Sunday, 2015-04-19

« Saturday, 2015-04-18 Index Monday, 2015-04-20 »
*** EricGonczer_ has quit IRC00:08
*** jamielennox|away is now known as jamielennox00:32
*** sdake_ has joined #heat01:16
*** sdake has quit IRC01:20
*** sdake_ has quit IRC01:31
*** sdake has joined #heat01:35
*** EricGonczer_ has joined #heat01:36
*** erkules_ has joined #heat01:41
*** erkules has quit IRC01:44
*** achanda has quit IRC01:49
*** mikeit has joined #heat02:47
*** achanda has joined #heat02:48
*** EricGonc_ has joined #heat02:56
*** EricGonczer_ has quit IRC02:59
*** EricGonc_ has quit IRC03:03
*** sdake has quit IRC03:36
*** ashishjain has joined #heat03:43
*** ashishjain has quit IRC03:55
*** achanda has quit IRC04:15
openstackgerritLimor Stotland proposed openstack/heat: Add restriction on the physical resource name length  https://review.openstack.org/17430904:41
*** LimorStotland has joined #heat04:41
*** jtomasek has joined #heat04:58
openstackgerritSirushti Murugesan proposed openstack/heat: Run heat_integrationtests in parallel  https://review.openstack.org/17233905:12
*** achanda has joined #heat05:15
*** achanda has quit IRC05:20
*** achanda has joined #heat05:25
*** mikeit has quit IRC05:41
*** sdake has joined #heat05:50
*** jamielennox is now known as jamielennox|away06:27
*** jamielennox|away is now known as jamielennox06:33
*** achanda has quit IRC06:35
*** achanda has joined #heat06:36
*** sabeen1 has joined #heat06:38
*** sdake has quit IRC06:55
*** sdake has joined #heat06:56
*** jamielennox is now known as jamielennox|away07:03
openstackgerritLimor Stotland proposed openstack/heat: Add restriction on the physical resource name length  https://review.openstack.org/17430907:11
*** hdd has joined #heat07:30
*** hdd has quit IRC07:36
*** noyitz has joined #heat07:44
*** noyitz_ has joined #heat08:04
*** noyitz has quit IRC08:04
*** hdd has joined #heat08:05
*** hdd has quit IRC08:10
*** sabeen1 has quit IRC08:20
*** LiJiansheng has joined #heat08:25
*** tspatzier has joined #heat08:27
*** tspatzier has quit IRC08:30
*** noyitz_ has quit IRC08:45
*** noyitz has joined #heat08:45
*** achanda has quit IRC08:50
*** noyitz has quit IRC09:00
*** pal has joined #heat09:00
*** noyitz has joined #heat09:01
*** pal has quit IRC09:02
*** noyitz_ has joined #heat09:08
*** noyitz has quit IRC09:08
*** smccully has quit IRC09:11
*** smccully has joined #heat09:11
*** sdake has quit IRC09:14
*** noyitz_ has quit IRC09:34
*** tspatzier has joined #heat09:41
*** noyitz has joined #heat09:47
*** noyitz has quit IRC10:14
*** tspatzier has quit IRC10:21
*** tspatzier has joined #heat10:28
*** jamielennox|away is now known as jamielennox10:42
*** jamielennox is now known as jamielennox|away11:12
*** noyitz has joined #heat11:15
*** noyitz has quit IRC11:19
*** Qiming has joined #heat11:26
*** serg_melikyan has joined #heat11:37
openstackgerritLimor Stotland proposed openstack/heat: Add restriction on the physical resource name length  https://review.openstack.org/17430911:42
*** serg_melikyan has quit IRC11:42
*** uros_ has quit IRC11:49
*** LiJiansheng has quit IRC12:01
*** noyitz has joined #heat12:12
*** noyitz_ has joined #heat12:16
*** noyitz has quit IRC12:16
openstackgerritQiming Teng proposed openstack/heat: Split engine service test case  https://review.openstack.org/16406212:39
*** noyitz_ has quit IRC12:40
*** noyitz has joined #heat12:45
*** EricGonczer_ has joined #heat12:51
*** Kennan has quit IRC12:53
*** Kennan has joined #heat12:54
*** EricGonczer_ has quit IRC13:05
*** Yanyanhu has joined #heat13:07
*** Yanyanhu has quit IRC13:07
*** ryansb has quit IRC13:13
*** noyitz has quit IRC13:14
*** noyitz has joined #heat13:16
*** ryansb has joined #heat13:16
*** ryansb has quit IRC13:16
*** ryansb has joined #heat13:16
*** signed8bit has joined #heat13:20
*** EricGonczer_ has joined #heat13:24
*** EricGonczer_ has quit IRC13:26
*** noyitz has quit IRC13:28
*** noyitz has joined #heat13:34
*** sdake has joined #heat13:36
*** signed8bit has quit IRC13:37
*** sdake has quit IRC14:17
*** noyitz has quit IRC14:28
*** noyitz has joined #heat14:29
*** signed8bit has joined #heat14:47
*** Qiming has quit IRC14:50
*** Qiming has joined #heat14:50
*** serg_melikyan has joined #heat14:53
*** serg_melikyan has quit IRC14:57
*** signed8bit has quit IRC15:03
*** EricGonczer_ has joined #heat15:19
*** sorantis has joined #heat15:40
*** Qiming has quit IRC15:42
*** EricGonczer_ has quit IRC15:51
*** alexpilotti has joined #heat16:05
*** sorantis has quit IRC16:12
*** alexpilotti has quit IRC16:22
*** tspatzier has joined #heat16:27
*** alexpilotti has joined #heat16:30
*** sorantis has joined #heat16:35
*** tspatzier has quit IRC16:42
*** sorantis has quit IRC16:48
*** LimorStotland has quit IRC16:52
*** tspatzier has joined #heat16:55
*** smccully has quit IRC16:58
*** jtomasek has quit IRC17:14
openstackgerritPavlo Shchelokovskyy proposed openstack/heat: Remove TaskRunner from Volume resources  https://review.openstack.org/15497717:16
*** MasterPiece has joined #heat17:23
*** MasterPiece has quit IRC17:27
*** MasterPiece has joined #heat17:27
*** MasterPiece has quit IRC17:29
*** MasterPiece has joined #heat17:30
*** noyitz has joined #heat17:30
*** MasterPiece has quit IRC17:33
*** MasterPiece has joined #heat17:33
*** MasterPiece has joined #heat17:33
*** noyitz has quit IRC17:35
*** MasterPiece has quit IRC17:35
*** MasterPiece has joined #heat17:36
*** serg_melikyan has joined #heat18:00
*** serg_melikyan has quit IRC18:07
*** serg_melikyan has joined #heat18:11
*** tspatzier has quit IRC18:13
*** tspatzier has joined #heat18:14
*** tspatzier has quit IRC18:14
*** serg_melikyan has quit IRC18:20
*** tspatzier has joined #heat18:20
*** serg_melikyan has joined #heat18:22
*** hdd has joined #heat18:25
*** tspatzier has quit IRC18:35
*** cody-somerville has joined #heat18:48
*** jtomasek has joined #heat18:52
*** hdd has quit IRC19:13
*** cdent has joined #heat19:16
*** cody-somerville has quit IRC19:26
*** achanda has joined #heat19:30
*** alex_xu has quit IRC19:33
*** alex_xu has joined #heat19:34
*** serg_melikyan has quit IRC19:34
*** serg_melikyan has joined #heat19:38
*** hdd has joined #heat19:50
*** serg_melikyan has quit IRC19:52
*** serg_melikyan has joined #heat19:54
*** sdake has joined #heat20:11
*** sdake_ has joined #heat20:14
*** sdake_ has quit IRC20:15
*** sdake has quit IRC20:17
*** hdd has quit IRC20:20
*** kevinbenton has quit IRC20:27
*** rdo has quit IRC20:36
*** rdo has joined #heat20:38
*** achanda has quit IRC20:54
*** cdent has quit IRC20:56
*** serg_melikyan has quit IRC21:10
*** asalkeld has joined #heat21:27
*** alexpilotti has quit IRC21:28
asalkeldmorning21:29
stevebakermorning21:30
asalkeldhi stevebaker anything happen that i need to know about?21:40
stevebakerasalkeld: I don't think so. I believe we're accumulating fixes in master in preperation for doing a bulk backport for an rc221:41
asalkeldok, are there lots?21:41
*** achanda has joined #heat21:48
stevebakerasalkeld: 9 fix-committed https://bugs.launchpad.net/heat/+bugs?field.tag=kilo-rc-potential21:52
asalkeldthanks stevebaker , lots of admin catchup :-O21:56
stevebakerheh, it seems to be reasonably under control.21:58
asalkeldstevebaker: (also non-heat upstream stuff)22:00
asalkeldga, emails...22:00
stevebakerstart digging!22:00
asalkeldi wonder how this "person-day" thing was calculated: http://stackalytics.com/?module=heat-group&metric=person-day22:09
asalkeldproposal bot putting some hard time in;)22:09
asalkeldelynn miguelgrinberg prazumovsky are you going to summit?22:15
*** uros_ has joined #heat22:18
flwangstevebaker: ping22:38
stevebakerflwang: hi22:39
flwangstevebaker: i'm feilong, and we have met many times though you may still can't remember who am i :)22:40
stevebakerflwang: lol, you can mock me in vancouver if you wish22:40
flwangcatalyst don't have more money for me to go to vancouver :D22:41
flwangmay be we can meet at the meetup of wellington22:41
flwangseriously, I would like to get some suggestion about the integration of heat and zaqar22:42
flwangstevebaker: now i'm reviewing the code of https://github.com/openstack/heat/blob/master/heat/engine/event.py22:42
stevebakerflwang: ah yes, I talked to Bruno last week22:42
flwangright22:42
stevebakerflwang: did you seen the thread on os-dev about a user notification API which asalkeld started?22:43
flwangstevebaker: it would be nice if you can share the link to remind my memory22:44
flwangit's still in weekend mode22:45
stevebakerflwang: the thread devolved a bit, but I think the upshot was that zaqar is still the most appropriate tool to give users notifications http://lists.openstack.org/pipermail/openstack-dev/2015-April/060748.html22:47
asalkeldnot totally happy about that22:48
asalkeldzaqar still has a lot to do22:48
flwangasalkeld: would you mind explaining more?22:48
asalkeldflwang: first read that thread22:49
flwangstevebaker: yep, for that thread, I almost went through it at all22:49
asalkeldmy main issue is mirantis is not super keen to support zaqar22:49
asalkeldlots of work to fully integrate a project, need to get resources on the project etc...22:50
asalkeldpita22:50
flwangI don't think Zaqar is installed in most installations (tho'please correct me here if this  is wrong)   ???22:50
asalkeldi'd guess it's available in rdo, but not much else22:51
flwangasalkeld: so you're afraid zaqar is not really ready, is it? just because Mirantis is not deploying it?22:52
asalkeldflwang: it needs other api too, something like atomhopper22:53
asalkeldflwang: it was very controversial during incubation22:54
asalkeldmany people are very sceptical about the aim and design22:54
stevebakerasalkeld: these are all features which can be added22:54
flwangasalkeld: yep, I can see your point, but we can may it configurable for heat22:55
asalkeldflwang: so what do we use if zaqar is not availble?22:55
stevebakerasalkeld: I think most of that was a misunderstanding of semantics22:55
flwangasalkeld: incubation is not really equal to not available, right?22:56
asalkeldflwang: of course22:56
flwanggiven the TC is considering the def core stuff22:56
asalkeldi find this big tent stuff dubious, in cases like this. how can we depend on something that is not install most of the time22:58
stevebakerI think zaqar would quickly find itself in the openstack namespace if it applied today. I'm sure Flavio will be doing that when he is ready22:58
flwangwe have to admit that the incubation status is blocking some potential originations to deploy it22:58
asalkeldhttps://github.com/openstack/zaqar22:58
asalkeldit is there now22:58
stevebakeroh, sweet22:58
asalkeldtho' that means nothing22:59
asalkeldi think we have just killed the meaning of https://github.com/openstack/22:59
asalkeldimho22:59
flwangasalkeld: so despite above, is there any feature you believe zaqar is missing to fit the requirement of heat?23:00
flwangasalkeld: seems we're mostly talking about the stuff should be worried by TC :)23:00
stevebakerif a project has benefits for cloud users, it will get deployed in clouds. If it has benefits for us then we should use it (optionally)23:00
asalkeldflwang: just finding the link23:01
flwangasalkeld: you mean the link of the mail thread?23:05
asalkeldyeah23:05
flwangas for the cons of zaqar I just saw you wrote "[cons] I don't think Zaqar is installed in most installations (tho' please correct me here if this is wrong). I know Mirantis does not currently support Zaqar, so that would be a problem for me. There is not the level of external tooling like in option "1" (logstash and friends)23:06
flwangso based on my understanding, one thing is the matured status, and the other thing is if zaqar can provide better log mechanism, am i missing anything?23:07
stevebakerwe built a list in an etherpad in paris, it would be good to dig that up23:11
flwangstevebaker: do you think it's reasonable if we just make zaqar optional for now? and replace current event mechanism when it's fully ready?23:12
stevebakerflwang: definately optional for now23:12
asalkeldhttp://lists.openstack.org/pipermail/openstack-dev/2015-April/061224.html23:13
stevebakerhere we go https://etherpad.openstack.org/p/kilo-zaqar-summit-integration-with-services23:13
asalkeldCreate a micro-service whose sole purpose in life is to filter  notifications and redirect to zaqar.23:13
asalkeldstevebaker: is etherpad working for you?23:13
stevebakeryep23:13
stevebakertry refresh23:14
asalkeldUncaught Error: Attempt to load undefined module. in https://etherpad.openstack.org/static/js/require-kernel.js (line 1)23:14
asalkeldok, seems to work now23:14
*** achanda has quit IRC23:15
flwangasalkeld: are you aware if there any existing 'micro-service' we can leverage or we have to cook one in openstack kitchen?23:17
asalkeldflwang: i'd assume it would just be a new endpoint in zaqar/ceilometer23:18
flwangwill the 'micro-service' be a single/separated project?23:18
flwangasalkeld: if so, i think we do have a similar endpoint can be leveraged? wait a sec23:19
flwangasalkeld: https://review.openstack.org/#/c/146354/2/specs/kilo/approved/notification-api.rst23:20
flwangwe just implemented a new subscription endpoint for notification service, maybe we can use it as the 'micro-service'23:21
asalkeldso this is to pull rpc notifications into zaqar messages23:21
asalkeldlifeless: what's your opinion of zaqar and would HP run an instance of zaqar for user notifications?23:22
flwangnope, the endpoint will create a subscription for a queue/topic, so when there is a msg posted to the queue/topic, then all the subscribers will be notified23:22
asalkeldflwang: so that's what i mean, we need an endpoint that subscribes to rpc notifications and pulls them into a zaqar message for the user23:23
asalkeld(and sanitises them too)23:23
*** vkmc has joined #heat23:24
vkmchi o/23:24
lifelessasalkeld: uhm, I can only answer one of those23:24
asalkeld1 > 023:24
flwangvkmc: we're talking about if we can work out a 'micro-service',see http://lists.openstack.org/pipermail/openstack-dev/2015-April/061224.html23:25
*** xuhaiwei has quit IRC23:26
vkmcflwang, I see, thanks23:26
lifelessasalkeld: Underwhelmed23:26
*** xuhaiwei has joined #heat23:27
asalkeldlifeless: yeah , I am not totally happy with relying on it if people are not going to install it23:27
*** sdake has joined #heat23:28
flwangasalkeld: but we do need a notification/messaging service like zaqar, right? if there is no other perfect option, why not try zaqar can let us make it better?23:29
flwangs/can/and23:29
vkmcif Zaqar fits this use case and it makes things easier for Heat users, we are willing to help with the integration23:31
*** tochi has joined #heat23:34
stevebakerI think it is something worth persuing23:35
vkmcZaqar is part of OpenStack, so I don't see any reason people will decide not to install it... more if its needed for this Heat feature23:35
*** chlong has joined #heat23:35
flwangstevebaker: asalkeld: so may I get your opinion if i can propose patches to save events in zaqar, at https://github.com/openstack/heat/blob/master/heat/engine/event.py   and it's definitely configurable. so it won't break any existing functions23:38
asalkeldflwang: i don't think we want to do that23:39
flwangdoes that make any sense?23:39
asalkeldwe want to send rpc notifications, and then pull those into zaqar23:39
asalkeldflwang: this needs a lot more thought imo23:39
asalkelddon't rush with specs/patches just yet23:40
flwangasalkeld: my thoughts from here http://lists.openstack.org/pipermail/openstack-dev/2014-September/046361.html23:40
flwangasalkeld: but at here(https://github.com/openstack/heat/blob/master/heat/engine/event.py ), seems these events are saved into database23:41
stevebakerflwang: a solution pushing filtered rpc notifications to zaqar would benefit all event-producing projects, not just heat23:41
lifelessvkmc: 'part of openstack' doesn't mean 'everyone will install it'23:43
lifelessvkmc: the oldest example of this is swift - many folk deploy ceph instead23:43
vkmclifeless, I agree, but it seems more reasonable to add a messaging solution within OpenStack than a non OpenStack alternative23:44
vkmclifeless, of course that customers will choose whatever fits their needs better23:44
flwangso, seems the way we're going to follow is avoiding changing current code, but work out a central endpoint to be the proxy to forward the message to zaqar, is it?23:44
asalkeldflwang: we can add that subscription api as a genenic entity maybe23:45
asalkeldso you don't have to have zaqar as the implementation23:45
asalkeldbut could have yagi/attomhopper instead23:46
flwangand IMO, it's just like asalkeld mentioned above, it's most like a new endpoint in zaqar or ceilometer which will be the 'proxy'23:46
flwangright?23:46
asalkeldyeah23:47
*** EricGonczer_ has joined #heat23:47
flwangand given ceilometer has already subscribe all the notifications/rpc from all the components, so it could be best candidate to be the proxy23:47
lifelessvkmc: non openstack being e.g. zmq or amqp ?23:48
flwanglifeless: IMHO, zmq/amqp is not on the same layer with zaqar23:48
flwangvkmc: pls correct me23:48
*** achanda has joined #heat23:50
vkmcflwang, lifeless, yeah, zmq/amqp would be more comparable to Oslo messaging23:50
vkmcflwang, lifeless, Zaqar is more comparable to StormMQ23:51
*** jamielennox|away is now known as jamielennox23:51
lifelessvkmc: you know stormmq uses amqp1.023:54
lifelessvkmc: right?23:54
vkmclifeless, I know, but its for cloud services23:55
*** EricGonczer_ has quit IRC23:56
vkmclifeless, plain amqp 1.0 can be used for the same purpose but the use cases are wider23:56
lifelessvkmc: I think we're going round in circles here.23:56
*** achanda has quit IRC23:56
*** EricGonczer_ has joined #heat23:56
lifelessThe proposition 'in openstack so people will deploy it' is clearly false.23:56
vkmclifeless, sorry, I don't want to generate confusion23:57
lifelessThe argument 'choosing a substrate that is 'in openstack because that means it will be deployed' is thus false23:57
lifelessAligning with other projects that are working with OpenStack is good because we all want the same things.23:58
*** EricGonczer_ has quit IRC23:58
lifelessBut if deployability is a concern, assess that directly, on its merits.23:58
asalkeldlifeless: big tent failure IMHO23:58
lifelessasalkeld: I agree.23:58
vkmcmaybe I just added a personal judgment to my statement23:58
vkmcI *personally* will choose an project in OpenStack instead of something outside OpenStack23:59
vkmcand I agree that its not true for everyone23:59
flwangasalkeld: lifeless: did you guys mean the big tent failed? :)23:59
*** EricGonczer_ has joined #heat23:59
lifelessflwang: I think the big tent initiative is a rather huge problem. We'll see how it pans out.23:59
« Saturday, 2015-04-18 Index Monday, 2015-04-20 »

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