add build team availability alerts #353
Conversation
tnevrlka
force-pushed
the
add-build-alerts
branch
from
3f5cb93
to
1cc4b27
Compare
- Add GitHubAppFailureAlert for build-service - Add QuayFailureAlert for image-controller Signed-off-by: Tomas Nevrlka <[email protected]>
tnevrlka
force-pushed
the
add-build-alerts
branch
from
1cc4b27
to
05ff02e
Compare
|
Just for confirmation, both alerts should only be |
That was my intention, yes. Do you think it shouldn't be a warning? |
|
That is not a problem per se, but maybe it is worth checking what those alerts are trying to achieve. Essentially, non-SLO, warning alerts won't get follow up actions from SREs, for example. If that is the intention, then we are fine. On the other hand, if you intend for those alerts to be actionable, they should be critical SLO alerts with corresponding actionable SOPs. I hope this broader context brings a little bit of clarity. That said, whether the alerts should be SLO or not, critical or warnings, is up to the build team. |
|
Thanks. Since there's not really anything for SREs to do, I think warning is fine |
Jira: STONEBLD-2651
Add availability metrics for the build team which fire after five minutes of not having the
konflux_upparameterThe alerts include
runbook_urlto currently non-existing SOPs. There is a MR open which adds the SOPs, but it's not yet merged (will mark this PR ready for review after it's merged)