Skip to content

Fix tango admin failure issue

Thomas Ives requested to merge fix-tango-admin-timeout into main

The recent CI failures with tango_admin --ping_database 30 timing out are the result of the mysql container being slow to start. This seems to happen when the runners are under load.

Here are some logs demonstrating the issue: https://gitlab.com/tango-controls/cppTango/-/jobs/4486278690/artifacts/browse/build/tests/results/cxx_always_hook_20230616.075749.795665020/

In this MR I have tried to solve the issue by increasing the timeout. Note that this also requires increasing the timeout for the tango-db container. See tango-controls/docker/tango-db!4 (merged).

I have included some changes I made to get the docker logs output for the mysql and tango-db containers. These might be useful in the future. I have also done a bit of tidying up so that we handle this failure case more gracefully.

TODO:

  • Update the tango-db container we use to a tagged version rather than my merge branch

Closes #1128 (closed)

Edited by Thomas Ives

Merge request reports