=========================================== #opendaylight-meeting: Kernel projects call =========================================== Meeting started by rovarga at 17:06:01 UTC. The full logs are available at http://meetings.opendaylight.org/opendaylight-meeting/2019/kernel_projects_call/opendaylight-meeting-kernel_projects_call.2019-01-08-17.06.log.html . Meeting summary --------------- * jmorvay asks about CDS persistence: can journal/snapshots be used to replicate data to another cluster instance? (rovarga, 17:09:09) * that should be doable, assuming source and target clusters match in configuration (member names, etc.) (rovarga, 17:09:48) * Tom mentions an online backup facility is available through cluster-admin service (rovarga, 17:11:15) * jmorvay asks if snapshots contain the data tree (rovarga, 17:12:03) * snapshots contain data tree contents plus all backend metadata -- all raft state, transaction state, etc. (rovarga, 17:12:31) * Fluorine SR2 readiness (for 2/7/2019) (rovarga, 17:14:14) * odlparent-3.1.6 needs a multipatch test, which should be done tomorrow (rovarga, 17:14:46) * then we need yangtools bump, which should take about a day to release (rovarga, 17:15:04) * Neon code freeze readiness (for 1/24/2019) (rovarga, 17:16:14) * LINK: https://jenkins.opendaylight.org/releng/view/netconf/job/netconf-csit-1node-userfeatures-all-neon/ (LuisGomez, 17:16:38) * build #140, #142 and #144-148 are completely failing (rovarga, 17:18:00) * jmorvay mentions we have had some code movement, but nothing that would be expected to regress (rovarga, 17:18:33) * ACTION: jmorvay to check what is going on (rovarga, 17:19:40) * LINK: https://jenkins.opendaylight.org/releng/view/netconf/job/netconf-csit-1node-userfeatures-all-fluorine/ (LuisGomez, 17:19:44) * LINK: https://logs.opendaylight.org/releng/vex-yul-odl-jenkins-1/netconf-csit-1node-userfeatures-all-neon/148/odl_1/odl1_karaf.log.gz (rovarga, 17:20:54) * Caused by: org.w3c.dom.DOMException: NAMESPACE_ERR: An attempt is made to create or change an object in a way which is incorrect with regard to namespaces. (rovarga, 17:21:57) * this looks like we are getting xerces implementation when we should not (rovarga, 17:22:14) * LINK: https://git.opendaylight.org/gerrit/#/c/77471/ (jmorvay, 17:23:30) * LINK: https://git.opendaylight.org/gerrit/#/c/77471/ (rovarga, 17:23:57) * that should help the situation (rovarga, 17:24:03) * we need odlparent-4.0.8 and the usual yangtools/mdsal propagation (rovarga, 17:25:40) * reverted spotbugs upgrade, because it breaks downstreams (rovarga, 17:27:40) * LINK: https://git.opendaylight.org/gerrit/#/c/79317/ is the final bit, awaiting review (rovarga, 17:28:08) * we should be ready to release tomorrow (rovarga, 17:28:21) * LINK: https://git.opendaylight.org/gerrit/#/q/topic:odlparent-4.0.7 is the ongoing bump (rovarga, 17:29:11) * LINK: https://logs.opendaylight.org/releng/vex-yul-odl-jenkins-1/openflowplugin-csit-1node-flow-services-all-oxygen/491/odl_1/odl1_karaf.log.gz (LuisGomez, 17:31:50) * skitt will updated the outstanding patches to 4.0.8 (rovarga, 17:31:52) * and push all the other necessary patches (skitt, 17:32:08) * oxygen is in a busted state, as it did not complete odlparent-3.1.5 bump, which was needed to fix the xtend issues (rovarga, 17:32:26) * odlparent-3.1.6 includes jackson-databind fix, which has a couple CVEs attached to it (rovarga, 17:33:19) * might be a cause to release Oxygen SR5 (rovarga, 17:33:36) * ACTION: rovarga to raise this at security@/tsc@ (rovarga, 17:33:50) * mdsal-3.0.4 has two more patches outstanding (rovarga, 17:34:58) * LINK: https://git.opendaylight.org/gerrit/79051 to add aggregate javadocs (for javadoc.io integration) (rovarga, 17:35:13) * LINK: https://git.opendaylight.org/gerrit/75569 is optional, as it just brings about mdsal-trace without adding a feature (rovarga, 17:36:06) * we still need to figure out how mdsal's odl-mdsal-trace and controller's odl-mdsal-trace will work together (rovarga, 17:37:51) * rovarga suspects this will need the code to be published as artifacts from mdsal and then controller will do *something* (rovarga, 17:38:22) * blueprint will need some more work, and should be fair for mdsal-3.0.5 (rovarga, 17:42:49) * mdsal-trace should go into 3.0.4, though (rovarga, 17:43:03) * mdsal/controller blueprint structure is TBD, as we may just keep the forked copy (rovarga, 17:50:40) * ACTION: rovarga to ask around how controller->mdsal migration is going in downstreams (rovarga, 17:52:07) * tell-based CSIT (rovarga, 17:52:50) * LINK: https://logs.opendaylight.org/releng/vex-yul-odl-jenkins-1/controller-csit-3node-ddb-expl-lead-movement-longevity-only-neon/21/robot-plugin/log.html.gz (rovarga, 17:53:56) * we are getting a DeadTransactionException (rovarga, 17:54:12) * which really should not be there, as we are expecting a stateful switchover (rovarga, 17:54:52) * the transaction is in purged transactions (rovarga, 17:56:10) * the transaction is from the default history, i.e. not chained (rovarga, 18:00:43) * ACTION: Tom to file an issue to track this (rovarga, 18:01:09) Meeting ended at 18:02:42 UTC. Action items, by person ----------------------- * jmorvay * jmorvay to check what is going on * rovarga * rovarga to raise this at security@/tsc@ * rovarga to ask around how controller->mdsal migration is going in downstreams People present (lines said) --------------------------- * rovarga (45) * LuisGomez (6) * odl_meetbot (3) * skitt (1) * jmorvay (1) Generated by `MeetBot`_ 0.1.4