#opendaylight-integration: integration

Meeting started by dfarrell07 at 17:01:06 UTC (full logs).

Meeting summary

    1. LuisGomez creating a custom distribution where openflowplugin versions are swapped. (jamoluhrsen, 17:05:07)
    2. the plan is to test this distribution with CSIT and let projects test manually and see how badly things break (jamoluhrsen, 17:07:30)
    3. so far VTN and VPNservice have noted that their projects depend heavily on their current plugin so we should expect issues with them switching (jamoluhrsen, 17:08:10)
    4. LuisGomez found issues when swapping plugins for l2switch (jamoluhrsen, 17:08:37)
    5. jamoluhrsen says worse case is that we can just document which features cannot coexist. (jamoluhrsen, 17:12:04)
    6. anipbu noting that the documentation route needs to be last resort (jamoluhrsen, 17:12:19)
    7. we need a process for preventing these multi-impl feature problem. (jamoluhrsen, 17:18:40)
    8. snbi has a patch to add to distribution (jamoluhrsen, 17:22:30)
    9. https://lists.opendaylight.org/pipermail/snbi-dev/2016-January/000626.html (anipbu, 17:22:50)
    10. https://git.opendaylight.org/gerrit/33217 (anipbu, 17:23:05)
    11. faas wants back in distribution (jamoluhrsen, 17:24:04)
    12. dfarrell07 intern project posted, lots of interest so far (jamoluhrsen, 17:24:41)
    13. new pre-release expiremental rpm released (consumed by opnfv installers) (jamoluhrsen, 17:25:37)
    14. question about these experimental autorelease builds. do they happen? (jamoluhrsen, 17:25:59)
    15. zxiiro says it's not happening, but we can probably trigger it. (jamoluhrsen, 17:26:32)
    16. zxiiro says it *IS* triggering distribution-test. (jamoluhrsen, 17:27:05)
    17. hackfest coming at the end of Feb. 29th and march 1st. (jamoluhrsen, 17:28:40)
    18. rackspace was hitting a bug in neutron. (if slave comes up in error, it does not release ports). (jamoluhrsen, 17:29:27)
    19. so over time, we ran out of ports and no VMs could get networks so most jobs were aborting. (jamoluhrsen, 17:29:51)
    20. rackspace cleared the stuck ports yesterday. going forward we have a monitor script to let us know if we have stuck ports. (jamoluhrsen, 17:30:18)
    21. we have some runway to allow us to have some stuck ports and still run, just need to get them cleared eventually. (jamoluhrsen, 17:30:43)
    22. there was talk about trying to cut down the number of CSIT jobs, but maybe things have settled down and we dont need to. (jamoluhrsen, 17:32:13)
    23. robot VMs in releng are now 2G (jamoluhrsen, 17:33:16)
    24. we need to figure out if we can reduce mem usage on other VMs (e.g. mininet vm) (jamoluhrsen, 17:33:55)
    25. zxiiro playing around with zuul and hoping we can think about moving to it in Boron timeframe (jamoluhrsen, 17:34:42)
    26. vratko noting that with jobs failing this week, verify jobs were failing so it was hard to merge things. (jamoluhrsen, 17:36:23)
    27. zxiiro one workaround is to uncheck jenkins user from the review and then manually verify it and then merge (jamoluhrsen, 17:36:47)
    28. shall we run java 8 for Boron, and then 7 for stable/beryllium (jamoluhrsen, 17:39:03)
    29. zxiiro notes that if you build with 8, you are no longer compatible with 7 (jamoluhrsen, 17:40:32)
    30. autorelease is building with 8. snapshot builds are done with 7. projects like controller are using 7 for merge job (jamoluhrsen, 17:41:45)
    31. the question is what JRE should we run for the CSIT jobs? (jamoluhrsen, 17:44:46)
    32. anipbu mentions that we wont be able to run Boron CSIT with openjdk7 since some java 8 features will be there that cannot work in 7 (jamoluhrsen, 17:45:48)
    33. https://git.opendaylight.org/gerrit/#/c/32811/ (anipbu, 17:46:13)
    34. https://git.opendaylight.org/gerrit/#/q/topic:jdk8only (zxiiro, 17:46:31)
    35. patch 33101 seems to have it as we are agreeing. we'll need test-distribution for berylliyum csit to use openjdk8 (jamoluhrsen, 17:46:44)
    36. https://lists.opendaylight.org/pipermail/release/2016-January/005215.html (anipbu, 17:48:01)
    37. from weekly Be sync, there is a list of blocking bugs (jamoluhrsen, 17:51:52)
    38. https://bugs.opendaylight.org/buglist.cgi?bug_severity=blocker&list_id=45711&query_format=advanced&resolution=--- (anipbu, 17:52:48)
    39. vratko sending advertisement to community for more system test reviewers (jamoluhrsen, 17:56:07)
    40. lots of system test patches that need review (jamoluhrsen, 17:56:52)


Meeting ended at 18:01:11 UTC (full logs).

Action items

  1. (none)


People present (lines said)

  1. jamoluhrsen (37)
  2. anipbu (7)
  3. odl_meetbot (4)
  4. icbts (2)
  5. dfarrell07 (2)
  6. zxiiro (1)
  7. odp-gerritbot (1)


Generated by MeetBot 0.1.4.