#fdio-meeting: FD.io CSIT Project Meeting
Meeting started by mackonstan at 15:05:38 UTC
(full logs).
Meeting summary
-
- Maciek Konstantynowicz (mackonstan,
15:05:46)
- Vratko Polak. (vrpolak,
15:05:54)
- Agenda bashing (mackonstan, 15:05:55)
- Tibor Frank (tifrank,
15:06:20)
- Inputs from projects - VPP (mackonstan, 15:07:44)
- Vratko - yday vpp-dev call, Lijian/Arm team are
working on crypto acceleration, expecting 20-30% improvement for
openssl library(?), will be pushing a patch to gerrit shortly
(mackonstan,
15:09:21)
- Inputs from projects - TSC (mackonstan, 15:10:09)
- Maciek - DaveW submitted lab extension budget
request for cy2024 - see list of equipment noted in one of the
previous meetings (mackonstan,
15:12:00)
- Intel HW accelerator status (mackonstan, 15:12:51)
- SPR QAT2.0 - Kai Peter updating - Peter it is
enabled on 2n-spr and can test initialization and uni-dir tests.
3na-spr and 3nb-spr require work due to the use of containers to run
VPP instances per NUMA. (mackonstan,
15:17:21)
- Dave Wallace (dwallacelf,
15:19:02)
- SNR QAT2.0 - Peter - CPU replacement units
arrived, Ali/Vexxhost replaced them, and working on getting them to
boot successfully. (mackonstan,
15:20:48)
- aside - ICX-D installation - Peter Maciek Dave
- Ali/Vexxhost had to use cables from Vexxhost stock, we should
replenish "used 2X 4 ft DAC SFP28cables brand 10GTEK", $15 on
amazon. (mackonstan,
15:22:43)
- agreement - Dave Maciek Peter - ask
Ali/Vexxhost to invoice LFN for those cables (mackonstan,
15:24:33)
- agreement2 - Peter Dave Maciek - agreed to
involve Vexxhost team in reviewing the new equipment shipping list
and highlight any gaps (e.g. rail kits, cables) before the shipment
is executed (mackonstan,
15:26:20)
- SPR DSA - Peter - BIOS v1.4 - updated on all
devices 2n-, 3n-spr. 1n-spr to be updated. DSA enabled in
BIOS. (mackonstan,
15:27:37)
- 1n-spr work (mackonstan, 15:29:18)
- Peter - interfaces vanishing issue fixed,
currently only IP timeouts happening sporadically (mackonstan,
15:30:11)
- back to hw accelerator - E810 GTPU offload -
Kai - intel team are looking at the results (mackonstan,
15:31:37)
- CSIT-2310 (mackonstan, 15:31:56)
- testing - Tibor - all testing done incl.
coverage, last testbed was 3nb-spr, finished earlier today. okay for
LFN IT Vanessa to go ahead with maintenance of: Production Jenkins,
Sandbox Jenkins, Gerrit, Jira. (mackonstan,
15:33:37)
- Tibor - 23.10 rls was very smooth, few hiccups
fixed by Peter. no unexpected bugs or issues from Tibor's
perspective. (mackonstan,
15:34:38)
- Tibor - latest coverage results will be
published in csit.fd.io by tomorrow, as it takes up to 24hrs to
process all results. (mackonstan,
15:35:25)
- Tibor - during rls testing implemented some
improvements based on Vratko's feedback: i) more info for
off-canvas, ii) more links, few more (total of 5 small
tweaks). (mackonstan,
15:36:50)
- Tibor - added GSO tests to iterative tests -
see rls report data - box graphs, comparison. (mackonstan,
15:37:36)
- Vratko - in documentation rls notes - new
issues got added - total of 8 issues (mackonstan,
15:39:10)
- https://csit.fd.io/cdocs/release_notes/current/vpp_performance/#new
(mackonstan,
15:39:22)
- CSIT-1935 rls 2310: Zero traffic reported in
udpquic tests (mackonstan,
15:41:36)
- https://jira.fd.io/browse/CSIT-1935
(mackonstan,
15:41:44)
- CSIT-1936 TRex occasionally sees link down in
L2 perf tests (mackonstan,
15:42:43)
- https://jira.fd.io/browse/CSIT-1936
(mackonstan,
15:42:53)
- CSIT-1937 Small but frequent loss in ASTF UDP
on cx7 mlx5 (mackonstan,
15:45:27)
- https://jira.fd.io/browse/CSIT-1937
(mackonstan,
15:45:34)
- CSIT-1938 3n-alt: high scale ipsec policy tests
may crash VPP (mackonstan,
15:45:40)
- https://jira.fd.io/browse/CSIT-1938
(mackonstan,
15:45:45)
- CSIT-1939 3na-spr: VPP fails to start in first
two test cases (mackonstan,
15:45:52)
- https://jira.fd.io/browse/CSIT-1939
(mackonstan,
15:45:58)
- CSIT-1940 hardware acceleration does not work
yet (mackonstan,
15:46:05)
- https://jira.fd.io/browse/CSIT-1940
(mackonstan,
15:46:12)
- CSIT-1941 TRex may wrongly detect link
bandwidth (mackonstan,
15:46:17)
- https://jira.fd.io/browse/CSIT-1941
(mackonstan,
15:46:23)
- CSIT-1942 3nb-spr hoststack: interface not up
after first test (mackonstan,
15:46:30)
- RCAs - Vratko - 2 RCAs (mackonstan,
15:48:00)
- https://csit.fd.io/cdocs/release_notes/current/vpp_performance/#root-cause-analysis-for-regressions
(mackonstan,
15:48:02)
- Virtual Infrastructure, CI/CD (mackonstan, 15:49:10)
- maintenance tomorrow - see note above
(mackonstan,
15:49:17)
- Environment, Framework (mackonstan, 15:49:38)
- 39549: interface dpdk avf: introducing setting RSS hash key
feature | https://gerrit.fd.io/r/c/vpp/+/39549
(vrpolak,
15:50:28)
- Vratko - someone added an VPP API call to set
the RSS key (mackonstan,
15:50:40)
- 39549: interface dpdk avf: introducing setting RSS hash key
feature | https://gerrit.fd.io/r/c/vpp/+/39549
(mackonstan,
15:50:42)
- Vratko - running gerrit triggered perf tests
that are still to complete (mackonstan,
15:51:20)
- re CLI without API - Vratko - some virtio vhost
tests failing due to typo - fix is easy, see jira ticket
(mackonstan,
15:52:57)
- https://jira.fd.io/browse/VPP-2088
virtio: bad CLI argument parsing introduced with tx-queue-size (mackonstan,
15:53:37)
- Trending and Test Failures (mackonstan, 15:54:11)
- Tibor - we are back to normal trending
cadence (mackonstan,
15:54:37)
- Vratko - still working through the coverage
results, will move to trending next (mackonstan,
15:55:43)
- Vratko - for the first time no difference
observed between release and trending testing. (mackonstan,
15:56:23)
Meeting ended at 15:56:43 UTC
(full logs).
Action items
- (none)
People present (lines said)
- mackonstan (56)
- collab-meetbot (3)
- vrpolak (2)
- tifrank (1)
- dwallacelf (1)
Generated by MeetBot 0.1.4.