]>
Commit | Line | Data |
---|---|---|
d8742b0c DM |
1 | ifdef::manvolnum[] |
2 | PVE({manvolnum}) | |
3 | ================ | |
4 | include::attributes.txt[] | |
5 | ||
6 | NAME | |
7 | ---- | |
8 | ||
74026b8f | 9 | pvecm - Proxmox VE Cluster Manager |
d8742b0c DM |
10 | |
11 | SYNOPSYS | |
12 | -------- | |
13 | ||
14 | include::pvecm.1-synopsis.adoc[] | |
15 | ||
16 | DESCRIPTION | |
17 | ----------- | |
18 | endif::manvolnum[] | |
19 | ||
20 | ifndef::manvolnum[] | |
21 | Cluster Manager | |
22 | =============== | |
23 | include::attributes.txt[] | |
24 | endif::manvolnum[] | |
25 | ||
8c1189b6 FG |
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 | |
8a865621 | 28 | http://www.corosync.org[Corosync Cluster Engine] for reliable group |
5eba0743 | 29 | communication, and such clusters can consist of up to 32 physical nodes |
8a865621 DM |
30 | (probably more, dependent on network latency). |
31 | ||
8c1189b6 | 32 | `pvecm` can be used to create a new cluster, join nodes to a cluster, |
8a865621 | 33 | leave the cluster, get status information and do various other cluster |
e300cf7d FG |
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 | |
8a865621 DM |
36 | nodes. |
37 | ||
38 | Grouping nodes into a cluster has the following advantages: | |
39 | ||
40 | * Centralized, web based management | |
41 | ||
5eba0743 | 42 | * Multi-master clusters: each node can do all management task |
8a865621 | 43 | |
8c1189b6 FG |
44 | * `pmxcfs`: database-driven file system for storing configuration files, |
45 | replicated in real-time on all nodes using `corosync`. | |
8a865621 | 46 | |
5eba0743 | 47 | * Easy migration of virtual machines and containers between physical |
8a865621 DM |
48 | hosts |
49 | ||
50 | * Fast deployment | |
51 | ||
52 | * Cluster-wide services like firewall and HA | |
53 | ||
54 | ||
55 | Requirements | |
56 | ------------ | |
57 | ||
8c1189b6 | 58 | * All nodes must be in the same network as `corosync` uses IP Multicast |
8a865621 | 59 | to communicate between nodes (also see |
ceabe189 | 60 | http://www.corosync.org[Corosync Cluster Engine]). Corosync uses UDP |
ff72a2ba | 61 | ports 5404 and 5405 for cluster communication. |
ceabe189 DM |
62 | + |
63 | NOTE: Some switches do not support IP multicast by default and must be | |
64 | manually enabled first. | |
8a865621 DM |
65 | |
66 | * Date and time have to be synchronized. | |
67 | ||
ceabe189 | 68 | * SSH tunnel on TCP port 22 between nodes is used. |
8a865621 | 69 | |
ceabe189 DM |
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 | |
72 | same version. | |
8a865621 DM |
73 | |
74 | * We recommend a dedicated NIC for the cluster traffic, especially if | |
75 | you use shared storage. | |
76 | ||
77 | NOTE: It is not possible to mix Proxmox VE 3.x and earlier with | |
ceabe189 | 78 | Proxmox VE 4.0 cluster nodes. |
8a865621 DM |
79 | |
80 | ||
ceabe189 DM |
81 | Preparing Nodes |
82 | --------------- | |
8a865621 DM |
83 | |
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. | |
87 | ||
88 | Currently the cluster creation has to be done on the console, so you | |
8c1189b6 | 89 | need to login via `ssh`. |
8a865621 | 90 | |
8a865621 | 91 | Create the Cluster |
ceabe189 | 92 | ------------------ |
8a865621 | 93 | |
8c1189b6 FG |
94 | Login via `ssh` to the first {pve} node. Use a unique name for your cluster. |
95 | This name cannot be changed later. | |
8a865621 DM |
96 | |
97 | hp1# pvecm create YOUR-CLUSTER-NAME | |
98 | ||
63f956c8 DM |
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. | |
102 | ||
8a865621 DM |
103 | To check the state of your cluster use: |
104 | ||
105 | hp1# pvecm status | |
106 | ||
107 | ||
108 | Adding Nodes to the Cluster | |
ceabe189 | 109 | --------------------------- |
8a865621 | 110 | |
8c1189b6 | 111 | Login via `ssh` to the node you want to add. |
8a865621 DM |
112 | |
113 | hp2# pvecm add IP-ADDRESS-CLUSTER | |
114 | ||
115 | For `IP-ADDRESS-CLUSTER` use the IP from an existing cluster node. | |
116 | ||
5eba0743 | 117 | CAUTION: A new node cannot hold any VMs, because you would get |
7980581f | 118 | conflicts about identical VM IDs. Also, all existing configuration in |
8c1189b6 FG |
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 | |
7980581f | 121 | adding the node to the cluster. |
8a865621 DM |
122 | |
123 | To check the state of cluster: | |
124 | ||
125 | # pvecm status | |
126 | ||
ceabe189 | 127 | .Cluster status after adding 4 nodes |
8a865621 DM |
128 | ---- |
129 | hp2# pvecm status | |
130 | Quorum information | |
131 | ~~~~~~~~~~~~~~~~~~ | |
132 | Date: Mon Apr 20 12:30:13 2015 | |
133 | Quorum provider: corosync_votequorum | |
134 | Nodes: 4 | |
135 | Node ID: 0x00000001 | |
136 | Ring ID: 1928 | |
137 | Quorate: Yes | |
138 | ||
139 | Votequorum information | |
140 | ~~~~~~~~~~~~~~~~~~~~~~ | |
141 | Expected votes: 4 | |
142 | Highest expected: 4 | |
143 | Total votes: 4 | |
144 | Quorum: 2 | |
145 | Flags: Quorate | |
146 | ||
147 | Membership information | |
148 | ~~~~~~~~~~~~~~~~~~~~~~ | |
149 | Nodeid Votes Name | |
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 | |
154 | ---- | |
155 | ||
156 | If you only want the list of all nodes use: | |
157 | ||
158 | # pvecm nodes | |
159 | ||
5eba0743 | 160 | .List nodes in a cluster |
8a865621 DM |
161 | ---- |
162 | hp2# pvecm nodes | |
163 | ||
164 | Membership information | |
165 | ~~~~~~~~~~~~~~~~~~~~~~ | |
166 | Nodeid Votes Name | |
167 | 1 1 hp1 | |
168 | 2 1 hp2 (local) | |
169 | 3 1 hp3 | |
170 | 4 1 hp4 | |
171 | ---- | |
172 | ||
e4ec4154 TL |
173 | Adding Nodes With Separated Cluster Network |
174 | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ | |
175 | ||
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: | |
178 | ||
179 | [source,bash] | |
180 | pvecm add IP-ADDRESS-CLUSTER -ring0_addr IP-ADDRESS-RING0 | |
181 | ||
182 | If you want to use the Redundant Ring Protocol you will also want to pass the | |
183 | 'ring1_addr' parameter. | |
184 | ||
8a865621 DM |
185 | |
186 | Remove a Cluster Node | |
ceabe189 | 187 | --------------------- |
8a865621 DM |
188 | |
189 | CAUTION: Read carefully the procedure before proceeding, as it could | |
190 | not be what you want or need. | |
191 | ||
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. | |
194 | ||
8c1189b6 | 195 | Log in to one remaining node via ssh. Issue a `pvecm nodes` command to |
7980581f | 196 | identify the node ID: |
8a865621 DM |
197 | |
198 | ---- | |
199 | hp1# pvecm status | |
200 | ||
201 | Quorum information | |
202 | ~~~~~~~~~~~~~~~~~~ | |
203 | Date: Mon Apr 20 12:30:13 2015 | |
204 | Quorum provider: corosync_votequorum | |
205 | Nodes: 4 | |
206 | Node ID: 0x00000001 | |
207 | Ring ID: 1928 | |
208 | Quorate: Yes | |
209 | ||
210 | Votequorum information | |
211 | ~~~~~~~~~~~~~~~~~~~~~~ | |
212 | Expected votes: 4 | |
213 | Highest expected: 4 | |
214 | Total votes: 4 | |
215 | Quorum: 2 | |
216 | Flags: Quorate | |
217 | ||
218 | Membership information | |
219 | ~~~~~~~~~~~~~~~~~~~~~~ | |
220 | Nodeid Votes Name | |
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 | |
225 | ---- | |
226 | ||
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 | |
229 | is. | |
230 | ||
231 | ---- | |
232 | hp1# pvecm nodes | |
233 | ||
234 | Membership information | |
235 | ~~~~~~~~~~~~~~~~~~~~~~ | |
236 | Nodeid Votes Name | |
237 | 1 1 hp1 (local) | |
238 | 2 1 hp2 | |
239 | 3 1 hp3 | |
240 | 4 1 hp4 | |
241 | ---- | |
242 | ||
243 | Log in to one remaining node via ssh. Issue the delete command (here | |
8c1189b6 | 244 | deleting node `hp4`): |
8a865621 DM |
245 | |
246 | hp1# pvecm delnode hp4 | |
247 | ||
248 | If the operation succeeds no output is returned, just check the node | |
8c1189b6 | 249 | list again with `pvecm nodes` or `pvecm status`. You should see |
8a865621 DM |
250 | something like: |
251 | ||
252 | ---- | |
253 | hp1# pvecm status | |
254 | ||
255 | Quorum information | |
256 | ~~~~~~~~~~~~~~~~~~ | |
257 | Date: Mon Apr 20 12:44:28 2015 | |
258 | Quorum provider: corosync_votequorum | |
259 | Nodes: 3 | |
260 | Node ID: 0x00000001 | |
261 | Ring ID: 1992 | |
262 | Quorate: Yes | |
263 | ||
264 | Votequorum information | |
265 | ~~~~~~~~~~~~~~~~~~~~~~ | |
266 | Expected votes: 3 | |
267 | Highest expected: 3 | |
268 | Total votes: 3 | |
269 | Quorum: 3 | |
270 | Flags: Quorate | |
271 | ||
272 | Membership information | |
273 | ~~~~~~~~~~~~~~~~~~~~~~ | |
274 | Nodeid Votes Name | |
275 | 0x00000001 1 192.168.15.90 (local) | |
276 | 0x00000002 1 192.168.15.91 | |
277 | 0x00000003 1 192.168.15.92 | |
278 | ---- | |
279 | ||
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. | |
283 | ||
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. | |
286 | ||
287 | If, for whatever reason, you want that this server joins the same | |
288 | cluster again, you have to | |
289 | ||
26ca7ff5 | 290 | * reinstall {pve} on it from scratch |
8a865621 DM |
291 | |
292 | * then join it, as explained in the previous section. | |
d8742b0c | 293 | |
555e966b TL |
294 | Separate A Node Without Reinstalling |
295 | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ | |
296 | ||
297 | CAUTION: This is *not* the recommended method, proceed with caution. Use the | |
298 | above mentioned method if you're unsure. | |
299 | ||
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. | |
305 | ||
306 | Move the guests which you want to keep on this node now, after the removal you | |
307 | can do this only via backup and restore. Its suggested that you create a new | |
308 | storage where only the node which you want to separate has access. This can be | |
309 | an new export on your NFS or a new Ceph pool, to name a few examples. Its just | |
310 | important that the exact same storage does not gets accessed by multiple | |
311 | clusters. After setting this storage up move all data from the node and its VMs | |
312 | to it. Then you are ready to separate the node from the cluster. | |
313 | ||
314 | WARNING: Ensure all shared resources are cleanly separated! You will run into | |
315 | conflicts and problems else. | |
316 | ||
317 | First stop the corosync and the pve-cluster services on the node: | |
318 | [source,bash] | |
319 | systemctl stop pve-cluster | |
320 | systemctl stop corosync | |
321 | ||
322 | Start the cluster filesystem again in local mode: | |
323 | [source,bash] | |
324 | pmxcfs -l | |
325 | ||
326 | Delete the corosync configuration files: | |
327 | [source,bash] | |
328 | rm /etc/pve/corosync.conf | |
329 | rm /etc/corosync/* | |
330 | ||
331 | You can now start the filesystem again as normal service: | |
332 | [source,bash] | |
333 | killall pmxcfs | |
334 | systemctl start pve-cluster | |
335 | ||
336 | The node is now separated from the cluster. You can deleted it from a remaining | |
337 | node of the cluster with: | |
338 | [source,bash] | |
339 | pvecm delnode oldnode | |
340 | ||
341 | If the command failed, because the remaining node in the cluster lost quorum | |
342 | when the now separate node exited, you may set the expected votes to 1 as a workaround: | |
343 | [source,bash] | |
344 | pvecm expected 1 | |
345 | ||
346 | And the repeat the 'pvecm delnode' command. | |
347 | ||
348 | Now switch back to the separated node, here delete all remaining files left | |
349 | from the old cluster. This ensures that the node can be added to another | |
350 | cluster again without problems. | |
351 | ||
352 | [source,bash] | |
353 | rm /var/lib/corosync/* | |
354 | ||
355 | As the configuration files from the other nodes are still in the cluster | |
356 | filesystem you may want to clean those up too. Remove simply the whole | |
357 | directory recursive from '/etc/pve/nodes/NODENAME', but check three times that | |
358 | you used the correct one before deleting it. | |
359 | ||
360 | CAUTION: The nodes SSH keys are still in the 'authorized_key' file, this means | |
361 | the nodes can still connect to each other with public key authentication. This | |
362 | should be fixed by removing the respective keys from the | |
363 | '/etc/pve/priv/authorized_keys' file. | |
d8742b0c | 364 | |
806ef12d DM |
365 | Quorum |
366 | ------ | |
367 | ||
368 | {pve} use a quorum-based technique to provide a consistent state among | |
369 | all cluster nodes. | |
370 | ||
371 | [quote, from Wikipedia, Quorum (distributed computing)] | |
372 | ____ | |
373 | A quorum is the minimum number of votes that a distributed transaction | |
374 | has to obtain in order to be allowed to perform an operation in a | |
375 | distributed system. | |
376 | ____ | |
377 | ||
378 | In case of network partitioning, state changes requires that a | |
379 | majority of nodes are online. The cluster switches to read-only mode | |
5eba0743 | 380 | if it loses quorum. |
806ef12d DM |
381 | |
382 | NOTE: {pve} assigns a single vote to each node by default. | |
383 | ||
e4ec4154 TL |
384 | Cluster Network |
385 | --------------- | |
386 | ||
387 | The cluster network is the core of a cluster. All messages sent over it have to | |
388 | be delivered reliable to all nodes in their respective order. In {pve} this | |
389 | part is done by corosync, an implementation of a high performance low overhead | |
390 | high availability development toolkit. It serves our decentralized | |
391 | configuration file system (`pmxcfs`). | |
392 | ||
393 | [[cluster-network-requirements]] | |
394 | Network Requirements | |
395 | ~~~~~~~~~~~~~~~~~~~~ | |
396 | This needs a reliable network with latencies under 2 milliseconds (LAN | |
397 | performance) to work properly. While corosync can also use unicast for | |
398 | communication between nodes its **highly recommended** to have a multicast | |
399 | capable network. The network should not be used heavily by other members, | |
400 | ideally corosync runs on its own network. | |
401 | *never* share it with network where storage communicates too. | |
402 | ||
403 | Before setting up a cluster it is good practice to check if the network is fit | |
404 | for that purpose. | |
405 | ||
406 | * Ensure that all nodes are in the same subnet. This must only be true for the | |
407 | network interfaces used for cluster communication (corosync). | |
408 | ||
409 | * Ensure all nodes can reach each other over those interfaces, using `ping` is | |
410 | enough for a basic test. | |
411 | ||
412 | * Ensure that multicast works in general and a high package rates. This can be | |
413 | done with the `omping` tool. The final "%loss" number should be < 1%. | |
414 | [source,bash] | |
415 | ---- | |
416 | omping -c 10000 -i 0.001 -F -q NODE1-IP NODE2-IP ... | |
417 | ---- | |
418 | ||
419 | * Ensure that multicast communication works over an extended period of time. | |
420 | This covers up problems where IGMP snooping is activated on the network but | |
421 | no multicast querier is active. This test has a duration of around 10 | |
422 | minutes. | |
423 | [source,bash] | |
424 | omping -c 600 -i 1 -q NODE1-IP NODE2-IP ... | |
425 | ||
426 | Your network is not ready for clustering if any of these test fails. Recheck | |
427 | your network configuration. Especially switches are notorious for having | |
428 | multicast disabled by default or IGMP snooping enabled with no IGMP querier | |
429 | active. | |
430 | ||
431 | In smaller cluster its also an option to use unicast if you really cannot get | |
432 | multicast to work. | |
433 | ||
434 | Separate Cluster Network | |
435 | ~~~~~~~~~~~~~~~~~~~~~~~~ | |
436 | ||
437 | When creating a cluster without any parameters the cluster network is generally | |
438 | shared with the Web UI and the VMs and its traffic. Depending on your setup | |
439 | even storage traffic may get sent over the same network. Its recommended to | |
440 | change that, as corosync is a time critical real time application. | |
441 | ||
442 | Setting Up A New Network | |
443 | ^^^^^^^^^^^^^^^^^^^^^^^^ | |
444 | ||
445 | First you have to setup a new network interface. It should be on a physical | |
446 | separate network. Ensure that your network fulfills the | |
447 | <<cluster-network-requirements,cluster network requirements>>. | |
448 | ||
449 | Separate On Cluster Creation | |
450 | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^ | |
451 | ||
452 | This is possible through the 'ring0_addr' and 'bindnet0_addr' parameter of | |
453 | the 'pvecm create' command used for creating a new cluster. | |
454 | ||
455 | If you have setup a additional NIC with a static address on 10.10.10.1/25 | |
456 | and want to send and receive all cluster communication over this interface | |
457 | you would execute: | |
458 | ||
459 | [source,bash] | |
460 | pvecm create test --ring0_addr 10.10.10.1 --bindnet0_addr 10.10.10.0 | |
461 | ||
462 | To check if everything is working properly execute: | |
463 | [source,bash] | |
464 | systemctl status corosync | |
465 | ||
466 | [[separate-cluster-net-after-creation]] | |
467 | Separate After Cluster Creation | |
468 | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ | |
469 | ||
470 | You can do this also if you have already created a cluster and want to switch | |
471 | its communication to another network, without rebuilding the whole cluster. | |
472 | This change may lead to short durations of quorum loss in the cluster, as nodes | |
473 | have to restart corosync and come up one after the other on the new network. | |
474 | ||
475 | Check how to <<edit-corosync-conf,edit the corosync.conf file>> first. | |
476 | The open it and you should see a file similar to: | |
477 | ||
478 | ---- | |
479 | logging { | |
480 | debug: off | |
481 | to_syslog: yes | |
482 | } | |
483 | ||
484 | nodelist { | |
485 | ||
486 | node { | |
487 | name: due | |
488 | nodeid: 2 | |
489 | quorum_votes: 1 | |
490 | ring0_addr: due | |
491 | } | |
492 | ||
493 | node { | |
494 | name: tre | |
495 | nodeid: 3 | |
496 | quorum_votes: 1 | |
497 | ring0_addr: tre | |
498 | } | |
499 | ||
500 | node { | |
501 | name: uno | |
502 | nodeid: 1 | |
503 | quorum_votes: 1 | |
504 | ring0_addr: uno | |
505 | } | |
506 | ||
507 | } | |
508 | ||
509 | quorum { | |
510 | provider: corosync_votequorum | |
511 | } | |
512 | ||
513 | totem { | |
514 | cluster_name: thomas-testcluster | |
515 | config_version: 3 | |
516 | ip_version: ipv4 | |
517 | secauth: on | |
518 | version: 2 | |
519 | interface { | |
520 | bindnetaddr: 192.168.30.50 | |
521 | ringnumber: 0 | |
522 | } | |
523 | ||
524 | } | |
525 | ---- | |
526 | ||
527 | The first you want to do is add the 'name' properties in the node entries if | |
528 | you do not see them already. Those *must* match the node name. | |
529 | ||
530 | Then replace the address from the 'ring0_addr' properties with the new | |
531 | addresses. You may use plain IP addresses or also hostnames here. If you use | |
532 | hostnames ensure that they are resolvable from all nodes. | |
533 | ||
534 | In my example I want to switch my cluster communication to the 10.10.10.1/25 | |
535 | network. So I replace all 'ring0_addr' respectively. I also set the bindetaddr | |
536 | in the totem section of the config to an address of the new network. It can be | |
537 | any address from the subnet configured on the new network interface. | |
538 | ||
539 | After you increased the 'config_version' property the new configuration file | |
540 | should look like: | |
541 | ||
542 | ---- | |
543 | ||
544 | logging { | |
545 | debug: off | |
546 | to_syslog: yes | |
547 | } | |
548 | ||
549 | nodelist { | |
550 | ||
551 | node { | |
552 | name: due | |
553 | nodeid: 2 | |
554 | quorum_votes: 1 | |
555 | ring0_addr: 10.10.10.2 | |
556 | } | |
557 | ||
558 | node { | |
559 | name: tre | |
560 | nodeid: 3 | |
561 | quorum_votes: 1 | |
562 | ring0_addr: 10.10.10.3 | |
563 | } | |
564 | ||
565 | node { | |
566 | name: uno | |
567 | nodeid: 1 | |
568 | quorum_votes: 1 | |
569 | ring0_addr: 10.10.10.1 | |
570 | } | |
571 | ||
572 | } | |
573 | ||
574 | quorum { | |
575 | provider: corosync_votequorum | |
576 | } | |
577 | ||
578 | totem { | |
579 | cluster_name: thomas-testcluster | |
580 | config_version: 4 | |
581 | ip_version: ipv4 | |
582 | secauth: on | |
583 | version: 2 | |
584 | interface { | |
585 | bindnetaddr: 10.10.10.1 | |
586 | ringnumber: 0 | |
587 | } | |
588 | ||
589 | } | |
590 | ---- | |
591 | ||
592 | Now after a final check whether all changed information is correct we save it | |
593 | and see again the <<edit-corosync-conf,edit corosync.conf file>> section to | |
594 | learn how to bring it in effect. | |
595 | ||
596 | As our change cannot be enforced live from corosync we have to do an restart. | |
597 | ||
598 | On a single node execute: | |
599 | [source,bash] | |
600 | systemctl restart corosync | |
601 | ||
602 | Now check if everything is fine: | |
603 | ||
604 | [source,bash] | |
605 | systemctl status corosync | |
606 | ||
607 | If corosync runs again correct restart corosync also on all other nodes. | |
608 | They will then join the cluster membership one by one on the new network. | |
609 | ||
610 | Redundant Ring Protocol | |
611 | ~~~~~~~~~~~~~~~~~~~~~~~ | |
612 | To avoid a single point of failure you should implement counter measurements. | |
613 | This can be on the hardware and operating system level through network bonding. | |
614 | ||
615 | Corosync itself offers also a possibility to add redundancy through the so | |
616 | called 'Redundant Ring Protocol'. This protocol allows running a second totem | |
617 | ring on another network, this network should be physically separated from the | |
618 | other rings network to actually increase availability. | |
619 | ||
620 | RRP On Cluster Creation | |
621 | ~~~~~~~~~~~~~~~~~~~~~~~ | |
622 | ||
623 | The 'pvecm create' command provides the additional parameters 'bindnetX_addr', | |
624 | 'ringX_addr' and 'rrp_mode', can be used for RRP configuration. | |
625 | ||
626 | NOTE: See the <<corosync-conf-glossary,glossary>> if you do not know what each parameter means. | |
627 | ||
628 | So if you have two networks, one on the 10.10.10.1/24 and the other on the | |
629 | 10.10.20.1/24 subnet you would execute: | |
630 | ||
631 | [source,bash] | |
632 | pvecm create CLUSTERNAME -bindnet0_addr 10.10.10.1 -ring0_addr 10.10.10.1 \ | |
633 | -bindnet1_addr 10.10.20.1 -ring1_addr 10.10.20.1 | |
634 | ||
635 | RRP On A Created Cluster | |
636 | ~~~~~~~~~~~~~~~~~~~~~~~~ | |
637 | ||
638 | When enabling an already running cluster to use RRP you will take similar steps | |
639 | as describe in <<separate-cluster-net-after-creation,separating the cluster | |
640 | network>>. You just do it on another ring. | |
641 | ||
642 | First add a new `interface` subsection in the `totem` section, set its | |
643 | `ringnumber` property to `1`. Set the interfaces `bindnetaddr` property to an | |
644 | address of the subnet you have configured for your new ring. | |
645 | Further set the `rrp_mode` to `passive`, this is the only stable mode. | |
646 | ||
647 | Then add to each node entry in the `nodelist` section its new `ring1_addr` | |
648 | property with the nodes additional ring address. | |
649 | ||
650 | So if you have two networks, one on the 10.10.10.1/24 and the other on the | |
651 | 10.10.20.1/24 subnet, the final configuration file should look like: | |
652 | ||
653 | ---- | |
654 | totem { | |
655 | cluster_name: tweak | |
656 | config_version: 9 | |
657 | ip_version: ipv4 | |
658 | rrp_mode: passive | |
659 | secauth: on | |
660 | version: 2 | |
661 | interface { | |
662 | bindnetaddr: 10.10.10.1 | |
663 | ringnumber: 0 | |
664 | } | |
665 | interface { | |
666 | bindnetaddr: 10.10.20.1 | |
667 | ringnumber: 1 | |
668 | } | |
669 | } | |
670 | ||
671 | nodelist { | |
672 | node { | |
673 | name: pvecm1 | |
674 | nodeid: 1 | |
675 | quorum_votes: 1 | |
676 | ring0_addr: 10.10.10.1 | |
677 | ring1_addr: 10.10.20.1 | |
678 | } | |
679 | ||
680 | node { | |
681 | name: pvecm2 | |
682 | nodeid: 2 | |
683 | quorum_votes: 1 | |
684 | ring0_addr: 10.10.10.2 | |
685 | ring1_addr: 10.10.20.2 | |
686 | } | |
687 | ||
688 | [...] # other cluster nodes here | |
689 | } | |
690 | ||
691 | [...] # other remaining config sections here | |
692 | ||
693 | ---- | |
694 | ||
695 | Bring it in effect like described in the <<edit-corosync-conf,edit the | |
696 | corosync.conf file>> section. | |
697 | ||
698 | This is a change which cannot take live in effect and needs at least a restart | |
699 | of corosync. Recommended is a restart of the whole cluster. | |
700 | ||
701 | If you cannot reboot the whole cluster ensure no High Availability services are | |
702 | configured and the stop the corosync service on all nodes. After corosync is | |
703 | stopped on all nodes start it one after the other again. | |
704 | ||
705 | Corosync Configuration | |
706 | ---------------------- | |
707 | ||
708 | The `/ect/pve/corosync.conf` file plays a central role in {pve} cluster. It | |
709 | controls the cluster member ship and its network. | |
710 | For reading more about it check the corosync.conf man page: | |
711 | [source,bash] | |
712 | man corosync.conf | |
713 | ||
714 | For node membership you should always use the `pvecm` tool provided by {pve}. | |
715 | You may have to edit the configuration file manually for other changes. | |
716 | Here are a few best practice tips for doing this. | |
717 | ||
718 | [[edit-corosync-conf]] | |
719 | Edit corosync.conf | |
720 | ~~~~~~~~~~~~~~~~~~ | |
721 | ||
722 | Editing the corosync.conf file can be not always straight forward. There are | |
723 | two on each cluster, one in `/etc/pve/corosync.conf` and the other in | |
724 | `/etc/corosync/corosync.conf`. Editing the one in our cluster file system will | |
725 | propagate the changes to the local one, but not vice versa. | |
726 | ||
727 | The configuration will get updated automatically as soon as the file changes. | |
728 | This means changes which can be integrated in a running corosync will take | |
729 | instantly effect. So you should always make a copy and edit that instead, to | |
730 | avoid triggering some unwanted changes by an in between safe. | |
731 | ||
732 | [source,bash] | |
733 | cp /etc/pve/corosync.conf /etc/pve/corosync.conf.new | |
734 | ||
735 | Then open the Config file with your favorite editor, `nano` and `vim.tiny` are | |
736 | preinstalled on {pve} for example. | |
737 | ||
738 | NOTE: Always increment the 'config_version' number on configuration changes, | |
739 | omitting this can lead to problems. | |
740 | ||
741 | After making the necessary changes create another copy of the current working | |
742 | configuration file. This serves as a backup if the new configuration fails to | |
743 | apply or makes problems in other ways. | |
744 | ||
745 | [source,bash] | |
746 | cp /etc/pve/corosync.conf /etc/pve/corosync.conf.bak | |
747 | ||
748 | Then move the new configuration file over the old one: | |
749 | [source,bash] | |
750 | mv /etc/pve/corosync.conf.new /etc/pve/corosync.conf | |
751 | ||
752 | You may check with the commands | |
753 | [source,bash] | |
754 | systemctl status corosync | |
755 | journalctl -b -u corosync | |
756 | ||
757 | If the change could applied automatically. If not you may have to restart the | |
758 | corosync service via: | |
759 | [source,bash] | |
760 | systemctl restart corosync | |
761 | ||
762 | On errors check the troubleshooting section below. | |
763 | ||
764 | Troubleshooting | |
765 | ~~~~~~~~~~~~~~~ | |
766 | ||
767 | Issue: 'quorum.expected_votes must be configured' | |
768 | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ | |
769 | ||
770 | When corosync starts to fail and you get the following message in the system log: | |
771 | ||
772 | ---- | |
773 | [...] | |
774 | corosync[1647]: [QUORUM] Quorum provider: corosync_votequorum failed to initialize. | |
775 | corosync[1647]: [SERV ] Service engine 'corosync_quorum' failed to load for reason | |
776 | 'configuration error: nodelist or quorum.expected_votes must be configured!' | |
777 | [...] | |
778 | ---- | |
779 | ||
780 | It means that the hostname you set for corosync 'ringX_addr' in the | |
781 | configuration could not be resolved. | |
782 | ||
783 | ||
784 | Write Configuration When Not Quorate | |
785 | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ | |
786 | ||
787 | If you need to change '/etc/pve/corosync.conf' on an node with no quorum, and you | |
788 | know what you do, use: | |
789 | [source,bash] | |
790 | pvecm expected 1 | |
791 | ||
792 | This sets the expected vote count to 1 and makes the cluster quorate. You can | |
793 | now fix your configuration, or revert it back to the last working backup. | |
794 | ||
795 | This is not enough if corosync cannot start anymore. Here its best to edit the | |
796 | local copy of the corosync configuration in '/etc/corosync/corosync.conf' so | |
797 | that corosync can start again. Ensure that on all nodes this configuration has | |
798 | the same content to avoid split brains. If you are not sure what went wrong | |
799 | it's best to ask the Proxmox Community to help you. | |
800 | ||
801 | ||
802 | [[corosync-conf-glossary]] | |
803 | Corosync Configuration Glossary | |
804 | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ | |
805 | ||
806 | ringX_addr:: | |
807 | This names the different ring addresses for the corosync totem rings used for | |
808 | the cluster communication. | |
809 | ||
810 | bindnetaddr:: | |
811 | Defines to which interface the ring should bind to. It may be any address of | |
812 | the subnet configured on the interface we want to use. In general its the | |
813 | recommended to just use an address a node uses on this interface. | |
814 | ||
815 | rrp_mode:: | |
816 | Specifies the mode of the redundant ring protocol and may be passive, active or | |
817 | none. Note that use of active is highly experimental and not official | |
818 | supported. Passive is the preferred mode, it may double the cluster | |
819 | communication throughput and increases availability. | |
820 | ||
806ef12d DM |
821 | |
822 | Cluster Cold Start | |
823 | ------------------ | |
824 | ||
825 | It is obvious that a cluster is not quorate when all nodes are | |
826 | offline. This is a common case after a power failure. | |
827 | ||
828 | NOTE: It is always a good idea to use an uninterruptible power supply | |
8c1189b6 | 829 | (``UPS'', also called ``battery backup'') to avoid this state, especially if |
806ef12d DM |
830 | you want HA. |
831 | ||
8c1189b6 FG |
832 | On node startup, service `pve-manager` is started and waits for |
833 | quorum. Once quorate, it starts all guests which have the `onboot` | |
612417fd DM |
834 | flag set. |
835 | ||
836 | When you turn on nodes, or when power comes back after power failure, | |
837 | it is likely that some nodes boots faster than others. Please keep in | |
838 | mind that guest startup is delayed until you reach quorum. | |
806ef12d DM |
839 | |
840 | ||
d8742b0c DM |
841 | ifdef::manvolnum[] |
842 | include::pve-copyright.adoc[] | |
843 | endif::manvolnum[] |