Hi Christian,
Thank you for the helping hand .
My prometheus.yml file looks like this :
global:
external_labels:
dc: europe
replica: primary
alerting:
alertmanagers:
- static_configs:
- targets:
- am:9093
relabel_configs:
- action: drop
source_labels: [replica]
regex: (.*)
rule_configs:
- prom_rules.yml
scrap_configs:
- job_name: 'prometheus'
.
.
.
For other server, replica is secondary
I want to drop only label in alert manager , but it drops entire alert in
alert_manager.
In Alert-Manager , I see details in boxes like as given
dc=europe instance=servername replica=primary
dc=europe instance=servername replica=secondary
I want the output like below, to avoid duplicate entry from secondary server
dc=europe instance=servername
Please suggest
Thanks
On Sat, Apr 4, 2020 at 3:04 AM Christian Hoffmann <
[email protected]> wrote:
> Hi Sunil,
>
> On 4/3/20 6:00 PM, Sagar wrote:
> > For more information , there are more global labels (such as
> > environment, cluster , etc) apart from replica .
> >
> > Do I need to remove other global variables and maintain only one label
> > as replica .
> You need to drop those labels which are used to distinguish different
> Prometheus instances which scrape the same targets.
>
> Some example:
> If you've got two Prometheus servers scraping all your database servers
> and are adding external labels such as cluster=database and replica=A
> for the one and replica=B for the other and if you've got another pair
> of Prometheus servers for your web servers with external labels
> cluster=web and replica=A and replica=B again, then you would only drop
> the replica label, not the cluster label.
>
> > I also tried with options given below ( from the link
> > : https://github.com/prometheus/prometheus/issues/3239 )
> >
> > |alerting: alert_relabel_configs:- action: drop source_labels:
> > [replica] regex: '1'|
> >
> > Please advise , thanks
> I don't think you should have regex: 1 here? Just drop the regex line so
> that it will default to '(.*)' which matches all values.
>
> If this doesn't help, can you share your actual config + the behaviour
> you are seeing?
>
> > Is it possible to query alertmanager alerts like metrics in prometheus ?
> > Like http:/server:9093/mertrics or something else .
> Alertmanager does have metrics, but it does not export alerts there.
> However, there is an API for reading them programatically.
>
>
> https://petstore.swagger.io/?url=https://raw.githubusercontent.com/prometheus/alertmanager/master/api/v2/openapi.yaml#/alert
>
>
> Kind regards,
> Christian
>
--
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 [email protected].
To view this discussion on the web visit
https://groups.google.com/d/msgid/prometheus-users/CANegE51_xR_HkC4GOG%2BNpJTWSJSy%2BWZ%3DpfO8DFXsKt2Ob_%2Br6w%40mail.gmail.com.
