15:00:19 #startmeeting FD.io CSIT project meeting 15:00:19 Meeting started Wed Nov 18 15:00:19 2020 UTC and is due to finish in 60 minutes. The chair is mackonstan. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:19 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:19 The meeting name has been set to 'fd_io_csit_project_meeting' 15:00:34 #info Dave Wallace 15:00:34 #info Juraj Linkeš 15:00:41 #info Tibor Frank 15:00:46 #info Vratko Polak. 15:00:48 #chair 15:00:48 Current chairs: mackonstan 15:01:06 #chair dwallacelf 15:01:06 Warning: Nick not in channel: dwallacelf 15:01:06 Current chairs: dwallacelf mackonstan 15:01:14 #info Jan Gelety 15:01:22 #chair tifrank 15:01:22 Warning: Nick not in channel: tifrank 15:01:22 Current chairs: dwallacelf mackonstan tifrank 15:01:37 #chair jgelety 15:01:37 Warning: Nick not in channel: jgelety 15:01:37 Current chairs: dwallacelf jgelety mackonstan tifrank 15:01:45 #chair vrpolak 15:01:45 Warning: Nick not in channel: vrpolak 15:01:45 Current chairs: dwallacelf jgelety mackonstan tifrank vrpolak 15:01:52 #chair pmikus 15:01:52 Warning: Nick not in channel: pmikus 15:01:52 Current chairs: dwallacelf jgelety mackonstan pmikus tifrank vrpolak 15:02:37 #topic Agenda bashing 15:03:34 #info Juraj, want to talk about perf tests on Arm/Tishan, and how to make them execute faster. 15:04:03 #topic Inputs from projects - VPP 15:04:34 #info Andrew, reminder sent F0 in 3 weeks 15:07:50 #info Maciek, questions from community about vpp memory/heap/hugepages configuration, asking for clearer usage guidelines. Looking for volunteers to work with me on that. 15:08:09 #topic Inputs from projects - TSC 15:08:29 #info Vratko, CSIT master branch is using GPL license in GPL directory. 15:09:55 #info Vratko, VPP master is still in works, RayK holds the pen to complete drawing the plan to conclude the Scapy driven part of make test GPL re-license. 15:10:28 #topic Releases CSIT-2101 15:11:22 #info F0 milestone 2020-12-09 15:12:03 #info Vratko, 2101 is the first rls with new VPP API deprecation process kicks in 15:13:23 #info Vratko, one safety net is voting vpp-csit-device job in vpp 15:16:05 #info Andrew, here is the pointer to **all** APIs that will get removed in v21.01, https://docs.fd.io/vpp/20.09/de/d1e/release_notes_2009.html > Newly deprecated API messages 15:17:11 #info Maciek, Jan, obvious CSIT test exposure: ipsec_tunnel_if_add_del 15:17:48 #info Maciek, Jan, Vratko to skin this cat. 15:18:32 #topic Releases CSIT-2009 15:19:05 #info Tibor, no new data/content today, so no report update 15:20:15 #info Tibor, nat44 test results and methodology, are ready to be pushed into rls2009 report. Data is already published in the master version. 15:21:47 #info Tibor, will wait till tomorrow evening to see if the gerrit methodology secions' review is complete, and if it is, publish the report 15:24:43 #info Tibor, started to test 2n-zn2 … 15:25:05 #info Peter, BUT, there is an issue with uploading artefacts to Nexus 15:26:32 #info Dave, discussed with Vanessa, unclear where the problem lies 15:30:58 #info Dave, jenkins CI is now down for all vpp and csit jobs for 24hrs 15:31:27 #info Dave, will bring it up on TSC call tomorrow 15:31:47 #info Vratko, sandbox Jenkins works though 15:35:25 #info Dave, will ask for a full visibility into Jenkins configs (on TSC call tomorrow), as currently blind 15:35:44 #topic Physical Infrastructure - Xeon, Atom 15:36:38 #info Maciek, Peter, will discuss offline best way to install 6x E810 NICs in 2n-clx testbeds 15:37:00 #topic Physical Infrastructure - Arm testbeds 15:37:45 #info Juraj, why do perf tests twice as long to execute on Arm/Tishan vs Xeon, and how to make them execute faster. 15:38:14 #info Juraj, one possible reason is that they are running on Arm executors that are slower 15:38:42 #info Juraj, but switching to ThunderX2s, fast servers, the jobs still take longer 15:39:18 #info Juraj, but turning off SMT on ThunderX2 executors, brought the execution time in line with Xeons 15:40:09 #info Juraj, root caused to slower cores on ThunderX1s. ThunderX2s are fine. 15:41:27 #info Juraj, two options to address this going forward: i) use Xeon executors, would need to add Arm arch support; ii) use ThunderX2 servers as executors, but there are only two servers. 15:43:56 #info Vratko, Peter, clearly prefer option ii) due to its simplicity, e.g. avoids cross-compiling 15:46:29 #info Juraj, email csit-dev to drive consensus 15:47:42 #topic Physical Infrastructure - AMD 15:49:02 #info Maciek, looking good. 15:55:30 #info Maciek, re FD.io lab extension in 2021, would like to propose to set up TWS starting next week 15:55:51 #info Maciek, expecting data points re existing costs for next week 15:56:05 #topic Virtual Infrastructure, CI/CD 15:56:29 #info Juraj, arm-based vpp-csit-device jobs are in and working, but not voting 15:57:54 #info Maciek, let's wait for a good green period before enabling voting 15:58:51 #info Agreed to add this as a standing point to agenda 15:59:02 #topic Environment, Framework 15:59:16 #info Maciek, ip6 latency tests broken, TRex ver at fault 15:59:45 #info Vratko, Peter, TRex 2.86 works, so moving to that 16:01:57 #topic Performance Tests 16:02:04 #info Vratko, nothing new since last week 16:02:24 #endmeeting