]>
Commit | Line | Data |
---|---|---|
80c0adcb | 1 | [[chapter_ha_manager]] |
22653ac8 | 2 | ifdef::manvolnum[] |
b2f242ab DM |
3 | ha-manager(1) |
4 | ============= | |
5f09af76 DM |
5 | :pve-toplevel: |
6 | ||
22653ac8 DM |
7 | NAME |
8 | ---- | |
9 | ||
734404b4 | 10 | ha-manager - Proxmox VE HA Manager |
22653ac8 | 11 | |
49a5e11c | 12 | SYNOPSIS |
22653ac8 DM |
13 | -------- |
14 | ||
15 | include::ha-manager.1-synopsis.adoc[] | |
16 | ||
17 | DESCRIPTION | |
18 | ----------- | |
19 | endif::manvolnum[] | |
22653ac8 DM |
20 | ifndef::manvolnum[] |
21 | High Availability | |
22 | ================= | |
5f09af76 | 23 | :pve-toplevel: |
194d2f29 | 24 | endif::manvolnum[] |
b5266e9f DM |
25 | |
26 | Our modern society depends heavily on information provided by | |
27 | computers over the network. Mobile devices amplified that dependency, | |
28 | because people can access the network any time from anywhere. If you | |
29 | provide such services, it is very important that they are available | |
30 | most of the time. | |
31 | ||
049fc557 | 32 | We can mathematically define the availability as the ratio of (A), the |
b5266e9f | 33 | total time a service is capable of being used during a given interval |
049fc557 | 34 | to (B), the length of the interval. It is normally expressed as a |
b5266e9f DM |
35 | percentage of uptime in a given year. |
36 | ||
37 | .Availability - Downtime per Year | |
38 | [width="60%",cols="<d,d",options="header"] | |
39 | |=========================================================== | |
40 | |Availability % |Downtime per year | |
41 | |99 |3.65 days | |
42 | |99.9 |8.76 hours | |
43 | |99.99 |52.56 minutes | |
44 | |99.999 |5.26 minutes | |
45 | |99.9999 |31.5 seconds | |
46 | |99.99999 |3.15 seconds | |
47 | |=========================================================== | |
48 | ||
04bde502 DM |
49 | There are several ways to increase availability. The most elegant |
50 | solution is to rewrite your software, so that you can run it on | |
049fc557 DW |
51 | several hosts at the same time. The software itself needs to have a way |
52 | to detect errors and do failover. If you only want to serve read-only | |
53 | web pages, then this is relatively simple. However, this is generally complex | |
54 | and sometimes impossible, because you cannot modify the software yourself. The | |
55 | following solutions works without modifying the software: | |
04bde502 | 56 | |
8c1189b6 | 57 | * Use reliable ``server'' components |
fd9e8984 | 58 | + |
049fc557 | 59 | NOTE: Computer components with the same functionality can have varying |
2af6af05 | 60 | reliability numbers, depending on the component quality. Most vendors |
8c1189b6 | 61 | sell components with higher reliability as ``server'' components - |
04bde502 | 62 | usually at higher price. |
b5266e9f DM |
63 | |
64 | * Eliminate single point of failure (redundant components) | |
8c1189b6 FG |
65 | ** use an uninterruptible power supply (UPS) |
66 | ** use redundant power supplies on the main boards | |
67 | ** use ECC-RAM | |
68 | ** use redundant network hardware | |
69 | ** use RAID for local storage | |
70 | ** use distributed, redundant storage for VM data | |
b5266e9f DM |
71 | |
72 | * Reduce downtime | |
8c1189b6 FG |
73 | ** rapidly accessible administrators (24/7) |
74 | ** availability of spare parts (other nodes in a {pve} cluster) | |
75 | ** automatic error detection (provided by `ha-manager`) | |
76 | ** automatic failover (provided by `ha-manager`) | |
b5266e9f | 77 | |
5771d9b0 | 78 | Virtualization environments like {pve} make it much easier to reach |
8c1189b6 | 79 | high availability because they remove the ``hardware'' dependency. They |
049fc557 DW |
80 | also support the setup and use of redundant storage and network |
81 | devices, so if one host fails, you can simply start those services on | |
43da8322 DM |
82 | another host within your cluster. |
83 | ||
049fc557 | 84 | Better still, {pve} provides a software stack called `ha-manager`, |
43da8322 DM |
85 | which can do that automatically for you. It is able to automatically |
86 | detect errors and do automatic failover. | |
87 | ||
8c1189b6 | 88 | {pve} `ha-manager` works like an ``automated'' administrator. First, you |
43da8322 | 89 | configure what resources (VMs, containers, ...) it should |
049fc557 | 90 | manage. Then, `ha-manager` observes the correct functionality, and handles |
8c1189b6 | 91 | service failover to another node in case of errors. `ha-manager` can |
43da8322 DM |
92 | also handle normal user requests which may start, stop, relocate and |
93 | migrate a service. | |
04bde502 DM |
94 | |
95 | But high availability comes at a price. High quality components are | |
049fc557 | 96 | more expensive, and making them redundant doubles the costs at |
04bde502 DM |
97 | least. Additional spare parts increase costs further. So you should |
98 | carefully calculate the benefits, and compare with those additional | |
99 | costs. | |
100 | ||
101 | TIP: Increasing availability from 99% to 99.9% is relatively | |
d5c3a54a | 102 | simple. But increasing availability from 99.9999% to 99.99999% is very |
8c1189b6 | 103 | hard and costly. `ha-manager` has typical error detection and failover |
43da8322 DM |
104 | times of about 2 minutes, so you can get no more than 99.999% |
105 | availability. | |
b5266e9f | 106 | |
823fa863 | 107 | |
5bd515d4 DM |
108 | Requirements |
109 | ------------ | |
3810ae1e | 110 | |
823fa863 DM |
111 | You must meet the following requirements before you start with HA: |
112 | ||
5bd515d4 | 113 | * at least three cluster nodes (to get reliable quorum) |
43da8322 | 114 | |
5bd515d4 | 115 | * shared storage for VMs and containers |
43da8322 | 116 | |
5bd515d4 | 117 | * hardware redundancy (everywhere) |
3810ae1e | 118 | |
823fa863 DM |
119 | * use reliable “server” components |
120 | ||
5bd515d4 | 121 | * hardware watchdog - if not available we fall back to the |
8c1189b6 | 122 | linux kernel software watchdog (`softdog`) |
3810ae1e | 123 | |
5bd515d4 | 124 | * optional hardware fencing devices |
3810ae1e | 125 | |
3810ae1e | 126 | |
80c0adcb | 127 | [[ha_manager_resources]] |
5bd515d4 DM |
128 | Resources |
129 | --------- | |
130 | ||
8c1189b6 FG |
131 | We call the primary management unit handled by `ha-manager` a |
132 | resource. A resource (also called ``service'') is uniquely | |
5bd515d4 | 133 | identified by a service ID (SID), which consists of the resource type |
049fc557 | 134 | and a type specific ID, for example `vm:100`. That example would be a |
8c1189b6 | 135 | resource of type `vm` (virtual machine) with the ID 100. |
5bd515d4 DM |
136 | |
137 | For now we have two important resources types - virtual machines and | |
138 | containers. One basic idea here is that we can bundle related software | |
a35aad4a | 139 | into such a VM or container, so there is no need to compose one big |
049fc557 | 140 | service from other services, as was done with `rgmanager`. In |
4c34defd | 141 | general, a HA managed resource should not depend on other resources. |
3810ae1e | 142 | |
22653ac8 | 143 | |
d4642672 DM |
144 | Management Tasks |
145 | ---------------- | |
146 | ||
147 | This section provides a short overview of common management tasks. The | |
148 | first step is to enable HA for a resource. This is done by adding the | |
149 | resource to the HA resource configuration. You can do this using the | |
150 | GUI, or simply use the command line tool, for example: | |
151 | ||
152 | ---- | |
153 | # ha-manager add vm:100 | |
154 | ---- | |
155 | ||
049fc557 | 156 | The HA stack now tries to start the resources and keep them |
d4642672 | 157 | running. Please note that you can configure the ``requested'' |
a35aad4a | 158 | resources state. For example you may want the HA stack to stop the |
d4642672 DM |
159 | resource: |
160 | ||
161 | ---- | |
162 | # ha-manager set vm:100 --state stopped | |
163 | ---- | |
164 | ||
165 | and start it again later: | |
166 | ||
167 | ---- | |
168 | # ha-manager set vm:100 --state started | |
169 | ---- | |
170 | ||
171 | You can also use the normal VM and container management commands. They | |
172 | automatically forward the commands to the HA stack, so | |
173 | ||
174 | ---- | |
175 | # qm start 100 | |
176 | ---- | |
177 | ||
049fc557 | 178 | simply sets the requested state to `started`. The same applies to `qm |
d4642672 DM |
179 | stop`, which sets the requested state to `stopped`. |
180 | ||
181 | NOTE: The HA stack works fully asynchronous and needs to communicate | |
049fc557 | 182 | with other cluster members. Therefore, it takes some seconds until you see |
d4642672 DM |
183 | the result of such actions. |
184 | ||
185 | To view the current HA resource configuration use: | |
186 | ||
187 | ---- | |
188 | # ha-manager config | |
189 | vm:100 | |
190 | state stopped | |
191 | ---- | |
192 | ||
193 | And you can view the actual HA manager and resource state with: | |
194 | ||
195 | ---- | |
196 | # ha-manager status | |
197 | quorum OK | |
198 | master node1 (active, Wed Nov 23 11:07:23 2016) | |
199 | lrm elsa (active, Wed Nov 23 11:07:19 2016) | |
200 | service vm:100 (node1, started) | |
201 | ---- | |
202 | ||
203 | You can also initiate resource migration to other nodes: | |
204 | ||
205 | ---- | |
206 | # ha-manager migrate vm:100 node2 | |
207 | ---- | |
208 | ||
209 | This uses online migration and tries to keep the VM running. Online | |
210 | migration needs to transfer all used memory over the network, so it is | |
049fc557 | 211 | sometimes faster to stop the VM, then restart it on the new node. This can be |
d4642672 DM |
212 | done using the `relocate` command: |
213 | ||
214 | ---- | |
215 | # ha-manager relocate vm:100 node2 | |
216 | ---- | |
217 | ||
218 | Finally, you can remove the resource from the HA configuration using | |
219 | the following command: | |
220 | ||
221 | ---- | |
222 | # ha-manager remove vm:100 | |
223 | ---- | |
224 | ||
225 | NOTE: This does not start or stop the resource. | |
226 | ||
a35aad4a | 227 | But all HA related tasks can be done in the GUI, so there is no need to |
d4642672 DM |
228 | use the command line at all. |
229 | ||
230 | ||
2b52e195 | 231 | How It Works |
22653ac8 DM |
232 | ------------ |
233 | ||
c7470421 DM |
234 | This section provides a detailed description of the {PVE} HA manager |
235 | internals. It describes all involved daemons and how they work | |
236 | together. To provide HA, two daemons run on each node: | |
3810ae1e | 237 | |
8c1189b6 | 238 | `pve-ha-lrm`:: |
3810ae1e | 239 | |
1600c60a DM |
240 | The local resource manager (LRM), which controls the services running on |
241 | the local node. It reads the requested states for its services from | |
242 | the current manager status file and executes the respective commands. | |
3810ae1e | 243 | |
8c1189b6 | 244 | `pve-ha-crm`:: |
3810ae1e | 245 | |
60ed554f | 246 | The cluster resource manager (CRM), which makes the cluster-wide |
1600c60a DM |
247 | decisions. It sends commands to the LRM, processes the results, |
248 | and moves resources to other nodes if something fails. The CRM also | |
249 | handles node fencing. | |
250 | ||
3810ae1e TL |
251 | |
252 | .Locks in the LRM & CRM | |
253 | [NOTE] | |
254 | Locks are provided by our distributed configuration file system (pmxcfs). | |
a35aad4a | 255 | They are used to guarantee that each LRM is active once and working. As an |
3821ecaf | 256 | LRM only executes actions when it holds its lock, we can mark a failed node |
049fc557 | 257 | as fenced if we can acquire its lock. This then lets us recover any failed |
5eba0743 | 258 | HA services securely without any interference from the now unknown failed node. |
049fc557 | 259 | This all gets supervised by the CRM which currently holds the manager master |
3810ae1e TL |
260 | lock. |
261 | ||
c7470421 DM |
262 | |
263 | Service States | |
264 | ~~~~~~~~~~~~~~ | |
265 | ||
049fc557 DW |
266 | The CRM uses a service state enumeration to record the current service |
267 | state. This state is displayed on the GUI and can be queried using | |
c7470421 DM |
268 | the `ha-manager` command line tool: |
269 | ||
270 | ---- | |
271 | # ha-manager status | |
272 | quorum OK | |
273 | master elsa (active, Mon Nov 21 07:23:29 2016) | |
274 | lrm elsa (active, Mon Nov 21 07:23:22 2016) | |
275 | service ct:100 (elsa, stopped) | |
276 | service ct:102 (elsa, started) | |
277 | service vm:501 (elsa, started) | |
278 | ---- | |
279 | ||
280 | Here is the list of possible states: | |
281 | ||
282 | stopped:: | |
283 | ||
284 | Service is stopped (confirmed by LRM). If the LRM detects a stopped | |
285 | service is still running, it will stop it again. | |
286 | ||
287 | request_stop:: | |
288 | ||
289 | Service should be stopped. The CRM waits for confirmation from the | |
290 | LRM. | |
291 | ||
1cd01666 DM |
292 | stopping:: |
293 | ||
294 | Pending stop request. But the CRM did not get the request so far. | |
295 | ||
c7470421 DM |
296 | started:: |
297 | ||
298 | Service is active an LRM should start it ASAP if not already running. | |
299 | If the Service fails and is detected to be not running the LRM | |
300 | restarts it | |
301 | (see xref:ha_manager_start_failure_policy[Start Failure Policy]). | |
302 | ||
1cd01666 DM |
303 | starting:: |
304 | ||
305 | Pending start request. But the CRM has not got any confirmation from the | |
306 | LRM that the service is running. | |
307 | ||
c7470421 DM |
308 | fence:: |
309 | ||
fa94c2b3 TL |
310 | Wait for node fencing as the service node is not inside the quorate cluster |
311 | partition (see xref:ha_manager_fencing[Fencing]). | |
312 | As soon as node gets fenced successfully the service will be placed into the | |
313 | recovery state. | |
314 | ||
315 | recovery:: | |
316 | ||
317 | Wait for recovery of the service. The HA manager tries to find a new node where | |
318 | the service can run on. This search depends not only on the list of online and | |
319 | quorate nodes, but also if the service is a group member and how such a group | |
320 | is limited. | |
321 | As soon as a new available node is found, the service will be moved there and | |
322 | initially placed into stopped state. If it's configured to run the new node | |
323 | will do so. | |
c7470421 DM |
324 | |
325 | freeze:: | |
326 | ||
327 | Do not touch the service state. We use this state while we reboot a | |
328 | node, or when we restart the LRM daemon | |
329 | (see xref:ha_manager_package_updates[Package Updates]). | |
330 | ||
581f2240 TL |
331 | ignored:: |
332 | ||
fb29acdd | 333 | Act as if the service were not managed by HA at all. |
fa94c2b3 TL |
334 | Useful, when full control over the service is desired temporarily, without |
335 | removing it from the HA configuration. | |
581f2240 | 336 | |
c7470421 DM |
337 | migrate:: |
338 | ||
339 | Migrate service (live) to other node. | |
340 | ||
341 | error:: | |
342 | ||
343 | Service is disabled because of LRM errors. Needs manual intervention | |
344 | (see xref:ha_manager_error_recovery[Error Recovery]). | |
345 | ||
1cd01666 DM |
346 | queued:: |
347 | ||
348 | Service is newly added, and the CRM has not seen it so far. | |
349 | ||
350 | disabled:: | |
351 | ||
352 | Service is stopped and marked as `disabled` | |
353 | ||
c7470421 | 354 | |
3810ae1e TL |
355 | Local Resource Manager |
356 | ~~~~~~~~~~~~~~~~~~~~~~ | |
357 | ||
8c1189b6 | 358 | The local resource manager (`pve-ha-lrm`) is started as a daemon on |
60ed554f | 359 | boot and waits until the HA cluster is quorate and thus cluster-wide |
3810ae1e TL |
360 | locks are working. |
361 | ||
362 | It can be in three states: | |
363 | ||
b8663359 | 364 | wait for agent lock:: |
e1ea726a FG |
365 | |
366 | The LRM waits for our exclusive lock. This is also used as idle state if no | |
367 | service is configured. | |
368 | ||
b8663359 | 369 | active:: |
e1ea726a FG |
370 | |
371 | The LRM holds its exclusive lock and has services configured. | |
372 | ||
b8663359 | 373 | lost agent lock:: |
e1ea726a FG |
374 | |
375 | The LRM lost its lock, this means a failure happened and quorum was lost. | |
3810ae1e TL |
376 | |
377 | After the LRM gets in the active state it reads the manager status | |
8c1189b6 | 378 | file in `/etc/pve/ha/manager_status` and determines the commands it |
2af6af05 | 379 | has to execute for the services it owns. |
a35aad4a | 380 | For each command a worker gets started, these workers are running in |
5eba0743 | 381 | parallel and are limited to at most 4 by default. This default setting |
8c1189b6 | 382 | may be changed through the datacenter configuration key `max_worker`. |
2af6af05 TL |
383 | When finished the worker process gets collected and its result saved for |
384 | the CRM. | |
3810ae1e | 385 | |
5eba0743 | 386 | .Maximum Concurrent Worker Adjustment Tips |
3810ae1e | 387 | [NOTE] |
5eba0743 | 388 | The default value of at most 4 concurrent workers may be unsuited for |
049fc557 | 389 | a specific setup. For example, 4 live migrations may occur at the same |
3810ae1e | 390 | time, which can lead to network congestions with slower networks and/or |
049fc557 DW |
391 | big (memory wise) services. Also, ensure that in the worst case, congestion is |
392 | at a minimum, even if this means lowering the `max_worker` value. On the | |
393 | contrary, if you have a particularly powerful, high-end setup you may also want | |
394 | to increase it. | |
3810ae1e | 395 | |
049fc557 DW |
396 | Each command requested by the CRM is uniquely identifiable by a UID. When |
397 | the worker finishes, its result will be processed and written in the LRM | |
8c1189b6 | 398 | status file `/etc/pve/nodes/<nodename>/lrm_status`. There the CRM may collect |
049fc557 | 399 | it and let its state machine - respective to the commands output - act on it. |
3810ae1e TL |
400 | |
401 | The actions on each service between CRM and LRM are normally always synced. | |
a35aad4a | 402 | This means that the CRM requests a state uniquely marked by a UID, the LRM |
049fc557 | 403 | then executes this action *one time* and writes back the result, which is also |
3810ae1e | 404 | identifiable by the same UID. This is needed so that the LRM does not |
a35aad4a | 405 | execute an outdated command. |
049fc557 DW |
406 | The only exceptions to this behaviour are the `stop` and `error` commands; |
407 | these two do not depend on the result produced and are executed | |
3810ae1e TL |
408 | always in the case of the stopped state and once in the case of |
409 | the error state. | |
410 | ||
411 | .Read the Logs | |
412 | [NOTE] | |
413 | The HA Stack logs every action it makes. This helps to understand what | |
414 | and also why something happens in the cluster. Here its important to see | |
415 | what both daemons, the LRM and the CRM, did. You may use | |
416 | `journalctl -u pve-ha-lrm` on the node(s) where the service is and | |
417 | the same command for the pve-ha-crm on the node which is the current master. | |
418 | ||
419 | Cluster Resource Manager | |
420 | ~~~~~~~~~~~~~~~~~~~~~~~~ | |
22653ac8 | 421 | |
8c1189b6 | 422 | The cluster resource manager (`pve-ha-crm`) starts on each node and |
22653ac8 DM |
423 | waits there for the manager lock, which can only be held by one node |
424 | at a time. The node which successfully acquires the manager lock gets | |
3810ae1e TL |
425 | promoted to the CRM master. |
426 | ||
2af6af05 | 427 | It can be in three states: |
3810ae1e | 428 | |
b8663359 | 429 | wait for agent lock:: |
e1ea726a | 430 | |
97ae300a | 431 | The CRM waits for our exclusive lock. This is also used as idle state if no |
e1ea726a FG |
432 | service is configured |
433 | ||
b8663359 | 434 | active:: |
e1ea726a | 435 | |
97ae300a | 436 | The CRM holds its exclusive lock and has services configured |
e1ea726a | 437 | |
b8663359 | 438 | lost agent lock:: |
e1ea726a | 439 | |
97ae300a | 440 | The CRM lost its lock, this means a failure happened and quorum was lost. |
3810ae1e | 441 | |
a35aad4a | 442 | Its main task is to manage the services which are configured to be highly |
4c34defd TL |
443 | available and try to always enforce the requested state. For example, a |
444 | service with the requested state 'started' will be started if its not | |
445 | already running. If it crashes it will be automatically started again. | |
a35aad4a | 446 | Thus the CRM dictates the actions the LRM needs to execute. |
22653ac8 | 447 | |
049fc557 DW |
448 | When a node leaves the cluster quorum, its state changes to unknown. |
449 | If the current CRM can then secure the failed node's lock, the services | |
22653ac8 DM |
450 | will be 'stolen' and restarted on another node. |
451 | ||
452 | When a cluster member determines that it is no longer in the cluster | |
453 | quorum, the LRM waits for a new quorum to form. As long as there is no | |
454 | quorum the node cannot reset the watchdog. This will trigger a reboot | |
049fc557 | 455 | after the watchdog times out (this happens after 60 seconds). |
22653ac8 | 456 | |
85363588 | 457 | |
b8633a34 RV |
458 | HA Simulator |
459 | ------------ | |
460 | ||
461 | [thumbnail="screenshot/gui-ha-manager-status.png"] | |
462 | ||
463 | By using the HA simulator you can test and learn all functionalities of the | |
464 | Proxmox VE HA solutions. | |
465 | ||
3c5584e9 TL |
466 | By default, the simulator allows you to watch and test the behaviour of a |
467 | real-world 3 node cluster with 6 VMs. You can also add or remove additional VMs | |
468 | or Container. | |
b8633a34 RV |
469 | |
470 | You do not have to setup or configure a real cluster, the HA simulator runs out | |
471 | of the box. | |
472 | ||
473 | Install with apt: | |
474 | ||
475 | ---- | |
476 | apt install pve-ha-simulator | |
477 | ---- | |
478 | ||
049fc557 | 479 | You can even install the package on any Debian-based system without any |
b8633a34 RV |
480 | other Proxmox VE packages. For that you will need to download the package and |
481 | copy it to the system you want to run it on for installation. When you install | |
482 | the package with apt from the local file system it will also resolve the | |
483 | required dependencies for you. | |
484 | ||
485 | ||
049fc557 | 486 | To start the simulator on a remote machine you must have an X11 redirection to |
b8633a34 RV |
487 | your current system. |
488 | ||
489 | If you are on a Linux machine you can use: | |
490 | ||
491 | ---- | |
3c5584e9 | 492 | ssh root@<IPofPVE> -Y |
b8633a34 RV |
493 | ---- |
494 | ||
049fc557 | 495 | On Windows it works with https://mobaxterm.mobatek.net/[mobaxterm]. |
b8633a34 | 496 | |
049fc557 DW |
497 | After connecting to an existing {pve} with the simulator installed or |
498 | installing it on your local Debian-based system manually, you can try it out as | |
3c5584e9 TL |
499 | follows. |
500 | ||
049fc557 DW |
501 | First you need to create a working directory where the simulator saves its |
502 | current state and writes its default config: | |
b8633a34 RV |
503 | |
504 | ---- | |
505 | mkdir working | |
506 | ---- | |
507 | ||
049fc557 | 508 | Then, simply pass the created directory as a parameter to 'pve-ha-simulator': |
b8633a34 RV |
509 | |
510 | ---- | |
511 | pve-ha-simulator working/ | |
512 | ---- | |
513 | ||
3c5584e9 TL |
514 | You can then start, stop, migrate the simulated HA services, or even check out |
515 | what happens on a node failure. | |
b8633a34 | 516 | |
2b52e195 | 517 | Configuration |
22653ac8 DM |
518 | ------------- |
519 | ||
85363588 DM |
520 | The HA stack is well integrated into the {pve} API. So, for example, |
521 | HA can be configured via the `ha-manager` command line interface, or | |
522 | the {pve} web interface - both interfaces provide an easy way to | |
523 | manage HA. Automation tools can use the API directly. | |
524 | ||
525 | All HA configuration files are within `/etc/pve/ha/`, so they get | |
526 | automatically distributed to the cluster nodes, and all nodes share | |
527 | the same HA configuration. | |
528 | ||
206c2476 | 529 | |
4c34defd | 530 | [[ha_manager_resource_config]] |
206c2476 DM |
531 | Resources |
532 | ~~~~~~~~~ | |
533 | ||
1ff5e4e8 | 534 | [thumbnail="screenshot/gui-ha-manager-status.png"] |
863a8f3a | 535 | |
4d63b3cc | 536 | |
85363588 DM |
537 | The resource configuration file `/etc/pve/ha/resources.cfg` stores |
538 | the list of resources managed by `ha-manager`. A resource configuration | |
a35aad4a | 539 | inside that list looks like this: |
85363588 DM |
540 | |
541 | ---- | |
8bdc398c | 542 | <type>: <name> |
85363588 DM |
543 | <property> <value> |
544 | ... | |
545 | ---- | |
546 | ||
698e5dd2 DM |
547 | It starts with a resource type followed by a resource specific name, |
548 | separated with colon. Together this forms the HA resource ID, which is | |
549 | used by all `ha-manager` commands to uniquely identify a resource | |
a9c77fec DM |
550 | (example: `vm:100` or `ct:101`). The next lines contain additional |
551 | properties: | |
85363588 DM |
552 | |
553 | include::ha-resources-opts.adoc[] | |
554 | ||
8bdc398c | 555 | Here is a real world example with one VM and one container. As you see, |
470d4313 | 556 | the syntax of those files is really simple, so it is even possible to |
8bdc398c DM |
557 | read or edit those files using your favorite editor: |
558 | ||
e7b9b0ac | 559 | .Configuration Example (`/etc/pve/ha/resources.cfg`) |
8bdc398c DM |
560 | ---- |
561 | vm: 501 | |
562 | state started | |
563 | max_relocate 2 | |
564 | ||
565 | ct: 102 | |
a319e18b DM |
566 | # Note: use default settings for everything |
567 | ---- | |
568 | ||
1ff5e4e8 | 569 | [thumbnail="screenshot/gui-ha-manager-add-resource.png"] |
4d63b3cc | 570 | |
049fc557 | 571 | The above config was generated using the `ha-manager` command line tool: |
a319e18b DM |
572 | |
573 | ---- | |
574 | # ha-manager add vm:501 --state started --max_relocate 2 | |
575 | # ha-manager add ct:102 | |
8bdc398c DM |
576 | ---- |
577 | ||
85363588 | 578 | |
1acab952 | 579 | [[ha_manager_groups]] |
206c2476 DM |
580 | Groups |
581 | ~~~~~~ | |
582 | ||
1ff5e4e8 | 583 | [thumbnail="screenshot/gui-ha-manager-groups-view.png"] |
4d63b3cc | 584 | |
85363588 DM |
585 | The HA group configuration file `/etc/pve/ha/groups.cfg` is used to |
586 | define groups of cluster nodes. A resource can be restricted to run | |
206c2476 DM |
587 | only on the members of such group. A group configuration look like |
588 | this: | |
85363588 | 589 | |
206c2476 DM |
590 | ---- |
591 | group: <group> | |
592 | nodes <node_list> | |
593 | <property> <value> | |
594 | ... | |
595 | ---- | |
85363588 | 596 | |
206c2476 | 597 | include::ha-groups-opts.adoc[] |
22653ac8 | 598 | |
1ff5e4e8 | 599 | [thumbnail="screenshot/gui-ha-manager-add-group.png"] |
4d63b3cc | 600 | |
e60ce90c | 601 | A common requirement is that a resource should run on a specific |
1acab952 DM |
602 | node. Usually the resource is able to run on other nodes, so you can define |
603 | an unrestricted group with a single member: | |
604 | ||
605 | ---- | |
606 | # ha-manager groupadd prefer_node1 --nodes node1 | |
607 | ---- | |
608 | ||
609 | For bigger clusters, it makes sense to define a more detailed failover | |
610 | behavior. For example, you may want to run a set of services on | |
611 | `node1` if possible. If `node1` is not available, you want to run them | |
049fc557 | 612 | equally split on `node2` and `node3`. If those nodes also fail, the |
1acab952 DM |
613 | services should run on `node4`. To achieve this you could set the node |
614 | list to: | |
615 | ||
616 | ---- | |
617 | # ha-manager groupadd mygroup1 -nodes "node1:2,node2:1,node3:1,node4" | |
618 | ---- | |
619 | ||
620 | Another use case is if a resource uses other resources only available | |
621 | on specific nodes, lets say `node1` and `node2`. We need to make sure | |
622 | that HA manager does not use other nodes, so we need to create a | |
623 | restricted group with said nodes: | |
624 | ||
625 | ---- | |
626 | # ha-manager groupadd mygroup2 -nodes "node1,node2" -restricted | |
627 | ---- | |
628 | ||
049fc557 | 629 | The above commands created the following group configuration file: |
1acab952 DM |
630 | |
631 | .Configuration Example (`/etc/pve/ha/groups.cfg`) | |
632 | ---- | |
633 | group: prefer_node1 | |
634 | nodes node1 | |
635 | ||
636 | group: mygroup1 | |
637 | nodes node2:1,node4,node1:2,node3:1 | |
638 | ||
639 | group: mygroup2 | |
640 | nodes node2,node1 | |
641 | restricted 1 | |
642 | ---- | |
643 | ||
644 | ||
645 | The `nofailback` options is mostly useful to avoid unwanted resource | |
e60ce90c | 646 | movements during administration tasks. For example, if you need to |
049fc557 DW |
647 | migrate a service to a node which doesn't have the highest priority in the |
648 | group, you need to tell the HA manager not to instantly move this service | |
649 | back by setting the `nofailback` option. | |
1acab952 DM |
650 | |
651 | Another scenario is when a service was fenced and it got recovered to | |
652 | another node. The admin tries to repair the fenced node and brings it | |
049fc557 DW |
653 | up online again to investigate the cause of failure and check if it runs |
654 | stably again. Setting the `nofailback` flag prevents the recovered services from | |
655 | moving straight back to the fenced node. | |
1acab952 | 656 | |
22653ac8 | 657 | |
80c0adcb | 658 | [[ha_manager_fencing]] |
3810ae1e TL |
659 | Fencing |
660 | ------- | |
661 | ||
0d427077 DM |
662 | On node failures, fencing ensures that the erroneous node is |
663 | guaranteed to be offline. This is required to make sure that no | |
664 | resource runs twice when it gets recovered on another node. This is a | |
049fc557 | 665 | really important task, because without this, it would not be possible to |
0d427077 DM |
666 | recover a resource on another node. |
667 | ||
bdfd4601 | 668 | If a node did not get fenced, it would be in an unknown state where |
0d427077 | 669 | it may have still access to shared resources. This is really |
049fc557 | 670 | dangerous! Imagine that every network but the storage one broke. Now, |
0d427077 DM |
671 | while not reachable from the public network, the VM still runs and |
672 | writes to the shared storage. | |
673 | ||
674 | If we then simply start up this VM on another node, we would get a | |
049fc557 DW |
675 | dangerous race condition, because we write from both nodes. Such |
676 | conditions can destroy all VM data and the whole VM could be rendered | |
677 | unusable. The recovery could also fail if the storage protects against | |
0d427077 DM |
678 | multiple mounts. |
679 | ||
5771d9b0 TL |
680 | |
681 | How {pve} Fences | |
0d427077 | 682 | ~~~~~~~~~~~~~~~~ |
5771d9b0 | 683 | |
61972f55 DM |
684 | There are different methods to fence a node, for example, fence |
685 | devices which cut off the power from the node or disable their | |
686 | communication completely. Those are often quite expensive and bring | |
687 | additional critical components into a system, because if they fail you | |
688 | cannot recover any service. | |
689 | ||
690 | We thus wanted to integrate a simpler fencing method, which does not | |
691 | require additional external hardware. This can be done using | |
692 | watchdog timers. | |
693 | ||
694 | .Possible Fencing Methods | |
695 | - external power switches | |
696 | - isolate nodes by disabling complete network traffic on the switch | |
697 | - self fencing using watchdog timers | |
698 | ||
049fc557 DW |
699 | Watchdog timers have been widely used in critical and dependable systems |
700 | since the beginning of microcontrollers. They are often simple, independent | |
701 | integrated circuits which are used to detect and recover from computer malfunctions. | |
61972f55 DM |
702 | |
703 | During normal operation, `ha-manager` regularly resets the watchdog | |
704 | timer to prevent it from elapsing. If, due to a hardware fault or | |
705 | program error, the computer fails to reset the watchdog, the timer | |
049fc557 | 706 | will elapse and trigger a reset of the whole server (reboot). |
61972f55 DM |
707 | |
708 | Recent server motherboards often include such hardware watchdogs, but | |
709 | these need to be configured. If no watchdog is available or | |
710 | configured, we fall back to the Linux Kernel 'softdog'. While still | |
711 | reliable, it is not independent of the servers hardware, and thus has | |
712 | a lower reliability than a hardware watchdog. | |
3810ae1e | 713 | |
a472fde8 | 714 | |
3810ae1e TL |
715 | Configure Hardware Watchdog |
716 | ~~~~~~~~~~~~~~~~~~~~~~~~~~~ | |
a472fde8 DM |
717 | |
718 | By default, all hardware watchdog modules are blocked for security | |
719 | reasons. They are like a loaded gun if not correctly initialized. To | |
720 | enable a hardware watchdog, you need to specify the module to load in | |
721 | '/etc/default/pve-ha-manager', for example: | |
722 | ||
723 | ---- | |
724 | # select watchdog module (default is softdog) | |
725 | WATCHDOG_MODULE=iTCO_wdt | |
726 | ---- | |
727 | ||
049fc557 | 728 | This configuration is read by the 'watchdog-mux' service, which loads |
a472fde8 DM |
729 | the specified module at startup. |
730 | ||
3810ae1e | 731 | |
2957ef80 TL |
732 | Recover Fenced Services |
733 | ~~~~~~~~~~~~~~~~~~~~~~~ | |
734 | ||
480e67e1 DM |
735 | After a node failed and its fencing was successful, the CRM tries to |
736 | move services from the failed node to nodes which are still online. | |
737 | ||
738 | The selection of nodes, on which those services gets recovered, is | |
739 | influenced by the resource `group` settings, the list of currently active | |
740 | nodes, and their respective active service count. | |
741 | ||
742 | The CRM first builds a set out of the intersection between user selected | |
743 | nodes (from `group` setting) and available nodes. It then choose the | |
744 | subset of nodes with the highest priority, and finally select the node | |
745 | with the lowest active service count. This minimizes the possibility | |
746 | of an overloaded node. | |
747 | ||
748 | CAUTION: On node failure, the CRM distributes services to the | |
049fc557 | 749 | remaining nodes. This increases the service count on those nodes, and |
480e67e1 DM |
750 | can lead to high load, especially on small clusters. Please design |
751 | your cluster so that it can handle such worst case scenarios. | |
2957ef80 | 752 | |
22653ac8 | 753 | |
c7470421 | 754 | [[ha_manager_start_failure_policy]] |
a3189ad1 TL |
755 | Start Failure Policy |
756 | --------------------- | |
757 | ||
049fc557 | 758 | The start failure policy comes into effect if a service failed to start on a |
a35aad4a | 759 | node one or more times. It can be used to configure how often a restart |
a3189ad1 | 760 | should be triggered on the same node and how often a service should be |
049fc557 | 761 | relocated, so that it has an attempt to be started on another node. |
a3189ad1 TL |
762 | The aim of this policy is to circumvent temporary unavailability of shared |
763 | resources on a specific node. For example, if a shared storage isn't available | |
049fc557 DW |
764 | on a quorate node anymore, for instance due to network problems, but is still |
765 | available on other nodes, the relocate policy allows the service to start | |
766 | nonetheless. | |
a3189ad1 TL |
767 | |
768 | There are two service start recover policy settings which can be configured | |
22653ac8 DM |
769 | specific for each resource. |
770 | ||
771 | max_restart:: | |
772 | ||
049fc557 | 773 | Maximum number of attempts to restart a failed service on the actual |
22653ac8 DM |
774 | node. The default is set to one. |
775 | ||
776 | max_relocate:: | |
777 | ||
049fc557 | 778 | Maximum number of attempts to relocate the service to a different node. |
22653ac8 DM |
779 | A relocate only happens after the max_restart value is exceeded on the |
780 | actual node. The default is set to one. | |
781 | ||
0abc65b0 | 782 | NOTE: The relocate count state will only reset to zero when the |
22653ac8 | 783 | service had at least one successful start. That means if a service is |
4c34defd | 784 | re-started without fixing the error only the restart policy gets |
22653ac8 DM |
785 | repeated. |
786 | ||
c7470421 DM |
787 | |
788 | [[ha_manager_error_recovery]] | |
2b52e195 | 789 | Error Recovery |
22653ac8 DM |
790 | -------------- |
791 | ||
049fc557 DW |
792 | If, after all attempts, the service state could not be recovered, it gets |
793 | placed in an error state. In this state, the service won't get touched | |
c5bca1ae | 794 | by the HA stack anymore. The only way out is disabling a service: |
d02982f7 | 795 | |
c5bca1ae TL |
796 | ---- |
797 | # ha-manager set vm:100 --state disabled | |
798 | ---- | |
d02982f7 | 799 | |
c5bca1ae TL |
800 | This can also be done in the web interface. |
801 | ||
802 | To recover from the error state you should do the following: | |
22653ac8 | 803 | |
c5bca1ae TL |
804 | * bring the resource back into a safe and consistent state (e.g.: |
805 | kill its process if the service could not be stopped) | |
22653ac8 | 806 | |
c5bca1ae | 807 | * disable the resource to remove the error flag |
22653ac8 DM |
808 | |
809 | * fix the error which led to this failures | |
810 | ||
4c34defd | 811 | * *after* you fixed all errors you may request that the service starts again |
22653ac8 DM |
812 | |
813 | ||
26513dae DM |
814 | [[ha_manager_package_updates]] |
815 | Package Updates | |
816 | --------------- | |
817 | ||
049fc557 | 818 | When updating the ha-manager, you should do one node after the other, never |
26513dae | 819 | all at once for various reasons. First, while we test our software |
049fc557 | 820 | thoroughly, a bug affecting your specific setup cannot totally be ruled out. |
d5c3a54a FE |
821 | Updating one node after the other and checking the functionality of each node |
822 | after finishing the update helps to recover from eventual problems, while | |
823 | updating all at once could result in a broken cluster and is generally not | |
26513dae DM |
824 | good practice. |
825 | ||
826 | Also, the {pve} HA stack uses a request acknowledge protocol to perform | |
827 | actions between the cluster and the local resource manager. For restarting, | |
828 | the LRM makes a request to the CRM to freeze all its services. This prevents | |
049fc557 DW |
829 | them from getting touched by the Cluster during the short time the LRM is restarting. |
830 | After that, the LRM may safely close the watchdog during a restart. | |
7dd7a0b7 TL |
831 | Such a restart happens normally during a package update and, as already stated, |
832 | an active master CRM is needed to acknowledge the requests from the LRM. If | |
fb29acdd FG |
833 | this is not the case the update process can take too long which, in the worst |
834 | case, may result in a reset triggered by the watchdog. | |
26513dae DM |
835 | |
836 | ||
a9023144 DM |
837 | Node Maintenance |
838 | ---------------- | |
52a75187 | 839 | |
049fc557 DW |
840 | It is sometimes necessary to shutdown or reboot a node to do maintenance tasks, |
841 | such as to replace hardware, or simply to install a new kernel image. This is | |
842 | also true when using the HA stack. The behaviour of the HA stack during a | |
843 | shutdown can be configured. | |
a9023144 | 844 | |
a4a67cdb TL |
845 | [[ha_manager_shutdown_policy]] |
846 | Shutdown Policy | |
847 | ~~~~~~~~~~~~~~~ | |
a9023144 | 848 | |
a4a67cdb TL |
849 | Below you will find a description of the different HA policies for a node |
850 | shutdown. Currently 'Conditional' is the default due to backward compatibility. | |
049fc557 | 851 | Some users may find that 'Migrate' behaves more as expected. |
a9023144 | 852 | |
a4a67cdb TL |
853 | Migrate |
854 | ^^^^^^^ | |
a9023144 | 855 | |
a4a67cdb | 856 | Once the Local Resource manager (LRM) gets a shutdown request and this policy |
049fc557 | 857 | is enabled, it will mark itself as unavailable for the current HA manager. |
a4a67cdb | 858 | This triggers a migration of all HA Services currently located on this node. |
049fc557 DW |
859 | The LRM will try to delay the shutdown process, until all running services get |
860 | moved away. But, this expects that the running services *can* be migrated to | |
861 | another node. In other words, the service must not be locally bound, for example | |
862 | by using hardware passthrough. As non-group member nodes are considered as | |
863 | runnable target if no group member is available, this policy can still be used | |
864 | when making use of HA groups with only some nodes selected. But, marking a group | |
865 | as 'restricted' tells the HA manager that the service cannot run outside of the | |
866 | chosen set of nodes. If all of those nodes are unavailable, the shutdown will | |
867 | hang until you manually intervene. Once the shut down node comes back online | |
868 | again, the previously displaced services will be moved back, if they were not | |
869 | already manually migrated in-between. | |
a9023144 | 870 | |
a4a67cdb TL |
871 | NOTE: The watchdog is still active during the migration process on shutdown. |
872 | If the node loses quorum it will be fenced and the services will be recovered. | |
a9023144 | 873 | |
e9833be4 TL |
874 | If you start a (previously stopped) service on a node which is currently being |
875 | maintained, the node needs to be fenced to ensure that the service can be moved | |
049fc557 | 876 | and started on another available node. |
e9833be4 | 877 | |
a4a67cdb TL |
878 | Failover |
879 | ^^^^^^^^ | |
880 | ||
881 | This mode ensures that all services get stopped, but that they will also be | |
882 | recovered, if the current node is not online soon. It can be useful when doing | |
049fc557 DW |
883 | maintenance on a cluster scale, where live-migrating VMs may not be possible if |
884 | too many nodes are powered off at a time, but you still want to ensure HA | |
a4a67cdb TL |
885 | services get recovered and started again as soon as possible. |
886 | ||
887 | Freeze | |
888 | ^^^^^^ | |
889 | ||
890 | This mode ensures that all services get stopped and frozen, so that they won't | |
891 | get recovered until the current node is online again. | |
892 | ||
893 | Conditional | |
894 | ^^^^^^^^^^^ | |
895 | ||
3dc611ff TL |
896 | The 'Conditional' shutdown policy automatically detects if a shutdown or a |
897 | reboot is requested, and changes behaviour accordingly. | |
898 | ||
a4a67cdb TL |
899 | .Shutdown |
900 | ||
049fc557 DW |
901 | A shutdown ('poweroff') is usually done if it is planned for the node to stay |
902 | down for some time. The LRM stops all managed services in this case. This means | |
903 | that other nodes will take over those services afterwards. | |
a4a67cdb TL |
904 | |
905 | NOTE: Recent hardware has large amounts of memory (RAM). So we stop all | |
906 | resources, then restart them to avoid online migration of all that RAM. If you | |
907 | want to use online migration, you need to invoke that manually before you | |
908 | shutdown the node. | |
909 | ||
910 | ||
911 | .Reboot | |
a9023144 | 912 | |
a4a67cdb TL |
913 | Node reboots are initiated with the 'reboot' command. This is usually done |
914 | after installing a new kernel. Please note that this is different from | |
915 | ``shutdown'', because the node immediately starts again. | |
a9023144 | 916 | |
a4a67cdb TL |
917 | The LRM tells the CRM that it wants to restart, and waits until the CRM puts |
918 | all resources into the `freeze` state (same mechanism is used for | |
049fc557 DW |
919 | xref:ha_manager_package_updates[Package Updates]). This prevents those resources |
920 | from being moved to other nodes. Instead, the CRM starts the resources after the | |
921 | reboot on the same node. | |
a9023144 DM |
922 | |
923 | ||
924 | Manual Resource Movement | |
3dc611ff | 925 | ^^^^^^^^^^^^^^^^^^^^^^^^ |
a9023144 | 926 | |
049fc557 | 927 | Last but not least, you can also manually move resources to other nodes, before |
a4a67cdb TL |
928 | you shutdown or restart a node. The advantage is that you have full control, |
929 | and you can decide if you want to use online migration or not. | |
a9023144 DM |
930 | |
931 | NOTE: Please do not 'kill' services like `pve-ha-crm`, `pve-ha-lrm` or | |
049fc557 | 932 | `watchdog-mux`. They manage and use the watchdog, so this can result in an |
a4a67cdb | 933 | immediate node reboot or even reset. |
52a75187 DM |
934 | |
935 | ||
221a0d1a | 936 | [[ha_manager_crs]] |
57aa03c7 TL |
937 | Cluster Resource Scheduling |
938 | --------------------------- | |
7210615d FE |
939 | |
940 | The scheduler mode controls how HA selects nodes for the recovery of a service | |
941 | as well as for migrations that are triggered by a shutdown policy. The default | |
942 | mode is `basic`, you can change it in `datacenter.cfg`: | |
943 | ||
944 | ---- | |
945 | crs: ha=static | |
946 | ---- | |
947 | ||
0031ebd8 FE |
948 | The change will be in effect starting with the next manager round (after a few |
949 | seconds). | |
7210615d FE |
950 | |
951 | For each service that needs to be recovered or migrated, the scheduler | |
952 | iteratively chooses the best node among the nodes with the highest priority in | |
953 | the service's group. | |
954 | ||
955 | NOTE: There are plans to add modes for (static and dynamic) load-balancing in | |
956 | the future. | |
957 | ||
958 | Basic | |
336ed7a0 | 959 | ~~~~~ |
7210615d | 960 | |
0940b99b | 961 | The number of active HA services on each node is used to choose a recovery node. |
d68bea00 | 962 | Non-HA-managed services are currently not counted. |
7210615d FE |
963 | |
964 | Static | |
336ed7a0 | 965 | ~~~~~~ |
7210615d | 966 | |
57aa03c7 | 967 | IMPORTANT: The static mode is still a technology preview. |
6d085cd5 | 968 | |
0940b99b | 969 | Static usage information from HA services on each node is used to choose a |
d68bea00 | 970 | recovery node. Usage of non-HA-managed services is currently not considered. |
7210615d FE |
971 | |
972 | For this selection, each node in turn is considered as if the service was | |
973 | already running on it, using CPU and memory usage from the associated guest | |
974 | configuration. Then for each such alternative, CPU and memory usage of all nodes | |
975 | are considered, with memory being weighted much more, because it's a truly | |
976 | limited resource. For both, CPU and memory, highest usage among nodes (weighted | |
977 | more, as ideally no node should be overcommitted) and average usage of all nodes | |
978 | (to still be able to distinguish in case there already is a more highly | |
979 | committed node) are considered. | |
980 | ||
57aa03c7 TL |
981 | IMPORTANT: The more services the more possible combinations there are, so it's |
982 | currently not recommended to use it if you have thousands of HA managed | |
983 | services. | |
984 | ||
985 | ||
22653ac8 DM |
986 | ifdef::manvolnum[] |
987 | include::pve-copyright.adoc[] | |
988 | endif::manvolnum[] | |
989 |