14:00:40 <mackonstan1> #startmeeting 7-Jun FD.io CSIT Project Meeting 14:00:40 <collabot> Meeting started Wed Jun 7 14:00:40 2017 UTC. The chair is mackonstan1. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:00:40 <collabot> Useful Commands: #action #agreed #help #info #idea #link #topic. 14:00:40 <collabot> The meeting name has been set to '7_jun_fd_io_csit_project_meeting' 14:00:54 <jgelety> #info Jan Gelety 14:00:57 <selias> #info Samuel Elias 14:01:10 <pmikus1> #info Peter Mikus 14:04:05 <dwallacelf> #info Dave Wallace 14:04:30 <Tibor_> #info Tibor Frank 14:05:47 <mackonstan1> #topic Agenda bashing 14:10:10 <mackonstan1> #topic VPP 14:10:25 <mackonstan1> #info Dave - reporting - VPP API freeze tomorrow 14:11:08 <mackonstan1> #info DBarach raised a point re make test flow exporter tests taking time - Maciek will send email asking for suggestions 14:12:56 <mackonstan1> #info also spotted vpp-verify repeatedly failing - Matus reported Klement is looking at addressing the issues 14:13:04 <mackonstan1> #topic • VPP: adding VPP debug image tests to CSIT 14:13:19 <mackonstan1> #info Dave Wallace - will send proposal to csit-dev 14:13:34 <mackonstan1> #topic FD.io CSIT operational status 14:16:38 <mackonstan1> #topic LF infra, testing jobs: all working fine, incl. upgraded tb4-virl1 intro'ed yday into vpp-csit verify jobs 14:16:49 <mackonstan1> #topic open: un-pegging dpdk dev package version from CSIT, track vpp master instead? (email thread) 14:17:54 <mackonstan1> #info Sergio, Damjan, Maciek - suggesting to use make verify to address un-pegging dpdk dev package version from CSIT 14:18:11 <mackonstan1> #info Maciek - make test is not required, but can be masked in make verify 14:21:28 <mackonstan1> #info Peter - the issue is that make verify also installs newly build vpp on jenkins slave - CSIT needs to install newly built vpp into CSIT environment - VIRL or phy perf 14:23:59 <mackonstan1> #action Peter, Dave, Maciek - follow-up on un-pegging dpdk dev package from csit, use csit-dev thread by Ed Warnicke 14:24:01 <mackonstan1> #link https://lists.fd.io/pipermail/csit-dev/2017-May/002006.html 14:24:50 <mackonstan1> #topicTest environment updates 14:24:55 <mackonstan1> #topic Test environment updates 14:25:24 <mackonstan1> #topic adding test equipment: Netgate 14:25:34 <mackonstan1> #info no progress, need to poke on csit-dev 14:27:16 <mackonstan1> #topic VIRL platform 14:29:53 <mackonstan1> #topic VIRL-GetWellPlan, VIRL-Optimizations - see jira links 14:29:59 <mackonstan1> #info see jira links 14:30:58 <mackonstan1> #topic VIRL hosts upgrade - tb4-virl1 in production, remaining hosts: tb4-virl2, tb4-virl3 14:31:18 <mackonstan1> #info Peter updating - expecting tb4-virl2 to move into upgrade phase this week 14:31:40 <mackonstan1> #topic Centos7 VIRL tests - updates? 14:32:39 <mackonstan1> #info Jan updating - Centos7 qemu start issue - will raise with Thomas F Herbert 14:33:55 <mackonstan1> #topic CSIT jobs reorgs 14:34:28 <mackonstan1> #info Tibor update - pdrchk for triggered vpp-csit perf tests failiing - need to bisect the issue 14:37:04 <mackonstan1> #info Tibor to ping pointer to issues to Fabien, Maciek 14:37:17 <mackonstan1> #topic CSIT design optimizations 14:37:32 <mackonstan1> #link https://wiki.fd.io/view/CSIT/Design_Optimizations 14:38:26 <mackonstan1> #link https://gerrit.fd.io/r/#/c/6968/ 14:39:51 <mackonstan1> #info Tibor - csit-deve email for - remove unused and not needed KWs. 14:40:13 <mackonstan1> #topic csit1704 report 14:40:44 <mackonstan1> #topic csit1704 known issues resolution - updates? 14:41:13 <mackonstan1> #info no major updates - issues with i40e driver performance still persist 14:41:55 <mackonstan1> #action Maciek - send a TPL level email summary to csit-dev - asking for status update on all csit1704 known issues 14:45:25 <mackonstan1> #topic csit1707 plan 14:48:55 <mackonstan1> #info tracked on CSIT wiki high-level 14:49:10 <mackonstan1> #info real tracking on jira as part of epics: 1707-* 14:49:42 <mackonstan1> #info Matus to help coordinate 14:50:29 <mackonstan1> #topic VPP functional tests - virl 14:50:42 <mackonstan1> #info Sam adding number of HC performance tests 14:52:00 <mackonstan1> #info Peter comment - need to schedule dry run pef testing the HC CRUD API performance in physical machines, due their different nature vs. all other perf tests done with VPP that are focusing on data plane 14:52:24 <mackonstan1> #endmeeting