15:07:05 <faseelak_> #startmeeting Genius Weekly 01/10/2019
15:07:05 <odl_meetbot> Meeting started Thu Jan 10 15:07:05 2019 UTC.  The chair is faseelak_. Information about MeetBot at http://ci.openstack.org/meetbot.html.
15:07:05 <odl_meetbot> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:07:05 <odl_meetbot> The meeting name has been set to 'genius_weekly_01_10_2019'
15:07:15 <faseelak_> #topic Roll call and agenda bashing (please #info <your-nick>)
15:07:32 <faseelak_> #link https://jira.opendaylight.org/browse/GENIUS-264
15:07:41 <shague> #info shague
15:08:08 <faseelak_> #topic Genius CSIT 3 node
15:08:24 <faseelak_> #info Karthika has finished 3 node CSIT patch for genius
15:09:08 <faseelak_> #link https://git.opendaylight.org/gerrit/#/c/71990/
15:09:13 <Dimpl> #info
15:10:27 <Hema> #info Hema
15:11:20 <faseelak_> #topic idmanager NoShardLeaderException
15:11:33 <faseelak_> #link https://lists.opendaylight.org/pipermail/genius-dev/2019-January/003554.html
15:15:18 <faseelak_> #info faseelak_ asks if databroker inject can be delayed till all shards have leaders
15:15:35 <faseelak_> #info tom indicates applications should still handle failures of datastore operations
15:17:28 <faseelak_> #info datastore operations should have retries on failures
15:17:37 <faseelak_> #info how long should the applications retry
15:21:11 <faseelak_> #info tom pantelis says it should be infinite to handle the worst case scenarios
15:27:29 <faseelak_> #info Hema says a variant of the problem is seen in Boron, where listener registrations before shards are UP, results in DTCNs not being fired
15:31:46 <faseelak_> #info tom says we should discuss this in next kernel call
15:42:38 <faseelak_> #topic OFTunnels
15:43:10 <faseelak_> #info OFTunnels with ITM Direct Tunnels needed for ITM scaling to meet COE requirements
15:43:23 <faseelak_> #info Dimple has started pushing patches, and planning to finish this soon
15:45:58 <faseelak_> #topic genius APEX jobs
15:46:13 <faseelak_> #info vthapar says it would be good to make APEX CSIT jobs voting
15:46:40 <faseelak_> #info there is one consistent failure in genius CSIT right now with auto tunnels, so would be better to make the change once this issue is fixed
15:53:31 <faseelak_> #info shague says this is not genius CSIT, and netvirt CSIT is what is going to be run
15:53:41 <faseelak_> #info agrees to enable this for genius
15:54:03 <faseelak_> #endmeeting