GitLab

Use IP address for controller1 to perform initial configuration.

Having our middleware client hit a socket timeout usually means that
the test VM is seriously broken. We should swiftly abort test run
in this case.

This API call needs to be made via the ws_client fixture rather
than normal middleware call test asset.

* simplify to 1 nvme function

* simplify and clean-up jbof/crud.py

A recent change (dade5605) disallowed a hostname from being supplied
as the ip parameter to runtest.py.  Work around this new restriction.
(Only useful in limited circumstances e.g. running tests locally)

* Added a 'quick' method to generate a single drive partition table
Fixed bug (introduced with Electric Eel) where we failed to update the kernel partition table
Added CI test to prevent regression on the partition table update.

* Changed the method by which we request the partition table update.  We now use the `BLKRRPART` ioctl call from `retaste.py`

If networking is not fully configured, kinit attempt may fail with
EAGAIN. Retry several times before giving up. Since the actual kinit
has a timeout set for 30 seconds, this loop may end up being quite
long-running and so convert directoryservices.initialize into a
middleware job. The only direct caller for this private endpoint
is smb.configure which is itself a job and so we don't have to worry
about any potential timeouts here.

* add zpool disk read/write/chksum errors

* update docstring in webui.enclosure.dashboard

* standardize key placement

* add SUPPORTS_IDENTIFY_KEY to constants.py

* add SUPPORTS_IDENTIFY_KEY to nvme enclosure mapping

* add docstring about new key

* initial logic for all platforms

* add the identify key to array device elements

* update docstring in webui.enclosure.dashboard

* update enclosure2.set_slot_status to work with new key

* update unit tests

* fix typos

Allow API keys to toggle services. We only need to check whether
this is a user session since API keys don't support roles. This
means that once an API key is within the permissions check function
we already know that it has been explicitly granted access to this
endpoint via its allowlist.

This particular check never quite worked correctly in that our
filter for whether the dataset was in boot pool was a de-facto
no op (since we were not populating the pool name before filtering).

This issue was not detected due to a compensating bug in filter_list
where missing keys were not being handled correctly either.

* Drives are now always partitioned to maximize the size of the data partition.
- Remove the data partition size parameter
- Improve input validation
- Improve start and stop sector calculation

Added more rigor to the CI tests.

* Fix drive name usage.

* Improved 'QUICK' disk wipe with cleaning of the partition gap areas
In format, added dependency check on 'grainSize'

CI tests:
* Moved the module 'global' stuff to a pytest fixture.
* Added a test for the new 'middle' partition cleaning

Earlier versions of filter_list would crash on handling None
type with `in` filter when an invalid key was specified for the filter.
Since fix in NAS-127480 we short-circuit when specified key doesn't
exist. This commit adds an additional test for this specific situation.

When key is absent we should omit from result rather than substitute
a None value otherwise it's not possible to differentiate between
case where value is explicitly NULL as opposed to where key is simply
absent.

This validation is perhaps overly restrictive as it prevents us
from automatically registering IPv6 and public IPv4 addresses in
Active Directory.

* Improve speed of jbof.create and jbof.delete APIs

Improve the speed of these APIs, generally by performing certain
operations in parallel rather than in series.

Also add some additional logging.

NAS-127151 / 24.10 / Make sure there is no sensitive information in apps/tc/vm events plugins (#13183)

* Make sure there is no sensitive information in apps/vm events plugin

This commit adds changes to make sure we don't expose any private info in apps/vm plugin events.

* Add a private endpoint to retrieve event config for tc

* Redact apit key from tc plugin events

* Add truecommand read/write roles

* Specify roles for truecommand plugin

* Add integration test for truecommand roles

* Fix no auth req endpoint roles

This commit updates AMD k8s device plugin image tag as we have a user whose gpu is not being recognized by the tag we had earlier. However there is still breakage expected for other users as we had bumped this earlier as well but we had quite a few users for whom AMD gpu was not being recognized anymore and hence we went back to the older version in https://github.com/truenas/middleware/pull/10312. Idea is to see that if things have improved as it's been a while since we reverted back and based on feedback we can re-evaluate.

* Improve local_configure_interface

- Do not perform configuration changes if none are necessary
- When clearing the address, set the interface state to DOWN

* Add private API jbof.configure_job

Add a private API that will bring up any previously configured JBOFs
in parallel, using the also newly added configure_jbof.

* Call jbof.configure_job from vrrp_master amd vrrp_backup

* Add jbof.configure_job to pool.import_on_boot, only on iX hardware

The legacy ixautomation script used to initiate runtest.py in
jenkins may fail to retrieve our interface name from console
logs (due to changed logging settings). This commit instead
gathers the interface name from the server instead of relying
on it being provided.

* stop the vrrp service in vrrp_backup

* add global lock for vrrp_master/backup

* add failover.vrrp.get_priority

* change keepalived.conf

* reconfigure keepalived on master event

* fix flake8 while im here

* short-circuit if failover.status is MASTER

* missing comma