Add support for user-events monitoring#520
Conversation
|
This looks great! Took a quick scan; will go through more carefully next week. TravisCI is failing because the prometheus-server pod is going into a crash loop backoff. Unfortunately the upload of logs from TravisCI to box hasn't been working for a while, so if you can't recreate locally you may need to debug by hacking up the script that runs the test to do some additional |
|
Thx for the tip @dgrove-oss, I was able to reproduce the problem locally. Prometheus deploy should be working fine now with or without persistence. |
dgrove-oss
left a comment
dgrove-oss
left a comment
There was a problem hiding this comment.
Looks good to me.
Left a couple of comments, but none of them are blocking.
Should we hold merging this until the main repo PR is merged (which allows removal of some or all of the adobe-apigateway github references) or merge independent of main repo PR and cleanup later?
selfxp
left a comment
selfxp
left a comment
There was a problem hiding this comment.
great, thx @dgrove-oss, yes let's hold of the merge until the user-events has been added to ow main repo, and all adobe references have been updated. Will ping you again when the reviewing process is done.
|
@selfxp -- should we go ahead and merge this now? |
|
@dgrove-oss yes please, it has all the discussed changes in place |
2 participants
Provides monitoring capabilities for all user specific metrics by deploying
user-events,prometheusandgrafanaon the cluster in caseuserMetricsEnabledis set totrue.It exposes a
/monitoringendpoint for accessing the Grafana dashboards.All system metrics will also be stored inside prometheus in case
prometheusEnabled=true.More on the
user-eventsservice that reads the events from Kafka on this OW PR.Fixes #519