Have a look at the repository here: https://gitlab.com/havilland/prometheus-ocp . This ansible role does exactly that and comes with some extra features.
On Fri, Jan 18, 2019 at 11:56 AM Simon Pasquier <[email protected]> wrote: > > > On Fri, Jan 18, 2019 at 10:12 AM Alexander Bartilla < > [email protected]> wrote: > >> Hi Simon, >> >> as far as I know you won't even be able to deploy prometheus instances >> via the prometheus-operator in the openshift-monitoring namespace, as it's >> namespaced through an argument in its deployment config. >> > > You're correct. I meant deploying another Prometheus operator in a > namespace you "own" which will deploy other Prometheus servers. > > >> >> But if you have another grafana dashboard deployed you should be able to >> add the k8s-prometheus as a datasource. To get the credentials through >> commandline run this command: >> >> oc get secret grafana-datasources -n openshift-monitoring -o yaml | grep >> prometheus.yaml | cut -d':' -f2 | cut -d' ' -f2 | base64 -d >> >> Alexander >> >> On Fri, Jan 18, 2019 at 9:45 AM Simon Pasquier <[email protected]> >> wrote: >> >>> On Thu, Jan 17, 2019 at 6:27 PM Pavel Sapozhnikov >>> <[email protected]> wrote: >>> > >>> > Hello folks >>> > >>> > We are using OpenShift 3.11 and monitoring that comes with that >>> package. Our application pods expose prometheus endpoints and we are >>> looking to dashboard our own metrics withing the Grafana that is in the >>> monitoring project. >>> > >>> > So far what we've been seeing is that there's no way to even create a >>> new Grafana dashboard and also I haven't been able to find any >>> documentation on how to do it. It seems like the Grafana that is in >>> monitoring project is locked down in some way. >>> > >>> > Perhaps that Grafana isn't meant to be edited like that because its a >>> cluster Grafana? However, I see no harm in creating a new custom dashboard >>> for our application. >>> >>> This. What runs in the openshift-monitoring namespace is dedicated for >>> the cluster monitoring which is why the Grafana instance is read-only >>> [1]. >>> You can still leverage the Prometheus operator to deploy additional >>> Prometheus instances (+ service monitors) in your namespace(s). >>> There are plans to refine and improve this situation but nothing acted >>> yet. >>> Simon >>> >>> [1] https://bugzilla.redhat.com/show_bug.cgi?id=1652536 >>> >>> >>> >>> > >>> > If this is the case, what that leaves us is installing our own >>> prometheus and Grafana instances inside our project - we were hoping to >>> leverage existing monitoring capabilities to look at our own custom metrics. >>> > >>> > If there's is documentation somewhere on how to create new Dashboards >>> in included Grafana could anyone point us in a right direction please? >>> > >>> > Otherwise, could anyone address our expectations and recommend a good >>> course of action. >>> > >>> > Thanks >>> > Pavel >>> > _______________________________________________ >>> > users mailing list >>> > [email protected] >>> > http://lists.openshift.redhat.com/openshiftmm/listinfo/users >>> >>> _______________________________________________ >>> users mailing list >>> [email protected] >>> http://lists.openshift.redhat.com/openshiftmm/listinfo/users >>> >> >> >> -- >> >> Alexander Bartilla >> >> IT-Consultant >> >> Cloudwerkstatt GmbH - Lassallestraße 7b – A-1020 Wien >> >> +43-660-8989058 >> >> [email protected] >> >> >> *[image: id:[email protected]]* >> >> ------------------------------ >> >> Cloudwerkstatt GmbH - Lassallestraße 7b - A-1020 Wien - ATU68384759 - >> FN408516i - Handelsgericht Wien >> >> >> -- Alexander Bartilla IT-Consultant Cloudwerkstatt GmbH - Lassallestraße 7b – A-1020 Wien +43-660-8989058 [email protected] *[image: id:[email protected]]* -- Cloudwerkstatt GmbH - Lassallestraße 7b - A-1020 Wien - ATU68384759 - FN408516i - Handelsgericht Wien
_______________________________________________ users mailing list [email protected] http://lists.openshift.redhat.com/openshiftmm/listinfo/users
