16:02:17 #startmeeting Kernel Projects Call 16:02:17 Meeting started Tue Jul 2 16:02:17 2019 UTC. The chair is rovarga. Information about MeetBot at http://ci.openstack.org/meetbot.html. 16:02:17 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 16:02:17 The meeting name has been set to 'kernel_projects_call' 16:02:23 #chair skitt rovarga 16:02:23 Current chairs: rovarga skitt 16:02:48 #topic agenda 16:03:49 #info Actions support update 16:04:40 #info odlparent-6 (i.e. Magnesium) 16:04:51 #info stable/neon MRI update 16:07:19 #info reinstall clustering from scratch 16:07:38 #info administrative things 16:08:20 #topic Actions support update 16:08:37 #info we have two trains: controller and netconf (in restconf) 16:08:52 #info reviewers requested 16:09:08 #action rovarga to add people on review 16:10:50 #action rovarga to send out the 'busy committers like this' rant 16:12:12 #info restconf patch by end of week 16:13:00 #info testtool magic 16:15:02 #link rovarga recently came across https://github.com/remkop/picocli 16:16:49 #info LuisGomez notes that *some* parameters are working, but not all and it takes some doing 16:18:13 #info we need a docs update to spell out how system tools are setting up testtool 16:21:34 #topic odlparent-6 16:21:51 #info java-11 required 16:21:57 #undo 16:21:57 Removing item from minutes: 16:22:10 #info java-11 required? pending the TSC's final decision 16:22:39 #info probably will have karaf-4.3.0 with OSGi R7 16:22:45 #undo 16:22:45 Removing item from minutes: 16:22:51 #info probably will have karaf-4.3.0 with OSGi R7, do we want that? 16:23:24 #info scala-2.13.0 available (with akka 2.5.x, 2.6.x) 16:24:47 #info forcing guava-28.0 is not fair 16:24:55 #Info checkstyle bump? 16:25:47 #info checkstyle/spotbugs are usually easily fixed up minor version bump changes 16:27:46 #agree update odlparent policy to say that minor patch can break build, but tests need to be kept working 16:29:24 #info removing jsr305 completely 16:30:04 #info we down to 4/5 downstreams using it, before Sodium there should be at most 2 16:30:40 #action create a 6.0.0 release in JIRA and will open issues for the changes we want to make 16:32:39 #info karaf-4.3.0 is probably going to be painful 16:32:52 #topic stable/neon MRI update 16:33:26 #info odlparent-4.0.11, yangtools-2.1.11, mdsal-3.0.9 is out 16:34:59 #link https://git.opendaylight.org/gerrit/#/q/topic:mri-neon-sr2 16:36:20 #info odlparent-5.0.2 would be nice 16:36:48 https://git.opendaylight.org/gerrit/73686 16:37:46 #info yangtools is still doing some work for the next release, will roll out either 5.0.1 or 5.0.2 dependending on readiness 16:38:13 #topic reinstall clustering from scratch 16:38:40 #info it's just a reinstall of a single node 16:38:42 #link https://jira.opendaylight.org/browse/CONTROLLER-1626 16:40:45 #link https://docs.opendaylight.org/en/stable-neon/getting-started-guide/clustering.html#geo-distributed-active-backup-setup 16:41:02 POST /restconf/operations/cluster-admin:add-replicas-for-all-shards 16:43:00 #info as a workaround, is it possible to add the old *system* as a new *cluster node*? it should be possible to add a new cluster member and delete the old one -- it's just that the reinstalled node would have a different name in ODL cluster 16:44:35 #topic general discussion 16:47:01 #info https://docs.google.com/spreadsheets/d/1CN-ACdIekRKMYRIuREIRALl6S8az7quAcQlRb3KeMJ8/edit#gid=650373173 looks a bit worrying 16:47:28 #link https://docs.google.com/spreadsheets/d/1CN-ACdIekRKMYRIuREIRALl6S8az7quAcQlRb3KeMJ8/edit#gid=650373173 looks a bit worrying 16:47:34 #link https://docs.google.com/spreadsheets/d/1CN-ACdIekRKMYRIuREIRALl6S8az7quAcQlRb3KeMJ8/edit#gid=588630221 16:48:48 #action rovarga to rever bgpcep/osgi patch 16:48:54 #undo 16:48:54 Removing item from minutes: 16:48:57 #action rovarga to revert bgpcep/osgi patch 16:49:37 #info LuisGomez needs to refactor bgp tests 16:51:02 #link https://jenkins.opendaylight.org/releng/view/bgpcep/job/bgpcep-csit-1node-userfeatures-all-sodium/ 16:52:36 #link https://jenkins.opendaylight.org/releng/view/bgpcep/job/bgpcep-csit-1node-throughpcep-all-sodium/ 16:53:09 #info for the first link it should be just the patch revert 16:53:30 #info for the second link I have to go deep in the suite 16:54:57 #info last success was https://logs.opendaylight.org/releng/vex-yul-odl-jenkins-1/bgpcep-csit-1node-throughpcep-all-sodium/65/ 16:55:15 #info which was Mar 28, which is a loong time ago 16:55:29 Counter({'status: 204 content: ': 65535})' does not contain 'Counter({'pass': 65535})' 16:57:10 #link https://git.opendaylight.org/gerrit/#/c/64782/ was changed on Mar 26nd in bgpceep 16:58:07 #link https://logs.opendaylight.org/releng/vex-yul-odl-jenkins-1/bgpcep-csit-1node-throughpcep-all-sodium/215/odl_1/odl1_karaf.log.gz 16:58:19 2019-07-01T15:26:51,544 | WARN | pool-24-thread-1 | TopologyStatsProviderImpl | 259 - org.opendaylight.bgpcep.pcep-topology-stats - 0.12.0 | Failed to prepare Tx for BGP stats update 16:58:20 java.util.ConcurrentModificationException: null 16:58:22 at java.util.HashMap$HashIterator.nextNode(HashMap.java:1445) ~[?:?] 16:58:23 at java.util.HashMap$ValueIterator.next(HashMap.java:1474) ~[?:?] 16:58:25 at java.util.AbstractCollection.toArray(AbstractCollection.java:141) ~[?:?] 16:58:26 at com.google.common.collect.ImmutableList.copyOf(ImmutableList.java:258) ~[36:com.google.guava:27.1.0.jre] 16:58:28 at org.opendaylight.bgpcep.pcep.topology.provider.AbstractTopologySessionListener.getDelegatedLspsCount(AbstractTopologySessionListener.java:596) ~[?:?] 16:58:42 which looks like a locking problem 16:58:48 or nesting problem 16:58:55 test runs this command: bash -c "cd /tmp; taskset 0x00000001 python updater.py --workers '64' --odladdress '10.30.170.69' --user 'admin' --password 'admin' --scope '' --pccaddress '10.30.170.38' --pccs '1' --lsps '65535' --hop '8.8.8.8/32' --timeout '900' --refresh '0.1' --reuse 'False' 2>&1" 16:59:07 and it returns: 16:59:08 took 153.47883296 16:59:09 Counter({'status: 204 content: ': 65535}) 17:02:01 #link https://jira.opendaylight.org/browse/BGPCEP-875 tracks the CME 17:02:11 #endmeeting