1 <?xml version="1.0" encoding="iso-8859-1"?>
3 PUBLIC "-//OASIS//DTD DocBook XML V4.2//EN"
4 "http://www.oasis-open.org/docbook/xml/4.2/docbookx.dtd">
8 <refentrytitle>ctdb</refentrytitle>
9 <manvolnum>1</manvolnum>
10 <refmiscinfo class="source">ctdb</refmiscinfo>
11 <refmiscinfo class="manual">CTDB - clustered TDB database</refmiscinfo>
16 <refname>ctdb</refname>
17 <refpurpose>clustered tdb database management utility</refpurpose>
22 <command>ctdb [ OPTIONS ] COMMAND ...</command>
26 <command>ctdb</command>
27 <arg choice="opt">-n <node></arg>
28 <arg choice="opt">-Y</arg>
29 <arg choice="opt">-t <timeout></arg>
30 <arg choice="opt">-T <timelimit></arg>
31 <arg choice="opt">-? --help</arg>
32 <arg choice="opt">--usage</arg>
33 <arg choice="opt">-d --debug=<INTEGER></arg>
34 <arg choice="opt">--socket=<filename></arg>
35 <arg choice="opt">--print-emptyrecords</arg>
36 <arg choice="opt">--print-datasize</arg>
37 <arg choice="opt">--print-lmaster</arg>
38 <arg choice="opt">--print-hash</arg>
39 <arg choice="opt">--print-recordflags</arg>
44 <refsect1><title>DESCRIPTION</title>
46 ctdb is a utility to view and manage a ctdb cluster.
52 <title>OPTIONS</title>
55 <varlistentry><term>-n <pnn></term>
58 This specifies the physical node number on which to execute the
59 command. Default is to run the command on the daemon running on
63 The physical node number is an integer that describes the node in the
64 cluster. The first node has physical node number 0.
69 <varlistentry><term>-Y</term>
72 Produce output in machine readable form for easier parsing by scripts. Not all commands support this option.
77 <varlistentry><term>-t <timeout></term>
80 How long should ctdb wait for the local ctdb daemon to respond to a command before timing out. Default is 3 seconds.
85 <varlistentry><term>-T <timelimit></term>
88 A limit on how long the ctdb command will run for before it will
89 be aborted. When this timelimit has been exceeded the ctdb command will
95 <varlistentry><term>-? --help</term>
98 Print some help text to the screen.
103 <varlistentry><term>--usage</term>
106 Print useage information to the screen.
111 <varlistentry><term>-d --debug=<debuglevel></term>
114 Change the debug level for the command. Default is 0.
119 <varlistentry><term>--socket=<filename></term>
122 Specify the socketname to use when connecting to the local ctdb
123 daemon. The default is /tmp/ctdb.socket .
126 You only need to specify this parameter if you run multiple ctdb
127 daemons on the same physical host and thus can not use the default
128 name for the domain socket.
133 <varlistentry><term>--print-emptyrecords</term>
136 This enables printing of empty records when dumping databases
137 with the catdb, cattbd and dumpdbbackup commands. Records with
138 empty data segment are considered deleted by ctdb and cleaned
139 by the vacuuming mechanism, so this switch can come in handy for
140 debugging the vacuuming behaviour.
145 <varlistentry><term>--print-datasize</term>
148 This lets database dumps (catdb, cattdb, dumpdbbackup) print the
149 size of the record data instead of dumping the data contents.
154 <varlistentry><term>--print-lmaster</term>
157 This lets catdb print the lmaster for each record.
162 <varlistentry><term>--print-hash</term>
165 This lets database dumps (catdb, cattdb, dumpdbbackup) print the
166 hash for each record.
171 <varlistentry><term>--print-recordflags</term>
174 This lets catdb and dumpdbbackup print the
175 record flags for each record. Note that cattdb always
185 <refsect1><title>Administrative Commands</title>
187 These are commands used to monitor and administrate a CTDB cluster.
190 <refsect2><title>pnn</title>
192 This command displays the pnn of the current node.
196 <refsect2><title>status</title>
198 This command shows the current status of the ctdb node.
201 <refsect3><title>node status</title>
203 Node status reflects the current status of the node. There are five possible states:
206 OK - This node is fully functional.
209 DISCONNECTED - This node could not be connected through the network and is currently not participating in the cluster. If there is a public IP address associated with this node it should have been taken over by a different node. No services are running on this node.
212 DISABLED - This node has been administratively disabled. This node is still functional and participates in the CTDB cluster but its IP addresses have been taken over by a different node and no services are currently being hosted.
215 UNHEALTHY - A service provided by this node is malfunctioning and should be investigated. The CTDB daemon itself is operational and participates in the cluster. Its public IP address has been taken over by a different node and no services are currnetly being hosted. All unhealthy nodes should be investigated and require an administrative action to rectify.
218 BANNED - This node failed too many recovery attempts and has been banned from participating in the cluster for a period of RecoveryBanPeriod seconds. Any public IP address has been taken over by other nodes. This node does not provide any services. All banned nodes should be investigated and require an administrative action to rectify. This node does not perticipate in the CTDB cluster but can still be communicated with. I.e. ctdb commands can be sent to it.
221 STOPPED - A node that is stopped does not host any public ip addresses,
222 nor is it part of the VNNMAP. A stopped node can not become LVSMASTER,
224 This node does not perticipate in the CTDB cluster but can still be
225 communicated with. I.e. ctdb commands can be sent to it.
228 PARTIALLYONLINE - A node that is partially online participates
229 in a cluster like a node that is ok. Some interfaces to serve
230 public ip addresses are down, but at least one interface is up.
231 See also "ctdb ifaces".
235 <refsect3><title>generation</title>
237 The generation id is a number that indicates the current generation
238 of a cluster instance. Each time a cluster goes through a
239 reconfiguration or a recovery its generation id will be changed.
242 This number does not have any particular meaning other than to keep
243 track of when a cluster has gone through a recovery. It is a random
244 number that represents the current instance of a ctdb cluster
246 CTDBD uses this number internally to be able to tell when commands
247 to operate on the cluster and the databases was issued in a different
248 generation of the cluster, to ensure that commands that operate
249 on the databases will not survive across a cluster database recovery.
250 After a recovery, all old outstanding commands will automatically
254 Sometimes this number will be shown as "INVALID". This only means that
255 the ctdbd daemon has started but it has not yet merged with the cluster through a recovery.
256 All nodes start with generation "INVALID" and are not assigned a real
257 generation id until they have successfully been merged with a cluster
262 <refsect3><title>VNNMAP</title>
264 The list of Virtual Node Numbers. This is a list of all nodes that actively participates in the cluster and that share the workload of hosting the Clustered TDB database records.
265 Only nodes that are participating in the vnnmap can become lmaster or dmaster for a database record.
269 <refsect3><title>Recovery mode</title>
271 This is the current recovery mode of the cluster. There are two possible modes:
274 NORMAL - The cluster is fully operational.
277 RECOVERY - The cluster databases have all been frozen, pausing all services while the cluster awaits a recovery process to complete. A recovery process should finish within seconds. If a cluster is stuck in the RECOVERY state this would indicate a cluster malfunction which needs to be investigated.
280 Once the recovery master detects an inconsistency, for example a node
281 becomes disconnected/connected, the recovery daemon will trigger a
282 cluster recovery process, where all databases are remerged across the
283 cluster. When this process starts, the recovery master will first
284 "freeze" all databases to prevent applications such as samba from
285 accessing the databases and it will also mark the recovery mode as
289 When CTDBD starts up, it will start in RECOVERY mode.
290 Once the node has been merged into a cluster and all databases
291 have been recovered, the node mode will change into NORMAL mode
292 and the databases will be "thawed", allowing samba to access the
297 <refsect3><title>Recovery master</title>
299 This is the cluster node that is currently designated as the recovery master. This node is responsible of monitoring the consistency of the cluster and to perform the actual recovery process when reqired.
302 Only one node at a time can be the designated recovery master. Which
303 node is designated the recovery master is decided by an election
304 process in the recovery daemons running on each node.
311 <para>Example output:</para>
312 <screen format="linespecific">
314 pnn:0 11.1.2.200 OK (THIS NODE)
318 Generation:1362079228
324 Recovery mode:NORMAL (0)
329 <refsect2><title>nodestatus [<nodespec>]</title>
331 This command is similar to the <command>status</command>
332 command. It displays the "node status" subset of output. The
333 main differences are:
339 The exit code is the bitwise-OR of the flags for each
340 specified node, while <command>ctdb status</command> exits
341 with 0 if it was able to retrieve status for all nodes.
347 <command>ctdb status</command> provides status information
348 for all nodes. <command>ctdb nodestatus</command>
349 defaults to providing status for only the current node.
350 If <nodespec> is provided then status is given for
351 the indicated node(s).
355 By default, <command>ctdb nodestatus</command> gathers
356 status from the local node. However, if invoked with "-n
357 all" (or similar) then status is gathered from the given
358 node(s). In particular <command>ctdb nodestatus
359 all</command> and <command>ctdb nodestatus -n
360 all</command> will produce different output. It is
361 possible to provide 2 different nodespecs (with and
362 without "-n") but the output is usually confusing!
368 A common invocation in scripts is <command>ctdb nodestatus
369 all</command> to check whether all nodes in a cluster are
374 Example: ctdb nodestatus
376 <para>Example output:</para>
377 <screen format="linespecific">
379 pnn:0 10.0.0.30 OK (THIS NODE)
383 Example: ctdb nodestatus all
385 <para>Example output:</para>
386 <screen format="linespecific">
387 # ctdb nodestatus all
389 pnn:0 10.0.0.30 OK (THIS NODE)
394 <refsect2><title>recmaster</title>
396 This command shows the pnn of the node which is currently the recmaster.
400 <refsect2><title>uptime</title>
402 This command shows the uptime for the ctdb daemon. When the last recovery or ip-failover completed and how long it took. If the "duration" is shown as a negative number, this indicates that there is a recovery/failover in progress and it started that many seconds ago.
408 <para>Example output:</para>
409 <screen format="linespecific">
410 Current time of node : Thu Oct 29 10:38:54 2009
411 Ctdbd start time : (000 16:54:28) Wed Oct 28 17:44:26 2009
412 Time of last recovery/failover: (000 16:53:31) Wed Oct 28 17:45:23 2009
413 Duration of last recovery/failover: 2.248552 seconds
417 <refsect2><title>listnodes</title>
419 This command shows lists the ip addresses of all the nodes in the cluster.
423 Example: ctdb listnodes
425 <para>Example output:</para>
426 <screen format="linespecific">
434 <refsect2><title>ping</title>
436 This command will "ping" specified CTDB node in the cluster to
437 verify that they are running.
440 Example: ctdb ping -n all
445 <screen format="linespecific">
446 response from 0 time=0.000054 sec (3 clients)
447 response from 1 time=0.000144 sec (2 clients)
448 response from 2 time=0.000105 sec (2 clients)
449 response from 3 time=0.000114 sec (2 clients)
453 <refsect2><title>runstate [setup|first_recovery|startup|running]</title>
455 Print the runstate of the specified node. Runstates are used
456 to serialise important state transitions in CTDB, particularly
460 If one or more optional runstate arguments are specified then
461 the node must be in one of these runstates for the command to
465 Example: ctdb runstate
470 <screen format="linespecific">
475 <refsect2><title>ifaces</title>
477 This command will display the list of network interfaces, which could
478 host public addresses, along with their status.
486 <screen format="linespecific">
488 name:eth5 link:up references:2
489 name:eth4 link:down references:0
490 name:eth3 link:up references:1
491 name:eth2 link:up references:1
494 Example: ctdb ifaces -Y
499 <screen format="linespecific">
500 :Name:LinkStatus:References:
508 <refsect2><title>setifacelink <iface> <status></title>
510 This command will set the status of a network interface.
511 The status needs to be "up" or "down". This is typically
512 used in the 10.interfaces script in the "monitor" event.
515 Example: ctdb setifacelink eth0 up
519 <refsect2><title>ip</title>
521 This command will display the list of public addresses that are provided by the cluster and which physical node is currently serving this ip. By default this command will ONLY show those public addresses that are known to the node itself. To see the full list of all public ips across the cluster you must use "ctdb ip -n all".
529 <screen format="linespecific">
531 172.31.91.82 node[1] active[] available[eth2,eth3] configured[eth2,eth3]
532 172.31.91.83 node[0] active[eth3] available[eth2,eth3] configured[eth2,eth3]
533 172.31.91.84 node[1] active[] available[eth2,eth3] configured[eth2,eth3]
534 172.31.91.85 node[0] active[eth2] available[eth2,eth3] configured[eth2,eth3]
535 172.31.92.82 node[1] active[] available[eth5] configured[eth4,eth5]
536 172.31.92.83 node[0] active[eth5] available[eth5] configured[eth4,eth5]
537 172.31.92.84 node[1] active[] available[eth5] configured[eth4,eth5]
538 172.31.92.85 node[0] active[eth5] available[eth5] configured[eth4,eth5]
546 <screen format="linespecific">
547 :Public IP:Node:ActiveInterface:AvailableInterfaces:ConfiguredInterfaces:
548 :172.31.91.82:1::eth2,eth3:eth2,eth3:
549 :172.31.91.83:0:eth3:eth2,eth3:eth2,eth3:
550 :172.31.91.84:1::eth2,eth3:eth2,eth3:
551 :172.31.91.85:0:eth2:eth2,eth3:eth2,eth3:
552 :172.31.92.82:1::eth5:eth4,eth5:
553 :172.31.92.83:0:eth5:eth5:eth4,eth5:
554 :172.31.92.84:1::eth5:eth4,eth5:
555 :172.31.92.85:0:eth5:eth5:eth4,eth5:
559 <refsect2><title>ipinfo <ip></title>
561 This command will display details about the specified public addresses.
564 Example: ctdb ipinfo 172.31.92.85
569 <screen format="linespecific">
570 Public IP[172.31.92.85] info on node 0
574 Interface[1]: Name:eth4 Link:down References:0
575 Interface[2]: Name:eth5 Link:up References:2 (active)
579 <refsect2><title>scriptstatus</title>
581 This command displays which scripts where run in the previous monitoring cycle and the result of each script. If a script failed with an error, causing the node to become unhealthy, the output from that script is also shown.
584 Example: ctdb scriptstatus
589 <screen format="linespecific">
590 7 scripts were executed last monitoring cycle
591 00.ctdb Status:OK Duration:0.056 Tue Mar 24 18:56:57 2009
592 10.interface Status:OK Duration:0.077 Tue Mar 24 18:56:57 2009
593 11.natgw Status:OK Duration:0.039 Tue Mar 24 18:56:57 2009
594 20.multipathd Status:OK Duration:0.038 Tue Mar 24 18:56:57 2009
595 31.clamd Status:DISABLED
596 40.vsftpd Status:OK Duration:0.045 Tue Mar 24 18:56:57 2009
597 41.httpd Status:OK Duration:0.039 Tue Mar 24 18:56:57 2009
598 50.samba Status:ERROR Duration:0.082 Tue Mar 24 18:56:57 2009
599 OUTPUT:ERROR: Samba tcp port 445 is not responding
603 <refsect2><title>disablescript <script></title>
605 This command is used to disable an eventscript.
608 This will take effect the next time the eventscripts are being executed so it can take a short while until this is reflected in 'scriptstatus'.
612 <refsect2><title>enablescript <script></title>
614 This command is used to enable an eventscript.
617 This will take effect the next time the eventscripts are being executed so it can take a short while until this is reflected in 'scriptstatus'.
621 <refsect2><title>getvar <name></title>
623 Get the runtime value of a tuneable variable.
626 Example: ctdb getvar MaxRedirectCount
631 <screen format="linespecific">
636 <refsect2><title>setvar <name> <value></title>
638 Set the runtime value of a tuneable variable.
641 Example: ctdb setvar MaxRedirectCount 5
645 <refsect2><title>listvars</title>
647 List all tuneable variables, except the values of the obsolete tunables
648 like VacuumMinInterval. The obsolete tunables can be retrieved only
649 explicitly with the "ctdb getvar" command.
652 Example: ctdb listvars
657 <screen format="linespecific">
659 SeqnumInterval = 1000
662 KeepaliveInterval = 5
669 TickleUpdateInterval = 20
670 EventScriptTimeout = 30
671 EventScriptTimeoutCount = 1
672 RecoveryGracePeriod = 120
673 RecoveryBanPeriod = 300
674 DatabaseHashSize = 100001
676 RerecoveryTimeout = 10
680 ReclockPingPeriod = 60
682 DisableIPFailover = 0
683 VerboseMemoryNames = 0
687 RecLockLatencyMs = 1000
688 RecoveryDropAllIPs = 120
689 VerifyRecoveryLock = 1
691 VacuumMaxRunTime = 30
694 VacuumFastPathCount = 60
695 MaxQueueDropMsg = 1000000
697 AllowUnhealthyDBRead = 0
698 StatHistoryInterval = 1
699 DeferredAttachTO = 120
700 AllowClientDBAttach = 1
701 RecoverPDBBySeqNum = 0
705 <refsect2><title>lvsmaster</title>
707 This command shows which node is currently the LVSMASTER. The
708 LVSMASTER is the node in the cluster which drives the LVS system and
709 which receives all incoming traffic from clients.
712 LVS is the mode where the entire CTDB/Samba cluster uses a single
713 ip address for the entire cluster. In this mode all clients connect to
714 one specific node which will then multiplex/loadbalance the clients
715 evenly onto the other nodes in the cluster. This is an alternative to using
716 public ip addresses. See the manpage for ctdbd for more information
721 <refsect2><title>lvs</title>
723 This command shows which nodes in the cluster are currently active in the
724 LVS configuration. I.e. which nodes we are currently loadbalancing
725 the single ip address across.
729 LVS will by default only loadbalance across those nodes that are both
730 LVS capable and also HEALTHY. Except if all nodes are UNHEALTHY in which
731 case LVS will loadbalance across all UNHEALTHY nodes as well.
732 LVS will never use nodes that are DISCONNECTED, STOPPED, BANNED or
739 <screen format="linespecific">
747 <refsect2><title>getcapabilities</title>
749 This command shows the capabilities of the current node.
750 Please see manpage for ctdbd for a full list of all capabilities and
751 more detailed description.
755 RECMASTER and LMASTER capabilities are primarily used when CTDBD
756 is used to create a cluster spanning across WAN links. In which case
757 ctdbd acts as a WAN accelerator.
761 LVS capabile means that the node is participating in LVS, a mode
762 where the entire CTDB cluster uses one single ip address for the
763 entire cluster instead of using public ip address failover.
764 This is an alternative to using a loadbalancing layer-4 switch.
770 <screen format="linespecific">
778 <refsect2><title>statistics</title>
780 Collect statistics from the CTDB daemon about how many calls it has served.
783 Example: ctdb statistics
788 <screen format="linespecific">
793 client_packets_sent 360489
794 client_packets_recv 360466
795 node_packets_sent 480931
796 node_packets_recv 240120
797 keepalive_packets_sent 4
798 keepalive_packets_recv 3
819 pending_lockwait_calls 0
822 max_call_latency 4.948321 sec
823 max_lockwait_latency 0.000000 sec
827 <refsect2><title>statisticsreset</title>
829 This command is used to clear all statistics counters in a node.
832 Example: ctdb statisticsreset
836 <refsect2><title>getreclock</title>
838 This command is used to show the filename of the reclock file that is used.
844 <screen format="linespecific">
845 Reclock file:/gpfs/.ctdb/shared
850 <refsect2><title>setreclock [filename]</title>
852 This command is used to modify, or clear, the file that is used as the reclock file at runtime. When this command is used, the reclock file checks are disabled. To re-enable the checks the administrator needs to activate the "VerifyRecoveryLock" tunable using "ctdb setvar".
856 If run with no parameter this will remove the reclock file completely. If run with a parameter the parameter specifies the new filename to use for the recovery lock.
860 This command only affects the runtime settings of a ctdb node and will be lost when ctdb is restarted. For persistent changes to the reclock file setting you must edit /etc/sysconfig/ctdb.
866 <refsect2><title>getdebug</title>
868 Get the current debug level for the node. the debug level controls what information is written to the log file.
871 The debug levels are mapped to the corresponding syslog levels.
872 When a debug level is set, only those messages at that level and higher
873 levels will be printed.
876 The list of debug levels from highest to lowest are :
879 EMERG ALERT CRIT ERR WARNING NOTICE INFO DEBUG
883 <refsect2><title>setdebug <debuglevel></title>
885 Set the debug level of a node. This controls what information will be logged.
888 The debuglevel is one of EMERG ALERT CRIT ERR WARNING NOTICE INFO DEBUG
892 <refsect2><title>getpid</title>
894 This command will return the process id of the ctdb daemon.
898 <refsect2><title>disable</title>
900 This command is used to administratively disable a node in the cluster.
901 A disabled node will still participate in the cluster and host
902 clustered TDB records but its public ip address has been taken over by
903 a different node and it no longer hosts any services.
907 <refsect2><title>enable</title>
909 Re-enable a node that has been administratively disabled.
913 <refsect2><title>stop</title>
915 This command is used to administratively STOP a node in the cluster.
916 A STOPPED node is connected to the cluster but will not host any
917 public ip addresse, nor does it participate in the VNNMAP.
918 The difference between a DISABLED node and a STOPPED node is that
919 a STOPPED node does not host any parts of the database which means
920 that a recovery is required to stop/continue nodes.
924 <refsect2><title>continue</title>
926 Re-start a node that has been administratively stopped.
930 <refsect2><title>addip <public_ip/mask> <iface></title>
932 This command is used to add a new public ip to a node during runtime.
933 This allows public addresses to be added to a cluster without having
934 to restart the ctdb daemons.
937 Note that this only updates the runtime instance of ctdb. Any changes will be lost next time ctdb is restarted and the public addresses file is re-read.
938 If you want this change to be permanent you must also update the public addresses file manually.
942 <refsect2><title>delip <public_ip></title>
944 This command is used to remove a public ip from a node during runtime.
945 If this public ip is currently hosted by the node it being removed from, the ip will first be failed over to another node, if possible, before it is removed.
948 Note that this only updates the runtime instance of ctdb. Any changes will be lost next time ctdb is restarted and the public addresses file is re-read.
949 If you want this change to be permanent you must also update the public addresses file manually.
953 <refsect2><title>moveip <public_ip> <node></title>
955 This command can be used to manually fail a public ip address to a
959 In order to manually override the "automatic" distribution of public
960 ip addresses that ctdb normally provides, this command only works
961 when you have changed the tunables for the daemon to:
971 <refsect2><title>shutdown</title>
973 This command will shutdown a specific CTDB daemon.
977 <refsect2><title>recover</title>
979 This command will trigger the recovery daemon to do a cluster
984 <refsect2><title>ipreallocate</title>
986 This command will force the recovery master to perform a full ip reallocation process and redistribute all ip addresses. This is useful to "reset" the allocations back to its default state if they have been changed using the "moveip" command. While a "recover" will also perform this reallocation, a recovery is much more hevyweight since it will also rebuild all the databases.
990 <refsect2><title>setlmasterrole <on|off></title>
992 This command is used ot enable/disable the LMASTER capability for a node at runtime. This capability determines whether or not a node can be used as an LMASTER for records in the database. A node that does not have the LMASTER capability will not show up in the vnnmap.
996 Nodes will by default have this capability, but it can be stripped off nodes by the setting in the sysconfig file or by using this command.
999 Once this setting has been enabled/disabled, you need to perform a recovery for it to take effect.
1002 See also "ctdb getcapabilities"
1006 <refsect2><title>setrecmasterrole <on|off></title>
1008 This command is used ot enable/disable the RECMASTER capability for a node at runtime. This capability determines whether or not a node can be used as an RECMASTER for the cluster. A node that does not have the RECMASTER capability can not win a recmaster election. A node that already is the recmaster for the cluster when the capability is stripped off the node will remain the recmaster until the next cluster election.
1012 Nodes will by default have this capability, but it can be stripped off nodes by the setting in the sysconfig file or by using this command.
1015 See also "ctdb getcapabilities"
1019 <refsect2><title>killtcp [<srcip:port> <dstip:port>]</title>
1021 This command will kill the specified TCP connections by
1022 issuing a TCP RST to the srcip:port endpoint. A single
1023 connection can be specified on the command-line, otherwise
1024 connections are read one-per-line from standard input. This
1025 is a command used by the ctdb eventscripts.
1029 <refsect2><title>gratiousarp <ip> <interface></title>
1031 This command will send out a gratious arp for the specified interface
1032 through the specified interface. This command is mainly used by the
1037 <refsect2><title>reloadnodes</title>
1039 This command is used when adding new nodes, or removing existing nodes from an existing cluster.
1042 Procedure to add a node:
1045 1, To expand an existing cluster, first ensure with 'ctdb status' that
1046 all nodes are up and running and that they are all healthy.
1047 Do not try to expand a cluster unless it is completely healthy!
1050 2, On all nodes, edit /etc/ctdb/nodes and add the new node as the last
1051 entry to the file. The new node MUST be added to the end of this file!
1054 3, Verify that all the nodes have identical /etc/ctdb/nodes files after you edited them and added the new node!
1057 4, Run 'ctdb reloadnodes' to force all nodes to reload the nodesfile.
1060 5, Use 'ctdb status' on all nodes and verify that they now show the additional node.
1063 6, Install and configure the new node and bring it online.
1066 Procedure to remove a node:
1069 1, To remove a node from an existing cluster, first ensure with 'ctdb status' that
1070 all nodes, except the node to be deleted, are up and running and that they are all healthy.
1071 Do not try to remove a node from a cluster unless the cluster is completely healthy!
1074 2, Shutdown and poweroff the node to be removed.
1077 3, On all other nodes, edit the /etc/ctdb/nodes file and comment out the node to be removed. Do not delete the line for that node, just comment it out by adding a '#' at the beginning of the line.
1080 4, Run 'ctdb reloadnodes' to force all nodes to reload the nodesfile.
1083 5, Use 'ctdb status' on all nodes and verify that the deleted node no longer shows up in the list..
1090 <refsect2><title>reloadips</title>
1092 This command is used to reload the public addresses file and update the
1093 ip configuration of the running daemon.
1096 Procedure to update the public address configuration on a single node:
1099 1, Update the /etc/ctdb/public_addresses file on the node
1102 2, Run 'ctdb reloadips' on the node.
1105 The file will then be reloaded on the node and addresses will be added
1106 or removed as required to match the newly loaded file. When updating
1107 a single node it may take a little while before any newly added
1108 addresses are failed onto the node.
1115 Procedure to update the public address configuration on whole cluster:
1118 1, Update the /etc/ctdb/public_addresses file on all nodes
1121 2, Run 'ctdb reloadips -n all'.
1124 This command will then force all nodes to reload and update the
1125 addresses. This process is controlled and synchronized by the recovery
1126 master to ensure that all addresses are added to all nodes as one
1127 single operation, after which any required ip node rebalancing may
1134 <refsect2><title>tickle <srcip:port> <dstip:port></title>
1136 This command will will send a TCP tickle to the source host for the
1137 specified TCP connection.
1138 A TCP tickle is a TCP ACK packet with an invalid sequence and
1139 acknowledge number and will when received by the source host result
1140 in it sending an immediate correct ACK back to the other end.
1143 TCP tickles are useful to "tickle" clients after a IP failover has
1144 occured since this will make the client immediately recognize the
1145 TCP connection has been disrupted and that the client will need
1146 to reestablish. This greatly speeds up the time it takes for a client
1147 to detect and reestablish after an IP failover in the ctdb cluster.
1151 <refsect2><title>gettickles <ip></title>
1153 This command is used to show which TCP connections are registered with
1154 CTDB to be "tickled" if there is a failover.
1157 <refsect2><title>repack [max_freelist]</title>
1159 Over time, when records are created and deleted in a TDB, the TDB list of free space will become fragmented. This can lead to a slowdown in accessing TDB records.
1160 This command is used to defragment a TDB database and pruning the freelist.
1164 If [max_freelist] is specified, then a database will only be repacked if it has more than this number of entries in the freelist.
1167 During repacking of the database, the entire TDB database will be locked to prevent writes. If samba tries to write to a record in the database during a repack operation, samba will block until the repacking has completed.
1171 This command can be disruptive and can cause samba to block for the duration of the repack operation. In general, a repack operation will take less than one second to complete.
1175 A repack operation will only defragment the local TDB copy of the CTDB database. You need to run this command on all of the nodes to repack a CTDB database completely.
1179 Example: ctdb repack 1000
1183 By default, this operation is issued from the 00.ctdb event script every 5 minutes.
1188 <refsect2><title>vacuum [max_records]</title>
1190 Over time CTDB databases will fill up with empty deleted records which will lead to a progressive slow down of CTDB database access.
1191 This command is used to prune all databases and delete all empty records from the cluster.
1195 By default, vacuum will delete all empty records from all databases.
1196 If [max_records] is specified, the command will only delete the first
1197 [max_records] empty records for each database.
1201 Vacuum only deletes records where the local node is the lmaster.
1202 To delete all records from the entire cluster you need to run a vacuum from each node.
1204 This command is not disruptive. Samba is unaffected and will still be able to read/write records normally while the database is being vacuumed.
1208 Example: ctdb vacuum
1212 By default, this operation is issued from the 00.ctdb event script every 5 minutes.
1216 <refsect2><title>backupdb <dbname> <file></title>
1218 This command can be used to copy the entire content of a database out to a file. This file can later be read back into ctdb using the restoredb command.
1219 This is mainly useful for backing up persistent databases such as secrets.tdb and similar.
1223 <refsect2><title>restoredb <file> [<dbname>]</title>
1225 This command restores a persistent database that was previously backed up using backupdb.
1226 By default the data will be restored back into the same database as
1227 it was created from. By specifying dbname you can restore the data
1228 into a different database.
1232 <refsect2><title>wipedb <dbname></title>
1234 This command can be used to remove all content of a database.
1240 <refsect2><title>getlog [<level>] [recoverd]</title>
1242 In addition to the normal logging to a log file,
1243 CTDBD also keeps a in-memory ringbuffer containing the most recent
1244 log entries for all log levels (except DEBUG).
1247 This is useful since it allows for keeping continuous logs to a file
1248 at a reasonable non-verbose level, but shortly after an incident has
1249 occured, a much more detailed log can be pulled from memory. This
1250 can allow you to avoid having to reproduce an issue due to the
1251 on-disk logs being of insufficient detail.
1254 This command extracts all messages of level or lower log level
1255 from memory and prints it to the screen. The level is not
1256 specified it defaults to NOTICE.
1259 By default, logs are extracted from the main CTDB daemon. If
1260 the recoverd option is given then logs are extracted from the
1265 <refsect2><title>clearlog [recoverd]</title>
1267 This command clears the in-memory logging ringbuffer.
1270 By default, logs are cleared in the main CTDB daemon. If the
1271 recoverd option is given then logs are cleared in the recovery
1276 <refsect2><title>setdbreadonly <dbname|hash></title>
1278 This command will enable the ReadOnly record support for a database.
1279 This is an experimental feature to improve performance for contended
1280 records primarily in locking.tdb and brlock.tdb.
1281 When enabling this feature you must set it on all nodes in the cluster.
1282 For now, this feature requires a special patch to samba in order to
1288 <refsect1><title>Debugging Commands</title>
1290 These commands are primarily used for CTDB development and testing and
1291 should not be used for normal administration.
1293 <refsect2><title>process-exists <pid></title>
1295 This command checks if a specific process exists on the CTDB host. This is mainly used by Samba to check if remote instances of samba are still running or not.
1299 <refsect2><title>getdbmap</title>
1301 This command lists all clustered TDB databases that the CTDB daemon has attached to. Some databases are flagged as PERSISTENT, this means that the database stores data persistently and the data will remain across reboots. One example of such a database is secrets.tdb where information about how the cluster was joined to the domain is stored.
1304 If a PERSISTENT database is not in a healthy state the database is
1305 flagged as UNHEALTHY. If there's at least one completely healthy node running in
1306 the cluster, it's possible that the content is restored by a recovery
1307 run automaticly. Otherwise an administrator needs to analyze the
1311 See also "ctdb getdbstatus", "ctdb backupdb", "ctdb restoredb",
1312 "ctdb dumpbackup", "ctdb wipedb", "ctdb setvar AllowUnhealthyDBRead 1"
1313 and (if samba or tdb-utils are installed) "tdbtool check".
1316 Most databases are not persistent and only store the state information that the currently running samba daemons need. These databases are always wiped when ctdb/samba starts and when a node is rebooted.
1319 Example: ctdb getdbmap
1324 <screen format="linespecific">
1325 Number of databases:10
1326 dbid:0x435d3410 name:notify.tdb path:/var/ctdb/notify.tdb.0
1327 dbid:0x42fe72c5 name:locking.tdb path:/var/ctdb/locking.tdb.0
1328 dbid:0x1421fb78 name:brlock.tdb path:/var/ctdb/brlock.tdb.0
1329 dbid:0x17055d90 name:connections.tdb path:/var/ctdb/connections.tdb.0
1330 dbid:0xc0bdde6a name:sessionid.tdb path:/var/ctdb/sessionid.tdb.0
1331 dbid:0x122224da name:test.tdb path:/var/ctdb/test.tdb.0
1332 dbid:0x2672a57f name:idmap2.tdb path:/var/ctdb/persistent/idmap2.tdb.0 PERSISTENT
1333 dbid:0xb775fff6 name:secrets.tdb path:/var/ctdb/persistent/secrets.tdb.0 PERSISTENT
1334 dbid:0xe98e08b6 name:group_mapping.tdb path:/var/ctdb/persistent/group_mapping.tdb.0 PERSISTENT
1335 dbid:0x7bbbd26c name:passdb.tdb path:/var/ctdb/persistent/passdb.tdb.0 PERSISTENT
1338 Example output for an unhealthy database:
1340 <screen format="linespecific">
1341 Number of databases:1
1342 dbid:0xb775fff6 name:secrets.tdb path:/var/ctdb/persistent/secrets.tdb.0 PERSISTENT UNHEALTHY
1346 Example output for a healthy database as machinereadable output -Y:
1348 <screen format="linespecific">
1349 :ID:Name:Path:Persistent:Unhealthy:
1350 :0x7bbbd26c:passdb.tdb:/var/ctdb/persistent/passdb.tdb.0:1:0:
1354 <refsect2><title>getdbstatus <dbname></title>
1356 This command displays more details about a database.
1359 Example: ctdb getdbstatus test.tdb.0
1364 <screen format="linespecific">
1367 path: /var/ctdb/test.tdb.0
1372 Example: ctdb getdbstatus registry.tdb (with a corrupted TDB)
1377 <screen format="linespecific">
1380 path: /var/ctdb/persistent/registry.tdb.0
1382 HEALTH: NO-HEALTHY-NODES - ERROR - Backup of corrupted TDB in '/var/ctdb/persistent/registry.tdb.0.corrupted.20091208091949.0Z'
1386 <refsect2><title>catdb <dbname></title>
1388 This command will dump a clustered TDB database to the screen. This is a debugging command.
1392 <refsect2><title>cattdb <dbname></title>
1394 This command will dump the content of the local TDB database to the screen. This is a debugging command.
1398 <refsect2><title>dumpdbbackup <backup-file></title>
1400 This command will dump the content of database backup to the screen
1401 (similar to ctdb catdb). This is a debugging command.
1405 <refsect2><title>getmonmode</title>
1407 This command returns the monutoring mode of a node. The monitoring mode is either ACTIVE or DISABLED. Normally a node will continuously monitor that all other nodes that are expected are in fact connected and that they respond to commands.
1410 ACTIVE - This is the normal mode. The node is actively monitoring all other nodes, both that the transport is connected and also that the node responds to commands. If a node becomes unavailable, it will be marked as DISCONNECTED and a recovery is initiated to restore the cluster.
1413 DISABLED - This node is not monitoring that other nodes are available. In this mode a node failure will not be detected and no recovery will be performed. This mode is useful when for debugging purposes one wants to attach GDB to a ctdb process but wants to prevent the rest of the cluster from marking this node as DISCONNECTED and do a recovery.
1418 <refsect2><title>setmonmode <0|1></title>
1420 This command can be used to explicitly disable/enable monitoring mode on a node. The main purpose is if one wants to attach GDB to a running ctdb daemon but wants to prevent the other nodes from marking it as DISCONNECTED and issuing a recovery. To do this, set monitoring mode to 0 on all nodes before attaching with GDB. Remember to set monitoring mode back to 1 afterwards.
1424 <refsect2><title>attach <dbname> [persistent]</title>
1426 This is a debugging command. This command will make the CTDB daemon create a new CTDB database and attach to it.
1430 <refsect2><title>dumpmemory</title>
1432 This is a debugging command. This command will make the ctdb
1433 daemon to write a fill memory allocation map to standard output.
1437 <refsect2><title>rddumpmemory</title>
1439 This is a debugging command. This command will dump the talloc memory
1440 allocation tree for the recovery daemon to standard output.
1444 <refsect2><title>thaw</title>
1446 Thaw a previously frozen node.
1451 <refsect2><title>eventscript <arguments></title>
1453 This is a debugging command. This command can be used to manually
1454 invoke and run the eventscritps with arbitrary arguments.
1458 <refsect2><title>ban <bantime|0></title>
1460 Administratively ban a node for bantime seconds. A bantime of 0 means that the node should be permanently banned.
1463 A banned node does not participate in the cluster and does not host any records for the clustered TDB. Its ip address has been taken over by another node and no services are hosted.
1466 Nodes are automatically banned if they are the cause of too many
1470 This is primarily a testing command. Note that the recovery daemon controls the overall ban state and it may automatically unban nodes at will. Meaning that a node that has been banned by the administrator can and ofter are unbanned before the admin specifid timeout triggers. If wanting to "drop" a node out from the cluster for mainentance or other reasons, use the "stop" / "continue" commands instad of "ban" / "unban".
1474 <refsect2><title>unban</title>
1476 This command is used to unban a node that has either been
1477 administratively banned using the ban command or has been automatically
1478 banned by the recovery daemon.
1483 <refsect2><title>check_srvids <srvid> ...</title>
1485 This command checks whether a set of srvid message ports are registered on the
1486 node or not. The command takes a list of values to check.
1489 Example: ctdb check_srvids 1 2 3 14765
1494 <screen format="linespecific">
1495 Server id 0:1 does not exist
1496 Server id 0:2 does not exist
1497 Server id 0:3 does not exist
1498 Server id 0:14765 exists
1504 <refsect1><title>SEE ALSO</title>
1507 <ulink url="http://ctdb.samba.org/"/>
1510 <refsect1><title>COPYRIGHT/LICENSE</title>
1512 Copyright (C) Andrew Tridgell 2007
1513 Copyright (C) Ronnie sahlberg 2007
1515 This program is free software; you can redistribute it and/or modify
1516 it under the terms of the GNU General Public License as published by
1517 the Free Software Foundation; either version 3 of the License, or (at
1518 your option) any later version.
1520 This program is distributed in the hope that it will be useful, but
1521 WITHOUT ANY WARRANTY; without even the implied warranty of
1522 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
1523 General Public License for more details.
1525 You should have received a copy of the GNU General Public License
1526 along with this program; if not, see http://www.gnu.org/licenses/.