1 <html><head><meta http-equiv="Content-Type" content="text/html; charset=ISO-8859-1"><title>ctdbd</title><meta name="generator" content="DocBook XSL Stylesheets V1.73.2"></head><body bgcolor="white" text="black" link="#0000FF" vlink="#840084" alink="#0000FF"><div class="refentry" lang="en"><a name="ctdbd.1"></a><div class="titlepage"></div><div class="refnamediv"><h2>Name</h2><p>ctdbd — The CTDB cluster daemon</p></div><div class="refsynopsisdiv"><h2>Synopsis</h2><div class="cmdsynopsis"><p><code class="command">ctdbd</code> </p></div><div class="cmdsynopsis"><p><code class="command">ctdbd</code> [-? --help] [-d --debug=<INTEGER>] {--dbdir=<directory>} {--dbdir-persistent=<directory>} [--event-script-dir=<directory>] [-i --interactive] [--listen=<address>] [--logfile=<filename>] [--lvs] {--nlist=<filename>} [--no-lmaster] [--no-recmaster] [--nosetsched] {--notification-script=<filename>} [--public-addresses=<filename>] [--public-interface=<interface>] {--reclock=<filename>} [--single-public-ip=<address>] [--socket=<filename>] [--start-as-disabled] [--start-as-stopped] [--syslog] [--torture] [--transport=<STRING>] [--usage]</p></div></div><div class="refsect1" lang="en"><a name="id2479664"></a><h2>DESCRIPTION</h2><p>
2 ctdbd is the main ctdb daemon.
4 ctdbd provides a clustered version of the TDB database with automatic rebuild/recovery of the databases upon nodefailures.
6 Combined with a cluster filesystem ctdbd provides a full HA environment for services such as clustered Samba and NFS as well as other services.
8 ctdbd provides monitoring of all nodes in the cluster and automatically reconfigures the cluster and recovers upon node failures.
10 ctdbd is the main component in clustered Samba that provides a high-availability load-sharing CIFS server cluster.
11 </p></div><div class="refsect1" lang="en"><a name="id2479695"></a><h2>OPTIONS</h2><div class="variablelist"><dl><dt><span class="term">-? --help</span></dt><dd><p>
12 Print some help text to the screen.
13 </p></dd><dt><span class="term">-d --debug=<DEBUGLEVEL></span></dt><dd><p>
14 This option sets the debuglevel on the ctdbd daemon which controls what will be written to the logfile. The default is 0 which will only log important events and errors. A larger number will provide additional logging.
15 </p></dd><dt><span class="term">--dbdir=<directory></span></dt><dd><p>
16 This is the directory on local storage where ctdbd keeps the local
17 copy of the TDB databases. This directory is local for each node and should not be stored on the shared cluster filesystem.
19 This directory would usually be /var/ctdb .
20 </p></dd><dt><span class="term">--dbdir-persistent=<directory></span></dt><dd><p>
21 This is the directory on local storage where ctdbd keeps the local
22 copy of the persistent TDB databases. This directory is local for each node and should not be stored on the shared cluster filesystem.
24 This directory would usually be /etc/ctdb/persistent .
25 </p></dd><dt><span class="term">--event-script-dir=<directory></span></dt><dd><p>
26 This option is used to specify the directory where the CTDB event
29 This will normally be /etc/ctdb/events.d which is part of the ctdb distribution.
30 </p></dd><dt><span class="term">-i --interactive</span></dt><dd><p>
31 By default ctdbd will detach itself from the shell and run in
32 the background as a daemon. This option makes ctdbd to start in interactive mode.
33 </p></dd><dt><span class="term">--listen=<address></span></dt><dd><p>
34 This specifies which ip address ctdb will bind to. By default ctdbd will bind to the first address it finds in the /etc/ctdb/nodes file and which is also present on the local system in which case you do not need to provide this option.
36 This option is only required when you want to run multiple ctdbd daemons/nodes on the same physical host in which case there would be multiple entries in /etc/ctdb/nodes what would match a local interface.
37 </p></dd><dt><span class="term">--logfile=<filename></span></dt><dd><p>
38 This is the file where ctdbd will write its log. This is usually /var/log/log.ctdb .
39 </p></dd><dt><span class="term">--lvs</span></dt><dd><p>
40 This option is used to activate the LVS capability on a CTDB node.
41 Please see the LVS section.
42 </p></dd><dt><span class="term">--nlist=<filename></span></dt><dd><p>
43 This file contains a list of the private ip addresses of every node in the cluster. There is one line/ip address for each node. This file must be the same for all nodes in the cluster.
45 This file is usually /etc/ctdb/nodes .
46 </p></dd><dt><span class="term">--no-lmaster</span></dt><dd><p>
47 This argument specifies that this node can NOT become an lmaster
48 for records in the database. This means that it will never show up
49 in the vnnmap. This feature is primarily used for making a cluster
50 span across a WAN link and use CTDB as a WAN-accelerator.
52 Please see the "remote cluster nodes" section for more information.
53 </p></dd><dt><span class="term">--no-recmaster</span></dt><dd><p>
54 This argument specifies that this node can NOT become a recmaster
55 for the database. This feature is primarily used for making a cluster
56 span across a WAN link and use CTDB as a WAN-accelerator.
58 Please see the "remote cluster nodes" section for more information.
59 </p></dd><dt><span class="term">--nosetsched</span></dt><dd><p>
60 This is a ctdbd debugging option. this option is only used when
63 Normally ctdb will change its scheduler to run as a real-time
64 process. This is the default mode for a normal ctdbd operation
65 to gurarantee that ctdbd always gets the cpu cycles that it needs.
67 This option is used to tell ctdbd to NOT run as a real-time process
68 and instead run ctdbd as a normal userspace process.
69 This is useful for debugging and when you want to run ctdbd under
70 valgrind or gdb. (You dont want to attach valgrind or gdb to a
72 </p></dd><dt><span class="term">--notification-script=<filename></span></dt><dd><p>
73 This specifies a script which will be invoked by ctdb when certain
74 state changes occur in ctdbd and when you may want to trigger this
75 to run certain scripts.
77 This file is usually /etc/ctdb/notify.sh .
79 See the NOTIFICATION SCRIPT section below for more information.
80 </p></dd><dt><span class="term">--public_addresses=<filename></span></dt><dd><p>
81 When used with IP takeover this specifies a file containing the public ip addresses to use on the cluster. This file contains a list of ip addresses netmasks and interfaces. When ctdb is operational it will distribute these public ip addresses evenly across the available nodes.
83 This is usually the file /etc/ctdb/public_addresses
84 </p></dd><dt><span class="term">--public-interface=<interface></span></dt><dd><p>
85 This option tells ctdb which interface to attach public-addresses
86 to and also where to attach the single-public-ip when used.
88 This is only required when using public ip addresses and only when
89 you dont specify the interface explicitly in /etc/ctdb/public_addresses or when you are using --single-public-ip.
91 If you omit this argument when using public addresses or single public ip, ctdb will not be able to send out Gratious ARPs correctly or be able to kill tcp connections correctly which will lead to application failures.
92 </p></dd><dt><span class="term">--reclock=<filename></span></dt><dd><p>
93 This is the name of the lock file stored of the shared cluster filesystem that ctdbd uses to prevent split brains from occuring.
94 This file must be stored on shared storage.
96 It is possible to run CTDB without a reclock file, but then there
97 will be no protection against split brain if the network becomes
98 partitioned. Using CTDB without a reclock file is strongly
100 </p></dd><dt><span class="term">--socket=<filename></span></dt><dd><p>
101 This specifies the name of the domain socket that ctdbd will create. This socket is used for local clients to attach to and communicate with the ctdbd daemon.
103 The default is /tmp/ctdb.socket . You only need to use this option if you plan to run multiple ctdbd daemons on the same physical host.
104 </p></dd><dt><span class="term">--start-as-disabled</span></dt><dd><p>
105 This makes the ctdb daemon to be DISABLED when it starts up.
107 As it is DISABLED it will not get any of the public ip addresses
108 allocated to it, and thus this allow you to start ctdb on a node
109 without causing any ip address to failover from other nodes onto
112 When used, the administrator must keep track of when nodes start and
113 manually enable them again using the "ctdb enable" command, or else
114 the node will not host any services.
116 A node that is DISABLED will not host any services and will not be
117 reachable/used by any clients.
118 </p></dd><dt><span class="term">--start-as-stopped</span></dt><dd><p>
119 This makes the ctdb daemon to be STOPPED when it starts up.
121 A node that is STOPPED does not host any public addresses. It is not part of the VNNMAP so it does act as an LMASTER. It also has all databases locked in recovery mode until restarted.
123 To restart and activate a STOPPED node, the command "ctdb continue" is used.
125 A node that is STOPPED will not host any services and will not be
126 reachable/used by any clients.
127 </p></dd><dt><span class="term">--syslog</span></dt><dd><p>
128 Send all log messages to syslog instead of to the ctdb logfile.
129 </p></dd><dt><span class="term">--torture</span></dt><dd><p>
130 This option is only used for development and testing of ctdbd. It adds artificial errors and failures to the common codepaths in ctdbd to verify that ctdbd can recover correctly for failures.
132 You do NOT want to use this option unless you are developing and testing new functionality in ctdbd.
133 </p></dd><dt><span class="term">--transport=<STRING></span></dt><dd><p>
134 This option specifies which transport to use for ctdbd internode communications. The default is "tcp".
136 Currently only "tcp" is supported but "infiniband" might be
137 implemented in the future.
138 </p></dd><dt><span class="term">--usage</span></dt><dd><p>
139 Print useage information to the screen.
140 </p></dd></dl></div></div><div class="refsect1" lang="en"><a name="id2528852"></a><h2>Private vs Public addresses</h2><p>
141 When used for ip takeover in a HA environment, each node in a ctdb
142 cluster has multiple ip addresses assigned to it. One private and one or more public.
143 </p><div class="refsect2" lang="en"><a name="id2528863"></a><h3>Private address</h3><p>
144 This is the physical ip address of the node which is configured in
145 linux and attached to a physical interface. This address uniquely
146 identifies a physical node in the cluster and is the ip addresses
147 that ctdbd will use to communicate with the ctdbd daemons on the
148 other nodes in the cluster.
150 The private addresses are configured in /etc/ctdb/nodes
151 (unless the --nlist option is used) and contain one line for each
152 node in the cluster. Each line contains the private ip address for one
153 node in the cluster. This file must be the same on all nodes in the
156 Since the private addresses are only available to the network when the
157 corresponding node is up and running you should not use these addresses
158 for clients to connect to services provided by the cluster. Instead
159 client applications should only attach to the public addresses since
160 these are guaranteed to always be available.
162 When using ip takeover, it is strongly recommended that the private
163 addresses are configured on a private network physically separated
164 from the rest of the network and that this private network is dedicated
167 Example /etc/ctdb/nodes for a four node cluster:
173 </pre></div><div class="refsect2" lang="en"><a name="id2528911"></a><h3>Public address</h3><p>
174 A public address on the other hand is not attached to an interface.
175 This address is managed by ctdbd itself and is attached/detached to
176 a physical node at runtime.
178 The ctdb cluster will assign/reassign these public addresses across the
179 available healthy nodes in the cluster. When one node fails, its public address
180 will be migrated to and taken over by a different node in the cluster
181 to ensure that all public addresses are always available to clients as
182 long as there are still nodes available capable of hosting this address.
184 These addresses are not physically attached to a specific node.
185 The 'ctdb ip' command can be used to view the current assignment of
186 public addresses and which physical node is currently serving it.
188 On each node this file contains a list of the public addresses that
189 this node is capable of hosting.
190 The list also contain the netmask and the
191 interface where this address should be attached for the case where you
192 may want to serve data out through multiple different interfaces.
194 Example /etc/ctdb/public_addresses for a node that can host 4 public addresses:
201 In most cases this file would be the same on all nodes in a cluster but
202 there are exceptions when one may want to use different files
205 Example: 4 nodes partitioned into two subgroups :
207 Node 0:/etc/ctdb/public_addresses
211 Node 1:/etc/ctdb/public_addresses
215 Node 2:/etc/ctdb/public_addresses
219 Node 3:/etc/ctdb/public_addresses
223 In this example nodes 0 and 1 host two public addresses on the
224 10.1.1.x network while nodes 2 and 3 host two public addresses for the
227 Ip address 10.1.1.1 can be hosted by either of nodes 0 or 1 and will be
228 available to clients as long as at least one of these two nodes are
230 If both nodes 0 and node 1 become unavailable 10.1.1.1 also becomes
231 unavailable. 10.1.1.1 can not be failed over to node 2 or node 3 since
232 these nodes do not have this ip address listed in their public
234 </p></div></div><div class="refsect1" lang="en"><a name="id2528993"></a><h2>Node status</h2><p>
235 The current status of each node in the cluster can be viewed by the
236 'ctdb status' command.
238 There are five possible states for a node.
240 OK - This node is fully functional.
242 DISCONNECTED - This node could not be connected through the network
243 and is currently not particpating in the cluster. If there is a
244 public IP address associated with this node it should have been taken
245 over by a different node. No services are running on this node.
247 DISABLED - This node has been administratively disabled. This node is
248 still functional and participates in the CTDB cluster but its IP
249 addresses have been taken over by a different node and no services are
250 currently being hosted.
252 UNHEALTHY - A service provided by this node is malfunctioning and should
253 be investigated. The CTDB daemon itself is operational and participates
254 in the cluster. Its public IP address has been taken over by a different
255 node and no services are currently being hosted. All unhealthy nodes
256 should be investigated and require an administrative action to rectify.
258 BANNED - This node failed too many recovery attempts and has been banned
259 from participating in the cluster for a period of RecoveryBanPeriod
260 seconds. Any public IP address has been taken over by other nodes. This
261 node does not provide any services. All banned nodes should be
262 investigated and require an administrative action to rectify. This node
263 does not perticipate in the CTDB cluster but can still be communicated
264 with. I.e. ctdb commands can be sent to it.
266 STOPPED - A node that is stopped does not host any public ip addresses,
267 nor is it part of the VNNMAP. A stopped node can not become LVSMASTER,
269 This node does not perticipate in the CTDB cluster but can still be
270 communicated with. I.e. ctdb commands can be sent to it.
271 </p></div><div class="refsect1" lang="en"><a name="id2529059"></a><h2>PUBLIC TUNABLES</h2><p>
272 These are the public tuneables that can be used to control how ctdb behaves.
273 </p><div class="refsect2" lang="en"><a name="id2529070"></a><h3>KeepaliveInterval</h3><p>Default: 1</p><p>
274 How often should the nodes send keepalives to eachother.
275 </p></div><div class="refsect2" lang="en"><a name="id2529083"></a><h3>KeepaliveLimit</h3><p>Default: 5</p><p>
276 After how many keepalive intervals without any traffic should a node
277 wait until marking the peer as DISCONNECTED.
278 </p></div><div class="refsect2" lang="en"><a name="id2529098"></a><h3>MonitorInterval</h3><p>Default: 15</p><p>
279 How often should ctdb run the event scripts to check for a nodes health.
280 </p></div><div class="refsect2" lang="en"><a name="id2529112"></a><h3>TickleUpdateInterval</h3><p>Default: 20</p><p>
281 How often will ctdb record and store the "tickle" information used to
282 kickstart stalled tcp connections after a recovery.
283 </p></div><div class="refsect2" lang="en"><a name="id2529126"></a><h3>EventScriptTimeout</h3><p>Default: 20</p><p>
284 How long should ctdb let an event script run before aborting it and
285 marking the node unhealthy.
286 </p></div><div class="refsect2" lang="en"><a name="id2529140"></a><h3>RecoveryBanPeriod</h3><p>Default: 300</p><p>
287 If a node becomes banned causing repetitive recovery failures. The node will
288 eventually become banned from the cluster.
289 This controls how long the culprit node will be banned from the cluster
290 before it is allowed to try to join the cluster again.
291 Dont set to small. A node gets banned for a reason and it is usually due
292 to real problems with the node.
293 </p></div><div class="refsect2" lang="en"><a name="id2529160"></a><h3>DatabaseHashSize</h3><p>Default: 100000</p><p>
294 Size of the hash chains for the local store of the tdbs that ctdb manages.
295 </p></div><div class="refsect2" lang="en"><a name="id2529174"></a><h3>RerecoveryTimeout</h3><p>Default: 10</p><p>
296 Once a recovery has completed, no additional recoveries are permitted until this timeout has expired.
297 </p></div><div class="refsect2" lang="en"><a name="id2529188"></a><h3>EnableBans</h3><p>Default: 1</p><p>
298 When set to 0, this disables BANNING completely in the cluster and thus nodes can not get banned, even it they break. Dont set to 0.
299 </p></div><div class="refsect2" lang="en"><a name="id2529203"></a><h3>DeterministicIPs</h3><p>Default: 1</p><p>
300 When enabled, this tunable makes ctdb try to keep public IP addresses locked to specific nodes as far as possible. This makes it easier for debugging since you can know that as long as all nodes are healthy public IP X will always be hosted by node Y.
302 The cost of using deterministic IP address assignment is that it disables part of the logic where ctdb tries to reduce the number of public IP assignment changes in the cluster. This tunable may increase the number of IP failover/failbacks that are performed on the cluster by a small margin.
303 </p></div><div class="refsect2" lang="en"><a name="id2529228"></a><h3>DisableWhenUnhealthy</h3><p>Default: 0</p><p>
304 When set, As soon as a node becomes unhealthy, that node will also automatically become permanently DISABLED. Once a node is DISABLED, the only way to make it participate in the cluster again and host services is by manually enabling the node again using 'ctdb enable'.
306 This disables parts of the resilience and robustness of the cluster and should ONLY be used when the system administrator is actively monitoring the cluster, so that nodes can be enabled again.
307 </p></div><div class="refsect2" lang="en"><a name="id2529252"></a><h3>NoIPFailback</h3><p>Default: 0</p><p>
308 When set to 1, ctdb will not perform failback of IP addresses when a node becomes healthy. Ctdb WILL perform failover of public IP addresses when a node becomes UNHEALTHY, but when the node becomes HEALTHY again, ctdb will not fail the addresses back.
310 Use with caution! Normally when a node becomes available to the cluster
311 ctdb will try to reassign public IP addresses onto the new node as a way to distribute the workload evenly across the clusternode. Ctdb tries to make sure that all running nodes have approximately the same number of public addresses it hosts.
313 When you enable this tunable, CTDB will no longer attempt to rebalance the cluster by failing IP addresses back to the new nodes. An unbalanced cluster will therefore remain unbalanced until there is manual intervention from the administrator. When this parameter is set, you can manually fail public IP addresses over to the new node(s) using the 'ctdb moveip' command.
314 </p></div></div><div class="refsect1" lang="en"><a name="id2529288"></a><h2>LVS</h2><p>
315 LVS is a mode where CTDB presents one single IP address for the entire
316 cluster. This is an alternative to using public IP addresses and round-robin
317 DNS to loadbalance clients across the cluster.
319 This is similar to using a layer-4 loadbalancing switch but with some restrictions.
321 In this mode the cluster select a set of nodes in the cluster and loadbalance
322 all client access to the LVS address across this set of nodes. This set of nodes are all LVS capable nodes that are HEALTHY, or if no HEALTHY nodes exists
323 all LVS capable nodes regardless of health status.
324 LVS will however never loadbalance traffic to nodes that are BANNED,
325 STOPPED, DISABLED or DISCONNECTED. The "ctdb lvs" command is used to show
326 which nodes are currently load-balanced across.
328 One of the these nodes are elected as the LVSMASTER. This node receives all
329 traffic from clients coming in to the LVS address and multiplexes it
330 across the internal network to one of the nodes that LVS is using.
331 When responding to the client, that node will send the data back
332 directly to the client, bypassing the LVSMASTER node.
333 The command "ctdb lvsmaster" will show which node is the current
336 The path used for a client i/o is thus :
337 </p><pre class="screen">
338 (1) Client sends request packet to LVSMASTER
339 (2) LVSMASTER passes the request on to one node across the internal network.
340 (3) Selected node processes the request.
341 (4) Node responds back to client.
344 This means that all incoming traffic to the cluster will pass through
345 one physical node, which limits scalability. You can send more data to the
346 LVS address that one physical node can multiplex. This means that you
347 should not use LVS if your I/O pattern is write-intensive since you will be
348 limited in the available network bandwidth that node can handle.
349 LVS does work wery well for read-intensive workloads where only smallish
350 READ requests are going through the LVSMASTER bottleneck and the majority
351 of the traffic volume (the data in the read replies) goes straight from
352 the processing node back to the clients. For read-intensive i/o patterns you can acheive very high throughput rates in this mode.
354 Note: you can use LVS and public addresses at the same time.
355 </p><div class="refsect2" lang="en"><a name="id2529360"></a><h3>Configuration</h3><p>
356 To activate LVS on a CTDB node you must specify CTDB_PUBLIC_INTERFACE and
357 CTDB_LVS_PUBLIC_ADDRESS in /etc/sysconfig/ctdb.
359 You must also specify the "--lvs" command line argument to ctdbd to activete LVS as a capability of the node. This should be done automatically for you by the /etc/init.d/ctdb script.
362 </p><pre class="screen">
363 CTDB_PUBLIC_INTERFACE=eth0
364 CTDB_LVS_PUBLIC_IP=10.0.0.237
367 If you use LVS, you must still have a real/permanent address configured
368 for the public interface on each node. This address must be routable
369 and the cluster nodes must be configured so that all traffic back to client
370 hosts are routed through this interface. This is also required in order
371 to allow samba/winbind on the node to talk to the domain controller.
372 (we can not use the lvs IP address to initiate outgoing traffic)
374 I.e. make sure that you can "ping" both the domain controller and also
375 all of the clients from the node BEFORE you enable LVS. Also make sure
376 that when you ping these hosts that the traffic is routed out through the
378 </p></div><div class="refsect1" lang="en"><a name="id2529408"></a><h2>REMOTE CLUSTER NODES</h2><p>
379 It is possible to have a CTDB cluster that spans across a WAN link.
380 For example where you have a CTDB cluster in your datacentre but you also
381 want to have one additional CTDB node located at a remote branch site.
382 This is similar to how a WAN accelerator works but with the difference
383 that while a WAN-accelerator often acts as a Proxy or a MitM, in
384 the ctdb remote cluster node configuration the Samba instance at the remote site
385 IS the genuine server, not a proxy and not a MitM, and thus provides 100%
386 correct CIFS semantics to clients.
388 See the cluster as one single multihomed samba server where one of
389 the NICs (the remote node) is very far away.
391 NOTE: This does require that the cluster filesystem you use can cope
392 with WAN-link latencies. Not all cluster filesystems can handle
393 WAN-link latencies! Whether this will provide very good WAN-accelerator
394 performance or it will perform very poorly depends entirely
395 on how optimized your cluster filesystem is in handling high latency
396 for data and metadata operations.
398 To activate a node as being a remote cluster node you need to set
399 the following two parameters in /etc/sysconfig/ctdb for the remote node:
400 </p><pre class="screen">
401 CTDB_CAPABILITY_LMASTER=no
402 CTDB_CAPABILITY_RECMASTER=no
405 Verify with the command "ctdb getcapabilities" that that node no longer
406 has the recmaster or the lmaster capabilities.
407 </p></div><div class="refsect1" lang="en"><a name="id2529459"></a><h2>NAT-GW</h2><p>
408 Sometimes it is desireable to run services on the CTDB node which will
409 need to originate outgoing traffic to external servers. This might
410 be contacting NIS servers, LDAP servers etc. etc.
412 This can sometimes be problematic since there are situations when a
413 node does not have any public ip addresses assigned. This could
414 be due to the nobe just being started up and no addresses have been
415 assigned yet or it could be that the node is UNHEALTHY in which
416 case all public addresses have been migrated off.
418 If then the service status of CTDB depends on such services being
419 able to always being able to originate traffic to external resources
420 this becomes extra troublesome. The node might be UNHEALTHY because
421 the service can not be reached, and the service can not be reached
422 because the node is UNHEALTHY.
424 There are two ways to solve this problem. The first is by assigning a
425 static ip address for one public interface on every node which will allow
426 every node to be able to route traffic to the public network even
427 if there are no public addresses assigned to the node.
428 This is the simplest way but it uses up a lot of ip addresses since you
429 have to assign both static and also public addresses to each node.
430 </p><div class="refsect2" lang="en"><a name="id2529499"></a><h3>NAT-GW</h3><p>
431 A second way is to use the built in NAT-GW feature in CTDB.
432 With NAT-GW you assign one public NATGW address for each natgw group.
433 Each NATGW group is a set of nodes in the cluster that shares the same
434 NATGW address to talk to the outside world. Normally there would only be
435 one NATGW group spanning the entire cluster, but in situations where one
436 ctdb cluster spans multiple physical sites it is useful to have one
437 NATGW group for each of the two sites.
439 There can be multiple NATGW groups in one cluster but each node can only
440 be member of one NATGW group.
442 In each NATGW group, one of the nodes is designated the NAT Gateway
443 through which all traffic that is originated by nodes in this group
444 will be routed through if a public addresses are not available.
445 </p></div><div class="refsect2" lang="en"><a name="id2529529"></a><h3>Configuration</h3><p>
446 NAT-GW is configured in /etc/sysconfigctdb by setting the following
448 </p><pre class="screen">
449 # NAT-GW configuration
450 # Some services running on nthe CTDB node may need to originate traffic to
451 # remote servers before the node is assigned any IP addresses,
452 # This is problematic since before the node has public addresses the node might
453 # not be able to route traffic to the public networks.
454 # One solution is to have static public addresses assigned with routing
455 # in addition to the public address interfaces, thus guaranteeing that
456 # a node always can route traffic to the external network.
457 # This is the most simple solution but it uses up a large number of
458 # additional ip addresses.
460 # A more complex solution is NAT-GW.
461 # In this mode we only need one additional ip address for the cluster from
462 # the exsternal public network.
463 # One of the nodes in the cluster is elected to be hosting this ip address
464 # so it can reach the external services. This node is also configured
465 # to use NAT MASQUERADING for all traffic from the internal private network
466 # to the external network. This node is the NAT-GW node.
468 # All other nodes are set up with a default rote with a metric of 10 to point
469 # to the nat-gw node.
471 # The effect of this is that only when a node does not have a public address
472 # and thus no proper routes to the external world it will instead
473 # route all packets through the nat-gw node.
475 # CTDB_NATGW_NODES is the list of nodes that belong to this natgw group.
476 # You can have multiple natgw groups in one cluster but each node
477 # can only belong to one single natgw group.
479 # CTDB_NATGW_PUBLIC_IP=10.0.0.227/24
480 # CTDB_NATGW_PUBLIC_IFACE=eth0
481 # CTDB_NATGW_DEFAULT_GATEWAY=10.0.0.1
482 # CTDB_NATGW_PRIVATE_NETWORK=10.1.1.0/24
483 # CTDB_NATGW_NODES=/etc/ctdb/natgw_nodes
484 </pre></div><div class="refsect2" lang="en"><a name="id2529545"></a><h3>CTDB_NATGW_PUBLIC_IP</h3><p>
485 This is an ip address in the public network that is used for all outgoing
486 traffic when the public addresses are not assigned.
487 This address will be assigned to one of the nodes in the cluster which
488 will masquerade all traffic for the other nodes.
490 Format of this parameter is IPADDRESS/NETMASK
491 </p></div><div class="refsect2" lang="en"><a name="id2529563"></a><h3>CTDB_NATGW_PUBLIC_IFACE</h3><p>
492 This is the physical interface where the CTDB_NATGW_PUBLIC_IP will be
493 assigned to. This should be an interface connected to the public network.
495 Format of this parameter is INTERFACE
496 </p></div><div class="refsect2" lang="en"><a name="id2475745"></a><h3>CTDB_NATGW_DEFAULT_GATEWAY</h3><p>
497 This is the default gateway to use on the node that is elected to host
498 the CTDB_NATGW_PUBLIC_IP. This is the default gateway on the public network.
500 Format of this parameter is IPADDRESS
501 </p></div><div class="refsect2" lang="en"><a name="id2475761"></a><h3>CTDB_NATGW_PRIVATE_NETWORK</h3><p>
502 This is the network/netmask used for the interal private network.
504 Format of this parameter is IPADDRESS/NETMASK
505 </p></div><div class="refsect2" lang="en"><a name="id2475775"></a><h3>CTDB_NATGW_NODES</h3><p>
506 This is the list of all nodes that belong to the same NATGW group
507 as this node. The default is /etc/ctdb/natgw_nodes.
508 </p></div><div class="refsect2" lang="en"><a name="id2475786"></a><h3>Operation</h3><p>
509 When the NAT-GW functionality is used, one of the nodes is elected
510 to act as a NAT router for all the other nodes in the group when
511 they need to originate traffic to the external public network.
513 The NAT-GW node is assigned the CTDB_NATGW_PUBLIC_IP to the designated
514 interface and the provided default route. The NAT-GW is configured
515 to act as a router and to masquerade all traffic it receives from the
516 internal private network and which is destined to the external network(s).
518 All other nodes in the group are configured with a default route of
519 metric 10 pointing to the designated NAT GW node.
521 This is implemented in the 11.natgw eventscript. Please see the
522 eventscript for further information.
523 </p></div><div class="refsect2" lang="en"><a name="id2475818"></a><h3>Removing/Changing NATGW at runtime</h3><p>
524 The following are the procedures to change/remove a NATGW configuration
525 at runtime, without having to restart ctdbd.
527 If you want to remove NATGW completely from a node, use these steps:
528 </p><pre class="screen">
529 1, Run 'CTDB_BASE=/etc/ctdb /etc/ctdb/events.d/11.natgw removenatgw'
530 2, Then remove the configuration from /etc/sysconfig/ctdb
532 If you want to change the NATGW configuration on a node :
533 </p><pre class="screen">
534 1, Run 'CTDB_BASE=/etc/ctdb /etc/ctdb/events.d/11.natgw removenatgw'
535 2, Then change the configuration in /etc/sysconfig/ctdb
536 3, Run 'CTDB_BASE=/etc/ctdb /etc/ctdb/events.d/11.natgw updatenatgw'
537 </pre></div></div><div class="refsect1" lang="en"><a name="id2475856"></a><h2>NOTIFICATION SCRIPT</h2><p>
538 Notification scripts are used with ctdb to have a call-out from ctdb
539 to a user-specified script when certain state changes occur in ctdb.
540 This is commonly to set up either sending SNMP traps or emails
541 when a node becomes unhealthy and similar.
543 This is activated by setting CTDB_NOTIFY_SCRIPT=<your script> in the
544 sysconfig file, or by adding --notification-script=<your script>.
546 See /etc/ctdb/notify.sh for an example script.
548 CTDB currently generates notifications on these state changes:
549 </p><div class="refsect2" lang="en"><a name="id2475887"></a><h3>unhealthy</h3><p>
550 This call-out is triggered when the node changes to UNHEALTHY state.
551 </p></div><div class="refsect2" lang="en"><a name="id2475897"></a><h3>healthy</h3><p>
552 This call-out is triggered when the node changes to HEALTHY state.
553 </p></div><div class="refsect2" lang="en"><a name="id2475907"></a><h3>startup</h3><p>
554 This call-out is triggered when ctdb has started up and all managed services are up and running.
555 </p></div></div><div class="refsect1" lang="en"><a name="id2475919"></a><h2>ClamAV Daemon</h2><p>
556 CTDB has support to manage the popular anti-virus daemon ClamAV.
557 This support is implemented through the
558 eventscript : /etc/ctdb/events.d/31.clamd.
559 </p><div class="refsect2" lang="en"><a name="id2475930"></a><h3>Configuration</h3><p>
560 Start by configuring CLAMAV normally and test that it works. Once this is
561 done, copy the configuration files over to all the nodes so that all nodes
562 share identical CLAMAV configurations.
563 Once this is done you can proceed with the intructions below to activate
564 CTDB support for CLAMAV.
566 First, to activate CLAMAV support in CTDB, edit /etc/sysconfig/ctdb and add the two lines :
567 </p><pre class="screen">
568 CTDB_MANAGES_CLAMD=yes
569 CTDB_CLAMD_SOCKET="/path/to/clamd.socket"
571 Second, activate the eventscript
572 </p><pre class="screen">
573 ctdb enablescript 31.clamd
575 Third, CTDB will now be starting and stopping this service accordingly,
576 so make sure that the system is not configured to start/stop this service
578 On RedHat systems you can disable the system starting/stopping CLAMAV automatically by running :
579 </p><pre class="screen">
583 Once you have restarted CTDBD, use
584 </p><pre class="screen">
587 and verify that the 31.clamd eventscript is listed and that it was executed successfully.
588 </p></div></div><div class="refsect1" lang="en"><a name="id2475990"></a><h2>SEE ALSO</h2><p>
590 <a class="ulink" href="http://ctdb.samba.org/" target="_top">http://ctdb.samba.org/</a>
591 </p></div><div class="refsect1" lang="en"><a name="id2476003"></a><h2>COPYRIGHT/LICENSE</h2><div class="literallayout"><p><br>
592 Copyright (C) Andrew Tridgell 2007<br>
593 Copyright (C) Ronnie sahlberg 2007<br>
595 This program is free software; you can redistribute it and/or modify<br>
596 it under the terms of the GNU General Public License as published by<br>
597 the Free Software Foundation; either version 3 of the License, or (at<br>
598 your option) any later version.<br>
600 This program is distributed in the hope that it will be useful, but<br>
601 WITHOUT ANY WARRANTY; without even the implied warranty of<br>
602 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU<br>
603 General Public License for more details.<br>
605 You should have received a copy of the GNU General Public License<br>
606 along with this program; if not, see http://www.gnu.org/licenses/.<br>
607 </p></div></div></div></body></html>