================================================================== #opendaylight-openflowplugin: OpenFlow Plugin Meeting Aug 18, 2016 ================================================================== Meeting started by abhijitkumbhare at 15:10:33 UTC. The full logs are available at http://meetings.opendaylight.org/opendaylight-openflowplugin/2016/openflow_plugin_meeting_aug_18__2016/opendaylight-openflowplugin-openflow_plugin_meeting_aug_18__2016.2016-08-18-15.10.log.html . Meeting summary --------------- * Agenda (abhijitkumbhare, 15:10:45) * Documentation (abhijitkumbhare, 15:10:56) * Test report from LuisGomez (abhijitkumbhare, 15:11:36) * Bugs (abhijitkumbhare, 15:11:45) * Integration with Singleton app on the master Status (abhijitkumbhare, 15:12:21) * Bugzilla cleanup (abhijitkumbhare, 15:12:59) * Bugzilla cleanup (abhijitkumbhare, 15:13:59) * investigating, testing and closing bugs 6658, 5888, 6445, 2719, 1395, 4210, 1647, 3706, 4148, 2612, 1939, 1565, 521 (abhijitkumbhare, 15:14:45) * investigating bugs 1462, 3208, 1810 (abhijitkumbhare, 15:15:04) * Bugs (abhijitkumbhare, 15:16:15) * jbacigal looking into https://bugs.opendaylight.org/show_bug.cgi?id=6465 raised by NetVirt (abhijitkumbhare, 15:16:34) * Happens only on the master branch (abhijitkumbhare, 15:17:14) * Patch for table features switched off (abhijitkumbhare, 15:18:30) * LINK: https://git.opendaylight.org/gerrit/#/c/42821/ (andrejleitner, 15:18:39) * This patch for table features needs to be merged - already reviewed by jbacigal (abhijitkumbhare, 15:19:48) * LINK: https://bugs.opendaylight.org/show_bug.cgi?id=5173 (shuva, 15:22:06) * Do we need individual bugs for too many logs? (abhijitkumbhare, 15:23:34) * As per shuva, for bug 5173 with OVS 2.4 - by default table features switched off on master we will see less number of logs (abhijitkumbhare, 15:26:22) * We will reopen https://bugs.opendaylight.org/show_bug.cgi?id=5173 (too many Karaf logs when connecting to OVS 2.4) (abhijitkumbhare, 15:29:44) * LINK: https://bugs.opendaylight.org/show_bug.cgi?id=6146 - a blocker bug raised by NetVirt (abhijitkumbhare, 15:32:53) * andrejleitner thinks 6146 is fixed, and netvirt has a workaround already (abhijitkumbhare, 15:33:56) * ACTION: andrejleitner will email openflow plugin & netvirt mailing list about bug 6146 status (abhijitkumbhare, 15:34:36) * LINK: https://bugs.opendaylight.org/show_bug.cgi?id=6459 - critical - [Li plugin] Reconciliation not triggering after a single node reboot in a non OFHA scenario (abhijitkumbhare, 15:36:02) * Bug 6459 has been fixed in He design - needs to be fixed in the Li design (abhijitkumbhare, 15:39:09) * Ideally reconciliation is independent of He or Li (abhijitkumbhare, 15:39:48) * Test Report from LuisGomez (abhijitkumbhare, 15:41:34) * LINK: https://bugs.opendaylight.org/show_bug.cgi?id=6176 (abhijitkumbhare, 15:42:02) * Cbench throughput test does not work in Boron: (abhijitkumbhare, 15:42:09) * LINK: https://bugs.opendaylight.org/show_bug.cgi?id=6176 (abhijitkumbhare, 15:42:19) * jbacigal thinks this may be environment related (abhijitkumbhare, 15:42:37) * LuisGomez will try to replicate locally to check if it is environment related (move to the cloud) (abhijitkumbhare, 15:43:17) * He Plugin: Topology is not being discovered. (abhijitkumbhare, 15:43:43) * vishnoianil will take a quick look at this (abhijitkumbhare, 15:46:22) * LINK: https://bugs.opendaylight.org/show_bug.cgi?id=6456 (abhijitkumbhare, 15:46:38) * Li PLUGIN: Modified flow does not reconcile in configuration (alien flow ID) (abhijitkumbhare, 15:47:41) * LINK: https://bugs.opendaylight.org/show_bug.cgi?id=6458 (abhijitkumbhare, 15:48:38) * shuva has a patch related to this waiting for review - needs to be merged (abhijitkumbhare, 15:50:31) * Entity Owner not resolved properly after owner is isolated and rejoined (abhijitkumbhare, 15:52:07) * LINK: https://bugs.opendaylight.org/show_bug.cgi?id=6177 (abhijitkumbhare, 15:52:30) * Need to check if this issue happens on the master (abhijitkumbhare, 15:52:49) * For master there are 92 failures for clustering; for the Boron branch there are 16 failures (abhijitkumbhare, 15:57:15) * For master there are 92 failures for clustering; for the Boron branch there are 16 failures (abhijitkumbhare, 15:58:12) * Need to check if the tests need to be updated for the singleton and whether some of the failures are expected but not real failures (abhijitkumbhare, 15:58:41) * LuisGomez will disable tests that do not apply for singleton design (abhijitkumbhare, 15:59:43) * Switch inventory remains when switch goes off (after owner restarts) (abhijitkumbhare, 16:00:35) * This does not happen on the Carbon (master); but happens on Boron - singleton related (abhijitkumbhare, 16:01:58) * Switch Disconnect+Connect to Ownership Candidate fails (abhijitkumbhare, 16:02:14) * This also does not happen on the Carbon (master); but happens on Boron - singleton related (abhijitkumbhare, 16:02:40) * Unstable Switch scalability (100 to 400 switches) (abhijitkumbhare, 16:03:02) * LINK: https://bugs.opendaylight.org/show_bug.cgi?id=6178 Li plugin: Global peformance issues (abhijitkumbhare, 16:03:41) * Test seems to work in Carbon as well as the Helium design for Boron; just fails Li design on Boron (abhijitkumbhare, 16:05:30) Meeting ended at 16:12:10 UTC. Action items, by person ----------------------- * andrejleitner * andrejleitner will email openflow plugin & netvirt mailing list about bug 6146 status People present (lines said) --------------------------- * abhijitkumbhare (53) * shuva (4) * odl_meetbot (4) * hideyuki (2) * andrejleitner (1) Generated by `MeetBot`_ 0.1.4