4 include::attributes.txt[]
9 pvecm - Proxmox VE Cluster Manager
14 include::pvecm.1-synopsis.adoc[]
23 include::attributes.txt[]
26 The {PVE} cluster manager `pvecm` is a tool to create a group of
27 physical servers. Such a group is called a *cluster*. We use the
28 http://www.corosync.org[Corosync Cluster Engine] for reliable group
29 communication, and such clusters can consist of up to 32 physical nodes
30 (probably more, dependent on network latency).
32 `pvecm` can be used to create a new cluster, join nodes to a cluster,
33 leave the cluster, get status information and do various other cluster
34 related tasks. The **P**rox**m**o**x** **C**luster **F**ile **S**ystem (``pmxcfs'')
35 is used to transparently distribute the cluster configuration to all cluster
38 Grouping nodes into a cluster has the following advantages:
40 * Centralized, web based management
42 * Multi-master clusters: each node can do all management task
44 * `pmxcfs`: database-driven file system for storing configuration files,
45 replicated in real-time on all nodes using `corosync`.
47 * Easy migration of virtual machines and containers between physical
52 * Cluster-wide services like firewall and HA
58 * All nodes must be in the same network as `corosync` uses IP Multicast
59 to communicate between nodes (also see
60 http://www.corosync.org[Corosync Cluster Engine]). Corosync uses UDP
61 ports 5404 and 5405 for cluster communication.
63 NOTE: Some switches do not support IP multicast by default and must be
64 manually enabled first.
66 * Date and time have to be synchronized.
68 * SSH tunnel on TCP port 22 between nodes is used.
70 * If you are interested in High Availability, you need to have at
71 least three nodes for reliable quorum. All nodes should have the
74 * We recommend a dedicated NIC for the cluster traffic, especially if
75 you use shared storage.
77 NOTE: It is not possible to mix Proxmox VE 3.x and earlier with
78 Proxmox VE 4.0 cluster nodes.
84 First, install {PVE} on all nodes. Make sure that each node is
85 installed with the final hostname and IP configuration. Changing the
86 hostname and IP is not possible after cluster creation.
88 Currently the cluster creation has to be done on the console, so you
89 need to login via `ssh`.
94 Login via `ssh` to the first {pve} node. Use a unique name for your cluster.
95 This name cannot be changed later.
97 hp1# pvecm create YOUR-CLUSTER-NAME
99 CAUTION: The cluster name is used to compute the default multicast
100 address. Please use unique cluster names if you run more than one
101 cluster inside your network.
103 To check the state of your cluster use:
108 Adding Nodes to the Cluster
109 ---------------------------
111 Login via `ssh` to the node you want to add.
113 hp2# pvecm add IP-ADDRESS-CLUSTER
115 For `IP-ADDRESS-CLUSTER` use the IP from an existing cluster node.
117 CAUTION: A new node cannot hold any VMs, because you would get
118 conflicts about identical VM IDs. Also, all existing configuration in
119 `/etc/pve` is overwritten when you join a new node to the cluster. To
120 workaround, use `vzdump` to backup and restore to a different VMID after
121 adding the node to the cluster.
123 To check the state of cluster:
127 .Cluster status after adding 4 nodes
132 Date: Mon Apr 20 12:30:13 2015
133 Quorum provider: corosync_votequorum
139 Votequorum information
140 ~~~~~~~~~~~~~~~~~~~~~~
147 Membership information
148 ~~~~~~~~~~~~~~~~~~~~~~
150 0x00000001 1 192.168.15.91
151 0x00000002 1 192.168.15.92 (local)
152 0x00000003 1 192.168.15.93
153 0x00000004 1 192.168.15.94
156 If you only want the list of all nodes use:
160 .List nodes in a cluster
164 Membership information
165 ~~~~~~~~~~~~~~~~~~~~~~
173 Adding Nodes With Separated Cluster Network
174 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
176 When adding a node to a cluster with a separated cluster network you need to
177 use the 'ringX_addr' parameters to set the nodes address on those networks:
180 pvecm add IP-ADDRESS-CLUSTER -ring0_addr IP-ADDRESS-RING0
182 If you want to use the Redundant Ring Protocol you will also want to pass the
183 'ring1_addr' parameter.
186 Remove a Cluster Node
187 ---------------------
189 CAUTION: Read carefully the procedure before proceeding, as it could
190 not be what you want or need.
192 Move all virtual machines from the node. Make sure you have no local
193 data or backups you want to keep, or save them accordingly.
195 Log in to one remaining node via ssh. Issue a `pvecm nodes` command to
196 identify the node ID:
203 Date: Mon Apr 20 12:30:13 2015
204 Quorum provider: corosync_votequorum
210 Votequorum information
211 ~~~~~~~~~~~~~~~~~~~~~~
218 Membership information
219 ~~~~~~~~~~~~~~~~~~~~~~
221 0x00000001 1 192.168.15.91 (local)
222 0x00000002 1 192.168.15.92
223 0x00000003 1 192.168.15.93
224 0x00000004 1 192.168.15.94
227 IMPORTANT: at this point you must power off the node to be removed and
228 make sure that it will not power on again (in the network) as it
234 Membership information
235 ~~~~~~~~~~~~~~~~~~~~~~
243 Log in to one remaining node via ssh. Issue the delete command (here
244 deleting node `hp4`):
246 hp1# pvecm delnode hp4
248 If the operation succeeds no output is returned, just check the node
249 list again with `pvecm nodes` or `pvecm status`. You should see
257 Date: Mon Apr 20 12:44:28 2015
258 Quorum provider: corosync_votequorum
264 Votequorum information
265 ~~~~~~~~~~~~~~~~~~~~~~
272 Membership information
273 ~~~~~~~~~~~~~~~~~~~~~~
275 0x00000001 1 192.168.15.90 (local)
276 0x00000002 1 192.168.15.91
277 0x00000003 1 192.168.15.92
280 IMPORTANT: as said above, it is very important to power off the node
281 *before* removal, and make sure that it will *never* power on again
282 (in the existing cluster network) as it is.
284 If you power on the node as it is, your cluster will be screwed up and
285 it could be difficult to restore a clean cluster state.
287 If, for whatever reason, you want that this server joins the same
288 cluster again, you have to
290 * reinstall {pve} on it from scratch
292 * then join it, as explained in the previous section.
294 Separate A Node Without Reinstalling
295 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
297 CAUTION: This is *not* the recommended method, proceed with caution. Use the
298 above mentioned method if you're unsure.
300 You can also separate a node from a cluster without reinstalling it from
301 scratch. But after removing the node from the cluster it will still have
302 access to the shared storages! This must be resolved before you start removing
303 the node from the cluster. A {pve} cluster cannot share the exact same
304 storage with another cluster, as it leads to VMID conflicts.
306 Its suggested that you create a new storage where only the node which you want
307 to separate has access. This can be an new export on your NFS or a new Ceph
308 pool, to name a few examples. Its just important that the exact same storage
309 does not gets accessed by multiple clusters. After setting this storage up move
310 all data from the node and its VMs to it. Then you are ready to separate the
311 node from the cluster.
313 WARNING: Ensure all shared resources are cleanly separated! You will run into
314 conflicts and problems else.
316 First stop the corosync and the pve-cluster services on the node:
318 systemctl stop pve-cluster
319 systemctl stop corosync
321 Start the cluster filesystem again in local mode:
325 Delete the corosync configuration files:
327 rm /etc/pve/corosync.conf
330 You can now start the filesystem again as normal service:
333 systemctl start pve-cluster
335 The node is now separated from the cluster. You can deleted it from a remaining
336 node of the cluster with:
338 pvecm delnode oldnode
340 If the command failed, because the remaining node in the cluster lost quorum
341 when the now separate node exited, you may set the expected votes to 1 as a workaround:
345 And the repeat the 'pvecm delnode' command.
347 Now switch back to the separated node, here delete all remaining files left
348 from the old cluster. This ensures that the node can be added to another
349 cluster again without problems.
352 rm /var/lib/corosync/*
354 As the configuration files from the other nodes are still in the cluster
355 filesystem you may want to clean those up too. Remove simply the whole
356 directory recursive from '/etc/pve/nodes/NODENAME', but check three times that
357 you used the correct one before deleting it.
359 CAUTION: The nodes SSH keys are still in the 'authorized_key' file, this means
360 the nodes can still connect to each other with public key authentication. This
361 should be fixed by removing the respective keys from the
362 '/etc/pve/priv/authorized_keys' file.
367 {pve} use a quorum-based technique to provide a consistent state among
370 [quote, from Wikipedia, Quorum (distributed computing)]
372 A quorum is the minimum number of votes that a distributed transaction
373 has to obtain in order to be allowed to perform an operation in a
377 In case of network partitioning, state changes requires that a
378 majority of nodes are online. The cluster switches to read-only mode
381 NOTE: {pve} assigns a single vote to each node by default.
386 The cluster network is the core of a cluster. All messages sent over it have to
387 be delivered reliable to all nodes in their respective order. In {pve} this
388 part is done by corosync, an implementation of a high performance low overhead
389 high availability development toolkit. It serves our decentralized
390 configuration file system (`pmxcfs`).
392 [[cluster-network-requirements]]
395 This needs a reliable network with latencies under 2 milliseconds (LAN
396 performance) to work properly. While corosync can also use unicast for
397 communication between nodes its **highly recommended** to have a multicast
398 capable network. The network should not be used heavily by other members,
399 ideally corosync runs on its own network.
400 *never* share it with network where storage communicates too.
402 Before setting up a cluster it is good practice to check if the network is fit
405 * Ensure that all nodes are in the same subnet. This must only be true for the
406 network interfaces used for cluster communication (corosync).
408 * Ensure all nodes can reach each other over those interfaces, using `ping` is
409 enough for a basic test.
411 * Ensure that multicast works in general and a high package rates. This can be
412 done with the `omping` tool. The final "%loss" number should be < 1%.
415 omping -c 10000 -i 0.001 -F -q NODE1-IP NODE2-IP ...
418 * Ensure that multicast communication works over an extended period of time.
419 This covers up problems where IGMP snooping is activated on the network but
420 no multicast querier is active. This test has a duration of around 10
423 omping -c 600 -i 1 -q NODE1-IP NODE2-IP ...
425 Your network is not ready for clustering if any of these test fails. Recheck
426 your network configuration. Especially switches are notorious for having
427 multicast disabled by default or IGMP snooping enabled with no IGMP querier
430 In smaller cluster its also an option to use unicast if you really cannot get
433 Separate Cluster Network
434 ~~~~~~~~~~~~~~~~~~~~~~~~
436 When creating a cluster without any parameters the cluster network is generally
437 shared with the Web UI and the VMs and its traffic. Depending on your setup
438 even storage traffic may get sent over the same network. Its recommended to
439 change that, as corosync is a time critical real time application.
441 Setting Up A New Network
442 ^^^^^^^^^^^^^^^^^^^^^^^^
444 First you have to setup a new network interface. It should be on a physical
445 separate network. Ensure that your network fulfills the
446 <<cluster-network-requirements,cluster network requirements>>.
448 Separate On Cluster Creation
449 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
451 This is possible through the 'ring0_addr' and 'bindnet0_addr' parameter of
452 the 'pvecm create' command used for creating a new cluster.
454 If you have setup a additional NIC with a static address on 10.10.10.1/25
455 and want to send and receive all cluster communication over this interface
459 pvecm create test --ring0_addr 10.10.10.1 --bindnet0_addr 10.10.10.0
461 To check if everything is working properly execute:
463 systemctl status corosync
465 [[separate-cluster-net-after-creation]]
466 Separate After Cluster Creation
467 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
469 You can do this also if you have already created a cluster and want to switch
470 its communication to another network, without rebuilding the whole cluster.
471 This change may lead to short durations of quorum loss in the cluster, as nodes
472 have to restart corosync and come up one after the other on the new network.
474 Check how to <<edit-corosync-conf,edit the corosync.conf file>> first.
475 The open it and you should see a file similar to:
509 provider: corosync_votequorum
513 cluster_name: thomas-testcluster
519 bindnetaddr: 192.168.30.50
526 The first you want to do is add the 'name' properties in the node entries if
527 you do not see them already. Those *must* match the node name.
529 Then replace the address from the 'ring0_addr' properties with the new
530 addresses. You may use plain IP addresses or also hostnames here. If you use
531 hostnames ensure that they are resolvable from all nodes.
533 In my example I want to switch my cluster communication to the 10.10.10.1/25
534 network. So I replace all 'ring0_addr' respectively. I also set the bindetaddr
535 in the totem section of the config to an address of the new network. It can be
536 any address from the subnet configured on the new network interface.
538 After you increased the 'config_version' property the new configuration file
554 ring0_addr: 10.10.10.2
561 ring0_addr: 10.10.10.3
568 ring0_addr: 10.10.10.1
574 provider: corosync_votequorum
578 cluster_name: thomas-testcluster
584 bindnetaddr: 10.10.10.1
591 Now after a final check whether all changed information is correct we save it
592 and see again the <<edit-corosync-conf,edit corosync.conf file>> section to
593 learn how to bring it in effect.
595 As our change cannot be enforced live from corosync we have to do an restart.
597 On a single node execute:
599 systemctl restart corosync
601 Now check if everything is fine:
604 systemctl status corosync
606 If corosync runs again correct restart corosync also on all other nodes.
607 They will then join the cluster membership one by one on the new network.
609 Redundant Ring Protocol
610 ~~~~~~~~~~~~~~~~~~~~~~~
611 To avoid a single point of failure you should implement counter measurements.
612 This can be on the hardware and operating system level through network bonding.
614 Corosync itself offers also a possibility to add redundancy through the so
615 called 'Redundant Ring Protocol'. This protocol allows running a second totem
616 ring on another network, this network should be physically separated from the
617 other rings network to actually increase availability.
619 RRP On Cluster Creation
620 ~~~~~~~~~~~~~~~~~~~~~~~
622 The 'pvecm create' command provides the additional parameters 'bindnetX_addr',
623 'ringX_addr' and 'rrp_mode', can be used for RRP configuration.
625 NOTE: See the <<corosync-conf-glossary,glossary>> if you do not know what each parameter means.
627 So if you have two networks, one on the 10.10.10.1/24 and the other on the
628 10.10.20.1/24 subnet you would execute:
631 pvecm create CLUSTERNAME -bindnet0_addr 10.10.10.1 -ring0_addr 10.10.10.1 \
632 -bindnet1_addr 10.10.20.1 -ring1_addr 10.10.20.1
634 RRP On A Created Cluster
635 ~~~~~~~~~~~~~~~~~~~~~~~~
637 When enabling an already running cluster to use RRP you will take similar steps
638 as describe in <<separate-cluster-net-after-creation,separating the cluster
639 network>>. You just do it on another ring.
641 First add a new `interface` subsection in the `totem` section, set its
642 `ringnumber` property to `1`. Set the interfaces `bindnetaddr` property to an
643 address of the subnet you have configured for your new ring.
644 Further set the `rrp_mode` to `passive`, this is the only stable mode.
646 Then add to each node entry in the `nodelist` section its new `ring1_addr`
647 property with the nodes additional ring address.
649 So if you have two networks, one on the 10.10.10.1/24 and the other on the
650 10.10.20.1/24 subnet, the final configuration file should look like:
661 bindnetaddr: 10.10.10.1
665 bindnetaddr: 10.10.20.1
675 ring0_addr: 10.10.10.1
676 ring1_addr: 10.10.20.1
683 ring0_addr: 10.10.10.2
684 ring1_addr: 10.10.20.2
687 [...] # other cluster nodes here
690 [...] # other remaining config sections here
694 Bring it in effect like described in the <<edit-corosync-conf,edit the
695 corosync.conf file>> section.
697 This is a change which cannot take live in effect and needs at least a restart
698 of corosync. Recommended is a restart of the whole cluster.
700 If you cannot reboot the whole cluster ensure no High Availability services are
701 configured and the stop the corosync service on all nodes. After corosync is
702 stopped on all nodes start it one after the other again.
704 Corosync Configuration
705 ----------------------
707 The `/ect/pve/corosync.conf` file plays a central role in {pve} cluster. It
708 controls the cluster member ship and its network.
709 For reading more about it check the corosync.conf man page:
713 For node membership you should always use the `pvecm` tool provided by {pve}.
714 You may have to edit the configuration file manually for other changes.
715 Here are a few best practice tips for doing this.
717 [[edit-corosync-conf]]
721 Editing the corosync.conf file can be not always straight forward. There are
722 two on each cluster, one in `/etc/pve/corosync.conf` and the other in
723 `/etc/corosync/corosync.conf`. Editing the one in our cluster file system will
724 propagate the changes to the local one, but not vice versa.
726 The configuration will get updated automatically as soon as the file changes.
727 This means changes which can be integrated in a running corosync will take
728 instantly effect. So you should always make a copy and edit that instead, to
729 avoid triggering some unwanted changes by an in between safe.
732 cp /etc/pve/corosync.conf /etc/pve/corosync.conf.new
734 Then open the Config file with your favorite editor, `nano` and `vim.tiny` are
735 preinstalled on {pve} for example.
737 NOTE: Always increment the 'config_version' number on configuration changes,
738 omitting this can lead to problems.
740 After making the necessary changes create another copy of the current working
741 configuration file. This serves as a backup if the new configuration fails to
742 apply or makes problems in other ways.
745 cp /etc/pve/corosync.conf /etc/pve/corosync.conf.bak
747 Then move the new configuration file over the old one:
749 mv /etc/pve/corosync.conf.new /etc/pve/corosync.conf
751 You may check with the commands
753 systemctl status corosync
754 journalctl -b -u corosync
756 If the change could applied automatically. If not you may have to restart the
757 corosync service via:
759 systemctl restart corosync
761 On errors check the troubleshooting section below.
766 Issue: 'quorum.expected_votes must be configured'
767 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
769 When corosync starts to fail and you get the following message in the system log:
773 corosync[1647]: [QUORUM] Quorum provider: corosync_votequorum failed to initialize.
774 corosync[1647]: [SERV ] Service engine 'corosync_quorum' failed to load for reason
775 'configuration error: nodelist or quorum.expected_votes must be configured!'
779 It means that the hostname you set for corosync 'ringX_addr' in the
780 configuration could not be resolved.
783 Write Configuration When Not Quorate
784 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
786 If you need to change '/etc/pve/corosync.conf' on an node with no quorum, and you
787 know what you do, use:
791 This sets the expected vote count to 1 and makes the cluster quorate. You can
792 now fix your configuration, or revert it back to the last working backup.
794 This is not enough if corosync cannot start anymore. Here its best to edit the
795 local copy of the corosync configuration in '/etc/corosync/corosync.conf' so
796 that corosync can start again. Ensure that on all nodes this configuration has
797 the same content to avoid split brains. If you are not sure what went wrong
798 it's best to ask the Proxmox Community to help you.
801 [[corosync-conf-glossary]]
802 Corosync Configuration Glossary
803 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
806 This names the different ring addresses for the corosync totem rings used for
807 the cluster communication.
810 Defines to which interface the ring should bind to. It may be any address of
811 the subnet configured on the interface we want to use. In general its the
812 recommended to just use an address a node uses on this interface.
815 Specifies the mode of the redundant ring protocol and may be passive, active or
816 none. Note that use of active is highly experimental and not official
817 supported. Passive is the preferred mode, it may double the cluster
818 communication throughput and increases availability.
824 It is obvious that a cluster is not quorate when all nodes are
825 offline. This is a common case after a power failure.
827 NOTE: It is always a good idea to use an uninterruptible power supply
828 (``UPS'', also called ``battery backup'') to avoid this state, especially if
831 On node startup, service `pve-manager` is started and waits for
832 quorum. Once quorate, it starts all guests which have the `onboot`
835 When you turn on nodes, or when power comes back after power failure,
836 it is likely that some nodes boots faster than others. Please keep in
837 mind that guest startup is delayed until you reach quorum.
841 include::pve-copyright.adoc[]