14:02:42 <mackonstan1> #startmeeting FD.io CSIT Project Meeting 5-Apr
14:02:42 <collabot`> Meeting started Wed Apr  5 14:02:42 2017 UTC.  The chair is mackonstan1. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:02:42 <collabot`> Useful Commands: #action #agreed #help #info #idea #link #topic.
14:02:42 <collabot`> The meeting name has been set to 'fd_io_csit_project_meeting_5_apr'
14:02:48 <jgelety1> # info Jan Gelety
14:02:51 <mackonstan1> #chair
14:02:51 <collabot`> Current chairs: mackonstan1
14:02:53 <dwallacelf> #info Dave Wallace
14:02:55 <mklotton1> #info Matej Klotton
14:02:59 <mackonstan1> #topic agenda bashing
14:03:49 <mackonstan1> #topic Inputs from LF and FD.io projects
14:07:01 <mackonstan1> #info test escape affecting tunnel overlays - not caught by make test and by virl
14:07:33 <mackonstan1> #info one way to catch it is to run multi-thread setup in virl
14:12:23 <mackonstan1> #info the ask is to test vpp worker thread setup in virl as part of verify job
14:13:29 <Tibor_> #info Tibor Frank
14:14:56 <mackonstan1> #info need to re-check the max capacity of VIRL test executions in VIRL testbeds
14:15:23 <mackonstan1> #topic FD.io CSIT operational status
14:18:34 <mackonstan1> #info Peter noticed that VIRL simulations are not starting up every now and again - due to garbage leftovers
14:19:20 <mackonstan1> #info need to automate garbage collection on VIRL servers, as captured in the link belo
14:19:30 <mackonstan1> #link https://wiki.fd.io/view/CSIT/CSIT_LF_VIRL_testbed
14:23:22 <mackonstan1> #info agreement to plan for addressing the issue with garbage collection as part of VIRL system refresh, planned for post csit 1704 rls
14:23:49 <mackonstan1> #topic Centos7 VIRL tests instability - status
14:27:06 <tfherbert> #info Tom Herbert
14:28:20 <mackonstan1> #info Matej reporting - Centos7 based VM-SUT1 gets shutdown - still t/shooting - Tom+Matej to continue …
14:29:33 <mackonstan1> #topic csit1701 follow-ups
14:37:15 <mackonstan1> #topic csit1704_plan
14:42:58 <pmikus1> #link https://jenkins.fd.io/view/csit/job/csit-dpdk-perf-master-all/plot/
14:45:31 <mackonstan1> #info testpmd xl710 reporting NDR 16Mpps and PDR 37Mpps - NDR too low, indicating some issues with card, driver, or environnmet
15:02:28 <mackonstan1> #topic Perf jobs reorg
15:03:16 <mackonstan1> #info proposal out
15:03:27 <mackonstan1> https://lists.fd.io/pipermail/csit-dev/2017-April/001801.html
15:14:39 <mackonstan1> #info Dave confirmed that modularizing jobs is the right approach forwardd
15:24:38 <mackonstan1> #info more detailed discussion into the mailer csit-dev
15:24:42 <mackonstan1> #endmeeting