Repository navigation
UpdateCommandDistTest failing in release/26.6 #48568
Description
Before reporting an issue
- I have read and understood the above terms for submitting issues, and I understand that my issue may be closed without action if I do not follow them.
Area
ci
Describe the bug
UpdateCommandDistTest has multiple failures in release/26.6 branches. I've not seen this test failing in other branches.
Version
26.6
Regression
- The issue is a regression
Expected behavior
Test pass
Actual behavior
Seems to fail all the time, including with retries
How to Reproduce?
Run locally to check if it works, if it doesn't then reproduce on GA
Anything else?
https://github.com/keycloak/keycloak/actions/runs/25038482464/job/73384062285#logs
https://github.com/keycloak/keycloak-private/actions/runs/25046126762/job/73363774625#logs
https://github.com/keycloak/keycloak-private/actions/runs/25046126762
https://github.com/keycloak/keycloak-private/actions/runs/25047492213/job/73387492621?pr=539#logs
https://github.com/keycloak/keycloak-private/actions/runs/25047498320/job/73383885835?pr=543#logs
https://github.com/keycloak/keycloak-private/actions/runs/25047508289/job/73383986544?pr=581#logs
Activity
Pepo48 commentedon Apr 29, 2026
I think that the root cause could be introduced with PR #47416.
withStartupTimeout(120s)inDockerKeycloakDistribution.getKeycloakContainer()is called beforewaitingFor(new WaitAllStrategy(...)). ThewaitingFor()call replaces the wait strategy, discarding the 120s timeout. The newWaitAllStrategyfalls back to its default 30-second timeout (I checked this in test-containers code directly, see line 39 inWaitAllStrategy), which I guess is not enough for Keycloak container startup (augmentation + DB init + realm setup).Assuming that the private repo runners are slower, it probably consistently exceeds 30s.
Pepo48 commentedon Apr 29, 2026
I forgot to add that I was able to reproduce my theory locally with testcontainers 2.0.3 on Podman by artificially adding some sleeps.
Pepo48 commentedon Apr 30, 2026
Resolved in a duplicate #48589
Quarkus test fixes (#606)