1CTDBD(1) CTDBD(1)
2
6 ctdbd - The CTDB cluster daemon
7
9 ctdbd
10 ctdbd [-? --help] [-d --debug=<INTEGER>] {--dbdir=<directory>}
12 {--dbdir-persistent=<directory>} [--event-script-dir=<directory>]
13 [-i --interactive] [--listen=<address>] [--logfile=<filename>]
14 [--lvs] {--nlist=<filename>} [--no-lmaster] [--no-recmaster]
15 [--nosetsched] {--notification-script=<filename>}
16 [--public-addresses=<filename>] [--public-interface=<interface>]
17 {--reclock=<filename>} [--single-public-ip=<address>]
18 [--socket=<filename>] [--start-as-disabled] [--start-as-stopped]
19 [--syslog] [--torture] [--transport=<STRING>] [--usage]
20
22 ctdbd is the main ctdb daemon.
23 ctdbd provides a clustered version of the TDB database with automatic
25 rebuild/recovery of the databases upon nodefailures.
26 Combined with a cluster filesystem ctdbd provides a full HA environment
28 for services such as clustered Samba and NFS as well as other services.
29 ctdbd provides monitoring of all nodes in the cluster and automatically
31 reconfigures the cluster and recovers upon node failures.
32 ctdbd is the main component in clustered Samba that provides a
34 high-availability load-sharing CIFS server cluster.
35
37 -? --help
38 Print some help text to the screen.
39 -d --debug=<DEBUGLEVEL>
41 This option sets the debuglevel on the ctdbd daemon which controls
42 what will be written to the logfile. The default is 0 which will
43 only log important events and errors. A larger number will provide
44 additional logging.
45 --dbdir=<directory>
47 This is the directory on local storage where ctdbd keeps the local
48 copy of the TDB databases. This directory is local for each node
49 and should not be stored on the shared cluster filesystem.
50 This directory would usually be /var/ctdb .
52 --dbdir-persistent=<directory>
54 This is the directory on local storage where ctdbd keeps the local
55 copy of the persistent TDB databases. This directory is local for
56 each node and should not be stored on the shared cluster
57 filesystem.
58 This directory would usually be /etc/ctdb/persistent .
60 --event-script-dir=<directory>
62 This option is used to specify the directory where the CTDB event
63 scripts are stored.
64 This will normally be /etc/ctdb/events.d which is part of the ctdb
66 distribution.
67 -i --interactive
69 By default ctdbd will detach itself from the shell and run in the
70 background as a daemon. This option makes ctdbd to start in
71 interactive mode.
72 --listen=<address>
74 This specifies which ip address ctdb will bind to. By default ctdbd
75 will bind to the first address it finds in the /etc/ctdb/nodes file
76 and which is also present on the local system in which case you do
77 not need to provide this option.
78 This option is only required when you want to run multiple ctdbd
80 daemons/nodes on the same physical host in which case there would
81 be multiple entries in /etc/ctdb/nodes what would match a local
82 interface.
83 --logfile=<filename>
85 This is the file where ctdbd will write its log. This is usually
86 /var/log/log.ctdb .
87 --lvs
89 This option is used to activate the LVS capability on a CTDB node.
90 Please see the LVS section.
91 --nlist=<filename>
93 This file contains a list of the private ip addresses of every node
94 in the cluster. There is one line/ip address for each node. This
95 file must be the same for all nodes in the cluster.
96 This file is usually /etc/ctdb/nodes .
98 --no-lmaster
100 This argument specifies that this node can NOT become an lmaster
101 for records in the database. This means that it will never show up
102 in the vnnmap. This feature is primarily used for making a cluster
103 span across a WAN link and use CTDB as a WAN-accelerator.
104 Please see the "remote cluster nodes" section for more information.
106 --no-recmaster
108 This argument specifies that this node can NOT become a recmaster
109 for the database. This feature is primarily used for making a
110 cluster span across a WAN link and use CTDB as a WAN-accelerator.
111 Please see the "remote cluster nodes" section for more information.
113 --nosetsched
115 This is a ctdbd debugging option. this option is only used when
116 debugging ctdbd.
117 Normally ctdb will change its scheduler to run as a real-time
119 process. This is the default mode for a normal ctdbd operation to
120 gurarantee that ctdbd always gets the cpu cycles that it needs.
121 This option is used to tell ctdbd to NOT run as a real-time process
123 and instead run ctdbd as a normal userspace process. This is useful
124 for debugging and when you want to run ctdbd under valgrind or gdb.
125 (You dont want to attach valgrind or gdb to a real-time process.)
126 --notification-script=<filename>
128 This specifies a script which will be invoked by ctdb when certain
129 state changes occur in ctdbd and when you may want to trigger this
130 to run certain scripts.
131 This file is usually /etc/ctdb/notify.sh .
133 See the NOTIFICATION SCRIPT section below for more information.
135 --public_addresses=<filename>
137 When used with IP takeover this specifies a file containing the
138 public ip addresses to use on the cluster. This file contains a
139 list of ip addresses netmasks and interfaces. When ctdb is
140 operational it will distribute these public ip addresses evenly
141 across the available nodes.
142 This is usually the file /etc/ctdb/public_addresses
144 --public-interface=<interface>
146 This option tells ctdb which interface to attach public-addresses
147 to and also where to attach the single-public-ip when used.
148 This is only required when using public ip addresses and only when
150 you dont specify the interface explicitly in
151 /etc/ctdb/public_addresses or when you are using
152 --single-public-ip.
153 If you omit this argument when using public addresses or single
155 public ip, ctdb will not be able to send out Gratious ARPs
156 correctly or be able to kill tcp connections correctly which will
157 lead to application failures.
158 --reclock=<filename>
160 This is the name of the lock file stored of the shared cluster
161 filesystem that ctdbd uses to prevent split brains from occuring.
162 This file must be stored on shared storage.
163 It is possible to run CTDB without a reclock file, but then there
165 will be no protection against split brain if the network becomes
166 partitioned. Using CTDB without a reclock file is strongly
167 discouraged.
168 --socket=<filename>
170 This specifies the name of the domain socket that ctdbd will
171 create. This socket is used for local clients to attach to and
172 communicate with the ctdbd daemon.
173 The default is /tmp/ctdb.socket . You only need to use this option
175 if you plan to run multiple ctdbd daemons on the same physical
176 host.
177 --start-as-disabled
179 This makes the ctdb daemon to be DISABLED when it starts up.
180 As it is DISABLED it will not get any of the public ip addresses
182 allocated to it, and thus this allow you to start ctdb on a node
183 without causing any ip address to failover from other nodes onto
184 the new node.
185 When used, the administrator must keep track of when nodes start
187 and manually enable them again using the "ctdb enable" command, or
188 else the node will not host any services.
189 A node that is DISABLED will not host any services and will not be
191 reachable/used by any clients.
192 --start-as-stopped
194 This makes the ctdb daemon to be STOPPED when it starts up.
195 A node that is STOPPED does not host any public addresses. It is
197 not part of the VNNMAP so it does act as an LMASTER. It also has
198 all databases locked in recovery mode until restarted.
199 To restart and activate a STOPPED node, the command "ctdb continue"
201 is used.
202 A node that is STOPPED will not host any services and will not be
204 reachable/used by any clients.
205 --syslog
207 Send all log messages to syslog instead of to the ctdb logfile.
208 --torture
210 This option is only used for development and testing of ctdbd. It
211 adds artificial errors and failures to the common codepaths in
212 ctdbd to verify that ctdbd can recover correctly for failures.
213 You do NOT want to use this option unless you are developing and
215 testing new functionality in ctdbd.
216 --transport=<STRING>
218 This option specifies which transport to use for ctdbd internode
219 communications. The default is "tcp".
220 Currently only "tcp" is supported but "infiniband" might be
222 implemented in the future.
223 --usage
225 Print useage information to the screen.
226
228 When used for ip takeover in a HA environment, each node in a ctdb
229 cluster has multiple ip addresses assigned to it. One private and one
230 or more public.
231 Private address
233 This is the physical ip address of the node which is configured in
234 linux and attached to a physical interface. This address uniquely
235 identifies a physical node in the cluster and is the ip addresses that
236 ctdbd will use to communicate with the ctdbd daemons on the other nodes
237 in the cluster.
238 The private addresses are configured in /etc/ctdb/nodes (unless the
240 --nlist option is used) and contain one line for each node in the
241 cluster. Each line contains the private ip address for one node in the
242 cluster. This file must be the same on all nodes in the cluster.
243 Since the private addresses are only available to the network when the
245 corresponding node is up and running you should not use these addresses
246 for clients to connect to services provided by the cluster. Instead
247 client applications should only attach to the public addresses since
248 these are guaranteed to always be available.
249 When using ip takeover, it is strongly recommended that the private
251 addresses are configured on a private network physically separated from
252 the rest of the network and that this private network is dedicated to
253 CTDB traffic.
254 Example /etc/ctdb/nodes for a four node cluster:
256 10.1.1.1
259 10.1.1.2
260 10.1.1.3
261 10.1.1.4
262 Public address
265 A public address on the other hand is not attached to an interface.
266 This address is managed by ctdbd itself and is attached/detached to a
267 physical node at runtime.
268 The ctdb cluster will assign/reassign these public addresses across the
270 available healthy nodes in the cluster. When one node fails, its public
271 address will be migrated to and taken over by a different node in the
272 cluster to ensure that all public addresses are always available to
273 clients as long as there are still nodes available capable of hosting
274 this address.
275 These addresses are not physically attached to a specific node. The
277 ´ctdb ip´ command can be used to view the current assignment of public
278 addresses and which physical node is currently serving it.
279 On each node this file contains a list of the public addresses that
281 this node is capable of hosting. The list also contain the netmask and
282 the interface where this address should be attached for the case where
283 you may want to serve data out through multiple different interfaces.
284 Example /etc/ctdb/public_addresses for a node that can host 4
286 public addresses:
287 11.1.1.1/24 eth0
290 11.1.1.2/24 eth0
291 11.1.2.1/24 eth1
292 11.1.2.2/24 eth1
293 In most cases this file would be the same on all nodes in a cluster but
296 there are exceptions when one may want to use different files on
297 different nodes.
298 Example: 4 nodes partitioned into two subgroups :
300 Node 0:/etc/ctdb/public_addresses
302 10.1.1.1/24 eth0
303 10.1.1.2/24 eth0
304 Node 1:/etc/ctdb/public_addresses
306 10.1.1.1/24 eth0
307 10.1.1.2/24 eth0
308 Node 2:/etc/ctdb/public_addresses
310 10.2.1.1/24 eth0
311 10.2.1.2/24 eth0
312 Node 3:/etc/ctdb/public_addresses
314 10.2.1.1/24 eth0
315 10.2.1.2/24 eth0
316 In this example nodes 0 and 1 host two public addresses on the 10.1.1.x
319 network while nodes 2 and 3 host two public addresses for the 10.2.1.x
320 network.
321 Ip address 10.1.1.1 can be hosted by either of nodes 0 or 1 and will be
323 available to clients as long as at least one of these two nodes are
324 available. If both nodes 0 and node 1 become unavailable 10.1.1.1 also
325 becomes unavailable. 10.1.1.1 can not be failed over to node 2 or node
326 3 since these nodes do not have this ip address listed in their public
327 addresses file.
328
330 The current status of each node in the cluster can be viewed by the
331 ´ctdb status´ command.
332 There are five possible states for a node.
334 OK - This node is fully functional.
336 DISCONNECTED - This node could not be connected through the network and
338 is currently not particpating in the cluster. If there is a public IP
339 address associated with this node it should have been taken over by a
340 different node. No services are running on this node.
341 DISABLED - This node has been administratively disabled. This node is
343 still functional and participates in the CTDB cluster but its IP
344 addresses have been taken over by a different node and no services are
345 currently being hosted.
346 UNHEALTHY - A service provided by this node is malfunctioning and
348 should be investigated. The CTDB daemon itself is operational and
349 participates in the cluster. Its public IP address has been taken over
350 by a different node and no services are currently being hosted. All
351 unhealthy nodes should be investigated and require an administrative
352 action to rectify.
353 BANNED - This node failed too many recovery attempts and has been
355 banned from participating in the cluster for a period of
356 RecoveryBanPeriod seconds. Any public IP address has been taken over by
357 other nodes. This node does not provide any services. All banned nodes
358 should be investigated and require an administrative action to rectify.
359 This node does not perticipate in the CTDB cluster but can still be
360 communicated with. I.e. ctdb commands can be sent to it.
361 STOPPED - A node that is stopped does not host any public ip addresses,
363 nor is it part of the VNNMAP. A stopped node can not become LVSMASTER,
364 RECMASTER or NATGW. This node does not perticipate in the CTDB cluster
365 but can still be communicated with. I.e. ctdb commands can be sent to
366 it.
367
369 These are the public tuneables that can be used to control how ctdb
370 behaves.
371 KeepaliveInterval
373 Default: 1
374 How often should the nodes send keepalives to eachother.
376 KeepaliveLimit
378 Default: 5
379 After how many keepalive intervals without any traffic should a node
381 wait until marking the peer as DISCONNECTED.
382 MonitorInterval
384 Default: 15
385 How often should ctdb run the event scripts to check for a nodes
387 health.
388 TickleUpdateInterval
390 Default: 20
391 How often will ctdb record and store the "tickle" information used to
393 kickstart stalled tcp connections after a recovery.
394 EventScriptTimeout
396 Default: 20
397 How long should ctdb let an event script run before aborting it and
399 marking the node unhealthy.
400 RecoveryBanPeriod
402 Default: 300
403 If a node becomes banned causing repetitive recovery failures. The node
405 will eventually become banned from the cluster. This controls how long
406 the culprit node will be banned from the cluster before it is allowed
407 to try to join the cluster again. Dont set to small. A node gets banned
408 for a reason and it is usually due to real problems with the node.
409 DatabaseHashSize
411 Default: 100000
412 Size of the hash chains for the local store of the tdbs that ctdb
414 manages.
415 RerecoveryTimeout
417 Default: 10
418 Once a recovery has completed, no additional recoveries are permitted
420 until this timeout has expired.
421 EnableBans
423 Default: 1
424 When set to 0, this disables BANNING completely in the cluster and thus
426 nodes can not get banned, even it they break. Dont set to 0.
427 DeterministicIPs
429 Default: 1
430 When enabled, this tunable makes ctdb try to keep public IP addresses
432 locked to specific nodes as far as possible. This makes it easier for
433 debugging since you can know that as long as all nodes are healthy
434 public IP X will always be hosted by node Y.
435 The cost of using deterministic IP address assignment is that it
437 disables part of the logic where ctdb tries to reduce the number of
438 public IP assignment changes in the cluster. This tunable may increase
439 the number of IP failover/failbacks that are performed on the cluster
440 by a small margin.
441 DisableWhenUnhealthy
443 Default: 0
444 When set, As soon as a node becomes unhealthy, that node will also
446 automatically become permanently DISABLED. Once a node is DISABLED, the
447 only way to make it participate in the cluster again and host services
448 is by manually enabling the node again using ´ctdb enable´.
449 This disables parts of the resilience and robustness of the cluster and
451 should ONLY be used when the system administrator is actively
452 monitoring the cluster, so that nodes can be enabled again.
453 NoIPFailback
455 Default: 0
456 When set to 1, ctdb will not perform failback of IP addresses when a
458 node becomes healthy. Ctdb WILL perform failover of public IP addresses
459 when a node becomes UNHEALTHY, but when the node becomes HEALTHY again,
460 ctdb will not fail the addresses back.
461 Use with caution! Normally when a node becomes available to the cluster
463 ctdb will try to reassign public IP addresses onto the new node as a
464 way to distribute the workload evenly across the clusternode. Ctdb
465 tries to make sure that all running nodes have approximately the same
466 number of public addresses it hosts.
467 When you enable this tunable, CTDB will no longer attempt to rebalance
469 the cluster by failing IP addresses back to the new nodes. An
470 unbalanced cluster will therefore remain unbalanced until there is
471 manual intervention from the administrator. When this parameter is set,
472 you can manually fail public IP addresses over to the new node(s) using
473 the ´ctdb moveip´ command.
474
476 LVS is a mode where CTDB presents one single IP address for the entire
477 cluster. This is an alternative to using public IP addresses and
478 round-robin DNS to loadbalance clients across the cluster.
479 This is similar to using a layer-4 loadbalancing switch but with some
481 restrictions.
482 In this mode the cluster select a set of nodes in the cluster and
484 loadbalance all client access to the LVS address across this set of
485 nodes. This set of nodes are all LVS capable nodes that are HEALTHY, or
486 if no HEALTHY nodes exists all LVS capable nodes regardless of health
487 status. LVS will however never loadbalance traffic to nodes that are
488 BANNED, STOPPED, DISABLED or DISCONNECTED. The "ctdb lvs" command is
489 used to show which nodes are currently load-balanced across.
490 One of the these nodes are elected as the LVSMASTER. This node receives
492 all traffic from clients coming in to the LVS address and multiplexes
493 it across the internal network to one of the nodes that LVS is using.
494 When responding to the client, that node will send the data back
495 directly to the client, bypassing the LVSMASTER node. The command "ctdb
496 lvsmaster" will show which node is the current LVSMASTER.
497 The path used for a client i/o is thus :
499 (1) Client sends request packet to LVSMASTER
501 (2) LVSMASTER passes the request on to one node across the internal network.
502 (3) Selected node processes the request.
503 (4) Node responds back to client.
504 This means that all incoming traffic to the cluster will pass through
507 one physical node, which limits scalability. You can send more data to
508 the LVS address that one physical node can multiplex. This means that
509 you should not use LVS if your I/O pattern is write-intensive since you
510 will be limited in the available network bandwidth that node can
511 handle. LVS does work wery well for read-intensive workloads where only
512 smallish READ requests are going through the LVSMASTER bottleneck and
513 the majority of the traffic volume (the data in the read replies) goes
514 straight from the processing node back to the clients. For
515 read-intensive i/o patterns you can acheive very high throughput rates
516 in this mode.
517 Note: you can use LVS and public addresses at the same time.
519 Configuration
521 To activate LVS on a CTDB node you must specify CTDB_PUBLIC_INTERFACE
522 and CTDB_LVS_PUBLIC_ADDRESS in /etc/sysconfig/ctdb.
523 You must also specify the "--lvs" command line argument to ctdbd to
525 activete LVS as a capability of the node. This should be done
526 automatically for you by the /etc/init.d/ctdb script.
527 Example:
529 CTDB_PUBLIC_INTERFACE=eth0
531 CTDB_LVS_PUBLIC_IP=10.0.0.237
532 If you use LVS, you must still have a real/permanent address configured
535 for the public interface on each node. This address must be routable
536 and the cluster nodes must be configured so that all traffic back to
537 client hosts are routed through this interface. This is also required
538 in order to allow samba/winbind on the node to talk to the domain
539 controller. (we can not use the lvs IP address to initiate outgoing
540 traffic)
541 I.e. make sure that you can "ping" both the domain controller and also
543 all of the clients from the node BEFORE you enable LVS. Also make sure
544 that when you ping these hosts that the traffic is routed out through
545 the eth0 interface.
546
548 It is possible to have a CTDB cluster that spans across a WAN link. For
549 example where you have a CTDB cluster in your datacentre but you also
550 want to have one additional CTDB node located at a remote branch site.
551 This is similar to how a WAN accelerator works but with the difference
552 that while a WAN-accelerator often acts as a Proxy or a MitM, in the
553 ctdb remote cluster node configuration the Samba instance at the remote
554 site IS the genuine server, not a proxy and not a MitM, and thus
555 provides 100% correct CIFS semantics to clients.
556 See the cluster as one single multihomed samba server where one of the
558 NICs (the remote node) is very far away.
559 NOTE: This does require that the cluster filesystem you use can cope
561 with WAN-link latencies. Not all cluster filesystems can handle
562 WAN-link latencies! Whether this will provide very good WAN-accelerator
563 performance or it will perform very poorly depends entirely on how
564 optimized your cluster filesystem is in handling high latency for data
565 and metadata operations.
566 To activate a node as being a remote cluster node you need to set the
568 following two parameters in /etc/sysconfig/ctdb for the remote node:
569 CTDB_CAPABILITY_LMASTER=no
571 CTDB_CAPABILITY_RECMASTER=no
572 Verify with the command "ctdb getcapabilities" that that node no longer
575 has the recmaster or the lmaster capabilities.
576
578 Sometimes it is desireable to run services on the CTDB node which will
579 need to originate outgoing traffic to external servers. This might be
580 contacting NIS servers, LDAP servers etc. etc.
581 This can sometimes be problematic since there are situations when a
583 node does not have any public ip addresses assigned. This could be due
584 to the nobe just being started up and no addresses have been assigned
585 yet or it could be that the node is UNHEALTHY in which case all public
586 addresses have been migrated off.
587 If then the service status of CTDB depends on such services being able
589 to always being able to originate traffic to external resources this
590 becomes extra troublesome. The node might be UNHEALTHY because the
591 service can not be reached, and the service can not be reached because
592 the node is UNHEALTHY.
593 There are two ways to solve this problem. The first is by assigning a
595 static ip address for one public interface on every node which will
596 allow every node to be able to route traffic to the public network even
597 if there are no public addresses assigned to the node. This is the
598 simplest way but it uses up a lot of ip addresses since you have to
599 assign both static and also public addresses to each node.
600 NAT-GW
602 A second way is to use the built in NAT-GW feature in CTDB. With NAT-GW
603 you assign one public NATGW address for each natgw group. Each NATGW
604 group is a set of nodes in the cluster that shares the same NATGW
605 address to talk to the outside world. Normally there would only be one
606 NATGW group spanning the entire cluster, but in situations where one
607 ctdb cluster spans multiple physical sites it is useful to have one
608 NATGW group for each of the two sites.
609 There can be multiple NATGW groups in one cluster but each node can
611 only be member of one NATGW group.
612 In each NATGW group, one of the nodes is designated the NAT Gateway
614 through which all traffic that is originated by nodes in this group
615 will be routed through if a public addresses are not available.
616 Configuration
618 NAT-GW is configured in /etc/sysconfigctdb by setting the following
619 variables:
620 # NAT-GW configuration
622 # Some services running on nthe CTDB node may need to originate traffic to
623 # remote servers before the node is assigned any IP addresses,
624 # This is problematic since before the node has public addresses the node might
625 # not be able to route traffic to the public networks.
626 # One solution is to have static public addresses assigned with routing
627 # in addition to the public address interfaces, thus guaranteeing that
628 # a node always can route traffic to the external network.
629 # This is the most simple solution but it uses up a large number of
630 # additional ip addresses.
631 #
632 # A more complex solution is NAT-GW.
633 # In this mode we only need one additional ip address for the cluster from
634 # the exsternal public network.
635 # One of the nodes in the cluster is elected to be hosting this ip address
636 # so it can reach the external services. This node is also configured
637 # to use NAT MASQUERADING for all traffic from the internal private network
638 # to the external network. This node is the NAT-GW node.
639 #
640 # All other nodes are set up with a default rote with a metric of 10 to point
641 # to the nat-gw node.
642 #
643 # The effect of this is that only when a node does not have a public address
644 # and thus no proper routes to the external world it will instead
645 # route all packets through the nat-gw node.
646 #
647 # CTDB_NATGW_NODES is the list of nodes that belong to this natgw group.
648 # You can have multiple natgw groups in one cluster but each node
649 # can only belong to one single natgw group.
650 #
651 # CTDB_NATGW_PUBLIC_IP=10.0.0.227/24
652 # CTDB_NATGW_PUBLIC_IFACE=eth0
653 # CTDB_NATGW_DEFAULT_GATEWAY=10.0.0.1
654 # CTDB_NATGW_PRIVATE_NETWORK=10.1.1.0/24
655 # CTDB_NATGW_NODES=/etc/ctdb/natgw_nodes
656 CTDB_NATGW_PUBLIC_IP
659 This is an ip address in the public network that is used for all
660 outgoing traffic when the public addresses are not assigned. This
661 address will be assigned to one of the nodes in the cluster which will
662 masquerade all traffic for the other nodes.
663 Format of this parameter is IPADDRESS/NETMASK
665 CTDB_NATGW_PUBLIC_IFACE
667 This is the physical interface where the CTDB_NATGW_PUBLIC_IP will be
668 assigned to. This should be an interface connected to the public
669 network.
670 Format of this parameter is INTERFACE
672 CTDB_NATGW_DEFAULT_GATEWAY
674 This is the default gateway to use on the node that is elected to host
675 the CTDB_NATGW_PUBLIC_IP. This is the default gateway on the public
676 network.
677 Format of this parameter is IPADDRESS
679 CTDB_NATGW_PRIVATE_NETWORK
681 This is the network/netmask used for the interal private network.
682 Format of this parameter is IPADDRESS/NETMASK
684 CTDB_NATGW_NODES
686 This is the list of all nodes that belong to the same NATGW group as
687 this node. The default is /etc/ctdb/natgw_nodes.
688 Operation
690 When the NAT-GW functionality is used, one of the nodes is elected to
691 act as a NAT router for all the other nodes in the group when they need
692 to originate traffic to the external public network.
693 The NAT-GW node is assigned the CTDB_NATGW_PUBLIC_IP to the designated
695 interface and the provided default route. The NAT-GW is configured to
696 act as a router and to masquerade all traffic it receives from the
697 internal private network and which is destined to the external
698 network(s).
699 All other nodes in the group are configured with a default route of
701 metric 10 pointing to the designated NAT GW node.
702 This is implemented in the 11.natgw eventscript. Please see the
704 eventscript for further information.
705 Removing/Changing NATGW at runtime
707 The following are the procedures to change/remove a NATGW configuration
708 at runtime, without having to restart ctdbd.
709 If you want to remove NATGW completely from a node, use these steps:
711 1, Run ´CTDB_BASE=/etc/ctdb /etc/ctdb/events.d/11.natgw removenatgw´
713 2, Then remove the configuration from /etc/sysconfig/ctdb
714 If you want to change the NATGW configuration on a node :
717 1, Run ´CTDB_BASE=/etc/ctdb /etc/ctdb/events.d/11.natgw removenatgw´
719 2, Then change the configuration in /etc/sysconfig/ctdb
720 3, Run ´CTDB_BASE=/etc/ctdb /etc/ctdb/events.d/11.natgw updatenatgw´
721
724 Notification scripts are used with ctdb to have a call-out from ctdb to
725 a user-specified script when certain state changes occur in ctdb. This
726 is commonly to set up either sending SNMP traps or emails when a node
727 becomes unhealthy and similar.
728 This is activated by setting CTDB_NOTIFY_SCRIPT=<your script> in the
730 sysconfig file, or by adding --notification-script=<your script>.
731 See /etc/ctdb/notify.sh for an example script.
733 CTDB currently generates notifications on these state changes:
735 unhealthy
737 This call-out is triggered when the node changes to UNHEALTHY state.
738 healthy
740 This call-out is triggered when the node changes to HEALTHY state.
741 startup
743 This call-out is triggered when ctdb has started up and all managed
744 services are up and running.
745
747 CTDB has support to manage the popular anti-virus daemon ClamAV. This
748 support is implemented through the eventscript :
749 /etc/ctdb/events.d/31.clamd.
750 Configuration
752 Start by configuring CLAMAV normally and test that it works. Once this
753 is done, copy the configuration files over to all the nodes so that all
754 nodes share identical CLAMAV configurations. Once this is done you can
755 proceed with the intructions below to activate CTDB support for CLAMAV.
756 First, to activate CLAMAV support in CTDB, edit /etc/sysconfig/ctdb and
758 add the two lines :
759 CTDB_MANAGES_CLAMD=yes
761 CTDB_CLAMD_SOCKET="/path/to/clamd.socket"
762 Second, activate the eventscript
764 ctdb enablescript 31.clamd
766 Third, CTDB will now be starting and stopping this service accordingly,
768 so make sure that the system is not configured to start/stop this
769 service automatically. On RedHat systems you can disable the system
770 starting/stopping CLAMAV automatically by running :
771 chkconfig clamd off
773 Once you have restarted CTDBD, use
775 ctdb scriptstatus
777 and verify that the 31.clamd eventscript is listed and that it was
779 executed successfully.
780
782 ctdb(1), onnode(1) http://ctdb.samba.org/
783
785 Copyright (C) Andrew Tridgell 2007
786 Copyright (C) Ronnie sahlberg 2007
787 This program is free software; you can redistribute it and/or modify
789 it under the terms of the GNU General Public License as published by
790 the Free Software Foundation; either version 3 of the License, or (at
791 your option) any later version.
792 This program is distributed in the hope that it will be useful, but
794 WITHOUT ANY WARRANTY; without even the implied warranty of
795 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
796 General Public License for more details.
797 You should have received a copy of the GNU General Public License
799 along with this program; if not, see http://www.gnu.org/licenses/.
800 12/04/2009 CTDBD(1)