#fdio-meeting: FD.io CSIT Project Meeting

Meeting started by mackonstan at 14:02:28 UTC (full logs).

Meeting summary

    1. mackonstan (mackonstan, 14:02:46)
    2. Juraj Linkeš (jlinkes, 14:02:49)
    3. viliam luc (viliamluc, 14:02:51)
    4. Vratko Polak. (vrpolak, 14:02:52)
    5. Tibor Frank (tifrank, 14:02:56)

  2. Agenda bashing (mackonstan, 14:03:31)
  3. Inputs from projects - VPP (mackonstan, 14:04:02)
    1. Andrew - vpp v22.10 rls images being built, expecting all artefacts to be published by tomorrow morning UTC (mackonstan, 14:04:42)
    2. Andrew - see rls notes (last vpp patch in stable/2210) for rls content (mackonstan, 14:05:15)
    3. Andrew - comments about rls - few late arrivals, should be avoided in the future, required manual tweaking of rls notes; (mackonstan, 14:06:11)

  4. Inputs from projects - TSC (mackonstan, 14:06:40)
  5. E810 flow "offload" tests (mackonstan, 14:09:10)
    1. Maciek: gtpu and ipsec policy tests - perf results should be checked by the team/authors of the tests - they are now using new TRex v3.00 DPDK ICE driver (mackonstan, 14:09:57)

  6. Scale tests API optimization (mackonstan, 14:10:16)
    1. Vratko: merge of new patch is postponed until after csit-2210 (mackonstan, 14:10:36)

  7. VPP and CSIT hot items (mackonstan, 14:11:56)
    1. Maciek: 4 topics as list per agendfa (mackonstan, 14:14:53)
    2. Georgii: re Spectre-BHB security vulnerability - recommend to use unmerged patch for CVE-2022-0001 and CVE-2022-0002 verification (mackonstan, 14:16:34)
    3. Peter: preference to stick to released version, otherwise risking having to run multiple versions of the scripting tool (mackonstan, 14:21:37)
    4. Maciek: propose to verify the status of spectre-meltdown-checker, revisit the use as appropriate, consider other tools for checking the CPU level vulnerabilities - Georgii, Juraj, Peter and Maciek to drive it (mackonstan, 14:23:01)

  8. release CSIT-2210 (mackonstan, 14:28:07)
    1. Viliam: rc2 perf testing is complete, no new issues, data is published in draft report in master (mackonstan, 14:29:00)
    2. Tibor - new ver of draft report generated this morning (mackonstan, 14:29:18)
    3. https://s3-docs.fd.io/csit/master/report/ (mackonstan, 14:29:40)
    4. Tibor - Peter fixed/update TRex loopback tests for both aws/ena-dpdk and icx/cvl-dpdk-2p100GbE (mackonstan, 14:30:26)
    5. Tibor - ready for rls testing now (just finishing dpdk on arm testbeds) (mackonstan, 14:31:26)
    6. Tibor: report TODO list sent to csit-dev - add TRex-loopback for icx and rls notes (mackonstan, 14:32:24)
    7. Vratko: rls comparison and regressions - no big movements identified apart from af-xdp (but with high stdev). (mackonstan, 14:35:21)
    8. Tibor: ready to start testing vpp v22.10 rls builds as of tomrorow (mackonstan, 14:36:02)

  9. Physical Infrastructure (mackonstan, 14:36:24)
    1. Peter: re 3n-snr - sporadic link flaps between on both links between snr and icx servers - next step is to collect low level info about pluggables and firmware, provide to Georgii and Intel/IPS team (mackonstan, 14:39:36)

  10. Virtual Infrastructure, CI/CD (mackonstan, 14:41:18)
  11. Trending (mackonstan, 14:41:54)
    1. Viliam: didn't update the list, but there are four issues fixed csit-1857, -1864, -1862, -1860 (mackonstan, 14:42:57)
    2. Vratko: aware of another issue not tracked on the list - ipsec avf related, similar to csit-1827 (mackonstan, 14:46:29)


Meeting ended at 14:46:57 UTC (full logs).

Action items

  1. (none)


People present (lines said)

  1. mackonstan (34)
  2. collab-meetbot` (9)
  3. jlinkes (1)
  4. viliamluc (1)
  5. vrpolak (1)
  6. tifrank (1)


Generated by MeetBot 0.1.4.