14:00:18 <acabot> #startmeeting watcher 14:00:18 <openstack> Meeting started Wed Mar 16 14:00:18 2016 UTC and is due to finish in 60 minutes. The chair is acabot. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:00:19 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:00:21 <dtardivel> hi 14:00:22 <openstack> The meeting name has been set to 'watcher' 14:00:29 <acabot> hi 14:00:35 <alexchadin> Hi! 14:01:15 <jed56> hello 14:01:23 <tkaczynski> hi 14:01:41 <gzhai> hello 14:01:42 <acabot> our agenda for today #link https://wiki.openstack.org/wiki/Watcher_Meeting_Agenda#03.2F16.2F2016 14:01:43 <vincentfrancoise> o/ 14:01:52 <sballe_> morning 14:02:47 <acabot> #topic Annoucements 14:02:48 <brunograz> hi 14:02:56 <acabot> #topic Announcements 14:03:13 <acabot> #info PTL election is now open to candidacy on the ML 14:03:52 <acabot> candidacy will end tomorrow 14:03:52 <sballe_> one quick tpic. I am looking into the meetup in July and it will probably be in Hillsboro, OR. ~ 20 mins from POrtland 14:04:26 <sballe_> I cannot host it in the Intel MA facility 14:04:31 <acabot> #info mid-cycle meetup for Newton will probably be in Hillsboro, OR. ~ 20 mins from POrtland 14:04:48 <gzhai> In july? 14:04:57 <acabot> sballe_ : is it still an Intel building in Hillsboro ? 14:05:08 <sballe_> yes 14:05:26 <sballe_> it is a large site with most of our OpeSNtack activities there 14:05:33 <acabot> gzhai : mid-cycle is planned for week 30 14:05:45 <gzhai> acabot:got it 14:05:56 <acabot> sballe_ : can we already fix the dates ? 14:06:38 <sballe_> no. also if we do it week 30 I won;t be able to attend given that I am going on vacation that weekend 14:07:03 <sballe_> but I am tryign for week 30 and week 29 14:07:37 <acabot> sballe_: ok 14:07:39 <sballe_> let's see what they tell me 14:08:11 <acabot> #action sballe_ ask for meeting room for the Newton mid-cycle 14:08:33 <tkaczynski> sballe_: it would be a weird meeting without you, our host :) 14:08:41 <sballe_> lol 14:08:57 <sballe_> I would find somebody t take over ;) 14:09:01 <gzhai> +1 14:09:05 <acabot> #info sballe_ shared a first draft of slides for our talk in Austin #link https://docs.google.com/presentation/d/1xzuIgmvRXiwTxv4R0al8DZoFd7u6OA6hx6aUrid7JPw/edit?usp=sharing 14:09:24 <acabot> feel free to add comments on the slideset, its open to editing 14:09:53 <acabot> any other announcement ? 14:10:31 <acabot> #topic Review action items 14:10:58 <acabot> #info we have many specs open that needs reviews #link https://review.openstack.org/#/q/project:openstack/watcher-specs+is:open 14:11:25 <acabot> sballe_ : we need to work on it asap as its becoming a bottleneck for the dev team 14:12:02 <dtardivel> review on #link https://review.openstack.org/#/c/287019/ done. tpeoples will propose another patch tomorrow 14:12:05 <acabot> some of them are targeted for RC-1 and RC-final and we wont be able to handle them if specs are not reviewed 14:12:08 <sballe_> I can work on some today. I looked at it last weekend 14:13:10 <acabot> #action sballe_ jwcroppe acabot review open specs https://review.openstack.org/#/q/project:openstack/watcher-specs+is:open 14:13:43 <tkaczynski> I can also have a look at some specs 14:13:45 <seanmurphy> i can review a couple 14:13:47 <seanmurphy> by EOW 14:14:12 <acabot> #action stkaczynski seanmurphy review open specs https://review.openstack.org/#/q/project:openstack/watcher-specs+is:open 14:14:17 <acabot> thx you guys 14:15:02 <seanmurphy> i will take https://review.openstack.org/#/c/286153/ and https://review.openstack.org/#/c/287019/ 14:15:10 <acabot> alexchadin : sorry for giving you late feedback, we will give more by EOW 14:15:46 <alexchadin> Nothing to worry, I'm still waiting for your feedback:) 14:16:31 <sballe_> Is http://review.penstack.org/#/c/275474 also in need of a review? or it is too low in the priority list? 14:16:45 <acabot> #info workload balance migration strategy spec has been merged and code is already available for review (thank you gzhai) https://review.openstack.org/#/c/292188/ 14:16:51 <sballe_> https://review.openstack.org/#/c/275474/ 14:17:50 <tkaczynski> fyi: I updated the spec for scoring module, would appreciate the feedback 14:18:05 <acabot> sballe_: you are right, it's low in the priority list 14:18:06 <sballe_> ok will review today 14:18:22 <acabot> sballe_: you can skip it 14:18:22 <jed56> tkaczynski i will sorry for the delay 14:18:45 <sballe_> ok thx 14:19:07 <tkaczynski> jed56 no worries, I made some good number of changes anyway. so it's a good moment to start :) 14:19:21 <jed56> great thanks 14:19:29 <dtardivel> tkaczynski: I will review it also 14:19:31 <acabot> #info on Watcher code, vincentfrancoise & dtardivel have implemented the purge DB command 14:20:02 <acabot> #action jed56 dtardivel review https://review.openstack.org/#/c/289880/ 14:20:30 <acabot> online doc has not been updated yet 14:21:16 <acabot> and we have 2 strategies submited by gzhai & brunograz open to reviews 14:22:03 <vincentfrancoise> acabot: I am currently reviewing the PS of brunograz 14:22:23 <acabot> #action vincentfrancoise review https://review.openstack.org/#/c/289259/ 14:23:00 <acabot> #topic Blueprint/Bug Review and Discussion 14:23:19 <acabot> #info RC-1 is targeted for March 18th #link https://launchpad.net/watcher/+milestone/mitaka-rc-1 14:23:32 <acabot> I think its time to postpone some BPs 14:24:08 <acabot> and bugs that cannot be fixed until friday 14:24:58 <acabot> gzhai : any update on the threshold implementation ? 14:25:13 <gzhai> acabot:I'm in debuging. 14:25:35 <acabot> ok so could you set it as "started" ? 14:25:42 <gzhai> acabot:sure. 14:26:00 <acabot> #action gzhai set https://blueprints.launchpad.net/watcher/+spec/optimization-threshold as started 14:26:33 <dtardivel> acabot: we received a new pathset from global requirement team one hour ago. We should work on it in priority 14:27:12 <acabot> #action dtardivel work on https://review.openstack.org/#/c/293432/ 14:28:00 <acabot> regarding the filtering of hosts for live migration, specs are in review so I will set it to rc-final 14:29:00 <dtardivel> acabot: we have 4 bug patchsets on review. Work on them also in priority before to introduce new ones ... 14:29:41 <acabot> seanmurphy brunograz: we will try to merge your strategy for rc-1 14:29:51 <seanmurphy> cool - thanks! 14:30:05 <brunograz> +1 14:30:28 <vtech> +1000 14:30:49 <gzhai> acabot: how about https://review.openstack.org/#/c/292188/? 14:30:56 <acabot> gzhai: I will set the threshold BP target to rc-final (as the code is not ready yet) 14:31:09 <gzhai> acabot: ok. 14:31:38 <acabot> gzhai : https://review.openstack.org/#/c/292188/ will be reviewed by the dev team asap 14:31:39 <gzhai> it should be ready soon. 14:31:52 <gzhai> acabot: thanks! 14:32:01 <acabot> #action dtardivel jed56 vincentfrancoise review https://review.openstack.org/#/c/292188/? 14:32:15 <acabot> #action dtardivel jed56 vincentfrancoise review https://review.openstack.org/#/c/292188/ 14:33:01 <acabot> any other thing to discuss in BPs and bugs ? 14:33:41 <acabot> #topic Open Discussion 14:34:39 <acabot> I'd like to share with you good practices on Launchpad 14:34:47 <sballe_> +1 14:34:54 <acabot> just to check that we are all clear about them 14:35:00 <seanmurphy> +1 14:35:05 <gzhai> +1 14:35:07 <acabot> regarding BPs #link https://blueprints.launchpad.net/watcher 14:35:24 <acabot> we will use the "design" field as this 14:35:50 <acabot> "approved" the BP has been discussed and specs are merged 14:36:38 <acabot> "Drafting" BP description has been reviewed, specs are needed and currently in review 14:37:10 <acabot> "Discussion" BP description has been reviewed and needs improvements 14:37:16 <tkaczynski> acabot sounds good 14:37:28 <sballe_> I just wanted to thank gzhai for giving a watcher presentation at the The 3rd China OpenStack Bug Smash. I am told he did a great job and it was well received. 14:37:28 <acabot> "New" BP description submited but not reviewed yet 14:37:35 <seanmurphy> who sets this - the author of the bp? 14:37:59 <acabot> seanmurphy : I will if you dont mind 14:38:12 <seanmurphy> of course not - it was just a q of where responsibility lies 14:38:20 <gzhai> sballe_: thank you. 14:38:28 <acabot> regarding bugs 14:38:55 <acabot> we will use the status "triaged" as a bug that can be taken by a developer 14:38:56 <vincentfrancoise> sballe_: +1 14:39:18 <acabot> so please do not take a bug in "new" state" 14:39:32 <acabot> as sometimes we dont want to fix it 14:39:57 <acabot> I will update the contrib doc accordingly 14:40:07 <jed56> gzhai great job 14:40:12 <acabot> everyone clear with these rules ? 14:40:18 <sballe_> acabot: +1 14:40:23 <gzhai> acabot: +1 14:40:24 <vincentfrancoise> sounds good to me 14:40:26 <jed56> acabot+³1 14:40:30 <brunograz> acabot: +1 14:40:38 <acabot> #action acabot update the contrib doc with status for BPs & bugs 14:40:46 <gzhai> acabot: if I report one bug, need wait for 'triage' state before fix? 14:40:57 <acabot> gzhai : yes please 14:41:11 <gzhai> acabot:ok 14:41:18 <acabot> gzhai : of course you can ask for a quick status on our irc channel 14:41:20 <dtardivel> acabot: "triaged" means that we accept the bug and we can reproduce it ? 14:41:26 <seanmurphy> rules sound good +1 14:41:59 <acabot> dtardivel : triaged means "we need to fix it", we can then set it to "confirmed" if someone reproduce it 14:42:00 <sballe_> acabot: did you submit the brown bag for Watche 14:42:06 <jed56> who is in charge of change the state to triage ? 14:42:32 <acabot> who wants to be the Watcher bug triager ? 14:42:46 <jed56> :) 14:42:53 <acabot> sballe_: no I didn't, I will look at it asap 14:42:58 <vincentfrancoise> I don't mind being one 14:43:00 <sballe_> acabot: thx 14:43:01 <jed56> IMHO, any core contributor 14:43:15 <acabot> #action acabot submit Watcher presentation to brown bag 14:43:19 <vincentfrancoise> jed56: +1 14:43:29 <hvprash> acabot: for reviewing any bugs or specs, is this something we need to agree upon in meetings or am i free to go ahead an review ? 14:43:56 <dtardivel> acabot: normally, it is the responsibility of the watcher drivers team, as explained on lauchpad 14:44:01 <acabot> what about every core contributor do 1 time per week ? 14:44:34 <acabot> hvprash: reviews are open to everyone so feel free to do it 14:44:37 <dtardivel> "What is a project driver?: The project driver is a person or team who is responsible for setting the direction for the project. The driver can set goals, approve bug targeting, or set backporting for any major series in the project. Alternatively, you can chose to leave the project driver unset and appoint a team for each series." 14:44:59 <hvprash> cool, thx :) was trying to understand the ettiquete 14:45:29 <dtardivel> acabot: I can do the triage 14:45:54 <acabot> #action dtardivel act as Watcher bug triager for mitaka release 14:45:55 <sballe_> dtardivel: +100 14:46:09 <vincentfrancoise> like dtardivel said, I guess it should be anyone in the watcher team on launchpad 14:46:28 <gzhai> dtardivel: +1000 14:46:39 <vincentfrancoise> just that a single bug cannot be reported and riged by the same person 14:46:43 <dtardivel> thanks :) 14:46:51 <vincentfrancoise> triged* 14:46:55 <vincentfrancoise> triaged* 14:46:57 <acabot> lets dtardivel do it until the summit, then we will decide to do another way 14:47:17 <tkaczynski> acabot +1 14:47:21 <vincentfrancoise> +1 14:47:26 <sballe_> +1 14:47:36 <acabot> gzhai : thx for promoting Watcher last week 14:47:47 <acabot> gzhai : did you get some feedback ? 14:48:02 <gzhai> acabot: yes. some company show interest in watcher 14:48:11 <gzhai> ZTE are interested in it. 14:48:20 <gzhai> Don't know if they are here:) 14:48:32 <jinquan> like me*^_^* 14:48:38 <acabot> I saw that someone from ZTE took a BP https://blueprints.launchpad.net/watcher/+spec/efficacy-indicator 14:48:51 <acabot> hello jinquan 14:49:05 <gzhai> jinquan: welcome. 14:49:06 <jinquan> I am come from eZTE 14:49:11 <sballe_> hi jinquan welcome 14:49:15 <jinquan> thanks 14:49:17 <jed56> welcome 14:49:30 <jinquan> I read the rst today 14:49:34 <gzhai> ZTE have some requirements for watcher: 14:49:39 <gzhai> strategy to offline percentage or number of nodes. 14:49:47 <gzhai> automatic execute action plan. Add a parameter when create audit, so it's automatic. 14:49:48 <jinquan> an setup the develop environment 14:49:53 <gzhai> one goal multiple strategy. 14:50:07 <gzhai> I know we have considered some of them. 14:50:15 <acabot> right 14:50:20 <jed56> what do you mean by multiple strategies ? 14:50:46 <gzhai> jed56: one goal can have different strategies. 14:50:56 <gzhai> e.g. some aggresive then others 14:51:01 <jinquan> I am chat on mobilephone reply maybe slow 14:51:21 <gzhai> I'm heard this in future plan. 14:51:32 <acabot> vmahe is working on it 14:51:43 <vmahe> I'm currently writing the spec explaining how goals will be returned by strategies 14:51:49 <vmahe> using a common base class 14:51:50 <jed56> :) 14:51:54 <gzhai> acabot:great! 14:51:55 <vincentfrancoise> #link https://blueprints.launchpad.net/watcher/+spec/get-goal-from-strategy 14:52:09 <vmahe> it will also be possible to get the list of available strategies for a given goal 14:52:44 <acabot> jinquan : do you plan to write specs on efficacy indicators ? 14:53:52 <acabot> jinquan : or to implement as soon as specs will be merged ? 14:54:11 <jinquan> I plan to write but maybe I need sometime learn watcher 14:54:23 <acabot> because there are specs open right now https://review.openstack.org/#/c/283449/ 14:54:48 <jinquan> I read it already 14:54:50 <acabot> the best would be to review the specs right now 14:55:03 <acabot> if ZTE has an interest on it 14:55:22 <acabot> or work with vmahe regarding multiple strategies per goal 14:55:36 <acabot> 5 minutes left 14:55:55 <acabot> our tempest gate is still not working 14:56:03 <acabot> the multi-node environment 14:56:22 <acabot> and it does not seems to be a high priority in the infra team 14:56:41 <acabot> does anyone knows how we can make it happen ? 14:57:00 <vincentfrancoise> FYI, I asked the infra team a couple of days ago about this problem ... and no one answered 14:57:29 <acabot> sballe_: any contact on the infra team that could help us ? 14:58:02 <jinquan> um， I will review the specs on efficacy indicators 14:58:05 <acabot> today we have no solution to test a live migration on the gate 14:58:06 <sballe_> we could contact monty 14:58:13 <vincentfrancoise> so basically I will push a PS onto openstack-infra/devstack-gate to add an extra env variable which would then allow us to fix the shell script used to run our tempest gate job in openstack-infra/project-config 14:58:26 <acabot> #action jinquan review https://review.openstack.org/#/c/283449/ 14:58:33 <sballe_> I would ask mordred 14:58:38 <jinquan> OK 14:58:58 <acabot> #action vincentfrancoise push a PS onto openstack-infra/devstack-gate to add an extra env variable which would then allow us to fix the shell script used to run our tempest gate job in openstack-infra/project-config 14:59:20 <dtardivel> brunograz: I did some comments about terms used in source code which are not correct 14:59:34 <acabot> #action sballe_ ask monty for helping us based on this PS 15:00:02 <sballe_> What the PS? 15:00:11 <acabot> I will have to close the meeting, dtardivel brunograz, please continue on openstack-watcher channel 15:00:23 <acabot> sballe_ : the PS from vincentfrancoise 15:00:23 <vincentfrancoise> by 15:00:25 <jinquan> I read irc log ，somethin will discuss at austin？ 15:00:26 <dtardivel> bye 15:00:32 <acabot> #endmeeting