#opendaylight-genius: Genius Weekly 01/17/19

Meeting started by faseelak at 15:04:16 UTC (full logs).

Meeting summary

    1. shague (shague, 15:04:21)

  2. Roll call and agenda bashing (please #info <your-nick>) (faseelak, 15:04:25)
    1. https://jira.opendaylight.org/browse/GENIUS-265 (faseelak, 15:04:37)

  3. odlparent and mdsal version bump status (faseelak, 15:06:21)
    1. Hema (Hema_, 15:06:50)
    2. https://jira.opendaylight.org/browse/TSC-186 (faseelak, 15:07:00)
    3. Vishal (vthapar, 15:07:18)
    4. genius CSIT seems to be passing today on master (faseelak, 15:07:19)
    5. skitt mentioned he will apply some workarounds in distribution so that projects can move forward with their development (faseelak, 15:08:26)
    6. openflowplugin CSIT is still broken (faseelak, 15:11:13)
    7. shague says some eod related errors are seen in netvirt CSIT as well (faseelak, 15:11:27)

  4. JIRA GENIUS-254 (faseelak, 15:14:49)
    1. faseelak has pushed a patch to fix the problem, but getting very difficult to reproduce the exact race condition (faseelak, 15:15:38)
    2. needs help from jamo and vthapar to run netvirt CSIT on this patch repeatedly to see if it solves the problem (faseelak, 15:16:18)

  5. enable itm-direct-tunnels by default (faseelak, 15:16:53)
    1. itm-direct-tunnels feature was introduced in oxygen to improve the scalability of ITM full-mesh of tunnels (faseelak, 15:17:58)
    2. this feature is disabled by default upstream (faseelak, 15:19:26)
    3. discussion is on whether it will be okay to enable by default (faseelak, 15:19:40)
    4. shague says the netvirt itm direct tunnels job fails 1/4 times (faseelak, 15:19:55)
    5. so if the CSIT job can be stabilized, we can consider making this default (faseelak, 15:20:20)
    6. netvirt itm-direct-tunnels job fails consistently so can't make it the default (shague, 15:22:18)
    7. https://jenkins.opendaylight.org/releng/view/netvirt-csit/job/netvirt-csit-1node-0cmb-1ctl-2cmp-openstack-queens-upstream-stateful-itm-direct-tunnels-neon/ (shague, 15:22:21)
    8. vthapar asks to check if it is anything wrong with odltools where tunnel verification is being done. (faseelak, 15:22:47)
    9. targeting the change for only Neon (faseelak, 15:24:35)
    10. neon might be Ok, fluorine had the issue with the itm-direct-tunnels (shague, 15:25:17)
    11. patches just cherry-picked to fluorine, need to watch fluorine now if stable (shague, 15:25:49)
    12. some of the Neon fixes for itm-direct-tunnels were cherry-picked to stable/fluorine only recently (faseelak, 15:26:44)
    13. need to closely monitor the stable/fluorine job to see if things are stable in the coming days (faseelak, 15:28:34)
    14. vthapar says we have to notify sfc also to see if they have any concerns (faseelak, 15:28:55)

  6. OFTunnels (faseelak, 15:30:09)
    1. code patches are up for review for non-BFD usecases (faseelak, 15:30:53)
    2. CSIT work is also in progress by an intern (faseelak, 15:31:02)
    3. COE will need OFTunnels for supporting the kubernetes scale (faseelak, 15:31:47)
    4. COE can continue in the short-term without needing oftunnels (shague, 15:32:24)

  7. genius CSIT for 3 OVS (faseelak, 15:33:38)
    1. shague and faseelak reviewed the patch, comments to be incorporated by Karthika (faseelak, 15:34:00)
    2. karthika to do a copy-logs on the patch (faseelak, 15:35:04)


Meeting ended at 15:36:47 UTC (full logs).

Action items

  1. (none)


People present (lines said)

  1. faseelak (34)
  2. shague (6)
  3. odl_meetbot (4)
  4. vorburger (2)
  5. vthapar (2)
  6. Hema_ (1)


Generated by MeetBot 0.1.4.