At Promcon [0] I talked with a guy who exposes some metrics (https connections, sql queries) in the prometheus format [1]. People also use JMX (for you that would probably be Jolokia?) and expose metrics that way[2]. That's how you should be able to reach the queue size etc. But I have to say, I haven't tried it myself.
But if you want automatic scaling, there would have to be support for that in openshift. AFAIK cpu/memory is supported at this time. [0] https://promcon.io/2017-munich/schedule/ [1] https://www.youtube.com/watch?v=ImxglcUIHZU https://github.com/fstab/promagent/ [2] https://github.com/prometheus/jmx_exporter On Sun, Sep 10, 2017 at 7:19 AM, Srikrishna Paparaju <[email protected]> wrote: > Hi, > > We are Fabric8 Analytics team, trying to scale workers (OSD containers) > up/down based on incoming user traffic. Based on research done so far, > there is no easy way to scale up / down (OpenShift) based on some metric > 'like number of messages in a queue'. > > Here is the issue being tracked: https://github.com/ > openshiftio/openshift.io/issues/668 > > Would like to your thoughts... > > Thanks, > SriKrishna > > > _______________________________________________ > dev mailing list > [email protected] > http://lists.openshift.redhat.com/openshiftmm/listinfo/dev > >
_______________________________________________ dev mailing list [email protected] http://lists.openshift.redhat.com/openshiftmm/listinfo/dev
