but it will be cause some problems, such as alerts is not work. when
project-prometheus has not metrics named alive_count, but my top-prometheus
still store it, so it can not trigger an alarm
在2021年3月4日星期四 UTC+8 下午8:35:24<Stuart Clark> 写道：
> On 04/03/2021 04:53, 'linlong Fu' via Prometheus Users wrote:
> > My case:
> > I have two prometheus one named top-prometheus, another named
> > project-prometheus.
> > and top-prometheus has federation configure (scrape metrics from
> > project-prometheus)
> > configure just like:
> >
> > - job_name: "project-prometheus"
> > scrape_interval: 15s
> > honor_labels: true
> > metrics_path: '/federate'
> > params:
> > 'match[]':
> > - '{job=~".+"}'
> > static_configs:
> > - targets:
> > - '22.2.1.2.1:9090'
> > labels:
> > project: test
> >
> > Then, I meet the following problems:
> > when I remove one metrics named alive_count in project-prometheus,
> > I see alive_count this metrics have no data from project-prometheus
> > web page, but from top-promethues web page, it retain this metics.
> > after 15 minutes it shows no data. (all prometheus scrape_interval is
> > 15 seconds )
> > why? federation will cache metrics?
>
> Federation is no different from any other scrape, so it fetches the
> metrics from the other Prometheus server and stores it within the TSDB.
> Therefore you will retain the metrics (and be able to query them) for
> however long your retention period is.
>
>
> --
> Stuart Clark
>
>
--
You received this message because you are subscribed to the Google Groups
"Prometheus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to prometheus-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit
https://groups.google.com/d/msgid/prometheus-users/aaac5d1d-18c3-4d2c-97e4-82c1bf04a5b7n%40googlegroups.com.
