15:02:26 #startmeeting NetVirt Weekly 10/24/17 15:02:26 Meeting started Tue Oct 24 15:02:26 2017 UTC. The chair is shague. Information about MeetBot at http://ci.openstack.org/meetbot.html. 15:02:26 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:02:26 The meeting name has been set to 'netvirt_weekly_10_24_17' 15:02:41 #topic Roll call and agenda bashing (please #info ) 15:02:43 #topic Review existing action items 15:02:51 #link https://meetings.opendaylight.org/opendaylight-netvirt/2017/netvirt_weekly_10_17_17/opendaylight-netvirt-netvirt_weekly_10_17_17.2017-10-17-15.01.html 15:03:18 #info [Review previous minutes](https://meetings.opendaylight.org/opendaylight-netvirt/2017/netvirt_weekly_10_17_17/opendaylight-netvirt-netvirt_weekly_10_17_17.2017-10-17-15.01.html) 15:03:59 #info Vishal 15:04:15 #info jamoluhrsen 15:04:22 #info Vic 15:05:26 #info Janki Chhatbar 15:05:28 #topics siva add bug if needed for ipv6 and clustering issues 15:05:32 #info Hanamantagoud 15:05:49 #info siva to report bug 15:05:52 #info jhershbe 15:06:33 #info sridharg to meet with e/// team to dicsuss changes 15:06:51 #topic shague to look at https://bugs.opendaylight.org/show_bug.cgi?id=9149 [jira](https://jira.opendaylight.org/browse/NETVIRT-905) - unable to reproduce in latest jobs 15:07:06 #info Sridhar Gaddam 15:07:33 #action jamoluhrsen to doublecheck 15:07:44 #info daya_k 15:07:45 #action jamoluhrsen to doublecheck bug 9149 15:07:46 #topic jamoluhrsen, shague, Hanamantagoud, kiran to start email with openflowplugin - [thread](https://lists.opendaylight.org/pipermail/openflowplugin-dev/2017-October/007746.html) 15:08:22 #topic shague initial triage at 3node tests - no progress yet 15:08:36 #topic janki need to look at tempest test case - initial triage and root cause found [thread](https://lists.opendaylight.org/pipermail/netvirt-dev/2017-October/005683.html) 15:09:22 #info janki found config issue needed for the one failing test 15:12:31 shague, janki: yes it's happening in queens: https://jenkins.opendaylight.org/releng/user/jluhrsen/my-views/view/netvirt%20csit/job/netvirt-csit-1node-openstack-queens-upstream-stateful-nitrogen/60/robot/ 15:13:36 #topic https://trello.com/c/iD2fOfF1/16-nitrogen-release-planning - SR1 10/26/17 15:13:58 jamoluhrsen, ack 15:14:08 #topic Jira migration is complete 15:20:24 #topic Spec reviews should include CSIT updates as mandatory. Code should not merge until CSIT tests are written. 15:21:41 #info by time code is to be merged, there should be a CSIT patch to verify the code 15:22:51 #info how to create CSIT tests for features spanning multiple patches 15:23:05 #info possibly create a feature specific job 15:23:34 #update CSIT to use a control+compute node and a compute node, rather than the current control node and two compute nodes 15:24:47 #info does this change infra 15:27:45 #topic bug scrub [jira query exception](https://jira.opendaylight.org/issues/?filter=10116) - [bugzilla query for exceptions](https://bugs.opendaylight.org/buglist.cgi?quicksearch=sw%3A%22csit%3Aexception%22&list_id=75946) - [jira query failures](https://jira.opendaylight.org/issues/?filter=10118) - [bugzilla query sporadic_failures](https://bugs.opendaylight.org/buglist.cgi?quicksearch=sw%3A%22csit%3Asporadic_failures%22&list_id=75946) 15:28:23 #topic [OPNFLWPLUG-858](https://jira.opendaylight.org/browse/OPNFLWPLUG-858) - [bug-7826](https://bugs.opendaylight.org/show_bug.cgi?id=7826) - openflowplugin patch was abandoned, new patch [63896](https://git.opendaylight.org/gerrit/63896) 15:28:46 #info sunil making progress, recreating locally to further debug 15:29:34 #topic [bug-9245]() - [NETVIRT-940](https://jira.opendaylight.org/browse/NETVIRT-940) - [gerrit 64382](https://git.opendaylight.org/gerrit/64382) 15:31:11 #info fix will cover all table 21 bugs, NETVIRT-940 and 946 15:32:05 #topic [bug-9297](https://bugs.opendaylight.org/show_bug.cgi?id=9297) - adjacency]} does not exist - [NETVIRT-959](https://jira.opendaylight.org/browse/NETVIRT-959) 15:32:45 #info Hanamantagoud fip is present but no adjacency is found 15:33:03 #info looks like fip is moving from node to node 15:33:57 #info Hanamantagoud requested help from NAT guys 15:36:14 #topic [bug-9298](https://bugs.opendaylight.org/show_bug.cgi?id=9298) - port-name=e21f56c9-1513-4753-bf40-f9d391c45658}] does not exist - [NETVIRT-960](https://jira.opendaylight.org/browse/NETVIRT-960) 15:37:02 #info Cheta is out will look when back 15:37:17 #topic [gerrit 64124](https://git.opendaylight.org/gerrit/64124) - Updated to use bind-service update instead of bind and unbind in Acl VPN - listener - doesn't seem to help with [NETVIRT-962](https://jira.opendaylight.org/browse/NETVIRT-962) or [NETVIRT-963](https://jira.opendaylight.org/browse/NETVIRT-963) - shold we still commit if change is useful? 15:38:39 #info Aswin doublecheck 15:38:52 #topic L2GW CSIT - Move BGP VPN service into it's own CSIT so we can make L2GW CSIT gating for HWVTEP and L2GW related patches 15:39:47 #info vpnservice suites are in both netvirt and l2gw - should it be removed from l2gw 15:40:01 #info should bgpvpn be moved to genius? 15:40:46 #info no, it is netvirt suite. 15:42:15 #info vthapar bgpvpn should not be in the netvirt or l2gw suite 15:42:53 #info daya_k bgpvpn is not decoupled from l3 so it will cause issues if not ran as part of the full suite 15:44:04 #info l2gw should not have bgpvpn 15:44:32 #info need to decide then where bgpvpn should be - in netvirt basic suite or in separate suite 15:45:42 #action jamoluhrsen to create a new job for the bgpvpn suite, remove from l2gw 15:46:22 #info use the new feature job 15:47:22 #info vpnservice suites should remain in netvirt. 15:47:40 #action vpickard to push patch to remove vpnservice suite from l2gw 15:49:00 #info vthapar should l2gw gates be ran on all patches? 15:50:04 #info all ovsdb hwvtep patches should run the l2gw gate 15:50:25 #info how to identify what netvirt patches should run l2gw gates 15:51:12 #info most code has l2gw or hwvtep in the patch so that can be used as a guide 15:52:26 #topic Auto transport zone, Auto tunnels, and L2GW https://bugs.opendaylight.org/show_bug.cgi?id=7743 15:54:50 #info l2gw does not full support the autotunnel feature 15:55:43 #info also seeing issues in basic netvirt around autotunnel, timing issues 15:56:33 #info should we disable autotunnel and use other mechanism to create the zone and full tunnel mesh 15:57:43 #info port-status has issues in clustered setup, this is needed for auto-tunnels to work 15:58:11 #info jhershbe patches have been pushed to fix port-status clustering issues. need to be verified 15:59:26 #info restconf is combersome 15:59:39 #info genius autotunnel is easier to use 16:01:17 #info look at genius spec for autotunnel 16:02:08 #info daya_k no work is needed in l2gw if the genius autotunnel is used 16:02:45 #action vpickard to vrify the genius autotunnel 16:02:46 jfyi, vthapar shared the following link regarding the genius auto-tunnel - http://docs.opendaylight.org/en/stable-nitrogen/submodules/genius/docs/specs/itm-tunnel-auto-config.html 16:03:04 #link http://docs.opendaylight.org/en/stable-nitrogen/submodules/genius/docs/specs/itm-tunnel-auto-config.html 16:06:25 #info after vpickard verifies genius autotunnel, need to continue discussion to decide if we can move to genius autotunnel 16:07:34 #topic clustering topics 16:07:45 #info daya_k to start email with ideas discussed at dDF 16:08:48 #endmeeting