#fdio-csit: FD.io CSIT project meeting
Meeting started by mackonstan at 15:02:25 UTC
(full logs).
Meeting summary
-
- Jan Gelety (jgelety,
15:02:48)
- FD.io CSIT lab infrastructure (mackonstan, 15:04:00)
- Arm testbeds (mackonstan, 15:04:19)
- ThunderX2 server - Mohammed - system unstable,
keeps power cycling, firmware to be updated. (mackonstan,
15:12:13)
- Atom testbeds (mackonstan, 15:12:44)
- Denverton servers - Mohammed - 4 servers
received SYS-E300-9A (mackonstan,
15:15:13)
- Yulong showing phy wiring diagram - 2n-dnv
testbed, 1 testbed. (mackonstan,
15:21:02)
- Yulong to add: need to complete wiring exam in
CSIT testbed wiki. (mackonstan,
15:26:09)
- Honeycomb update (mackonstan, 15:28:45)
- Michal over slack: The only update for
honeycomb is that currently we are testing the possibility to
migrate to Java 11 for JVPP, Honeycomb and HC2VPP. (mackonstan,
15:28:54)
- Operational status (mackonstan, 15:29:37)
- Jan - no issues. (mackonstan,
15:30:05)
- Juraj - arm update - Bionic Arm VPP verify and
VPP merge jobs are now working and voting. WOOT! (mackonstan,
15:31:39)
- VPP code performance (mackonstan, 15:33:07)
- some tests failed due to keyword name collision
with new VPP_Device keywords - fixed now (mackonstan,
15:34:36)
- Releases (mackonstan, 15:36:03)
- last maintenance report CSIT-1810.49
published (mackonstan,
15:37:16)
- updated CSIT-1901 plan in wiki (mackonstan,
15:40:22)
- https://wiki.fd.io/view/CSIT/csit1901_plan
(mackonstan,
15:40:25)
- Developments (mackonstan, 15:40:43)
- VAT to PAPI (mackonstan, 15:41:45)
- proposal to start with l2util library.. and
then take it from there (mackonstan,
15:42:00)
- Jan - based on current work progress impossible
to complete migration to PAPI in 1901 (mackonstan,
15:43:07)
- Test environments (mackonstan, 15:43:29)
- Jan - testbeds are up, base tests are in and
working, need to include them into semi-weekly CSIT jobs.
(mackonstan,
15:45:20)
- Performance trending and analytics (mackonstan, 15:46:19)
- Anomaly detection (mackonstan, 15:46:30)
- updated detection to BMRR with increased
precision, more green and red circles, new url to be provided over
email by Vratko (mackonstan,
15:49:10)
- per VPP patch - Vratko - still analysing the
data, delayed by KW naming collision (mackonstan,
15:49:46)
- housekeeping (mackonstan, 15:53:19)
- no call next week due to Holidays (mackonstan,
15:53:28)
- next meeting 02-Jan (mackonstan,
15:53:39)
Meeting ended at 15:53:42 UTC
(full logs).
Action items
- (none)
People present (lines said)
- mackonstan (33)
- collabot` (3)
- jgelety (1)
Generated by MeetBot 0.1.4.