Alert us via email or chat when there is any issue in the cluster

place-labs / k8s-helm

Helm charts for PlaceOS on Kubernetes

0 stars 2 forks source link

Closed w-le closed 2 years ago

w-le commented 2 years ago

Suggested ideas (open to others):

Notify us via the google chat channel "stderr" when:
- Any pods in the namespace 'placeos' end with a non 0 exit code
- Any pods in the namespace 'placeos' are rescheduled on to a different node
- Any pods in the namespace 'placeos' get high cpu or ram pressure (e.g. 95%?)
- Any nodes hit cpu or ram pressure (e.g. 95%?)

w-le commented 2 years ago

hey @jeremyw24 perhaps we can get Ali to work on this?

naqvis commented 2 years ago

@w-le Hi Will, can you please elaborate more? From the title doesn't know if things need to be looked at infra level or app level.

w-le commented 2 years ago

yep done, open to other suggestions for alerts. Do you have any @viv-4 ?