Open orfeas-k opened 7 months ago
Bumped into this in a PR that didn't update any code https://github.com/canonical/kfp-operators/pull/393 https://github.com/canonical/kfp-operators/actions/runs/10366700320/job/28696539467?pr=393
@orfeas-k do we know what causes this flakiness?
@kimwnasptd No, I haven't investigated this. IIRC, increasing the timeout (or the amount of retries) has resolved this in the past, so it could be caused from charms/integration not being ready yet.
Bug Description
kfp-api integration tests are flaky and sometimes fail in this part, when trying to assert response metrics with the following error
This time it happened in
track/2.0
but I think I 've seen it inmain
in the past as well.Cause
My guess is that this is due to resources and if we increase the retry attempts, it will be resolved.
To Reproduce
Run multiple times the integration CI.
Environment
Github runners,
LXD
juju (3.1/stable) 3.1.8 from Canonical installed charmcraft (candidate) 2.6.0 from Canonical installed microk8s (1.25-strict/stable) v1.25.16 from Canonical** installed
juju status
:Relevant Log Output
Additional Context
No response