=========================================
#fdio-meeting: FD.io CSIT Project Meeting
=========================================


Meeting started by mackonstan at 14:03:53 UTC.  The full logs are
available at
http://ircbot.wl.linuxfoundation.org/meetings/fdio-meeting/2022/fd_io_csit_project_meeting/fdio-meeting-fd_io_csit_project_meeting.2022-08-03-14.03.log.html
.



Meeting summary
---------------

* Tibor Frank  (tifrank, 14:04:06)
* Fan Zhang  (fan_zhang, 14:06:46)
* mackonstan  (mackonstan, 14:07:35)
* Agenda bashing  (mackonstan, 14:08:21)

* Inputs from projects - VPP  (mackonstan, 14:08:30)

* Inputs from projects - TSC  (mackonstan, 14:08:52)

* New or being refactored tests  (mackonstan, 14:09:04)
  * Tibor, both set of tests added to trending and report, gtpu and
    ipsec policy tests  (mackonstan, 14:10:14)
  * Fan, Yulong sent an email highlighting an issue with gtpuhw and
    gtpusw tests performance. Maciek, Peter, to following on the
    reported 3n-icx TRex issue for 25GbE gtpu tests.  (mackonstan,
    14:13:43)

* Scale tests revisited  (mackonstan, 14:14:21)
  * Vratko, 1st version of code working okay, but it's introducing some
    new issues: PAPI related failures, ipsec keyword issue
    (non-blocking, affecting only ipsec tests).  (mackonstan, 14:17:40)

* VPP and CSIT hot items  (mackonstan, 14:18:05)
  * Maciek, re Spectre-BHB security vulnerability, the PR we were
    tracking is not merged, may be as well de facto abandoned, albeit
    it's still listed as active  (mackonstan, 14:21:23)
  * Peter, CSIT should not be integrated potential abandoned patch of
    spectre-meltdown-checker  (mackonstan, 14:22:31)
  * Fan, will check internally if any guidance from the team how to
    proceed here  (mackonstan, 14:22:51)

* Previous Release - CSIT-2206  (mackonstan, 14:23:47)
  * Tibor, small addition to report: added graphs for 2n-tx2 container
    tests, as data available  (mackonstan, 14:24:23)

* Next Release - CSIT-2210  (mackonstan, 14:24:50)
  * LINK: https://wiki.fd.io/view/CSIT/csit2210_plan  (mackonstan,
    14:25:27)
  * Peter, Ubuntu 22.04 integration is in progress  (mackonstan,
    14:26:36)
  * Peter, will be starting with performance testbeds next week
    (mackonstan, 14:26:51)
  * Fan, known incompatibility with ubuntu 22.04 and vfio-pci binding to
    intel qat, impacts only 3n-snr testbed  (mackonstan, 14:29:48)
  * LINK: 36785: dpdk: bump to 22.07 |
    https://gerrit.fd.io/r/c/vpp/+/36785  (mackonstan, 14:30:56)
  * Peter, CSIT compatibility matrix will follow rls notes of dpdk
    v22.07  (mackonstan, 14:36:54)
  * LINK:
    https://github.com/FDio/csit/blob/master/fdio.infra.ansible/roles/intel/defaults/main.yaml
    (mackonstan, 14:36:58)
  * LINK:
    https://doc.dpdk.org/guides/rel_notes/release_22_07.html#tested-platforms
    (mackonstan, 14:38:00)
  * phasing of env upgrade in perf testbeds, ubuntu 22.04 first, dpdk
    next, have trending data available in-between  (mackonstan,
    14:39:21)

* Physical Infrastructure  (mackonstan, 14:39:37)
  * SNR testbed is still waiting for SFP28 25GbE pluggables
    (mackonstan, 14:40:28)

* Virtual Infrastructure, CI/CD  (mackonstan, 14:40:50)
  * Peter, working on upgrading Nomad cluster to ubuntu v22.04, incl. ip
    address renumbering and dns renaming  (mackonstan, 14:41:40)
  * Peter, 23 nodes in the cluster, some server being obsoloeted, so
    lots of work, almost done, Dave and Juraj involved  (mackonstan,
    14:43:15)

* Environment, Framework  (mackonstan, 14:44:12)
  * Vratko, preparing bump of VPP APIs for PAPI, planning this work for
    next week  (mackonstan, 14:44:57)

* Trending  (mackonstan, 14:45:24)
  * Viliam, re failures, there are 5 long-term issues (last 5 on the
    list on wiki)  (mackonstan, 14:46:10)
  * Viliam, re 2nd failure, CSIT-1850, 2n-dnv: sporadic 1518B tput tests
    failing to establish required sessions, Vratko - not considering it
    important.  (mackonstan, 14:48:26)
  * Viliam, re 3rd failure, CSIT-1827, 3n-icx, 3n-skx: all AVF crypto
    tests sporadically fail. 1518B with no traffic, Vratko - in some job
    runs it works fine, in some other it doesn't with all crypto tests
    failing in the affected run  (mackonstan, 14:51:08)



Meeting ended at 15:00:04 UTC.



People present (lines said)
---------------------------

* mackonstan (37)
* collab-meetbot` (3)
* tifrank (1)
* fan_zhang (1)
* dwallacelf (1)



Generated by `MeetBot`_ 0.1.4