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


Meeting started by mackonstan at 14:00:34 UTC.  The full logs are
available at
http://ircbot.wl.linuxfoundation.org/meetings/fdio-meeting/2023/fd_io_csit_project_meeting/fdio-meeting-fd_io_csit_project_meeting.2023-05-17-14.00.log.html
.



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

* Juraj Linkeš  (jlinkes, 14:02:05)
* Agenda bashing\  (mackonstan, 14:02:10)
  * Vratko Polak.  (vrpolak, 14:02:13)

* Inputs from projects  (mackonstan, 14:03:35)
  * mackonstan - on behalf of Dave W - observed intermittent failures,
    1-2 a day, of vpp-csit-device-verify jobs  (mackonstan, 14:05:06)
  * Peter - discussed with Dave - agreed to switch to non-critical the
    vhost tests that are intermittently failing, something to do with
    runtime memory allocation most probably  (mackonstan, 14:06:31)
  * Peter - failures are not observed in csit-vpp-verify-device tests
    (mackonstan, 14:06:52)
  * mackonstan - from TSC: Dave has submitted fd.io project summary
    readout slides to LFN last Thu  (mackonstan, 14:08:07)

* vpp bump to dpdk 23.03  (mackonstan, 14:09:06)
  * Peter - all infra work done per patches  (mackonstan, 14:09:30)
  * Peter - need to upgrade csit dpdk tests to use dpdk 23.03
    (mackonstan, 14:09:47)
  * Juraj - all arm infra also got upgraded, had to move NICs to x86
    machines to complete the upgrade, good coordination with vexxhost
    (ALi)  (mackonstan, 14:11:11)
  * Vratko - there were some intermittent failures during the upgrade
    process, now all good  (mackonstan, 14:11:52)
  * Vratko - however issues that were expected to get fixed, didn't
    (mackonstan, 14:12:16)

* CSIT-2306  (mackonstan, 14:13:33)
  * Vratko - reassembly tests, working through issues  (mackonstan,
    14:13:46)
  * Vratko - current vpp code not thread-safe, opened issue with vpp-dev
    (mackonstan, 14:14:19)
  * Vratko - also issues with ipsec policy mode, interface mode is fine
    (mackonstan, 14:14:55)
  * Vratko - mtu changes also ready  (mackonstan, 14:15:15)

* Physical Infrastructure  (mackonstan, 14:15:59)
  * Peter - 3vnb-spr ready and loaded into Jenkins CI, producing test
    data - see csit.fd.io  (mackonstan, 14:17:49)
  * Peter - 3na-spr being worked through  (mackonstan, 14:18:40)
  * Peter - 3na-spr - trex not complaining about NICs in different NUMAs
    (mackonstan, 14:19:20)
  * Peter - 3na-spr - trex support cx7 - compiled from source - testing
    it  (mackonstan, 14:19:52)
  * mackonstan - Hanoh from trex team confirmed that once Peter confirms
    cx7 works, trex will include it in the new rls version  (mackonstan,
    14:21:29)
  * Peter - re cx6 and remaining cx7 nics - waiting for inventory
    confirmation, and then will update the allocation per testbed
    (mackonstan, 14:23:24)
  * Juraj - cx6 nics into 3n-alt - see 38717: docs: 3n-alt mlx cx6 lab
    spec | https://gerrit.fd.io/r/c/csit/+/38717  (mackonstan, 14:25:04)
  * LINK: https://gerrit.fd.io/r/c/csit/+/38717  (mackonstan, 14:25:14)
  * Kai - discussion about enabling DSA in SPR using updated SuperMicro
    BIOS  (mackonstan, 14:28:30)
  * mackonstan - re ordering more SPRs and ICX-D - waiting for updates
    from Intel team  (mackonstan, 14:32:20)

* Trending and Test Failures  (mackonstan, 14:32:59)
  * Vratko - one new issue, one new & fixec  (mackonstan, 14:33:15)
  * Vratko - on 19 known issues  (mackonstan, 14:34:34)



Meeting ended at 14:35:06 UTC.



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

* mackonstan (34)
* collab-meetbot (9)
* jlinkes (1)
* vrpolak (1)
* dwallacelf (0)
* tifrank (0)



Generated by `MeetBot`_ 0.1.4