14:03:45 <mackonstan> #startmeeting FD.io CSIT project meeting 22-Aug
14:03:45 <collabot`> Meeting started Wed Aug 22 14:03:45 2018 UTC.  The chair is mackonstan. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:03:45 <collabot`> Useful Commands: #action #agreed #help #info #idea #link #topic.
14:03:45 <collabot`> The meeting name has been set to 'fd_io_csit_project_meeting_22_aug'
14:03:52 <mackonstan> #topic Agenda bashing
14:03:54 <mgradzki> #info Marek Gradzki
14:03:55 <mackonstan> #chair
14:03:55 <collabot`> Current chairs: mackonstan
14:03:59 <jlinkes> #info Juraj Linkeš
14:04:01 <mackonstan> #info Maciek
14:04:41 <mackonstan> #topic Inputs from LFN and FD.io projects
14:05:52 <jgelety> #info Jan Gelety
14:07:05 <mackonstan> #info VPP - 18.10 schedule confirmed, cmake introduced, DPDK ver bumped to 18.08.
14:12:51 <mackonstan> #info Dave Wallace asking about options for testing VPP stack, in a CI/CD fashion.
14:13:16 <mackonstan> #info VPP stack = incl. tcp stack and VCL.
14:14:56 <mackonstan> #into HoneyComb - Mark updating: discussing 18.10 rls plan, PTL change s/Marek Gradzki/Michal Cmarada/
14:15:27 <mackonstan> #info DMM - no udpate
14:15:39 <mackonstan> #info TSC - no update
14:16:01 <mackonstan> #topic CSIT operational status
14:17:50 <vrpolak> #info Package cloud link example: https://packagecloud.io/app/fdio/master/search?dist=ubuntu%2Fbionic
14:18:31 <mackonstan> #topic VPP code performance:
14:27:07 <mackonstan> #info lbdpdk tests keep failiing, need to add to the list of CSIT-18.07 known issues.
14:31:03 <jgelety> #info lbdpdk tests keep failing in case of x710 NICs, tests are OK in case of x520 NICs
14:31:40 <mackonstan> #info Skylake trending - trending jobs running and collecting data
14:33:29 <mackonstan> #info see csit-dev email from Tibor "Subject: [csit-dev] Proposed changes in Trending"
14:35:12 <mackonstan> info Vratko: we should add comparisons of trending data, similar to what's done in the last report e.g. 3n-skx vs. 3n-hsw, 3n-skx vs. 2n-skx, same test on different NICs, etc.
14:35:24 <mackonstan> #info Vratko: we should add comparisons of trending data, similar to what's done in the last report e.g. 3n-skx vs. 3n-hsw, 3n-skx vs. 2n-skx, same test on different NICs, etc.
14:36:01 <mackonstan> #topic Releases
14:39:41 <mackonstan> #info CSIT-18.07 known issues updates
14:40:13 <mackonstan> #info CSIT-1245 - now fixed by Peter.
14:40:42 <mackonstan> #info VPP-1361 - still open, waiting for feedback from vpp-dev.
14:46:45 <mackonstan> #info CSIT-18.10 plan - dates to follow VPP, report 2 weeks after vpp release, not one as before.
14:46:55 <mackonstan> #link https://wiki.fd.io/view/CSIT/csit1810_plan
14:47:15 <mackonstan> #info add clarify text to CSIT release "code freeze".
14:47:48 <mackonstan> #topic Framework
14:50:04 <mackonstan> #info VPP_Device 1-Node testbed design - next TWS call tomorrow
14:50:14 <mackonstan> #link https://wiki.fd.io/view/CSIT/VPP_Device_test_design
14:50:33 <mackonstan> #info Maciek to add logical drawings by COB today, before TWS call
14:50:56 <mackonstan> #info Ed, Peter making progress with Container orchestration POC.
14:51:30 <mackonstan> #info VPP_Path: CSIT_VIRL migration progress
14:51:55 <mackonstan> #info Ed disabled vpp-csit-verify-virl jobs not voting anymore in vpp project, but still running
14:52:21 <mackonstan> #info VIRL P2s are for grabs to port to make test
14:52:57 <mackonstan> #info 3-Node and 2-Node Skx testbeds - RMAed HW ETAs not known
14:54:00 <mackonstan> #info Arm performance testbed - Sirshak - testing Tishan testbeds (short testbed name). jumbo frame testbeds failing, others progressing..
14:54:50 <mackonstan> #info Sirshak to provide a wiki link with status summary, and jira tickets to be created and tracked
14:57:44 <mackonstan> #info Sirshak reported issues with VPP Mlx drivers for Arm
14:58:26 <mackonstan> #info Sirshak to send summary of issues to csit-dev
14:59:11 <mackonstan> #topic VPP code performance monitoring
14:59:29 <mackonstan> #info Per VPP patch vs. parent performance tests and voting - Vratko to send status update at some point
15:01:30 <mackonstan> #endmeeting