Patric Thanks for pointing Grafant plug-in for Hawkuarl. We already invested effort and time on TICK setup where metrics are collected via telegraf, InfluxDB and Grafant. This plug-in is nice on top of our Grafana setup. However, telegraf Docker plugin is much more advanced and have more metrics then Hawkular metrics.
Also, I noticed tags not working. Searching by name is pathetic and pretty slow. Only gauge working . Other 2 metrics types like count, and availability not working. Not sure why… Even Hawkular also not showing any information regarding PVC or PV. [cid:[email protected]] -- Srinivas Kotaru From: Patrick Tescher <[email protected]> Date: Wednesday, March 29, 2017 at 11:38 AM To: Srinivas Naga Kotaru <[email protected]> Cc: dev <[email protected]> Subject: Re: PV/PVC storage use tracking The system that would track this is Heapster but it doesn't appear to. There are a few open issues: https://github.com/kubernetes/heapster/issues/885 https://github.com/kubernetes/heapster/issues/1270 The other option would be Hawkular Openshift Agent: https://github.com/hawkular/hawkular-openshift-agent This means that the pod mounting your PVC needs to have some sort of agent that reports filesystem usage. Sometimes this makes sense. For instance I am using https://github.com/wrouesnel/postgres_exporter to export other PostgresSQL stats and one of those is filesystem usage. In other cases it may not make as much sense. Both of these are a little odd since they would track pod volume usage, not necessarily a specific PVC. It would be nice to somehow monitor usage of a PVC directly but that is hard since the PVC is not always mounted on any node. Lastly if you want to be able to display the stats generated by Heapster or Hawkular Openshift Agent you can set up https://github.com/hawkular/hawkular-grafana-datasource. -- Patrick Tescher On Mar 29, 2017, at 10:05 AM, Srinivas Naga Kotaru (skotaru) <[email protected]<mailto:[email protected]>> wrote: Does Openshift has any mechanism to track PV/PVC usage? PV/PVC are getting filled but there is no mechanism for us or our clients to track what is current utilization? One way to check is , platform teams or clients has to check usage by mounting the PV somewhere and check by using OS commands like du etc. wondering any better way to track each PVC usage and alert if they reach 90 % threshold etc. Our monitoring systems can handle alerts but looking for a API call or better simple way to check PVC usage Don’t’ want to use OS commands to check or track usage. -- Srinivas Kotaru _______________________________________________ dev mailing list [email protected]<mailto:[email protected]> http://lists.openshift.redhat.com/openshiftmm/listinfo/dev
_______________________________________________ dev mailing list [email protected] http://lists.openshift.redhat.com/openshiftmm/listinfo/dev
