mitigate CrashLoopBackoff in lifecycle test app #42
Assignees
Comments
This was referenced May 23, 2018
siggy
added a commit
that referenced
this issue
May 24, 2018
The lifecycle environment was testing service discovery via bb-terminus container exit. This did not play well with k8s's `CrashLoopBackoff`. Disable bb-terminus container exit in favor of a `redeployer` script, that deletes each bb-terminus pod once per minute. Fixes #42, relates to #43. Signed-off-by: Andrew Seigner <siggy@buoyant.io>
siggy
added a commit
that referenced
this issue
May 24, 2018
The lifecycle environment was testing service discovery via bb-terminus container exit. This did not play well with k8s's `CrashLoopBackoff`. Disable bb-terminus container exit in favor of a `redeployer` script, that deletes each bb-terminus pod once per minute. Also create `ServiceAccount`, `ClusterRole`, and `ClusterRoleBinding`, and install conduit into conduit-lifecycle namespace. Fixes #42, relates to #43. Signed-off-by: Andrew Seigner <siggy@buoyant.io>
9 tasks
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Follow up from #41. Containers in the lifecycle test app intentionally exit after a given number of requests. Per kubernetes/kubernetes#57291, Kubernetes puts these pods into an exponential
CrashLoopBackoff, with up to 5 minute backoff. Ideally for our testing, we need a faster restart.One option to mitigate this would be a script that periodically runs a
kubectl delete po/foocommand.The text was updated successfully, but these errors were encountered: