Commit Graph

443 Commits

Author SHA1 Message Date
Joshua Boniface a95e72008e Add size validations for volume clones
Adds the same validations as a volume add or resize to volume clones, to
ensure there is enough free space for them.
2024-02-02 11:37:29 -05:00
Joshua Boniface efc7434143 Add safety check for 80% full size
Adds a check that a volume creation or resize won't violate the 80% full
rule for the storage cluster. This ensures a cluster won't get too full
if a storage volume fills up.

Also adds a force flag throughout the pipeline to override this check,
should an administrator really want to do so.

Closes #177
2024-02-02 11:37:00 -05:00
Joshua Boniface 8419659e1b Ensure zkhandler is always cleaned up
Even if the subfunction of an API @ZKConnection call fails, the
zkhandler needs to terminate and clean up, or it leaves stuck threads
around.
2024-01-30 09:48:17 -05:00
Joshua Boniface d28fb71f57 Fix incorrect variable set 2024-01-24 14:40:40 -05:00
Joshua Boniface 09269f182c Add live migrate max downtime selector meta field
Adds a new flag to VM metadata to allow setting the VM live migration
max downtime. This will enable very busy VMs that hang live migration to
have this value changed.
2024-01-11 00:05:50 -05:00
Joshua Boniface 362edeed8c Add backup reporting and improve metrics
Major improvements to autobackup and backups, including additional
information/fields in the backup JSON itself, improved error handling,
and the ability to email reports of autobackups using a local sendmail
utility.
2024-01-10 14:18:44 -05:00
Joshua Boniface 39c8367723 Add additional metainfo to VM backups
Adds additional information about failures, runtime, file sizes, etc. to
the JSON output of a VM backup.

This helps enable additional reporting and summary information for
autobackup runs.
2024-01-10 10:37:29 -05:00
Joshua Boniface aac306c55b Fix missing stats on old Debians 2024-01-09 12:10:16 -05:00
Joshua Boniface c1ae571213 Add additional VM details to Prometheus 2023-12-29 14:09:39 -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 4969e90f8a Allow enable/disable of Prometheus endpoints
Since these are unauthenticated, it might be the case that an
administrator wishes to completely disable these metrics endpoints.
Provide that option via pvc.conf through pvc-ansible's existing
enable_prometheus_exporters option and the new enable_prometheus
configuration flag.

Defaults to "yes" to provide all functionality unless explicitly
disabled, as the author assumes that the PVC API is secured in other
ways as well and that metric information is not completely sensitive.
2023-12-29 09:25:10 -05:00
Joshua Boniface 3346ce9bb0 Add missing shutdown state from combinations 2023-12-27 13:40:30 -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 4375f66793 Use proper get() for invalid values 2023-12-27 12:03:48 -05:00
Joshua Boniface 3df3ca5b44 Fix value for OSD utilization
Ceph provides in KB; convert to bytes.
2023-12-27 11:56:50 -05:00
Joshua Boniface 431ee69620 Use proper percentage for pool util 2023-12-27 10:03:00 -05:00
Joshua Boniface 88f4d79d5a Handle invalid values on older Libvirt versions 2023-12-27 09:51:24 -05:00
Joshua Boniface 84d22751d8 Fix bad JSON data handler 2023-12-27 09:43:37 -05:00
Joshua Boniface 40ff005a09 Fix handling of Ceph OSD bytes 2023-12-26 12:43:51 -05:00
Joshua Boniface 9604f655d0 Improve node utilization metrics and fix bugs 2023-12-25 02:47:41 -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 d2d2a9c617 Include our newline atomically
Sometimes clashing log entries would print on the same line, likely due
to some sort of race condition in Python's print() built-in.

Instead, add a newline to our actual message and print without an end
character. This ensures atomic printing of our log messages.
2023-12-21 13:12:43 -05:00
Joshua Boniface 6ed4efad33 Add new network.stats key to nodes 2023-12-21 12:48:48 -05:00
Joshua Boniface 39f9f3640c Rename health metrics and add resource metrics 2023-12-21 09:40:49 -05:00
Joshua Boniface c64e888d30 Fix incorrect cast of None 2023-12-14 16:00:53 -05:00
Joshua Boniface f1249452e5 Fix bug if no nodes are present 2023-12-14 15:32:18 -05:00
Joshua Boniface f41c5176be Ensure health value is an int properly 2023-12-13 14:34:02 -05:00
Joshua Boniface ed9c37982a Move metric collection into daemon library 2023-12-11 19:20:30 -05:00
Joshua Boniface 57c28376a6 Port one final Ceph function to read_many 2023-12-11 10:25:36 -05:00
Joshua Boniface e781d742e6 Fix bug with volume and snapshot listing 2023-12-11 10:21:46 -05:00
Joshua Boniface 741dafb26b Port VM functions to read_many 2023-12-11 03:34:36 -05:00
Joshua Boniface 5d9e83e8ed Fix output bugs in VM information 2023-12-11 03:04:46 -05:00
Joshua Boniface 7c116b2fbc Ensure node health value is an int 2023-12-10 23:56:50 -05:00
Joshua Boniface 1023c55087 Fix bug in VM state list 2023-12-10 23:44:01 -05:00
Joshua Boniface 9235187c6f Port Ceph functions to read_many
Only ports getOSDInformation, as all the others feature 3 or less reads
which is acceptable sequentially.
2023-12-10 22:24:38 -05:00
Joshua Boniface 0c94f1b4f8 Port Network functions to read_many 2023-12-10 22:19:21 -05:00
Joshua Boniface 44a4f0e1f7 Use new info detail output instead of new lists
Avoids multiple additional ZK calls by using data that is now in the
status detail output.
2023-12-10 22:19:09 -05:00
Joshua Boniface 5d53a3e529 Add state and faults detail to cluster information
We already parse this information out anyways, so might as well add it
to the API output JSON. This can be leveraged by the Prometheus endpoint
as well to avoid duplicate listings.
2023-12-10 17:29:32 -05:00
Joshua Boniface 35e22cb50f Simplify cluster status handling
This significantly simplifies cluster state handling by removing most of
the superfluous get_list() calls, replacing them with basic child reads
since most of them are just for a count anyways. The ones that require
states simplify this down to a child read plus direct reads for the
exact items required while leveraging the new read_many() function.
2023-12-10 17:05:46 -05:00
Joshua Boniface a3171b666b Split node health into separate function 2023-12-10 16:52:10 -05:00
Joshua Boniface 48e41d7b05 Port Faults getFault and getAllFaults to read_many 2023-12-10 16:05:16 -05:00
Joshua Boniface d6aecf195e Port Node getNodeInformation to read_many 2023-12-10 15:53:28 -05:00
Joshua Boniface 9329784010 Implement async ZK read function
Adds a function, "read_many", which can take in multiple ZK keys and
return the values from all of them, using asyncio to avoid reading
sequentially.

Initial tests show a marked improvement in read performance of multiple
read()-heavy functions (e.g. "get_list()" functions) with this method.
2023-12-10 15:35:40 -05:00
Joshua Boniface b9fbfe2ed5 Improve fault ID format
Instead of using random hex characters from an md5sum, use a nice name
in all-caps similar to how Ceph does. This further helps prevent dupes
but also permits a changing health delta within a single event (which
would really only ever apply to plugin faults).
2023-12-09 16:48:14 -05:00
Joshua Boniface 7e6d922877 Improve fault detail handling further
Since we already had a "details" field, simply move where it gets added
to the message later, in generate_fault, after the main message value
was used to generate the ID.
2023-12-09 16:13:36 -05:00
Joshua Boniface 4003204f14 Remove bracketed text from fault_str
This ensures that certain faults e.g. Ceph status faults, will be
combined despite the added text in brackets, while still keeping them
mostly separate.

Also ensure the health text is updated each time to assist with this, as
this health text may now change independent of the fault ID.
2023-12-09 15:34:18 -05:00
Joshua Boniface 2bea78d25e Make all remaining limits optional 2023-12-09 13:43:58 -05:00
Joshua Boniface fd717b702d Use external list of fault states 2023-12-09 12:51:41 -05:00
Joshua Boniface 317ca4b98c Move defined state combinations into common 2023-12-09 12:36:32 -05:00
Joshua Boniface 0bda095571 Move libvirt_schema and fix other imports 2023-12-09 12:20:29 -05:00
Joshua Boniface 813aef1463 Fix incorrect UUID key name 2023-12-09 12:14:57 -05:00
Joshua Boniface 5a7ea25266 Fix incorrect database name entries 2023-12-09 12:12:00 -05:00
Joshua Boniface 61b39d0739 Fix incorrect cluster health calculation 2023-12-07 11:13:36 -05:00
Joshua Boniface 4bf80a5913 Fix missing datetime shrink 2023-12-06 17:15:36 -05:00
Joshua Boniface e0bf7f7d1a Fix bad ID values in acknowledge 2023-12-06 14:18:31 -05:00
Joshua Boniface 20acf3295f Add mass ack/delete of faults 2023-12-06 13:59:39 -05:00
Joshua Boniface d1e34e7333 Store fault times only to the second
Any more precision is unnecessary and saves 6 chars when displaying
these times elsewhere.
2023-12-06 13:20:18 -05:00
Joshua Boniface 79eb54d5da Move fault generation to common library 2023-12-06 13:17:10 -05:00
Joshua Boniface 2267a9c85d Improve output formatting for simplicity 2023-12-05 10:37:35 -05:00
Joshua Boniface 672e58133f Implement interfaces to faults 2023-12-04 01:37:54 -05:00
Joshua Boniface 3dc48c1783 Lower default monitoring interval to 15s
Faults are also reported on the monitoring interval, so 60s seems like
too long. Lower this to 15 seconds by default instead.
2023-12-01 17:38:28 -05:00
Joshua Boniface 9c2b1b29ee Add node health to fault states
Adjusts ordering and ensures that node health states are included in
faults if they are less than 50%.

Also adjusts fault ID generation and runs fault checks only coordinator
nodes to avoid too many runs.
2023-12-01 17:38:28 -05:00
Joshua Boniface 8594eb697f Add initial fault generation in pvchealthd
References: #164
2023-12-01 17:38:27 -05:00
Joshua Boniface 7cb9ebae6b Remove legacy configuration handler
This is not going to be needed.
2023-12-01 01:25:40 -05:00
Joshua Boniface 102c3c3106 Port all Celery worker functions to discrete pkg
Moves all tasks run by the Celery worker into a discrete package/module
for easier installation. Also adjusts several parameters throughout to
accomplish this.
2023-11-30 02:24:54 -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 11db3c5b20 Fix ordering during termination 2023-11-29 21:21:51 -05:00
Joshua Boniface fa12a3c9b1 Permit buffered log appending 2023-11-29 21:21:51 -05:00
Joshua Boniface 787f4216b3 Expand Zookeeper log daemon prefix to match 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 2e5958640a Remove erroneous time from message 2023-11-29 15:12:41 -05:00
Joshua Boniface 7abc697c8a Improve Zookeeper log handling
Ensures that messages are fully read before each append. Adds more
Zookeeper hits, but ensures logs won't be overwritten by multiple
daemons.

Also don't use a set on the client side, to avoid "removing duplicate"
entries erroneously.
2023-11-29 15:12:41 -05:00
Joshua Boniface dd6a38d5ea Properly pass the name of the exception 2023-11-16 18:05:52 -05:00
Joshua Boniface f50f170d4e Convert vmbuilder to use new Celery step structure 2023-11-16 16:08:49 -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 b522306f87 Increase Celery wait times
It's a bit inefficient, but provides nicer output and a bit of settling
time between each stage.
2023-11-09 23:54:05 -05:00
Joshua Boniface 07026efb63 Ensure OSD checks in before completing
Avoids issues where the new OSD doesn't check in; at least the
administrator will know.

Also fixes some issues with osd_db in removal.
2023-11-09 23:51:05 -05:00
Joshua Boniface 08411708f6 Clean up dangling references to cmd pipes
Also removes the schema references for these CMD pipes as they are no
longer required.
2023-11-09 23:28:14 -05:00
Joshua Boniface ce17c60a20 Port OSD on-node tasks to Celery worker system
Adds Celery versions of the osd_add, osd_replace, osd_refresh,
osd_remove, and osd_db_vg_add functions.
2023-11-09 23:28:08 -05:00
Joshua Boniface 89681d54b9 Port VM on-node tasks to Celery worker system
Adds Celery versions of the flush_locks, device_attach, and
device_detach functions.
2023-11-06 20:40:46 -05:00
Joshua Boniface a016337f57 Remove block verify in APi
This doesn't work right and is handled by the node anyways.
2023-11-04 02:45:10 -04:00
Joshua Boniface 7f5dd385b5 Use right key for FSID elsewhere 2023-11-03 23:51:01 -04:00
Joshua Boniface ec42b19d0e Send FSID to clients too 2023-11-03 16:37:55 -04:00
Joshua Boniface 64e37ae963 Update OSD replacement functionality
1. Simplify this by leveraging the existing remove_osd/add_osd
functions, since its task was functionally identical to those two in
sequential order.
2. Add support for split OSDs within the command (replacing all OSDs on
the block device(s) as required).
3. Add additional configurability and flexibility around the old device,
weight, and external DB LVs.
2023-11-03 01:45:49 -04:00
Joshua Boniface 980ea6a9e9 Adjust handling of ext_db and _count options
Avoid the use of superfluous flag options, default them to none, and add
support for fixed-size DB LVs.
2023-11-02 13:29:47 -04:00
Joshua Boniface 526a5f4a74 Add support for split OSD adds
Allows creating multiple OSDs on a single (NVMe) block device,
leveraging the "ceph-volume lvm batch" command. Replaces the previous
method of creating OSDs.

Also adds a new ZK item for each OSD indicating if it is split or not.
2023-11-01 21:31:35 -04:00
Joshua Boniface 35f80e544c Use more hierarchical backup path structure 2023-10-24 02:04:16 -04:00
Joshua Boniface 83b937654c Avoid removing nonexistent snapshots
Store retain_snapshot in JSON and use that to check during delete.
2023-10-24 01:35:00 -04:00
Joshua Boniface 714bde89e6 Fix incorrect variable ref 2023-10-24 01:25:01 -04:00
Joshua Boniface c87736eb0a Use consistent path name and format 2023-10-24 01:20:44 -04:00
Joshua Boniface 63d0a85e29 Add backup deletion command 2023-10-24 01:18:27 -04:00
Joshua Boniface 55ca131c2c Handle snapshots on restore and provide options
Also rename the retain option to remove superfluous plural.
2023-10-24 00:25:06 -04:00
Joshua Boniface 8d256a1737 Complete VM restore functionality 2023-10-23 22:23:17 -04:00
Joshua Boniface d3b3fdfc80 Revert "Export backup images to a tar archive"
This reverts commit 38abd078af.
2023-10-23 11:01:16 -04:00
Joshua Boniface f1b29ea94e Initial VM restore work 2023-10-23 11:00:54 -04:00
Joshua Boniface 38abd078af Export backup images to a tar archive
This helps ensure an easier restore as the tar archive(s) can be sent
directly to the API via the normal process of image uploading, instead
of individual disks.
2023-10-23 09:56:50 -04:00
Joshua Boniface fabb97cf48 Only split a command_string if its not a list 2023-10-23 09:50:58 -04:00
Joshua Boniface 68124db323 Remove extra spaces 2023-10-17 13:01:38 -04:00
Joshua Boniface 8921efd269 Fix incorrect tuple construct 2023-10-17 12:55:44 -04:00
Joshua Boniface 3d12915989 Further improve return messages 2023-10-17 12:53:08 -04:00