Asking for floating IPs on an undercloud results in a NotFoundException.
Make openstackclient handle it gracefully.
No test is added for this because it would need to be a scenario test -
deploying a cloud without the foalting IP extension loaded. I don't
think this edge case is worth an entire new job just to exercise it.
Change-Id: I73b544853376d98ab0dbb14e32fefc43c1a8a179
Story: 2006863
The name of the errors change and we need to move things around a
little, but it's otherwise a straight swap.
Change-Id: I0a19765ebeaa14c0534faa1542165b76ed2bf4e2
Signed-off-by: Stephen Finucane <stephenfin@redhat.com>
With `openstacksdk` 3.2.0 the `host` attribute of an Instance got added
to the `Server` class [0]. With that change, listing servers with the `host`
attribute leads to a query-filter for `compute_host` as expected, but
`openstacksdk` will also filter for the `host` attribute locally after
the results are returned. Since `compute_host` being
`OS-EXT-SRV-ATTR:host` is not the same as `host, this means no results
are returned.
Since we want to keep the old behaviour of filtering by `compute_host`
i.e. the service host name, we need to switch to filter for
`compute_host`. This is already supported in older versions of
`openstacksdk`.
[0] 0f311ff3e2
Change-Id: I0cd32c5b7d6d4d21194f3efdcfb9b205dea6a91e
Closes-bug: #2074200
Add file to the reno documentation build to show release notes for
stable/2024.2.
Use pbr instruction to increment the minor version number
automatically so that master versions are higher than the versions on
stable/2024.2.
Sem-Ver: feature
Change-Id: I5b2c70118df024dfc9236c7e1fa39554ffccf5e0
Well, mostly. We still keep our own flake8 hooks and the hacking hooks
enabled. Everything else can be handled by ruff.
Doing this enables a couple of hacking checks that were previously
unaddressed. It also highlights a few cases that flake8 missed. Both are
addressed.
Change-Id: If81c7055e9ef692425da2789bae18a96d04b104f
Signed-off-by: Stephen Finucane <stephenfin@redhat.com>
We also drop the default language setting: everything is Python 3
nowadays.
Change-Id: I9dc9573a86c93416d1bbbc782dac76ecdda6effd
Signed-off-by: Stephen Finucane <stephenfin@redhat.com>
We changed these in change I37d07a6c5cdc98680b8d65d596521cad2b049500 but
the '-c/--column' option provided by cliff currently requires an
explicit match on column names. Change them back for now. We can revert
this when cliff is a little bit smarter.
Change-Id: I9180922e9da5c22ae3d8878946d1bf1ec4b8c6e1
Signed-off-by: Stephen Finucane <stephenfin@redhat.com>
Closes-bug: #2076212
We changed these in change Iba3fee2672d32266623c6f367beaabe84bd3d24e but
the '-c/--column' option provided by cliff currently requires an
explicit match on column names. Change them back for now. We can revert
this when cliff is a little bit cleverer.
Change-Id: I6b4f1b793dc383856bfdf9a01514381be3cd2bf1
Signed-off-by: Stephen Finucane <stephenfin@redhat.com>
Related-bug: #2076212
In the event that a user provides a hypervisor name rather than an ID to
the 'hypervisor show' command, passing 'details=True' (the default) to
'find_hypervisor' will ensure we get the detailed response we need.
However, this comes at the cost of retrieving reams of additional
irrelevant data for all the other hypervisors. Rather than doing this,
use a summary view and then a second call to fetch only the hypervisor
we care about.
Change-Id: I92b53802e41a962c6f916c3a111dc2de7c12d0fc
Signed-off-by: Stephen Finucane <stephenfin@redhat.com>
Closes-bug: #2072965
In SDK change I3bc3150877c6c00aa9ec4355104308d7755aa1d4, we modified the
'rebuild_server' proxy method to use a sentinel 'unset' value so that we
could distinguish between fields that were not being changed and those
that were being changed to 'null'. However, we are currently passing the
'admin_password' field to the 'rebuild_server' SDK proxy command
regardless of whether it is set or not. Resolve this conflict.
Change-Id: If7b7585aadd43cdc6d2a9358f14223e43dc21a73
Signed-off-by: Stephen Finucane <stephenfin@redhat.com>
Closes-bug: #2076232
Unlike cinderclient, SDK attempts to connect to a service as soon as you
create a client. A keystoneauth1.exceptions.catalog.EndpointNotFound
exception can be raised if this service does not exist in the service
catalog. Avoid this for the quota and limits commands by first checking
if the service is enabled.
In the process, we rework the 'is_volume_endpoint_enabled' helper we are
using to check for the existence of the service to *not* require a
volume client, since this was causing a chicken and egg issue for us
(and was also pretty much unnecessary).
Change-Id: I56e68f00ea221d689eb7f668e9e5ffa7d1a20184
Signed-off-by: Stephen Finucane <stephenfin@redhat.com>
Closes-bug: #2076229
When running project cleanup and using the --skip-resource flag, this
flag is only used during the --dry-run stage. This change ensures it is
also used when deleting the resources which fixes the bug mentioned below.
Closes-Bug: #2074109
Change-Id: I8e50fd14dcde069b932cc8ec4bb4fc809f7d2eb7
I have seen a few cases where import errors (distutils - I am looking at you)
result in an extension not being available, but there is no indication why this
is the case. We do configure logging, but this happens too late (as part
of the 'cliff.app.App.run' call to execute a command, which calls
osc-lib's 'configure_logging' but which happens long after we've tried
to import our plugins) to be of any use. Instead, make use of a callback
to make it more obvious.
Change-Id: Id68b06161e445b79fe43f463e06cda3c4771ef02
Signed-off-by: Stephen Finucane <stephenfin@redhat.com>
