Commit Graph

58 Commits

Author SHA1 Message Date
Joshua Boniface a5763c9d25 Fix possible race condition applying schemas
Found an instance where two of these fired too close together, and
caused a fatal error. Use a write lock, and then catch the schema.apply
function in case it fails anyways.
2024-01-11 10:21:01 -05:00
Joshua Boniface 123c7ce857 Update copyright header on all files for 2024
Last release of 2023 is probably the best time to do this.
2023-12-29 11:16:59 -05:00
Joshua Boniface e654fbba08 Move debug condition handling to Logger
Avoids many dozens of conditionals sprinkled throughout the code by
centralizing this check into the main Logger instance.
2023-12-27 13:01:45 -05:00
Joshua Boniface 3e4cc53fdd Add node network statistics and utilization values
Adds a new physical network interface stats parser to the node
keepalives, and leverages this information to provide a network
utilization overview in the Prometheus metrics.
2023-12-21 15:45:01 -05:00
Joshua Boniface 0f24184b78 Explicitly clear resources of fenced node
This actually solves the bug originally "fixed" in
5f1432ccdd without breaking VM resource
allocations for working nodes.
2023-12-11 12:14:56 -05:00
Joshua Boniface 1ba37fe33d Restore VM resource allocation location
Commit 5f1432ccdd changed where these
happen due to a bug after fencing. However this completely broke node
resource reporting as only the final instance will be queried here.

Revert this change and look further into the original bug.
2023-12-11 11:52:59 -05:00
Joshua Boniface 1a05077b10 Fix missing fstring 2023-12-11 11:29:49 -05:00
Joshua Boniface 7bc0760b78 Add time to "starting keepalive" message
Matches the pvchealthd output and provides a useful message detail to
this otherwise contextless message.
2023-12-10 00:40:32 -05:00
Joshua Boniface 1fb0463dea Adjust daemon service startup
Add healthd, adjust workerd, lower waittime
2023-11-30 03:28:02 -05:00
Joshua Boniface 03a738f878 Move config parser into daemon_lib
And reformat/add config values for API.
2023-11-30 00:05:37 -05:00
Joshua Boniface 4a2eba0961 Improve node output messages (from pvchealthd)
1. Output startup "list" entries in cyan with s state
2. Add start of keepalive run message
2023-11-29 21:21:51 -05:00
Joshua Boniface 83ceb41138 Add daemon name to Logger entries 2023-11-29 15:18:37 -05:00
Joshua Boniface 2545a7b744 Allow similar for IPMI hostnames 2023-11-28 16:09:01 -05:00
Joshua Boniface ce907ff26a Allow specifying static IPs instead of a file 2023-11-28 15:28:31 -05:00
Joshua Boniface fc3d292081 Add missing subdirectory configs 2023-11-27 13:40:07 -05:00
Joshua Boniface eab1ae873b Ensure upstream_gateway key will exist 2023-11-27 13:37:57 -05:00
Joshua Boniface eaf93cdf96 Readd missing subsystem configurations 2023-11-27 13:33:41 -05:00
Joshua Boniface c8f4cbb39e Fix node entry keys 2023-11-27 13:24:01 -05:00
Joshua Boniface bcc57638a9 Refactor pvcnoded to use new configuration 2023-11-26 15:41:25 -05:00
Joshua Boniface 18e43a9377 Adjust name in worker log output 2023-11-16 02:25:14 -05:00
Joshua Boniface aef38639cf Rename pvcapid-worker to pvcworkerd 2023-11-15 20:31:39 -05:00
Joshua Boniface 5f1432ccdd Fix memory allocation updates and add more debug
Previously, we were assigning memalloc/memprov/vcpualloc during an
earlier phase using the main d_domain list. I'm not sure exactly why,
but this was throwing off stats after a fence. Instead, set these values
later on while parsing the actually-active VMs.
2023-11-10 10:29:32 -05:00
Joshua Boniface d6b8808448 Clean up fencing handler
1. Remove all format strings in favour of f-strings
2. Ensure all logger messages have a prefix
3. Add a few more logger messages for clarity
2023-11-10 10:09:54 -05:00
Joshua Boniface 83c4c6633d Readd RBD lock detection and clearing on startup
This is still needed due to the nature of the locks and freeing them on
startup, and to preserve lock=fail behaviour on VM startup.

Also fixes the fencing lock flush to directly use the client library
outside of Celery. I don't like this hack but it seems prudent until we
move fencing to the workers as well.
2023-11-10 01:33:48 -05:00
Joshua Boniface 2c15036f86 Add KeyDB to node startup services
Also ensure API worker starts on all nodes, not just coordinators.
2023-11-05 19:26:38 -05:00
Joshua Boniface 30d7e49401 Start API worker with node daemon on coordinators 2023-11-04 13:08:16 -04:00
Joshua Boniface 8b93f9a80e Handle OSD index errors during stats collection 2023-11-01 21:33:40 -04:00
Joshua Boniface 0769f1ea52 Increase service start time to 10s 2023-10-23 22:24:03 -04:00
Joshua Boniface 457b7bed3d Handle exceptions in fence migrations 2023-09-16 22:56:09 -04:00
Joshua Boniface 48662e90c1 Remove obsolete monitoring_instance passing 2023-09-15 22:47:45 -04:00
Joshua Boniface 079381c03e Move printing to end and add runtime 2023-09-15 22:40:09 -04:00
Joshua Boniface 4d51318a40 Make monitoring interval configurable 2023-09-15 16:54:51 -04:00
Joshua Boniface 254303b9d4 Use coordinator_state instead of router_state
Makes it much clearer what this variable represents.
2023-09-15 16:47:56 -04:00
Joshua Boniface 40b7d68853 Separate monitoring and move to 60s interval
Removes the dependency of the monitoring subsystem from the node
keepalives, and runs them at a 60s interval to avoid excessive backups
if a plugin takes too long.

Adds its own logs and related items as required.

Finally adds a new required argument to the run() of plugins, the
coordinator state, which can be used by a plugin to determine actions
based on whether the node is a primary, secondary, or non-coordinator.
2023-09-15 16:47:11 -04:00
Joshua Boniface cb413e5ce6 [Bookworm] Fix Ceph 16 OSD stat parsing 2023-08-31 00:45:03 -04:00
Joshua Boniface ed087d83c2 Found cpuload to 2 decimal places 2023-08-29 21:41:44 -04:00
Joshua Boniface 7c07fbefff Adjust keepalive health printing and ordering 2023-02-24 11:08:30 -05:00
Joshua Boniface f4eef30770 Add JSON health to cluster data 2023-02-15 15:26:57 -05:00
Joshua Boniface bc88d764b0 Add logging flag for montioring plugin output 2023-02-13 22:04:39 -05:00
Joshua Boniface 2ee52e44d3 Move Ceph cluster health reporting to plugin
Also removes several outputs from the normal keepalive that were
superfluous/static so that the main output fits on one line.
2023-02-13 12:13:56 -05:00
Joshua Boniface 3c742a827b Initial implementation of monitoring plugin system 2023-02-13 12:06:26 -05:00
Joshua Boniface 726d0a562b Update copyright header year 2022-10-06 11:55:27 -04:00
Joshua Boniface 5942aa50fc Avoid raise/handle deadlocks
Can cause log flooding in some edge cases and isn't really needed any
longer. Use a proper conditional followed by an actual error handler.
2022-10-03 14:04:12 -04:00
Joshua Boniface 8d0f26ff7a Add additional kb_ values to OSD stats
Allows for easier parsing later to get e.g. % values and more details on
the used amounts.
2022-08-11 11:06:36 -04:00
Joshua Boniface 23b1501f40 Fix linting error F541 f-string placeholders 2021-11-06 03:26:03 -04:00
Joshua Boniface c41664d2da Reformat code with Black code formatter
Unify the code style along PEP and Black principles using the tool.
2021-11-06 03:02:43 -04:00
Joshua Boniface 2e7b9b28b3 Add some delay and additional tries to fencing 2021-10-27 16:24:17 -04:00
Joshua Boniface 55f397a347 Fix bad location of config sets 2021-10-12 17:23:04 -04:00
Joshua Boniface fe73dfbdc9 Use current live value for bridge_mtu
This will ensure that upgrading without the bridge_mtu config key set
will keep things as they are.
2021-10-12 12:24:03 -04:00
Joshua Boniface 8f906c1f81 Use power off in fence instead of reset
Use a power off (and then make the power on a requirement) during a node
fence. Removes some potential ambiguity in the power state, since we
will know for certain if it is off.
2021-10-12 11:04:27 -04:00