Closed rgildein closed 4 months ago
I will run this change in one of the charms, with the CI. - https://review.opendev.org/c/openstack/charm-mysql-innodb-cluster/+/915994
@rgildein , can you take the chance of enabling juju 3.4 in the CI? - https://github.com/openstack-charmers/zaza/blob/master/.github/workflows/tox.yaml#L38 .
@rgildein , can you take the chance of enabling juju 3.4 in the CI? - https://github.com/openstack-charmers/zaza/blob/master/.github/workflows/tox.yaml#L38 .
Done
All modified and coverable lines are covered by tests :white_check_mark:
Project coverage is 89.53%. Comparing base (
b22c2ee
) to head (8b06c8e
). Report is 2 commits behind head on master.:exclamation: Current head 8b06c8e differs from pull request most recent head 14bab3a. Consider uploading reports for the commit 14bab3a to get more accurate results
:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.
LGTM. Thanks for the effort put into this. I'm just waiting for the CI results on this change - https://review.opendev.org/c/openstack/charm-mysql-innodb-cluster/+/915994 - to merge.
note: the ci failure in the 'third' test, it's not related to this change.
jammy https://openstack-ci-reports.ubuntu.com/artifacts/f20/915994/2/check/jammy/f2025fa/ : SUCCESS in 1h 18m 51s
the focal job failed, because an undercloud issue, so I won't block this PR on that.
The run_on_unit require to be awaited for juju 3+, and this fix proposes a solution to check status of action and wait for it if it's "pending". This was original described in issue #649 and marked as fixed by #650, however the action object is never awaitable as it can be seen in my examples below.
I tested in with this script
and here are the results for Juju 2.9
and results for Juju 3.4:
The example of usage on both 2.9 and 3.4 (see logs) can be found here. The func tests are failing, but it's not related with zaza. fixes: https://github.com/openstack-charmers/zaza/issues/649