 10.interface OK 0.037 Sat Dec 17 19:39:11 2016
371 11.natgw OK 0.011 Sat Dec 17 19:39:11 2016
372 11.routing OK 0.007 Sat Dec 17 19:39:11 2016
373 13.per_ip_routing OK 0.007 Sat Dec 17 19:39:11 2016
374 20.multipathd OK 0.007 Sat Dec 17 19:39:11 2016
375 31.clamd OK 0.007 Sat Dec 17 19:39:11 2016
376 40.vsftpd OK 0.013 Sat Dec 17 19:39:11 2016
377 41.httpd OK 0.018 Sat Dec 17 19:39:11 2016
378 49.winbind OK 0.023 Sat Dec 17 19:39:11 2016
379 50.samba OK 0.100 Sat Dec 17 19:39:12 2016
380 60.nfs OK 0.376 Sat Dec 17 19:39:12 2016
381 70.iscsi OK 0.009 Sat Dec 17 19:39:12 2016
382 91.lvs OK 0.007 Sat Dec 17 19:39:12 2016
383 script list COMPONENT
386 List the available event scripts in COMPONENT. Enabled scripts are
387 flagged with a '*'.
388 Generally, event scripts are provided by CTDB. However, local or
390 3rd party event scripts may also be available. These are shown in a
391 separate section after those provided by CTDB.
392 Example
394 # ctdb event script list legacy
396 * 00.ctdb
397 * 01.reclock
398 * 05.system
399 * 06.nfs
400 * 10.interface
401 11.natgw
402 11.routing
403 13.per_ip_routing
404 20.multipathd
405 31.clamd
406 40.vsftpd
407 41.httpd
408 * 49.winbind
409 * 50.samba
410 * 60.nfs
411 70.iscsi
412 91.lvs
413 * 02.local
415 script enable COMPONENT SCRIPT
418 Enable the specified event SCRIPT in COMPONENT. Only enabled
419 scripts will be executed when running any event.
420 script disable COMPONENT SCRIPT
422 Disable the specified event SCRIPT in COMPONENT. This will prevent
423 the script from executing when running any event.
424 scriptstatus
426 This command displays which event scripts where run in the previous
427 monitoring cycle and the result of each script. If a script failed with
428 an error, causing the node to become unhealthy, the output from that
429 script is also shown.
430 This command is deprecated. It's provided for backward compatibility.
432 In place of ctdb scriptstatus, use ctdb event status.
433 Example
435 # ctdb scriptstatus
436 00.ctdb OK 0.011 Sat Dec 17 19:40:46 2016
437 01.reclock OK 0.010 Sat Dec 17 19:40:46 2016
438 05.system OK 0.030 Sat Dec 17 19:40:46 2016
439 06.nfs OK 0.014 Sat Dec 17 19:40:46 2016
440 10.interface OK 0.041 Sat Dec 17 19:40:46 2016
441 11.natgw OK 0.008 Sat Dec 17 19:40:46 2016
442 11.routing OK 0.007 Sat Dec 17 19:40:46 2016
443 13.per_ip_routing OK 0.007 Sat Dec 17 19:40:46 2016
444 20.multipathd OK 0.007 Sat Dec 17 19:40:46 2016
445 31.clamd OK 0.007 Sat Dec 17 19:40:46 2016
446 40.vsftpd OK 0.013 Sat Dec 17 19:40:46 2016
447 41.httpd OK 0.015 Sat Dec 17 19:40:46 2016
448 49.winbind OK 0.022 Sat Dec 17 19:40:46 2016
449 50.samba ERROR 0.077 Sat Dec 17 19:40:46 2016
450 OUTPUT: ERROR: samba tcp port 445 is not responding
451 listvars
454 List all tuneable variables, except the values of the obsolete tunables
455 like VacuumMinInterval. The obsolete tunables can be retrieved only
456 explicitly with the "ctdb getvar" command.
457 Example
459 # ctdb listvars
460 SeqnumInterval = 1000
461 ControlTimeout = 60
462 TraverseTimeout = 20
463 KeepaliveInterval = 5
464 KeepaliveLimit = 5
465 RecoverTimeout = 120
466 RecoverInterval = 1
467 ElectionTimeout = 3
468 TakeoverTimeout = 9
469 MonitorInterval = 15
470 TickleUpdateInterval = 20
471 EventScriptTimeout = 30
472 MonitorTimeoutCount = 20
473 RecoveryGracePeriod = 120
474 RecoveryBanPeriod = 300
475 DatabaseHashSize = 100001
476 DatabaseMaxDead = 5
477 RerecoveryTimeout = 10
478 EnableBans = 1
479 NoIPFailback = 0
480 VerboseMemoryNames = 0
481 RecdPingTimeout = 60
482 RecdFailCount = 10
483 LogLatencyMs = 0
484 RecLockLatencyMs = 1000
485 RecoveryDropAllIPs = 120
486 VacuumInterval = 10
487 VacuumMaxRunTime = 120
488 RepackLimit = 10000
489 VacuumFastPathCount = 60
490 MaxQueueDropMsg = 1000000
491 AllowUnhealthyDBRead = 0
492 StatHistoryInterval = 1
493 DeferredAttachTO = 120
494 AllowClientDBAttach = 1
495 RecoverPDBBySeqNum = 1
496 DeferredRebalanceOnNodeAdd = 300
497 FetchCollapse = 1
498 HopcountMakeSticky = 50
499 StickyDuration = 600
500 StickyPindown = 200
501 NoIPTakeover = 0
502 DBRecordCountWarn = 100000
503 DBRecordSizeWarn = 10000000
504 DBSizeWarn = 100000000
505 PullDBPreallocation = 10485760
506 LockProcessesPerDB = 200
507 RecBufferSizeLimit = 1000000
508 QueueBufferSize = 1024
509 IPAllocAlgorithm = 2
510 getvar NAME
513 Get the runtime value of a tuneable variable.
514 Example
516 # ctdb getvar MonitorInterval
517 MonitorInterval = 15
518 setvar NAME VALUE
521 Set the runtime value of a tuneable variable.
522 Example
524 # ctdb setvar MonitorInterval 20
525 lvs {leader|list|status}
528 This command shows different aspects of LVS status. For an overview of
529 CTDB's LVS functionality please see the LVS section in ctdb(7).
530 leader
532 Shows the PNN of the current LVS leader node.
533 Example output:
535 2
537 list
540 Lists the currently usable LVS nodes.
541 Example output:
543 2 10.0.0.13
545 3 10.0.0.14
546 status
549 List the nodes in the current LVS group and their status.
550 Example output:
552 pnn:0 10.0.0.11 UNHEALTHY (THIS NODE)
554 pnn:1 10.0.0.12 UNHEALTHY
555 pnn:2 10.0.0.13 OK
556 pnn:3 10.0.0.14 OK
557 getcapabilities
560 This command shows the capabilities of the current node. See the
561 CAPABILITIES section in ctdb(7) for more details.
562 Example output:
564 LEADER: YES
566 LMASTER: YES
567 statistics
570 Collect statistics from the CTDB daemon about how many calls it has
571 served. Information about various fields in statistics can be found in
572 ctdb-statistics(7).
573 Example
575 # ctdb statistics
576 CTDB version 1
577 Current time of statistics : Tue Mar 8 15:18:51 2016
578 Statistics collected since : (003 21:31:32) Fri Mar 4 17:47:19 2016
579 num_clients 9
580 frozen 0
581 recovering 0
582 num_recoveries 2
583 client_packets_sent 8170534
584 client_packets_recv 7166132
585 node_packets_sent 16549998
586 node_packets_recv 5244418
587 keepalive_packets_sent 201969
588 keepalive_packets_recv 201969
589 node
590 req_call 26
591 reply_call 0
592 req_dmaster 9
593 reply_dmaster 12
594 reply_error 0
595 req_message 1339231
596 req_control 8177506
597 reply_control 6831284
598 client
599 req_call 15
600 req_message 334809
601 req_control 6831308
602 timeouts
603 call 0
604 control 0
605 traverse 0
606 locks
607 num_calls 8
608 num_current 0
609 num_pending 0
610 num_failed 0
611 total_calls 15
612 pending_calls 0
613 childwrite_calls 0
614 pending_childwrite_calls 0
615 memory_used 394879
616 max_hop_count 1
617 total_ro_delegations 0
618 total_ro_revokes 0
619 hop_count_buckets: 8 5 0 0 0 0 0 0 0 0 0 0 0 0 0 0
620 lock_buckets: 0 0 8 0 0 0 0 0 0 0 0 0 0 0 0 0
621 locks_latency MIN/AVG/MAX 0.010005/0.010418/0.011010 sec out of 8
622 reclock_ctdbd MIN/AVG/MAX 0.002538/0.002538/0.002538 sec out of 1
623 reclock_recd MIN/AVG/MAX 0.000000/0.000000/0.000000 sec out of 0
624 call_latency MIN/AVG/MAX 0.000044/0.002142/0.011702 sec out of 15
625 childwrite_latency MIN/AVG/MAX 0.000000/0.000000/0.000000 sec out of 0
626 statisticsreset
629 This command is used to clear all statistics counters in a node.
630 Example: ctdb statisticsreset
632 dbstatistics DB
634 Display statistics about the database DB. Information about various
635 fields in dbstatistics can be found in ctdb-statistics(7).
636 Example
638 # ctdb dbstatistics locking.tdb
639 DB Statistics: locking.tdb
640 ro_delegations 0
641 ro_revokes 0
642 locks
643 total 14356
644 failed 0
645 current 0
646 pending 0
647 hop_count_buckets: 28087 2 1 0 0 0 0 0 0 0 0 0 0 0 0 0
648 lock_buckets: 0 14188 38 76 32 19 3 0 0 0 0 0 0 0 0 0
649 locks_latency MIN/AVG/MAX 0.001066/0.012686/4.202292 sec out of 14356
650 vacuum_latency MIN/AVG/MAX 0.000472/0.002207/15.243570 sec out of 224530
651 Num Hot Keys: 1
652 Count:8 Key:ff5bd7cb3ee3822edc1f0000000000000000000000000000
653 getreclock
656 Show details of the recovery lock, if any.
657 Example output:
659 /clusterfs/.ctdb/recovery.lock
661 getdebug
664 Get the current debug level for the node. the debug level controls what
665 information is written to the log file.
666 The debug levels are mapped to the corresponding syslog levels. When a
668 debug level is set, only those messages at that level and higher levels
669 will be printed.
670 The list of debug levels from highest to lowest are :
672 ERROR WARNING NOTICE INFO DEBUG
674 setdebug DEBUGLEVEL
676 Set the debug level of a node. This controls what information will be
677 logged.
678 The debuglevel is one of ERROR WARNING NOTICE INFO DEBUG
680 getpid
682 This command will return the process id of the ctdb daemon.
683 disable
685 This command is used to administratively disable a node in the cluster.
686 A disabled node will still participate in the cluster and host
687 clustered TDB records but its public ip address has been taken over by
688 a different node and it no longer hosts any services.
689 enable
691 Re-enable a node that has been administratively disabled.
692 stop
694 This command is used to administratively STOP a node in the cluster. A
695 STOPPED node is connected to the cluster but will not host any public
696 ip addresse, nor does it participate in the VNNMAP. The difference
697 between a DISABLED node and a STOPPED node is that a STOPPED node does
698 not host any parts of the database which means that a recovery is
699 required to stop/continue nodes.
700 continue
702 Re-start a node that has been administratively stopped.
703 addip IPADDR/mask IFACE
705 This command is used to add a new public ip to a node during runtime.
706 It should be followed by a ctdb ipreallocate. This allows public
707 addresses to be added to a cluster without having to restart the ctdb
708 daemons.
709 Note that this only updates the runtime instance of ctdb. Any changes
711 will be lost next time ctdb is restarted and the public addresses file
712 is re-read. If you want this change to be permanent you must also
713 update the public addresses file manually.
714 delip IPADDR
716 This command flags IPADDR for deletion from a node at runtime. It
717 should be followed by a ctdb ipreallocate. If IPADDR is currently
718 hosted by the node it is being removed from, this ensures that the IP
719 will first be failed over to another node, if possible, and that it is
720 then actually removed.
721 Note that this only updates the runtime instance of CTDB. Any changes
723 will be lost next time CTDB is restarted and the public addresses file
724 is re-read. If you want this change to be permanent you must also
725 update the public addresses file manually.
726 moveip IPADDR PNN
728 This command can be used to manually fail a public ip address to a
729 specific node.
730 In order to manually override the "automatic" distribution of public ip
732 addresses that ctdb normally provides, this command only works when you
733 have changed the tunables for the daemon to:
734 IPAllocAlgorithm != 0
736 NoIPFailback = 1
738 shutdown
740 This command will shutdown a specific CTDB daemon.
741 setlmasterrole on|off
743 This command is used to enable/disable the LMASTER capability for a
744 node at runtime. This capability determines whether or not a node can
745 be used as an LMASTER for records in the database. A node that does not
746 have the LMASTER capability will not show up in the vnnmap.
747 Nodes will by default have this capability, but it can be stripped off
749 nodes by the setting in the sysconfig file or by using this command.
750 Once this setting has been enabled/disabled, you need to perform a
752 recovery for it to take effect.
753 See also "ctdb getcapabilities"
755 setleaderrole on|off
757 This command is used to enable/disable the LEADER capability for a node
758 at runtime. This capability determines whether or not a node can be
759 elected leader of the cluster. A node that does not have the LEADER
760 capability can not be elected leader. If the current leader has this
761 capability removed then an election will occur.
762 Nodes have this capability enabled by default, but it can be removed
764 via the cluster:leader capability configuration setting or by using
765 this command.
766 See also "ctdb getcapabilities"
768 reloadnodes
770 This command is used when adding new nodes, or removing existing nodes
771 from an existing cluster.
772 Procedure to add nodes:
774 1. To expand an existing cluster, first ensure with ctdb status that
776 all nodes are up and running and that they are all healthy. Do not
777 try to expand a cluster unless it is completely healthy!
778 2. On all nodes, edit /etc/ctdb/nodes and add the new nodes at the end
780 of this file.
781 3. Verify that all the nodes have identical /etc/ctdb/nodes files
783 after adding the new nodes.
784 4. Run ctdb reloadnodes to force all nodes to reload the nodes file.
786 5. Use ctdb status on all nodes and verify that they now show the
788 additional nodes.
789 6. Install and configure the new node and bring it online.
791 Procedure to remove nodes:
793 1. To remove nodes from an existing cluster, first ensure with ctdb
795 status that all nodes, except the node to be deleted, are up and
796 running and that they are all healthy. Do not try to remove nodes
797 from a cluster unless the cluster is completely healthy!
798 2. Shutdown and power off the node to be removed.
800 3. On all other nodes, edit the /etc/ctdb/nodes file and comment out
802 the nodes to be removed. Do not delete the lines for the deleted
803 nodes, just comment them out by adding a '#' at the beginning of
804 the lines.
805 4. Run ctdb reloadnodes to force all nodes to reload the nodes file.
807 5. Use ctdb status on all nodes and verify that the deleted nodes are
809 no longer listed.
810 reloadips [PNN-LIST]
812 This command reloads the public addresses configuration file on the
813 specified nodes. When it completes addresses will be reconfigured and
814 reassigned across the cluster as necessary.
815 This command is currently unable to make changes to the netmask or
817 interfaces associated with existing addresses. Such changes must be
818 made in 2 steps by deleting addresses in question and re-adding then.
819 Unfortunately this will disrupt connections to the changed addresses.
820 getdbmap
822 This command lists all clustered TDB databases that the CTDB daemon has
823 attached to. Some databases are flagged as PERSISTENT, this means that
824 the database stores data persistently and the data will remain across
825 reboots. One example of such a database is secrets.tdb where
826 information about how the cluster was joined to the domain is stored.
827 Some database are flagged as REPLICATED, this means that the data in
828 that database is replicated across all the nodes. But the data will not
829 remain across reboots. This type of database is used by CTDB to store
830 it's internal state.
831 If a PERSISTENT database is not in a healthy state the database is
833 flagged as UNHEALTHY. If there's at least one completely healthy node
834 running in the cluster, it's possible that the content is restored by a
835 recovery run automatically. Otherwise an administrator needs to analyze
836 the problem.
837 See also "ctdb getdbstatus", "ctdb backupdb", "ctdb restoredb", "ctdb
839 dumpbackup", "ctdb wipedb", "ctdb setvar AllowUnhealthyDBRead 1" and
840 (if samba or tdb-utils are installed) "tdbtool check".
841 Most databases are not persistent and only store the state information
843 that the currently running samba daemons need. These databases are
844 always wiped when ctdb/samba starts and when a node is rebooted.
845 Example
847 # ctdb getdbmap
848 Number of databases:10
849 dbid:0x435d3410 name:notify.tdb path:/var/lib/ctdb/notify.tdb.0
850 dbid:0x42fe72c5 name:locking.tdb path:/var/lib/ctdb/locking.tdb.0
851 dbid:0x1421fb78 name:brlock.tdb path:/var/lib/ctdb/brlock.tdb.0
852 dbid:0x17055d90 name:connections.tdb path:/var/lib/ctdb/connections.tdb.0
853 dbid:0xc0bdde6a name:sessionid.tdb path:/var/lib/ctdb/sessionid.tdb.0
854 dbid:0x122224da name:test.tdb path:/var/lib/ctdb/test.tdb.0
855 dbid:0x2672a57f name:idmap2.tdb path:/var/lib/ctdb/persistent/idmap2.tdb.0 PERSISTENT
856 dbid:0xb775fff6 name:secrets.tdb path:/var/lib/ctdb/persistent/secrets.tdb.0 PERSISTENT
857 dbid:0xe98e08b6 name:group_mapping.tdb path:/var/lib/ctdb/persistent/group_mapping.tdb.0 PERSISTENT
858 dbid:0x7bbbd26c name:passdb.tdb path:/var/lib/ctdb/persistent/passdb.tdb.0 PERSISTENT
859 # ctdb getdbmap # example for unhealthy database
861 Number of databases:1
862 dbid:0xb775fff6 name:secrets.tdb path:/var/lib/ctdb/persistent/secrets.tdb.0 PERSISTENT UNHEALTHY
863 # ctdb -X getdbmap
865 |ID|Name|Path|Persistent|Unhealthy|
866 |0x7bbbd26c|passdb.tdb|/var/lib/ctdb/persistent/passdb.tdb.0|1|0|
867 backupdb DB FILE
870 Copy the contents of database DB to FILE. FILE can later be read back
871 using restoredb. This is mainly useful for backing up persistent
872 databases such as secrets.tdb and similar.
873 restoredb FILE [DB]
875 This command restores a persistent database that was previously backed
876 up using backupdb. By default the data will be restored back into the
877 same database as it was created from. By specifying dbname you can
878 restore the data into a different database.
879 setdbreadonly DB
881 This command will enable the read-only record support for a database.
882 This is an experimental feature to improve performance for contended
883 records primarily in locking.tdb and brlock.tdb. When enabling this
884 feature you must set it on all nodes in the cluster.
885 setdbsticky DB
887 This command will enable the sticky record support for the specified
888 database. This is an experimental feature to improve performance for
889 contended records primarily in locking.tdb and brlock.tdb. When
890 enabling this feature you must set it on all nodes in the cluster.
891
893 Internal commands are used by CTDB's scripts and are not required for
894 managing a CTDB cluster. Their parameters and behaviour are subject to
895 change.
896 gettickles IPADDR
898 Show TCP connections that are registered with CTDB to be "tickled" if
899 there is a failover.
900 gratarp IPADDR INTERFACE
902 Send out a gratuitous ARP for the specified interface through the
903 specified interface. This command is mainly used by the ctdb
904 eventscripts.
905 pdelete DB KEY
907 Delete KEY from DB.
908 pfetch DB KEY
910 Print the value associated with KEY in DB.
911 pstore DB KEY FILE
913 Store KEY in DB with contents of FILE as the associated value.
914 ptrans DB [FILE]
916 Read a list of key-value pairs, one per line from FILE, and store them
917 in DB using a single transaction. An empty value is equivalent to
918 deleting the given key.
919 The key and value should be separated by spaces or tabs. Each key/value
921 should be a printable string enclosed in double-quotes.
922 runstate [setup|first_recovery|startup|running]
924 Print the runstate of the specified node. Runstates are used to
925 serialise important state transitions in CTDB, particularly during
926 startup.
927 If one or more optional runstate arguments are specified then the node
929 must be in one of these runstates for the command to succeed.
930 Example
932 # ctdb runstate
933 RUNNING
934 setifacelink IFACE up|down
937 Set the internal state of network interface IFACE. This is typically
938 used in the 10.interface script in the "monitor" event.
939 Example: ctdb setifacelink eth0 up
941 tickle
943 Read a list of TCP connections, one per line, from standard input and
944 send a TCP tickle to the source host for each connection. A connection
945 is specified as:
946 SRC-IPADDR:SRC-PORT DST-IPADDR:DST-PORT
948 A single connection can be specified on the command-line rather than on
951 standard input.
952 A TCP tickle is a TCP ACK packet with an invalid sequence and
954 acknowledge number and will when received by the source host result in
955 it sending an immediate correct ACK back to the other end.
956 TCP tickles are useful to "tickle" clients after a IP failover has
958 occurred since this will make the client immediately recognize the TCP
959 connection has been disrupted and that the client will need to
960 reestablish. This greatly speeds up the time it takes for a client to
961 detect and reestablish after an IP failover in the ctdb cluster.
962 version
964 Display the CTDB version.
965
967 These commands are primarily used for CTDB development and testing and
968 should not be used for normal administration.
969 OPTIONS
971 --print-emptyrecords
972 This enables printing of empty records when dumping databases with
973 the catdb, cattbd and dumpdbbackup commands. Records with empty
974 data segment are considered deleted by ctdb and cleaned by the
975 vacuuming mechanism, so this switch can come in handy for debugging
976 the vacuuming behaviour.
977 --print-datasize
979 This lets database dumps (catdb, cattdb, dumpdbbackup) print the
980 size of the record data instead of dumping the data contents.
981 --print-lmaster
983 This lets catdb print the lmaster for each record.
984 --print-hash
986 This lets database dumps (catdb, cattdb, dumpdbbackup) print the
987 hash for each record.
988 --print-recordflags
990 This lets catdb and dumpdbbackup print the record flags for each
991 record. Note that cattdb always prints the flags.
992 process-exists PID [SRVID]
994 This command checks if a specific process exists on the CTDB host. This
995 is mainly used by Samba to check if remote instances of samba are still
996 running or not. When the optional SRVID argument is specified, the
997 command check if a specific process exists on the CTDB host and has
998 registered for specified SRVID.
999 getdbstatus DB
1001 This command displays more details about a database.
1002 Example
1004 # ctdb getdbstatus test.tdb.0
1005 dbid: 0x122224da
1006 name: test.tdb
1007 path: /var/lib/ctdb/test.tdb.0
1008 PERSISTENT: no
1009 HEALTH: OK
1010 # ctdb getdbstatus registry.tdb # with a corrupted TDB
1012 dbid: 0xf2a58948
1013 name: registry.tdb
1014 path: /var/lib/ctdb/persistent/registry.tdb.0
1015 PERSISTENT: yes
1016 HEALTH: NO-HEALTHY-NODES - ERROR - Backup of corrupted TDB in '/var/lib/ctdb/persistent/registry.tdb.0.corrupted.20091208091949.0Z'
1017 catdb DB
1020 Print a dump of the clustered TDB database DB.
1021 cattdb DB
1023 Print a dump of the contents of the local TDB database DB.
1024 dumpdbbackup FILE
1026 Print a dump of the contents from database backup FILE, similar to
1027 catdb.
1028 wipedb DB
1030 Remove all contents of database DB.
1031 recover
1033 This command will trigger the recovery daemon to do a cluster recovery.
1034 ipreallocate, sync
1036 This command will force the leader to perform a full ip reallocation
1037 process and redistribute all ip addresses. This is useful to "reset"
1038 the allocations back to its default state if they have been changed
1039 using the "moveip" command. While a "recover" will also perform this
1040 reallocation, a recovery is much more hevyweight since it will also
1041 rebuild all the databases.
1042 attach DBNAME [persistent|replicated]
1044 Create a new CTDB database called DBNAME and attach to it on all nodes.
1045 detach DB-LIST
1047 Detach specified non-persistent database(s) from the cluster. This
1048 command will disconnect specified database(s) on all nodes in the
1049 cluster. This command should only be used when none of the specified
1050 database(s) are in use.
1051 All nodes should be active and tunable AllowClientDBAccess should be
1053 disabled on all nodes before detaching databases.
1054 dumpmemory
1056 This is a debugging command. This command will make the ctdb daemon to
1057 write a fill memory allocation map to standard output.
1058 rddumpmemory
1060 This is a debugging command. This command will dump the talloc memory
1061 allocation tree for the recovery daemon to standard output.
1062 ban BANTIME
1064 Administratively ban a node for BANTIME seconds. The node will be
1065 unbanned after BANTIME seconds have elapsed.
1066 A banned node does not participate in the cluster. It does not host any
1068 records for the clustered TDB and does not host any public IP
1069 addresses.
1070 Nodes are automatically banned if they misbehave. For example, a node
1072 may be banned if it causes too many cluster recoveries.
1073 To administratively exclude a node from a cluster use the stop command.
1075 unban
1077 This command is used to unban a node that has either been
1078 administratively banned using the ban command or has been automatically
1079 banned.
1080
1082 ctdbd(1), onnode(1), ctdb(7), ctdb-statistics(7), ctdb-tunables(7),
1083 http://ctdb.samba.org/
1084
1086 This documentation was written by Ronnie Sahlberg, Amitay Isaacs,
1087 Martin Schwenke
1088
1090 Copyright © 2007 Andrew Tridgell, Ronnie Sahlberg
1091 This program is free software; you can redistribute it and/or modify it
1093 under the terms of the GNU General Public License as published by the
1094 Free Software Foundation; either version 3 of the License, or (at your
1095 option) any later version.
1096 This program is distributed in the hope that it will be useful, but
1098 WITHOUT ANY WARRANTY; without even the implied warranty of
1099 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
1100 General Public License for more details.
1101 You should have received a copy of the GNU General Public License along
1103 with this program; if not, see http://www.gnu.org/licenses.
1104ctdb 01/26/2023 CTDB(1)