Scheduled workflow 'Example Charm Integration Tests' failed

[github-actions]

Scheduled workflow 'Example Charm Integration Tests' failed: https://github.com/canonical/operator/actions/runs/34991284777

[dwilding]

Looks to be a relatively "simple" timeout of the postgres deploy. Rerunning.

[dwilding]

The problem seems to be slow image pulls of the postgresql-k8s resource, which for some reason are taking longer than our 180s timeout today. I explored whether anything has changed recently that might have caused this. One possibility was a CoreDNS change in the k8s snap, but testing with an older snap revision was inconclusive.

Perhaps the pulls are just slower than normal today because of infra issues.

I'm not massively keen on bumping the PostgreSQL deploy timeouts in the integration tests. In the K8s tutorial we do advise to set timeout=10 * 60 if tests fail, but that's more of a backstop. Jubilant's default seems to usually be fine.

Let's keep this issue open and try rerunning after a couple of days.