Stan Kwong
Stan Kwong
@fredr Yes, they are definitely deleted.
Seeing another instance of this. These two metrics existed at the same time for the pod named `taskmanager-0`... the IP addresses differ because one IP is old one and the...
this can be closed, these two PRs are equivalent: https://github.com/grafana-tools/sdk/commit/e94e88581b1b06f156fff6b53cf69fd2d51909b5 https://github.com/grafana-tools/sdk/commit/4c823444ef117aac8ee610b6e75c983e39a45377
FYI you can use [AWS ES index state management](https://docs.aws.amazon.com/elasticsearch-service/latest/developerguide/ism.html) to do this, although it's time-based and not count-based as OP requests
> the pod object was created but wasn't yet assigned to a node and this should be reflected since it might be meaningful for debugging. That makes sense to me....
@roidelapluie thanks for your response! to answer your questions: - the 100gb -> 0 graph represents memory on the nodes. - here are the requested screenshots for our two Prometheus...
here are the tsdb logs from the same day (from 4:17 to 10:53) ``` level=info ts=2020-07-22T04:17:09.675Z caller=compact.go:495 component=tsdb msg="write block" mint=1595404800000 maxt=1595412000000 ulid=01EDV2W6W3FDJYEY40HE65FHS2 duration=17m9.67171137s level=info ts=2020-07-22T04:18:13.483Z caller=head.go:792 component=tsdb msg="Head GC...
> There are no gaps in the dashboards. Is that coming from the same prometheus server? Only the statsd jobs are missing datapoints? That's correct - we only noticed these...
> The metrics with gaps, do they come from recording rules? They do not.