add healthchecks and prepare monitoring #95
Conversation
| @@ -0,0 +1,15 @@ | |||
| {{- if .Values.monitoring.enabled -}} | |||
| apiVersion: monitoring.coreos.com/v1 | |||
There was a problem hiding this comment.
For this to work the Prometheus Operator needs to be installed on the Cluster, right? Shouldn't we mention that somewhere it the docs?
There was a problem hiding this comment.
Right, I'll add this to the docs
There was a problem hiding this comment.
I removed the ServiceMonitor from the chart because we should keep dependencies on non Kubernetes resources low. It can be easily added separately and I'll add this to the documentation.
There was a problem hiding this comment.
see also owncloud/docs-ocis#287
wkloucek
changed the title
|
@wkloucek To cleanup the commit history of this I'd like to rebase and probably squash a few commits. Is that ok for you. Otherwise I think this one is good now. |
IDM has a debug port with health endpoint: https://github.com/owncloud/ocis/blob/9193e6f5235761778223b41be00baa9fc3f52cd3/services/idm/pkg/server/debug/server.go#L29-L43 NATS has indeed no debug port. Which is a bug, because it has the config for it... |
Hm, I am pretty sure that that is dead code.
Yeah. There are bugs in both services with the debug endpoints. I'll open issues for both. For the time being I thing the TCP check for IDM is much more useful than checking a HTTP port for a service that does not even serve HTTP. |
IDM and NAT don't provide any debug service/port so use a simple TCP check on the main port. Notifications and Audit do currently not provide a debug port either. As they are not listening on any port by default we don't currently implement a useful liveness probe for them.
Description
adds healthchecks and prepares monitoring where applicable
Related Issue
Motivation and Context
How Has This Been Tested?
Screenshots (if appropriate):
Types of changes
Checklist: