===================
#opnfv-cperf: cperf
===================


Meeting started by mattw4 at 18:16:39 UTC.  The full logs are available
at
http://ircbot.wl.linuxfoundation.org/meetings/opnfv-cperf/2017/opnfv-cperf.2017-08-24-18.16.log.html
.



Meeting summary
---------------

* jamoluhrsen got upstream tweaks to run robot scale tests in cperf pod
  (mattw4, 18:18:48)
* sai started latest round of scale testing  (mattw4, 18:20:08)
* env is running with clustered ODL  (mattw4, 18:20:31)
* started with 3 controllers & 3 computes  (mattw4, 18:21:11)
* LINK:
  https://nexus.opendaylight.org/content/repositories/opendaylight-carbon-epel-7-x86_64-devel/org/opendaylight/integration-packaging/opendaylight/6.2.0-0.1.20170824rel1941.el7.noarch/opendaylight-6.2.0-0.1.20170824rel1941.el7.noarch.rpm
  Sai may want to consider using this RPM, it has no-IP fix Tim wanted
  for OPNFV  (dfarrell07, 18:21:18)
* much harder to hit OOM this time (with ODL defaults)  (mattw4,
  18:21:42)
* encountering bugs similar to memory leaks & Java failures seen
  previously  (mattw4, 18:22:45)
* LINK: https://bugs.opendaylight.org/show_bug.cgi?id=9034  (sai,
  18:22:51)
* ODL version == prerelease SR2 build  (mattw4, 18:23:02)
* ODL should have memory leak fix, but still may fail clustering
  (mattw4, 18:23:42)
* we likely have more memory leaks to fix  (mattw4, 18:24:39)
* second issue under investigation is networking-odl driver itself
  (mattw4, 18:25:43)
* journal driver getting hit hard  (mattw4, 18:25:55)
* workaround is to reduce number of worker processes  (mattw4, 18:26:23)
* mysql has several queries that are taking a long time and consuming a
  lot of resources  (mattw4, 18:27:04)
* browbeat tool helping to monitor resources, but missing java info -
  sai has added a plugin  (mattw4, 18:28:03)
* need deep-dive into java resource consumption to understand better
  (mattw4, 18:28:46)
* LINK:
  https://snapshot.raintank.io/dashboard/snapshot/YGlm4u58NFJzMzgNhLYDiYGHX2pFKUdy?orgId=2&panelId=226&fullscreen&from=1503579973482&to=1503583573482&var-Cloud=odl-osp12-non-containerized&var-NodeType=*&var-Node=overcloud-controller-2&var-Interface=interface-br-ex&var-Disk=disk-sda&var-OSD=All&var-Pool=All&var-cpus0=All&var-cpus00=All
  (sai, 18:29:13)
* LINK:
  https://snapshot.raintank.io/dashboard/snapshot/emsmNggId7uHKq9S1r2xX5qhBew2GCP6?orgId=2&panelId=69&fullscreen&from=1503595117077&to=1503598717077&var-Cloud=odl-osp12-non-containerized&var-NodeType=*&var-Node=overcloud-controller-0&var-Interface=interface-br-ex&var-Disk=disk-sda&var-OSD=All&var-Pool=All&var-cpus0=All&var-cpus00=All
  (sai, 18:29:33)
* sai shared lovely grafana dashboards showing resource consumption
  (mattw4, 18:30:41)
* sai still seeing lots of slow queries to mysql  (mattw4, 18:33:47)
* mattw4 to take AR to follow up with LF  (mattw4, 18:34:25)
* mattw4 needs to learn more about robot  (mattw4, 18:35:24)
* mattw4 to ping sai for help on monitoring  (mattw4, 18:37:38)



Meeting ended at 18:37:57 UTC.



People present (lines said)
---------------------------

* mattw4 (22)
* sai (4)
* collabot (3)
* dfarrell07 (1)



Generated by `MeetBot`_ 0.1.4