1CTDB(1) CTDB(1)
2
3
4
6 ctdb - clustered tdb database management utility
7
9 ctdb [ OPTIONS ] COMMAND ...
10
11 ctdb [-n <node>] [-Y] [-t <timeout>] [-T <timelimit>] [-? --help]
12 [--usage] [-d --debug=<INTEGER>] [--socket=<filename>]
13
15 ctdb is a utility to view and manage a ctdb cluster.
16
18 -n <pnn>
19 This specifies the physical node number on which to execute the
20 command. Default is to run the command on the deamon running on the
21 local host.
22
23 The physical node number is an integer that describes the node in
24 the cluster. The first node has physical node number 0.
25
26 -Y
27 Produce output in machine readable form for easier parsing by
28 scripts. Not all commands support this option.
29
30 -t <timeout>
31 How long should ctdb wait for the local ctdb daemon to respond to a
32 command before timing out. Default is 3 seconds.
33
34 -T <timelimit>
35 A limit on how long the ctdb command will run for before it will be
36 aborted. When this timelimit has been exceeded the ctdb command
37 will terminate.
38
39 -? --help
40 Print some help text to the screen.
41
42 --usage
43 Print useage information to the screen.
44
45 -d --debug=<debuglevel>
46 Change the debug level for the command. Default is 0.
47
48 --socket=<filename>
49 Specify the socketname to use when connecting to the local ctdb
50 daemon. The default is /tmp/ctdb.socket .
51
52 You only need to specify this parameter if you run multiple ctdb
53 daemons on the same physical host and thus can not use the default
54 name for the domain socket.
55
57 These are commands used to monitor and administrate a CTDB cluster.
58
59 pnn
60 This command displays the pnn of the current node.
61
62 status
63 This command shows the current status of the ctdb node.
64
65 node status
66 Node status reflects the current status of the node. There are five
67 possible states:
68
69 OK - This node is fully functional.
70
71 DISCONNECTED - This node could not be connected through the network
72 and is currently not participating in the cluster. If there is a
73 public IP address associated with this node it should have been
74 taken over by a different node. No services are running on this
75 node.
76
77 DISABLED - This node has been administratively disabled. This node
78 is still functional and participates in the CTDB cluster but its IP
79 addresses have been taken over by a different node and no services
80 are currently being hosted.
81
82 UNHEALTHY - A service provided by this node is malfunctioning and
83 should be investigated. The CTDB daemon itself is operational and
84 participates in the cluster. Its public IP address has been taken
85 over by a different node and no services are currnetly being
86 hosted. All unhealthy nodes should be investigated and require an
87 administrative action to rectify.
88
89 BANNED - This node failed too many recovery attempts and has been
90 banned from participating in the cluster for a period of
91 RecoveryBanPeriod seconds. Any public IP address has been taken
92 over by other nodes. This node does not provide any services. All
93 banned nodes should be investigated and require an administrative
94 action to rectify. This node does not perticipate in the CTDB
95 cluster but can still be communicated with. I.e. ctdb commands can
96 be sent to it.
97
98 STOPPED - A node that is stopped does not host any public ip
99 addresses, nor is it part of the VNNMAP. A stopped node can not
100 become LVSMASTER, RECMASTER or NATGW. This node does not
101 perticipate in the CTDB cluster but can still be communicated with.
102 I.e. ctdb commands can be sent to it.
103
104 generation
105 The generation id is a number that indicates the current generation
106 of a cluster instance. Each time a cluster goes through a
107 reconfiguration or a recovery its generation id will be changed.
108
109 This number does not have any particular meaning other than to keep
110 track of when a cluster has gone through a recovery. It is a random
111 number that represents the current instance of a ctdb cluster and
112 its databases. CTDBD uses this number internally to be able to tell
113 when commands to operate on the cluster and the databases was
114 issued in a different generation of the cluster, to ensure that
115 commands that operate on the databases will not survive across a
116 cluster database recovery. After a recovery, all old outstanding
117 commands will automatically become invalid.
118
119 Sometimes this number will be shown as "INVALID". This only means
120 that the ctdbd daemon has started but it has not yet merged with
121 the cluster through a recovery. All nodes start with generation
122 "INVALID" and are not assigned a real generation id until they have
123 successfully been merged with a cluster through a recovery.
124
125 VNNMAP
126 The list of Virtual Node Numbers. This is a list of all nodes that
127 actively participates in the cluster and that share the workload of
128 hosting the Clustered TDB database records. Only nodes that are
129 participating in the vnnmap can become lmaster or dmaster for a
130 database record.
131
132 Recovery mode
133 This is the current recovery mode of the cluster. There are two
134 possible modes:
135
136 NORMAL - The cluster is fully operational.
137
138 RECOVERY - The cluster databases have all been frozen, pausing all
139 services while the cluster awaits a recovery process to complete. A
140 recovery process should finish within seconds. If a cluster is
141 stuck in the RECOVERY state this would indicate a cluster
142 malfunction which needs to be investigated.
143
144 Once the recovery master detects an inconsistency, for example a
145 node becomes disconnected/connected, the recovery daemon will
146 trigger a cluster recovery process, where all databases are
147 remerged across the cluster. When this process starts, the recovery
148 master will first "freeze" all databases to prevent applications
149 such as samba from accessing the databases and it will also mark
150 the recovery mode as RECOVERY.
151
152 When CTDBD starts up, it will start in RECOVERY mode. Once the node
153 has been merged into a cluster and all databases have been
154 recovered, the node mode will change into NORMAL mode and the
155 databases will be "thawed", allowing samba to access the databases
156 again.
157
158 Recovery master
159 This is the cluster node that is currently designated as the
160 recovery master. This node is responsible of monitoring the
161 consistency of the cluster and to perform the actual recovery
162 process when reqired.
163
164 Only one node at a time can be the designated recovery master.
165 Which node is designated the recovery master is decided by an
166 election process in the recovery daemons running on each node.
167
168 Example: ctdb status
169
170 Example output:
171
172 Number of nodes:4
173 pnn:0 11.1.2.200 OK (THIS NODE)
174 pnn:1 11.1.2.201 OK
175 pnn:2 11.1.2.202 OK
176 pnn:3 11.1.2.203 OK
177 Generation:1362079228
178 Size:4
179 hash:0 lmaster:0
180 hash:1 lmaster:1
181 hash:2 lmaster:2
182 hash:3 lmaster:3
183 Recovery mode:NORMAL (0)
184 Recovery master:0
185
186
187 recmaster
188 This command shows the pnn of the node which is currently the
189 recmaster.
190
191 uptime
192 This command shows the uptime for the ctdb daemon. When the last
193 recovery or ip-failover completed and how long it took. If the
194 "duration" is shown as a negative number, this indicates that there is
195 a recovery/failover in progress and it started that many seconds ago.
196
197 Example: ctdb uptime
198
199 Example output:
200
201 Current time of node : Thu Oct 29 10:38:54 2009
202 Ctdbd start time : (000 16:54:28) Wed Oct 28 17:44:26 2009
203 Time of last recovery/failover: (000 16:53:31) Wed Oct 28 17:45:23 2009
204 Duration of last recovery/failover: 2.248552 seconds
205
206
207 listnodes
208 This command shows lists the ip addresses of all the nodes in the
209 cluster.
210
211 Example: ctdb listnodes
212
213 Example output:
214
215 10.0.0.71
216 10.0.0.72
217 10.0.0.73
218 10.0.0.74
219
220
221 ping
222 This command will "ping" all CTDB daemons in the cluster to verify that
223 they are processing commands correctly.
224
225 Example: ctdb ping
226
227 Example output:
228
229 response from 0 time=0.000054 sec (3 clients)
230 response from 1 time=0.000144 sec (2 clients)
231 response from 2 time=0.000105 sec (2 clients)
232 response from 3 time=0.000114 sec (2 clients)
233
234
235 ip
236 This command will display the list of public addresses that are
237 provided by the cluster and which physical node is currently serving
238 this ip. By default this command will ONLY show those public addresses
239 that are known to the node itself. To see the full list of all public
240 ips across the cluster you must use "ctdb ip -n all".
241
242 Example: ctdb ip
243
244 Example output:
245
246 Number of addresses:4
247 12.1.1.1 0
248 12.1.1.2 1
249 12.1.1.3 2
250 12.1.1.4 3
251
252
253 scriptstatus
254 This command displays which scripts where run in the previous
255 monitoring cycle and the result of each script. If a script failed with
256 an error, causing the node to become unhealthy, the output from that
257 script is also shown.
258
259 Example: ctdb scriptstatus
260
261 Example output:
262
263 7 scripts were executed last monitoring cycle
264 00.ctdb Status:OK Duration:0.056 Tue Mar 24 18:56:57 2009
265 10.interface Status:OK Duration:0.077 Tue Mar 24 18:56:57 2009
266 11.natgw Status:OK Duration:0.039 Tue Mar 24 18:56:57 2009
267 20.multipathd Status:OK Duration:0.038 Tue Mar 24 18:56:57 2009
268 31.clamd Status:DISABLED
269 40.vsftpd Status:OK Duration:0.045 Tue Mar 24 18:56:57 2009
270 41.httpd Status:OK Duration:0.039 Tue Mar 24 18:56:57 2009
271 50.samba Status:ERROR Duration:0.082 Tue Mar 24 18:56:57 2009
272 OUTPUT:ERROR: Samba tcp port 445 is not responding
273
274
275 disablescript <script>
276 This command is used to disable an eventscript.
277
278 This will take effect the next time the eventscripts are being executed
279 so it can take a short while until this is reflected in ´scriptstatus´.
280
281 enablescript <script>
282 This command is used to enable an eventscript.
283
284 This will take effect the next time the eventscripts are being executed
285 so it can take a short while until this is reflected in ´scriptstatus´.
286
287 getvar <name>
288 Get the runtime value of a tuneable variable.
289
290 Example: ctdb getvar MaxRedirectCount
291
292 Example output:
293
294 MaxRedirectCount = 3
295
296
297 setvar <name> <value>
298 Set the runtime value of a tuneable variable.
299
300 Example: ctdb setvar MaxRedirectCount 5
301
302 listvars
303 List all tuneable variables.
304
305 Example: ctdb listvars
306
307 Example output:
308
309 MaxRedirectCount = 3
310 SeqnumInterval = 1000
311 ControlTimeout = 60
312 TraverseTimeout = 20
313 KeepaliveInterval = 5
314 KeepaliveLimit = 5
315 MaxLACount = 7
316 RecoverTimeout = 20
317 RecoverInterval = 1
318 ElectionTimeout = 3
319 TakeoverTimeout = 5
320 MonitorInterval = 15
321 TickleUpdateInterval = 20
322 EventScriptTimeout = 30
323 EventScriptBanCount = 10
324 EventScriptUnhealthyOnTimeout = 0
325 RecoveryGracePeriod = 120
326 RecoveryBanPeriod = 300
327 DatabaseHashSize = 10000
328 DatabaseMaxDead = 5
329 RerecoveryTimeout = 10
330 EnableBans = 1
331 DeterministicIPs = 1
332 DisableWhenUnhealthy = 0
333 ReclockPingPeriod = 60
334 NoIPFailback = 0
335 VerboseMemoryNames = 0
336 RecdPingTimeout = 60
337 RecdFailCount = 10
338 LogLatencyMs = 0
339 RecLockLatencyMs = 1000
340 RecoveryDropAllIPs = 60
341 VerifyRecoveryLock = 1
342 VacuumDefaultInterval = 300
343 VacuumMaxRunTime = 30
344 RepackLimit = 10000
345 VacuumLimit = 5000
346 VacuumMinInterval = 60
347 VacuumMaxInterval = 600
348 MaxQueueDropMsg = 1000
349 UseStatusEvents = 0
350 AllowUnhealthyDBRead = 0
351
352
353 lvsmaster
354 This command shows which node is currently the LVSMASTER. The LVSMASTER
355 is the node in the cluster which drives the LVS system and which
356 receives all incoming traffic from clients.
357
358 LVS is the mode where the entire CTDB/Samba cluster uses a single ip
359 address for the entire cluster. In this mode all clients connect to one
360 specific node which will then multiplex/loadbalance the clients evenly
361 onto the other nodes in the cluster. This is an alternative to using
362 public ip addresses. See the manpage for ctdbd for more information
363 about LVS.
364
365 lvs
366 This command shows which nodes in the cluster are currently active in
367 the LVS configuration. I.e. which nodes we are currently loadbalancing
368 the single ip address across.
369
370 LVS will by default only loadbalance across those nodes that are both
371 LVS capable and also HEALTHY. Except if all nodes are UNHEALTHY in
372 which case LVS will loadbalance across all UNHEALTHY nodes as well. LVS
373 will never use nodes that are DISCONNECTED, STOPPED, BANNED or
374 DISABLED.
375
376 Example output:
377
378 2:10.0.0.13
379 3:10.0.0.14
380
381
382 getcapabilities
383 This command shows the capabilities of the current node. Please see
384 manpage for ctdbd for a full list of all capabilities and more detailed
385 description.
386
387 RECMASTER and LMASTER capabilities are primarily used when CTDBD is
388 used to create a cluster spanning across WAN links. In which case ctdbd
389 acts as a WAN accelerator.
390
391 LVS capabile means that the node is participating in LVS, a mode where
392 the entire CTDB cluster uses one single ip address for the entire
393 cluster instead of using public ip address failover. This is an
394 alternative to using a loadbalancing layer-4 switch.
395
396 Example output:
397
398 RECMASTER: YES
399 LMASTER: YES
400 LVS: NO
401
402
403 statistics
404 Collect statistics from the CTDB daemon about how many calls it has
405 served.
406
407 Example: ctdb statistics
408
409 Example output:
410
411 CTDB version 1
412 num_clients 3
413 frozen 0
414 recovering 0
415 client_packets_sent 360489
416 client_packets_recv 360466
417 node_packets_sent 480931
418 node_packets_recv 240120
419 keepalive_packets_sent 4
420 keepalive_packets_recv 3
421 node
422 req_call 2
423 reply_call 2
424 req_dmaster 0
425 reply_dmaster 0
426 reply_error 0
427 req_message 42
428 req_control 120408
429 reply_control 360439
430 client
431 req_call 2
432 req_message 24
433 req_control 360440
434 timeouts
435 call 0
436 control 0
437 traverse 0
438 total_calls 2
439 pending_calls 0
440 lockwait_calls 0
441 pending_lockwait_calls 0
442 memory_used 5040
443 max_hop_count 0
444 max_call_latency 4.948321 sec
445 max_lockwait_latency 0.000000 sec
446
447
448 statisticsreset
449 This command is used to clear all statistics counters in a node.
450
451 Example: ctdb statisticsreset
452
453 getreclock
454 This command is used to show the filename of the reclock file that is
455 used.
456
457 Example output:
458
459 Reclock file:/gpfs/.ctdb/shared
460
461
462 setreclock [filename]
463 This command is used to modify, or clear, the file that is used as the
464 reclock file at runtime. When this command is used, the reclock file
465 checks are disabled. To re-enable the checks the administrator needs to
466 activate the "VerifyRecoveryLock" tunable using "ctdb setvar".
467
468 If run with no parameter this will remove the reclock file completely.
469 If run with a parameter the parameter specifies the new filename to use
470 for the recovery lock.
471
472 This command only affects the runtime settings of a ctdb node and will
473 be lost when ctdb is restarted. For persistent changes to the reclock
474 file setting you must edit /etc/sysconfig/ctdb.
475
476 getdebug
477 Get the current debug level for the node. the debug level controls what
478 information is written to the log file.
479
480 The debug levels are mapped to the corresponding syslog levels. When a
481 debug level is set, only those messages at that level and higher levels
482 will be printed.
483
484 The list of debug levels from highest to lowest are :
485
486 EMERG ALERT CRIT ERR WARNING NOTICE INFO DEBUG
487
488 setdebug <debuglevel>
489 Set the debug level of a node. This controls what information will be
490 logged.
491
492 The debuglevel is one of EMERG ALERT CRIT ERR WARNING NOTICE INFO DEBUG
493
494 getpid
495 This command will return the process id of the ctdb daemon.
496
497 disable
498 This command is used to administratively disable a node in the cluster.
499 A disabled node will still participate in the cluster and host
500 clustered TDB records but its public ip address has been taken over by
501 a different node and it no longer hosts any services.
502
503 enable
504 Re-enable a node that has been administratively disabled.
505
506 stop
507 This command is used to administratively STOP a node in the cluster. A
508 STOPPED node is connected to the cluster but will not host any public
509 ip addresse, nor does it participate in the VNNMAP. The difference
510 between a DISABLED node and a STOPPED node is that a STOPPED node does
511 not host any parts of the database which means that a recovery is
512 required to stop/continue nodes.
513
514 continue
515 Re-start a node that has been administratively stopped.
516
517 addip <public_ip/mask> <iface>
518 This command is used to add a new public ip to a node during runtime.
519 This allows public addresses to be added to a cluster without having to
520 restart the ctdb daemons.
521
522 Note that this only updates the runtime instance of ctdb. Any changes
523 will be lost next time ctdb is restarted and the public addresses file
524 is re-read. If you want this change to be permanent you must also
525 update the public addresses file manually.
526
527 delip <public_ip>
528 This command is used to remove a public ip from a node during runtime.
529 If this public ip is currently hosted by the node it being removed
530 from, the ip will first be failed over to another node, if possible,
531 before it is removed.
532
533 Note that this only updates the runtime instance of ctdb. Any changes
534 will be lost next time ctdb is restarted and the public addresses file
535 is re-read. If you want this change to be permanent you must also
536 update the public addresses file manually.
537
538 moveip <public_ip> <node>
539 This command can be used to manually fail a public ip address to a
540 specific node.
541
542 In order to manually override the "automatic" distribution of public ip
543 addresses that ctdb normally provides, this command only works when you
544 have changed the tunables for the daemon to:
545
546 DeterministicIPs = 0
547
548 NoIPFailback = 1
549
550 shutdown
551 This command will shutdown a specific CTDB daemon.
552
553 recover
554 This command will trigger the recovery daemon to do a cluster recovery.
555
556 ipreallocate
557 This command will force the recovery master to perform a full ip
558 reallocation process and redistribute all ip addresses. This is useful
559 to "reset" the allocations back to its default state if they have been
560 changed using the "moveip" command. While a "recover" will also perform
561 this reallocation, a recovery is much more hevyweight since it will
562 also rebuild all the databases.
563
564 setlmasterrole <on|off>
565 This command is used ot enable/disable the LMASTER capability for a
566 node at runtime. This capability determines whether or not a node can
567 be used as an LMASTER for records in the database. A node that does not
568 have the LMASTER capability will not show up in the vnnmap.
569
570 Nodes will by default have this capability, but it can be stripped off
571 nodes by the setting in the sysconfig file or by using this command.
572
573 Once this setting has been enabled/disabled, you need to perform a
574 recovery for it to take effect.
575
576 See also "ctdb getcapabilities"
577
578 setrecmasterrole <on|off>
579 This command is used ot enable/disable the RECMASTER capability for a
580 node at runtime. This capability determines whether or not a node can
581 be used as an RECMASTER for the cluster. A node that does not have the
582 RECMASTER capability can not win a recmaster election. A node that
583 already is the recmaster for the cluster when the capability is
584 stripped off the node will remain the recmaster until the next cluster
585 election.
586
587 Nodes will by default have this capability, but it can be stripped off
588 nodes by the setting in the sysconfig file or by using this command.
589
590 See also "ctdb getcapabilities"
591
592 killtcp <srcip:port> <dstip:port>
593 This command will kill the specified TCP connection by issuing a TCP
594 RST to the srcip:port endpoint. This is a command used by the ctdb
595 eventscripts.
596
597 gratiousarp <ip> <interface>
598 This command will send out a gratious arp for the specified interface
599 through the specified interface. This command is mainly used by the
600 ctdb eventscripts.
601
602 reloadnodes
603 This command is used when adding new nodes, or removing existing nodes
604 from an existing cluster.
605
606 Procedure to add a node:
607
608 1, To expand an existing cluster, first ensure with ´ctdb status´ that
609 all nodes are up and running and that they are all healthy. Do not try
610 to expand a cluster unless it is completely healthy!
611
612 2, On all nodes, edit /etc/ctdb/nodes and add the new node as the last
613 entry to the file. The new node MUST be added to the end of this file!
614
615 3, Verify that all the nodes have identical /etc/ctdb/nodes files after
616 you edited them and added the new node!
617
618 4, Run ´ctdb reloadnodes´ to force all nodes to reload the nodesfile.
619
620 5, Use ´ctdb status´ on all nodes and verify that they now show the
621 additional node.
622
623 6, Install and configure the new node and bring it online.
624
625 Procedure to remove a node:
626
627 1, To remove a node from an existing cluster, first ensure with ´ctdb
628 status´ that all nodes, except the node to be deleted, are up and
629 running and that they are all healthy. Do not try to remove a node from
630 a cluster unless the cluster is completely healthy!
631
632 2, Shutdown and poerwoff the node to be removed.
633
634 3, On all other nodes, edit the /etc/ctdb/nodes file and comment out
635 the node to be removed. Do not delete the line for that node, just
636 comment it out by adding a ´#´ at the beginning of the line.
637
638 4, Run ´ctdb reloadnodes´ to force all nodes to reload the nodesfile.
639
640 5, Use ´ctdb status´ on all nodes and verify that the deleted node no
641 longer shows up in the list..
642
643 tickle <srcip:port> <dstip:port>
644 This command will will send a TCP tickle to the source host for the
645 specified TCP connection. A TCP tickle is a TCP ACK packet with an
646 invalid sequence and acknowledge number and will when received by the
647 source host result in it sending an immediate correct ACK back to the
648 other end.
649
650 TCP tickles are useful to "tickle" clients after a IP failover has
651 occured since this will make the client immediately recognize the TCP
652 connection has been disrupted and that the client will need to
653 reestablish. This greatly speeds up the time it takes for a client to
654 detect and reestablish after an IP failover in the ctdb cluster.
655
656 gettickles <ip>
657 This command is used to show which TCP connections are registered with
658 CTDB to be "tickled" if there is a failover.
659
660 repack [max_freelist]
661 Over time, when records are created and deleted in a TDB, the TDB list
662 of free space will become fragmented. This can lead to a slowdown in
663 accessing TDB records. This command is used to defragment a TDB
664 database and pruning the freelist.
665
666 If [max_freelist] is specified, then a database will only be repacked
667 if it has more than this number of entries in the freelist.
668
669 During repacking of the database, the entire TDB database will be
670 locked to prevent writes. If samba tries to write to a record in the
671 database during a repack operation, samba will block until the
672 repacking has completed.
673
674 This command can be disruptive and can cause samba to block for the
675 duration of the repack operation. In general, a repack operation will
676 take less than one second to complete.
677
678 A repack operation will only defragment the local TDB copy of the CTDB
679 database. You need to run this command on all of the nodes to repack a
680 CTDB database completely.
681
682 Example: ctdb repack 1000
683
684 By default, this operation is issued from the 00.ctdb event script
685 every 5 minutes.
686
687 vacuum [max_records]
688 Over time CTDB databases will fill up with empty deleted records which
689 will lead to a progressive slow down of CTDB database access. This
690 command is used to prune all databases and delete all empty records
691 from the cluster.
692
693 By default, vacuum will delete all empty records from all databases. If
694 [max_records] is specified, the command will only delete the first
695 [max_records] empty records for each database.
696
697 Vacuum only deletes records where the local node is the lmaster. To
698 delete all records from the entire cluster you need to run a vacuum
699 from each node. This command is not disruptive. Samba is unaffected and
700 will still be able to read/write records normally while the database is
701 being vacuumed.
702
703 Example: ctdb vacuum
704
705 By default, this operation is issued from the 00.ctdb event script
706 every 5 minutes.
707
708 backupdb <dbname> <file>
709 This command can be used to copy the entire content of a database out
710 to a file. This file can later be read back into ctdb using the
711 restoredb command. This is mainly useful for backing up persistent
712 databases such as secrets.tdb and similar.
713
714 restoredb <file>
715 This command restores a persistent database that was previously backed
716 up using backupdb.
717
718 wipedb <dbname>
719 This command can be used to remove all content of a database.
720
722 These commands are primarily used for CTDB development and testing and
723 should not be used for normal administration.
724
725 process-exists <pid>
726 This command checks if a specific process exists on the CTDB host. This
727 is mainly used by Samba to check if remote instances of samba are still
728 running or not.
729
730 getdbmap
731 This command lists all clustered TDB databases that the CTDB daemon has
732 attached to. Some databases are flagged as PERSISTENT, this means that
733 the database stores data persistently and the data will remain across
734 reboots. One example of such a database is secrets.tdb where
735 information about how the cluster was joined to the domain is stored.
736
737 If a PERSISTENT database is not in a healthy state the database is
738 flagged as UNHEALTHY. If there´s at least one completely healthy node
739 running in the cluster, it´s possible that the content is restored by a
740 recovery run automaticly. Otherwise an administrator needs to analyze
741 the problem.
742
743 See also "ctdb getdbstatus", "ctdb backupdb", "ctdb restoredb", "ctdb
744 dumpbackup", "ctdb wipedb", "ctdb setvar AllowUnhealthyDBRead 1" and
745 (if samba or tdb-utils are installed) "tdbtool check".
746
747 Most databases are not persistent and only store the state information
748 that the currently running samba daemons need. These databases are
749 always wiped when ctdb/samba starts and when a node is rebooted.
750
751 Example: ctdb getdbmap
752
753 Example output:
754
755 Number of databases:10
756 dbid:0x435d3410 name:notify.tdb path:/var/ctdb/notify.tdb.0
757 dbid:0x42fe72c5 name:locking.tdb path:/var/ctdb/locking.tdb.0 dbid:0x1421fb78 name:brlock.tdb path:/var/ctdb/brlock.tdb.0
758 dbid:0x17055d90 name:connections.tdb path:/var/ctdb/connections.tdb.0
759 dbid:0xc0bdde6a name:sessionid.tdb path:/var/ctdb/sessionid.tdb.0
760 dbid:0x122224da name:test.tdb path:/var/ctdb/test.tdb.0
761 dbid:0x2672a57f name:idmap2.tdb path:/var/ctdb/persistent/idmap2.tdb.0 PERSISTENT
762 dbid:0xb775fff6 name:secrets.tdb path:/var/ctdb/persistent/secrets.tdb.0 PERSISTENT
763 dbid:0xe98e08b6 name:group_mapping.tdb path:/var/ctdb/persistent/group_mapping.tdb.0 PERSISTENT
764 dbid:0x7bbbd26c name:passdb.tdb path:/var/ctdb/persistent/passdb.tdb.0 PERSISTENT
765
766
767 Example output for an unhealthy database:
768
769 Number of databases:1
770 dbid:0xb775fff6 name:secrets.tdb path:/var/ctdb/persistent/secrets.tdb.0 PERSISTENT UNHEALTHY
771
772
773 Example output for a healthy database as machinereadable output -Y:
774
775 :ID:Name:Path:Persistent:Unhealthy:
776 :0x7bbbd26c:passdb.tdb:/var/ctdb/persistent/passdb.tdb.0:1:0:
777
778
779 getdbstatus <dbname>
780 This command displays more details about a database.
781
782 Example: ctdb getdbstatus test.tdb.0
783
784 Example output:
785
786 dbid: 0x122224da
787 name: test.tdb
788 path: /var/ctdb/test.tdb.0
789 PERSISTENT: no
790 HEALTH: OK
791
792
793 Example: ctdb getdbstatus registry.tdb (with a corrupted TDB)
794
795 Example output:
796
797 dbid: 0xf2a58948
798 name: registry.tdb
799 path: /var/ctdb/persistent/registry.tdb.0
800 PERSISTENT: yes
801 HEALTH: NO-HEALTHY-NODES - ERROR - Backup of corrupted TDB in ´/var/ctdb/persistent/registry.tdb.0.corrupted.20091208091949.0Z´
802
803
804 catdb <dbname>
805 This command will dump a clustered TDB database to the screen. This is
806 a debugging command.
807
808 dumpdbbackup <backup-file>
809 This command will dump the content of database backup to the screen
810 (similar to ctdb catdb). This is a debugging command.
811
812 getmonmode
813 This command returns the monutoring mode of a node. The monitoring mode
814 is either ACTIVE or DISABLED. Normally a node will continously monitor
815 that all other nodes that are expected are in fact connected and that
816 they respond to commands.
817
818 ACTIVE - This is the normal mode. The node is actively monitoring all
819 other nodes, both that the transport is connected and also that the
820 node responds to commands. If a node becomes unavailable, it will be
821 marked as DISCONNECTED and a recovery is initiated to restore the
822 cluster.
823
824 DISABLED - This node is not monitoring that other nodes are available.
825 In this mode a node failure will not be detected and no recovery will
826 be performed. This mode is useful when for debugging purposes one wants
827 to attach GDB to a ctdb process but wants to prevent the rest of the
828 cluster from marking this node as DISCONNECTED and do a recovery.
829
830 setmonmode <0|1>
831 This command can be used to explicitely disable/enable monitoring mode
832 on a node. The main purpose is if one wants to attach GDB to a running
833 ctdb daemon but wants to prevent the other nodes from marking it as
834 DISCONNECTED and issuing a recovery. To do this, set monitoring mode to
835 0 on all nodes before attaching with GDB. Remember to set monitoring
836 mode back to 1 afterwards.
837
838 attach <dbname>
839 This is a debugging command. This command will make the CTDB daemon
840 create a new CTDB database and attach to it.
841
842 dumpmemory
843 This is a debugging command. This command will make the ctdb daemon to
844 write a fill memory allocation map to standard output.
845
846 rddumpmemory
847 This is a debugging command. This command will dump the talloc memory
848 allocation tree for the recovery daemon to standard output.
849
850 freeze
851 This command will lock all the local TDB databases causing clients that
852 are accessing these TDBs such as samba3 to block until the databases
853 are thawed.
854
855 This is primarily used by the recovery daemon to stop all samba daemons
856 from accessing any databases while the database is recovered and
857 rebuilt.
858
859 thaw
860 Thaw a previously frozen node.
861
862 eventscript <arguments>
863 This is a debugging command. This command can be used to manually
864 invoke and run the eventscritps with arbitrary arguments.
865
866 ban <bantime|0>
867 Administratively ban a node for bantime seconds. A bantime of 0 means
868 that the node should be permanently banned.
869
870 A banned node does not participate in the cluster and does not host any
871 records for the clustered TDB. Its ip address has been taken over by an
872 other node and no services are hosted.
873
874 Nodes are automatically banned if they are the cause of too many
875 cluster recoveries.
876
877 unban
878 This command is used to unban a node that has either been
879 administratively banned using the ban command or has been automatically
880 banned by the recovery daemon.
881
883 ctdbd(1), onnode(1) http://ctdb.samba.org/
884
886 Copyright (C) Andrew Tridgell 2007
887 Copyright (C) Ronnie sahlberg 2007
888
889 This program is free software; you can redistribute it and/or modify
890 it under the terms of the GNU General Public License as published by
891 the Free Software Foundation; either version 3 of the License, or (at
892 your option) any later version.
893
894 This program is distributed in the hope that it will be useful, but
895 WITHOUT ANY WARRANTY; without even the implied warranty of
896 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
897 General Public License for more details.
898
899 You should have received a copy of the GNU General Public License
900 along with this program; if not, see http://www.gnu.org/licenses/.
901
902
903
904 12/09/2009 CTDB(1)