14:00:05 <acabot> #startmeeting watcher
14:00:05 <openstack> Meeting started Wed May 18 14:00:05 2016 UTC and is due to finish in 60 minutes.  The chair is acabot. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:00:06 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
14:00:08 <openstack> The meeting name has been set to 'watcher'
14:00:13 <acabot> hi
14:00:17 <dtardivel> hi
14:00:17 <jed56> hi
14:00:19 <edleafe> \o
14:00:21 <ecelik> hi
14:00:27 <hvprash> hi
14:00:29 <acabot> agenda for today #link https://wiki.openstack.org/wiki/Watcher_Meeting_Agenda#05.2F18.2F2016
14:00:30 <jwcroppe> \o
14:00:46 <xrollla> hi
14:01:00 <vincentfrancoise> o/
14:01:13 <acabot> #topic Announcements
14:01:45 <seanmurphy> hi
14:02:02 <acabot> #info Craton project (fleet management) officially launched and will organize its mid-cycle at the same place & time than Watcher
14:03:01 <acabot> I had discussion with Jim about this project in Austin, they plan to build a tool to deploy services on very large scale infrastructures
14:03:06 <hvprash> acabot, i was in that meeting and suggested it
14:03:48 <sballe_> morning
14:03:56 <acabot> hvprash : yes, thank you for introducing me, I think this project is interesting and we need to follow where they want to go
14:04:21 <gzhai2> hi
14:04:43 <edleafe> So Watcher, Craton, and Nova all in Hillsboro at the same time?
14:04:51 <acabot> today, it is dedicated to very large scale infra but they are interested by the "audit part" of Watcher, I will plan sync time with Craton team in Hilsboro
14:04:52 <sballe_> acabot: what does deploy at large scale mean?
14:05:13 <acabot> sballe_: it means more than 1000 nodes
14:05:21 <sballe_> edleafe: I hadn't heard Craton but I knew that Water and nova
14:06:00 <acabot> sballe_ : https://etherpad.openstack.org/p/craton-meeting-2016-05-16
14:06:08 <sballe_> but what does deploy mean? manage? are they going to rewrite nova for that? nova doesn't scale to taht scale so are they looking at many noa islands
14:06:43 <hvprash> sballe_, one of the objective is to collect the metadata.. think like a CMDB
14:06:49 <acabot> sballe_ : you should look at the etherpad ;-)
14:07:04 <sballe_> perfect and that is what though. Deploy is just such a buzzword
14:07:45 <sballe_> I have read the etherpad... and to me it is a CMDB like hvprash points out
14:07:56 <sballe_> and so deploy is not really in play
14:07:57 <hvprash> basically collectiing all the info in a datacenter like network topology, how its connected to the tor etc
14:08:16 <sballe_> hvprash: and we need thta
14:08:48 <acabot> sballe_ : I think we need to discuss with them, I hope they will be in Hillsboro
14:08:52 <hvprash> sballe_, will be interesting to meet them in hillsboro
14:09:08 <jed56> +1
14:09:17 <sballe_> I agree. Again I hadn't heard they would be there too
14:09:38 <acabot> #action acabot sync with Craton to define a cross-project meeting during our myd-cycle meetup
14:09:51 <acabot> #info I wrote down a commit msg for the big tent that is open to reviews now https://etherpad.openstack.org/p/watcher-big-tent-commit-msg
14:10:39 <acabot> I plan to submit our project after next meeting (1 week to review the commit msg)
14:11:10 <acabot> #action acabot submit Watcher to the big tent after our next meeting
14:11:16 <sballe_> acabot: thx I will lok at it
14:11:38 <acabot> #action sballe_ jwcroppe review https://etherpad.openstack.org/p/watcher-big-tent-commit-msg
14:12:19 <edleafe> acabot: I'll look it over too. This is great progress!
14:12:23 <acabot> sync with congress, I wasn't able to attend their meeting last week, did someone attend ?
14:12:34 <acabot> #action edleafe review https://etherpad.openstack.org/p/watcher-big-tent-commit-msg
14:12:55 <jwcroppe> acabot: reading now
14:13:12 <acabot> #info Dont forget that meeting time has changed every 2 weeks starting next week
14:13:56 <acabot> any feedback about Congress delegation for VM placement ? #link https://docs.google.com/document/d/1ksDilJYXV-5AXWON8PLMedDKr9NpS8VbT0jIy_MIEtI/edit#
14:14:37 <jed56> acabot what kind of feedback do you want ?
14:15:20 <acabot> I think the TC will look at it regarding Congress/Watcher overlap so I'd like to have everybody clear the difference between the way we do optimization and the way Congress wants to do it
14:15:35 <acabot> jed56 : is it the same thing than Watcher ?
14:15:43 <sballe_> jed56 I looked at it and to me it is enforcing static policies a lot like what Nova should be doing if it had a good architecture
14:15:47 <acabot> jed56 : but I know your opinion ;-)
14:16:28 <jed56> Congres has not the same approach but the finality is the same
14:16:33 <acabot> sballe_ : I agree with you, just to be sure that we share the same opinion in the team
14:17:05 <acabot> #info Watcher puppet repository creation currently in review https://review.openstack.org/#/c/315431/
14:17:17 <acabot> any other announcements ?
14:17:21 <dtardivel> acabot: I'm pushing a new patchset
14:17:40 <dtardivel> for puppt
14:17:45 <dtardivel> for puppet-watcher
14:17:49 <acabot> dtardivel : ok thx
14:18:08 <acabot> #action dtardivel submit a new PS for https://review.openstack.org/#/c/315431/
14:18:19 <acabot> #topic Review Action Items
14:18:33 <jed56> acabot: regarding congress do we need to add some information in our documentation ?
14:19:15 <acabot> jed56 : for now, I think we need to clarify things in the commit msg for the big tent
14:19:30 <jed56> okay
14:19:33 <acabot> Watcher specs
14:20:05 <acabot> https://review.openstack.org/#/c/312143/ needs review from core before EOW
14:20:20 <acabot> #action acabot jwcroppe sballe_ review https://review.openstack.org/#/c/275474/
14:20:53 <acabot> Airflow event based migration strategy spec needs reviews from core #link https://review.openstack.org/#/c/275474/
14:21:35 <acabot> #action acabot jwcroppe sballe_ review https://review.openstack.org/#/c/275474/
14:21:44 <acabot> Graph model describing virtual and physical elements in a data center & Consolidation of multiple Action-plans into a single one & Add Watcher Continuosly Optimization spec need reviews
14:22:11 <acabot> we need more reviewers on specs
14:22:23 <acabot> does anyone wants to contribute on this ?
14:22:41 <gzhai2> I can review 312143
14:23:08 <hvprash> +1
14:23:24 <acabot> #action gzhai2 hvprash review https://review.openstack.org/#/c/312143/
14:23:50 <sballe_> +1
14:23:53 <jed56> acabot: i will read it
14:24:08 <xrollla> hi all, I and alexchadin are working on "https://review.openstack.org/#/c/315551/". We've already implemented this to the Watcher, python-watcherclient and are working on the tests at the moment. Watcher-dashboard changes will be done later.
14:24:48 <sballe_> xrollla: acabot I will take a look at 315551
14:25:11 <jinquan> i will review the Add Watcher Continuosly Optimization spec
14:25:36 <acabot> #action sballe_ jinquan acabot review https://review.openstack.org/#/c/315551/
14:25:38 <vincentfrancoise> I'll try to review 315551 by EOW
14:25:39 <jwcroppe> reviewing 275474
14:26:05 <acabot> #action vincentfrancoise review https://review.openstack.org/#/c/315551/
14:26:33 <acabot> hvprash : is https://review.openstack.org/#/c/317130/ still in WIP ?
14:26:50 <hvprash> acabot, Though the spec is currently WIP, can the core parallel review ? I may need some help understanding the code impact since this is first time I am doing this :)
14:27:00 <acabot> ok
14:27:18 <acabot> #action jwcroppe sballe_ acabot review https://review.openstack.org/#/c/317130/ in WIP
14:27:19 <hvprash> i have update the data model impact. was thinking to hhave an audit version table
14:27:22 <xrollla> sballe_: thanks :)
14:27:46 <hvprash> thx jwcroppe sballe_
14:27:49 <acabot> Cluster model objects wrapper work items are still missing #link https://review.openstack.org/#/c/287019/ (for a couple of weeks...)
14:28:44 <acabot> jwcroppe : any update on https://review.openstack.org/#/c/287019/ ?
14:30:00 <acabot> #info https://review.openstack.org/#/q/status:merged+project:openstack/watcher+branch:master+topic:bp/get-goal-from-strategy has been merged last week
14:30:35 <acabot> thx vincentfrancoise for this great contribution
14:30:51 <edleafe> acabot: on the commit message etherpad, how do you want suggested changes added? Strike out the old, and type in the new? Or some other way?
14:31:09 <sballe_> vincentfrancoise: +1
14:31:10 <acabot> edleafe : Strike out the old, and type in the new
14:31:20 <edleafe> acabot: ok
14:32:05 <acabot> 2 strategies need reviews https://review.openstack.org/#/c/292188/ & https://review.openstack.org/#/c/297590/
14:32:14 <acabot> who wants to review ?
14:32:22 <sballe_> me
14:32:34 <sballe_> I will sepnd the afternon at it
14:32:49 <jed56> i just finish reviewing 297590
14:32:50 <acabot> #action sballe_ review implemented strategies https://review.openstack.org/#/c/292188/ & https://review.openstack.org/#/c/297590/
14:32:57 <jed56> i can review 292188
14:33:13 <jinquan> ican review 297590
14:33:21 <acabot> #action jed56 review https://review.openstack.org/#/c/292188/
14:33:39 <acabot> #action jinquan review https://review.openstack.org/#/c/297590/
14:34:13 <acabot> we also need reviews on openstack-plugin #link https://review.openstack.org/#/q/status:open+project:openstack/python-watcherclient+branch:master+topic:bp/openstackclient-plugin
14:34:30 <acabot> for now, only vincentfrancoise & dtardivel have reviewed this code
14:35:41 <acabot> #topic Blueprint/Bug Review and Discussion
14:36:30 <acabot> #info Newton-1 release is due in 10 days so we need to start cleaning on-going BPs
14:36:44 <acabot> #link https://launchpad.net/watcher/+milestone/newton-1
14:37:18 <acabot> xrollla : I dont think https://blueprints.launchpad.net/watcher/+spec/continuously-optimization will be implemented for Newton-1 right ?
14:38:03 <acabot> vincentfrancoise : do you think we will be able to deliver https://blueprints.launchpad.net/watcher/+spec/efficacy-indicator in 10 days ?
14:38:47 <vincentfrancoise> https://blueprints.launchpad.net/watcher/+spec/efficacy-indicator is wal on its way but it depends on https://blueprints.launchpad.net/watcher/+spec/persistent-audit-parameters
14:38:51 <vincentfrancoise> well*
14:39:08 <vincentfrancoise> by EOW I should be able to push the code for review
14:39:34 <vincentfrancoise> but the las piece which consist of comparing the strategies cannot be finished because of the dependency
14:39:42 <acabot> ok so I will keep it in Newton-1 for now
14:40:03 <vincentfrancoise> but the rest can be merged without persistent-audit-parameters <https://blueprints.launchpad.net/watcher/+spec/persistent-audit-parameters>
14:40:16 <xrollla> acabot: I'll check with alexchadin now
14:40:21 <acabot> we really need more code reviewers until next week if we want to have 6 BPs implemented for Newton-1
14:41:15 <seanmurphy> i can help out
14:41:30 <seanmurphy> i can do one
14:41:34 <acabot> seanmurphy : do you want to review code as well ?
14:41:40 <seanmurphy> i can do tis
14:41:41 <seanmurphy> this
14:41:59 <dtardivel> acabot: I will review also this one "Added Goals and Strategies to Dashbooard" #link https://review.openstack.org/#/c/301516/
14:42:21 <acabot> #action seanmurphy look at the list of BPs that need reviews on https://launchpad.net/watcher/+milestone/newton-1
14:42:33 <sballe_> acabot: we just need more reviwer. The more people that give a +1 the less work the cores have to do to give it +2
14:42:43 <acabot> #action dtardivel review https://review.openstack.org/#/c/301516/
14:43:07 <acabot> sballe_ : yes, +1 is important !
14:43:11 <sballe_> that is how it worked in Neutron at some points. The cores would do a quick review and basedon all the +1s whould 2
14:43:26 <acabot> sballe_ : we need more +1 :-D
14:43:51 <vincentfrancoise> +1 ^
14:44:07 <acabot> seanmurphy : Voljtech has a bug in review https://bugs.launchpad.net/watcher/+bug/1564297 and there is really nothing to do to close it, can he work on it by EOW ?
14:44:09 <openstack> Launchpad bug 1564297 in watcher "Add missing cold (non-live) VM migration support" [Wishlist,Triaged] - Assigned to Vojtech Cima (cima-vojtech)
14:44:14 <xrollla> acabot: alexchadin considers that for review we will publish it to the next meeting
14:44:41 <acabot> xrolla : ok in this case I will postpone it to Newton-2
14:44:58 <acabot> #action acabot move https://blueprints.launchpad.net/watcher/+spec/continuously-optimization to Newton-2
14:45:01 <seanmurphy> @acabot - he cannot - he is on vacation right now - we can look at it, but no guarantees it will be done by EOW
14:45:33 <acabot> seanmurphy : please let me know if he can do it before end of next week (Newton-1 release)
14:45:40 <seanmurphy> ok - will do
14:45:55 <vincentfrancoise> seanmurphy: it's just some docstring to modify if I'm not mistaken
14:46:13 <acabot> gzhai2 : any update on https://blueprints.launchpad.net/watcher/+spec/optimization-threshold ?
14:46:30 <gzhai2> acabot: in rebase.
14:46:51 <acabot> gzhai2 : I see, can you rebase it ?
14:47:02 <gzhai2> yes.
14:47:04 <acabot> thx
14:47:13 <acabot> #action rebase https://review.openstack.org/#/c/297556/
14:47:19 <acabot> #action gzhai2 rebase https://review.openstack.org/#/c/297556/
14:47:43 <acabot> #topic Open Discussion
14:47:53 <acabot> which keyword should we use in OpenStack official CLI to identify Watcher service (currently infra-optim) ?
14:48:12 <acabot> dtardivel : can you give us more details ?
14:49:04 <gzhai2> acabot: Anybody request for OSIC nodes?
14:49:07 <dtardivel> with openstackcli, when we introduce  a new plugin, as watcher, we have to use a keyword to indicate command is linked to watcher
14:49:31 <vincentfrancoise> e.g.
14:49:39 <dtardivel> for now, if the patchset proposed by vincentfrancoise we have : openstack infraoptim audit list
14:49:49 <dtardivel> s/if/in
14:49:57 <edleafe> dtardivel: understood, but 'infra-optim' doesn't sound good
14:50:16 * edleafe needs to think of a better one
14:50:19 <vincentfrancoise> edleafe: agreed
14:50:32 <jinquan> edleafe +1
14:50:36 <jed56> +1
14:50:37 <dtardivel> the question is: which keywork could be better for watcher command line in openstack CLI.
14:50:47 <dtardivel> vincentfrancoise proposed 'optimize'
14:50:51 <hvprash> +1
14:51:00 <sballe_> +1
14:51:00 <vincentfrancoise> I would suggest "optmization" to be more precise
14:51:15 <edleafe> dtardivel: yeah, 'optimize' is too subjective
14:51:25 <acabot> so it would be "openstack optimize audit list" sounds weird no ?
14:51:25 <vincentfrancoise> so it would look like "openstack optimization audit list"
14:51:37 <edleafe> maybe 'balance'?
14:51:51 <jed56> openstack watcher audit list ?
14:52:03 <sballe_> balance is too general
14:52:10 <edleafe> jed56: you can't use code names
14:52:23 <jed56> edleafe: I didn't know :)
14:52:27 <edleafe> sballe_: so is optimize :-P
14:52:29 <hvprash> optimiz* is better i think
14:52:33 <dtardivel> congress did it :) openstack congress ....
14:52:41 <edleafe> really?
14:52:45 <sballe_> then I vote for watcher
14:53:00 <vincentfrancoise> congress is actually the only one to do so
14:53:05 <sballe_> we could quickly settle on optimize and trademark it
14:53:13 <vincentfrancoise> everyone else is referring to the service it delivers
14:53:16 <sballe_> so noboy else uses optimization or optimize
14:53:19 <jed56> sballe_: +2
14:53:29 <vincentfrancoise> sballe_: +2 that was my idea with optimization
14:53:37 <dtardivel> sballe_: +2
14:53:40 <hvprash> sballe_: +2
14:53:48 <edleafe> I would expect some pushback from others on 'optimize'
14:53:54 <sballe_> lol
14:54:16 <acabot> #startvote Keyword for OpenStack CLI optimize watcher optimization balance
14:54:17 <openstack> Unable to parse vote topic and options.
14:54:18 <edleafe> naming is hard
14:54:43 <acabot> #startvote Keyword for OpenStack CLI ? optimize, watcher, optimization, balance
14:54:43 <openstack> Begin voting on: Keyword for OpenStack CLI ? Valid vote options are optimize, watcher, optimization, balance.
14:54:44 <openstack> Vote using '#vote OPTION'. Only your last vote counts.
14:54:45 <vincentfrancoise> optimizatoin would be coherent with the remit stated in the big tent commit message
14:55:02 <acabot> #vote optimize
14:55:13 <sballe_> #vote optimize
14:55:13 <dtardivel> #vote optimization
14:55:15 <vincentfrancoise> #vote optimization
14:55:23 <jed56> #vote optimize
14:55:25 <jinquan> irc://chat.freenode.net:6667/#vote optimize
14:55:33 <hvprash> #vote optimize
14:55:41 <edleafe> #vote optimize
14:56:08 <acabot> #endvote
14:56:09 <openstack> Voted on "Keyword for OpenStack CLI ?" Results are
14:56:10 <openstack> optimization (2): vincentfrancoise, dtardivel
14:56:12 <openstack> optimize (5): hvprash, edleafe, jed56, acabot, sballe_
14:56:25 <acabot> go for "optimize"
14:56:31 <sballe_> +2
14:56:38 <seanmurphy> (we were just abt to vote optimize! ;-)
14:56:47 <acabot> #action vincentfrancoise update Watcher keywork in CLI by "optimize"
14:56:59 <acabot> seanmurphy : too slow ;-)
14:57:05 <sballe_> lol
14:57:07 <hvprash> :)
14:57:20 <acabot> should we already ask for an infra to OSIC before being part of the big tent ?
14:57:47 <acabot> I think we should wait until we submit Watcher to the big tent
14:58:01 <gzhai2> ok
14:58:01 <sballe_> just submit for the big tent and in parallel ask for the cluster
14:58:19 <acabot> but gzhai2 said that there are a lot of requests
14:58:33 <sballe_> gzhai2: has a request in for an internal cluster too
14:58:41 <sballe_> gzhai2: what's the status on that?
14:59:04 <gzhai2> pls. just give me test requirements and time.
14:59:08 <sballe_> acabot: if we submit the request now we get inthe queue
14:59:20 <gzhai2> I can try to ask for a 30+ nodes in internal cluster.
14:59:32 <acabot> yes but I cannt mention that we will be in the big tent soon in the form..
14:59:46 <acabot> 30 nodes would be great
14:59:57 <acabot> time to end the meeting
15:00:02 <acabot> thx
15:00:06 <vincentfrancoise> bye
15:00:09 <jinquan> bye
15:00:11 <acabot> #endmeeting