08:00:44 <ifat_afek> #startmeeting vitrage
08:00:45 <openstack> Meeting started Wed Jul  5 08:00:44 2017 UTC and is due to finish in 60 minutes.  The chair is ifat_afek. Information about MeetBot at http://wiki.debian.org/MeetBot.
08:00:46 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
08:00:48 <openstack> The meeting name has been set to 'vitrage'
08:00:49 <ifat_afek> Hi :-)
08:00:54 <eyalb> \o/
08:00:57 <yujunz> Hi
08:03:43 <yujunz> ping
08:03:55 <ifat_afek> Yes, let’s start :-)
08:04:11 <ifat_afek> Today’s agenda:
08:04:15 <ifat_afek> •	Sydney summit
08:04:20 <ifat_afek> •	Status and Updates
08:04:23 <ifat_afek> •	Open Discussion
08:04:35 <ifat_afek> #topic Sydney Summit
08:05:06 <ifat_afek> Sydney summit CFP:
08:05:13 <ifat_afek> #link https://www.openstack.org/summit/sydney-2017/call-for-presentations/
08:05:29 <adary> hi there
08:05:31 <ifat_afek> I update the etherpad with our current ideas. The deadline for submissions is July 14th. Feel free to edit or add to the etherpad.
08:05:38 <ifat_afek> #link https://etherpad.openstack.org/p/vitrage-sydney-cpf
08:06:32 <ifat_afek> #topic Status and Updates
08:06:41 <ifat_afek> An update about the release dates:
08:06:50 <ifat_afek> Client libraries final releases (i.e. python-vitrageclient): July 27
08:06:58 <ifat_afek> Pike-3 milestone (and Feature freeze): July 27
08:07:04 <ifat_afek> Final Pike release: August 30
08:07:16 <ifat_afek> The meaning is that if you want to add or modify the API, you must do it by the end of this month. Same goes for adding new features.
08:07:26 <ifat_afek> We can continue working on small improvements and bug fixes during August.
08:07:51 <ifat_afek> Two tasks that must be completed by July 27:
08:07:57 <ifat_afek> •	Full support for python 3.5 (in our tempest tests)
08:08:03 <ifat_afek> •	Update Vitrage version in the global-requirements
08:08:20 <ifat_afek> Another issue: Weekly IRC meetings are no longer required by the TC.
08:08:28 <ifat_afek> #link https://review.openstack.org/#/c/462077/
08:08:36 <ifat_afek> I mentioned it last week. Meanwhile, I’ve seen some email discussions about it, and got the impression that for now we better continue our regular meeting.
08:08:48 <ifat_afek> We can check in a while and see if other projects are canceling their own
08:08:56 <ifat_afek> If someone feels that we should stop these meetings, let’s discuss it.
08:09:09 <ifat_afek> My update: I’m implementing the Vitrage-Mistral integration, hope to finish it this week.
08:09:14 <yujunz> +2 for continuing regular meeting from my side
08:09:22 <ifat_afek> Ok, noted :-)
08:09:27 <ifat_afek> #link https://blueprints.launchpad.net/vitrage/+spec/integration-with-mistral
08:10:04 <ifat_afek> I ran into an unexpected problem: until today, all actions had an action-target, but execute-mistral doesn’t have it
08:10:20 <ifat_afek> Turns out that the action-target is a must for the subgraph matching
08:10:55 <ifat_afek> It is needed in order to determine the connected components or something like that. At least that’s what I understood from idan_hefetz and alexey_weyl
08:11:35 <ifat_afek> I’m thinking of solving it by saying that for Mistral the action-targed will be an arbitrary part of the condition
08:11:54 <ifat_afek> But - it must be a “positive” part of the condition, that doesn’t have a “not” before it
08:12:27 <ifat_afek> I understood that it *is* valid to have a condition that is only negative, like: not alarm_on_host
08:12:40 <ifat_afek> I’m not sure it makes sense; but at least for Mistral I can’t allow this
08:13:16 <ifat_afek> So I plan to change the validation for Mistral, and make sure there is at least one positive part in the condition. If everything works fine (and it must…), we will consider adding this restriction to all actions
08:13:20 <ifat_afek> That’s it for me
08:14:28 <yujunz> I'll update
08:15:03 <yujunz> Start working on osprofiler support. api integration is under review
08:15:26 <yujunz> Next would be tracing RPC and some classes
08:16:04 <yujunz> That's it
08:16:14 <ifat_afek> Sounds great
08:16:32 <ifat_afek> What should we expect to get once you finish? performance profiling? something else?
08:17:53 <yujunz> First benefit is user can use it on demand
08:17:58 <yujunz> to analysis performance issue
08:18:33 <yujunz> If there is some key performance indicator to monitor, we can consider add profiling in test
08:20:04 <idan_hefetz> yujunz: i've pushed a tool load_generator, to create many mock nitifications
08:20:42 <idan_hefetz> I've been using that to test Vitrage with a large entity graph, and to profile the evaluator
08:20:43 <yujunz> I was just about to mention it :-) Failed to find the link
08:21:19 <idan_hefetz> https://review.openstack.org/#/c/479746/
08:21:32 <idan_hefetz> It's already merged
08:21:51 <ifat_afek> If you write: #link https://… it will be part of the meeting summary
08:21:57 <yujunz> Do you plan to integrate it in automatic tests?
08:22:33 <idan_hefetz> No plans to do so, just a manual tool for now
08:22:41 <yujunz> I see. Thanks
08:23:14 <idan_hefetz> Ok, my update:
08:24:02 <idan_hefetz> First commit as part of the parallel evaluation blueprint, is a refactoring of the initialization step
08:24:02 <idan_hefetz> mostly to separate the evaluator from the consistency.
08:24:24 <idan_hefetz> #link https://review.openstack.org/#/c/479867/
08:25:20 <idan_hefetz> Besides that, there are a few challenges
08:26:16 <idan_hefetz> overlapping templates feature, how to assign tasks to processes
08:26:52 <idan_hefetz> We should also find a solution that wont cause an issue with future template crud
08:27:02 <alexey_weyl> Hello
08:27:22 <alexey_weyl> Privet :)
08:27:53 <danoffek> Kakdila Alosha ?
08:28:10 <ifat_afek> Is it English? ;-)
08:28:25 <danoffek> The letters are
08:28:41 <yujunz> I can read the letters
08:28:50 <danoffek> ':)
08:29:15 <alexey_weyl> Now put it in google translate, and check what is says in russian :)
08:29:37 <ifat_afek> alexey_weyl: maybe you have an update?
08:29:51 <alexey_weyl> I will update
08:30:02 <alexey_weyl> I continue to work on the DB Dao
08:30:14 <alexey_weyl> it is more complicated than I thought in the beggining
08:30:39 <alexey_weyl> part by part it will work :)
08:30:48 <alexey_weyl> step by step (I meant)
08:31:49 <alexey_weyl> Thats it
08:31:58 <ifat_afek> Thanks
08:32:06 <ifat_afek> Anyone else wants to update?
08:32:18 <danoffek> Niet
08:33:12 <yujunz> 没
08:33:25 <ifat_afek> And what does that mean?
08:33:34 <ifat_afek> I can’t even put it in google...
08:33:36 <alexey_weyl> Can you translate please?
08:33:49 <yujunz> It's "No"
08:34:11 <ifat_afek> I’ll try to remember it for next time ;-)
08:34:37 <ifat_afek> #topic Open Discussion
08:34:39 <danoffek> A PTL has to ....
08:34:45 <ifat_afek> Putting this topic just in case…
08:34:50 <ifat_afek> danoffek: what?
08:35:00 <ifat_afek> remember? don’t count on my memory
08:35:11 <danoffek> You need to remember 没 for next time
08:35:13 <yujunz> I want to ask the status of use case repository
08:35:26 <ifat_afek> No progress so far
08:35:33 <yujunz> OK
08:35:33 <ifat_afek> We are doing other tasks
08:35:39 <ifat_afek> Is it important to you?
08:36:22 <yujunz> To our company yes
08:36:54 <ifat_afek> Do you already know your own use cases? even if we contribute some use cases to the repository, they will depend on specific Zabbix alarms
08:38:32 <yujunz> I know a bit, curious on how vitrage is used in other projects
08:39:40 <ifat_afek> Makes sense. As we said before, this is on our roadmap, but I can’t tell you when exactly it will happen
08:39:52 <ifat_afek> We keep identifying other tasks that seem more urgent…
08:40:03 <yujunz> Understood
08:41:29 <ifat_afek> Any other issue someone would like to discuss?
08:41:37 <dwj> Hi, how is the progress of integration with doctor?
08:42:15 <dwj> As doctor team mentioned it in the weekly meeting. :)
08:42:18 <ifat_afek> Hmm… I’m supposed to work on it, plan to get back to it after I finish the integration with Mistral
08:42:30 <ifat_afek> What did they say?
08:42:56 <dwj> just ask about the progress.
08:43:12 <ifat_afek> I need to work on the tests for Vitrage on the inspector. And someone from Vitrage should work on the Apex installation, which is a bigger task
08:43:50 <dwj> BTW, doctor weekly meeting changed to  2PM UTC on Monday.
08:44:26 <ifat_afek> I know, I should make it easier for me to join
08:44:45 <ifat_afek> BTW, if you have any tips regarding Apex we will be happy to hear them
08:45:05 <ifat_afek> And I already know that once I start the with the tests I need to talk with you about the new structure :-)
08:47:03 <dwj> OK, I already post a patch about refactoring the sample inspector. It can be a refence
08:47:20 <ifat_afek> Great, thanks
08:47:30 <dwj> #link https://gerrit.opnfv.org/gerrit/#/c/36839/
08:48:45 <ifat_afek> Thanks
08:48:58 <ifat_afek> Any other issue someone would like to raise?
08:49:38 <yujunz> 没
08:49:54 <ifat_afek> I remember!
08:50:32 <ifat_afek> Bye then, see you next week
08:50:33 <yujunz> 再见
08:50:38 <eyalb> o/
08:50:39 <ifat_afek> Is this bye?
08:50:46 <dwj> 拜～
08:50:48 <yujunz> Correct !
08:50:54 <ifat_afek> So what did dwj write?
08:51:05 <dwj> another bye
08:51:08 <dwj> :)
08:51:09 <idan_hefetz> have a nice one!
08:51:19 <ifat_afek> And you expect me to remember all of these?? :-)
08:51:21 <yujunz> It pronounces as bye
08:51:36 <ifat_afek> ביי
08:51:56 <danoffek> Ifat, add 拜～  to your vocabulary, or we are voting for another PTL
08:52:19 <ifat_afek> danoffek: behave ;-)
08:53:37 <ifat_afek> Bye, see you next week
08:53:38 <ifat_afek> #endmeeting