Tuesday, 2017-03-07

*** ianychoi has quit IRC00:10
*** ianychoi has joined #openstack-searchlight00:11
*** shu-mutou has joined #openstack-searchlight00:27
*** zhurong has joined #openstack-searchlight01:05
*** TravT has quit IRC01:21
*** lei-zh has joined #openstack-searchlight01:31
openstackgerritOpenStack Proposal Bot proposed openstack/searchlight master: Updated from global requirements  https://review.openstack.org/42462502:06
*** lei-zh1 has joined #openstack-searchlight03:38
*** lei-zh has quit IRC03:40
*** lei-zh1 has quit IRC04:05
*** lei-zh has joined #openstack-searchlight05:38
*** Jeffrey4l has quit IRC05:52
*** Jeffrey4l has joined #openstack-searchlight05:56
*** pcaruana has joined #openstack-searchlight07:50
*** Jeffrey4l_ has joined #openstack-searchlight08:30
*** Jeffrey4l has quit IRC08:34
*** Jeffrey4l_ has quit IRC08:37
*** Jeffrey4l_ has joined #openstack-searchlight08:38
*** shu-mutou is now known as shu-mutou-AWAY09:49
*** lei-zh has quit IRC10:07
*** openstackgerrit has quit IRC10:33
*** Guest6667 is now known as Guest666610:43
*** catintheroof has joined #openstack-searchlight12:24
*** zhurong has quit IRC12:45
*** zhurong has joined #openstack-searchlight13:35
*** yingjun has joined #openstack-searchlight14:03
*** zhurong has quit IRC14:16
*** matt-borland has joined #openstack-searchlight14:23
*** TravT has joined #openstack-searchlight15:04
*** pcaruana has quit IRC15:09
*** yingjun has quit IRC15:11
*** pcaruana has joined #openstack-searchlight15:23
*** pcaruana has quit IRC16:26
*** david-lyle_ has joined #openstack-searchlight17:52
*** david-lyle has quit IRC17:54
*** david-lyle_ is now known as david-lyle17:55
*** david-lyle has quit IRC18:04
*** david-lyle has joined #openstack-searchlight18:05
*** david-lyle has quit IRC18:23
*** david-lyle has joined #openstack-searchlight18:25
*** david-lyle has quit IRC18:29
*** david-lyle has joined #openstack-searchlight18:36
*** mriedem has joined #openstack-searchlight18:37
mriedemsjmc7: do you know if searchlight supports / handles nova versioned notifications?18:38
mriedemi assumed it did when writing https://review.openstack.org/#/c/441692/ but wasn't actually sure18:38
sjmc7mriedem: currently we consume the old unversioned ones. we planned to switch over to the versioned ones in ocata but mainly because of peoples’ availability we didn’t get round to it. i don’t forsee it being a huge amount of work18:59
sjmc7and we need to do it anyway18:59
mriedemok, do you have a blueprint or something tracking that work?19:00
sjmc7yep, one sec19:00
sjmc7https://blueprints.launchpad.net/searchlight/+spec/nova-versioned-notifications19:00
*** david-lyle has quit IRC19:04
*** david-lyle has joined #openstack-searchlight19:06
mriedemsjmc7: random question, but do you know where the "compute." prefix gets into the event type?19:13
mriedemi.e. compute.instance.create.start19:13
mriedemi see where nova emits the instance.create.start, but not what puts the compute. prefix on that19:14
sjmc7no idea! possibly oslo?19:14
sjmc7since it matches the service type?19:14
sjmc7all the notifications seem prefixed that way (‘image’, ‘dns’ etc)19:15
mriedemfor unversioned i think it might be this https://github.com/openstack/nova/blob/master/nova/compute/utils.py#L32419:16
sjmc7ah, that’d be it. it may not have that prefix for versioned19:16
mriedemfor versioned it's this https://github.com/openstack/nova/blob/master/nova/compute/utils.py#L35619:16
mriedemright, i think versioned doesn't have the prefix19:17
sjmc7when we implemented versioned notifications for ironic it was quite different to the unversioned ones for other services19:17
sjmc7lemme check to see if they have a prefix19:17
sjmc7hm.. they have ‘baremetal'19:18
mriedemok. i don't know if dropping the prefix for the nova versioned notifications was intentional or not.19:22
mriedemi can look back at the original spec19:22
mriedemdid ironic use different notification topics?19:25
sjmc7they didn’t have unversioned ones19:27
mriedemoh19:27
mriedemi found https://specs.openstack.org/openstack/nova-specs/specs/newton/implemented/versioned-notification-transformation-newton.html#notifications-impact19:27
mriedem"As implemented in the versioned-notification-api bp the versioned notifications are always emitted to a different amqp topic called ‘versioned_notifications’ so the consumer can differentiate between the legacy and the new format by the topic."19:27
mriedembut i'm not sure that's true19:27
sjmc7it’d be a bit awkward if theyt go to the same one19:28
mriedemwell, i'm wondering if they are going to the same one and that's why the versioned notifications dropped the compute. prefix19:29
sjmc7hmmm. maybe one for Balazs19:29
mriedemoh i see it19:31
mriedemhttps://github.com/openstack/nova/blob/master/nova/rpc.py#L9219:31
sjmc7ok. that’s good19:32
mriedemso we send to different topics with different event types19:32
mriedemlegacy has the compute. prefix, versioned doesn't19:32
sjmc7that’s probably ok. the handling is sufficiently different that changing the name wouldn’t be a big deal19:33
*** Jeffrey4l_ has quit IRC21:34
*** Jeffrey4l_ has joined #openstack-searchlight21:37
*** matt-borland has quit IRC22:35
*** catintheroof has quit IRC22:41
*** catintheroof has joined #openstack-searchlight22:42
*** TravT has quit IRC22:43
*** catintheroof has quit IRC22:47
*** TravT has joined #openstack-searchlight23:05
*** catintheroof has joined #openstack-searchlight23:32
*** mriedem has quit IRC23:36
*** yingjun has joined #openstack-searchlight23:59

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!