08:00:21 <ifat_afek> #startmeeting vitrage
08:00:22 <openstack> Meeting started Wed Jan  4 08:00:21 2017 UTC and is due to finish in 60 minutes.  The chair is ifat_afek. Information about MeetBot at http://wiki.debian.org/MeetBot.
08:00:23 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
08:00:25 <openstack> The meeting name has been set to 'vitrage'
08:00:29 <ifat_afek> Hi :-)
08:00:39 <danoffek> Hi guys
08:00:52 <nbloom> Hello!!!
08:00:57 <eyalb> hi
08:01:53 <yujunz> hi
08:02:49 <ifat_afek> #topic Status and Updates
08:03:13 <ifat_afek> Ocata development is about to end this month. The feature freeze is on January 27.
08:03:22 <ifat_afek> #link https://releases.openstack.org/ocata/schedule.html
08:03:46 <ifat_afek> Boston CFP is now open. Let’s start thinking about ideas for sessions. We can submit proposals until February 6th.
08:03:54 <ifat_afek> #link https://www.openstack.org/summit/boston-2017/call-for-presentations/
08:04:06 <yujunz> What features are ready for ocata at the moment?
08:04:33 <elisha_r> hello
08:04:53 <ifat_afek> I will finish the Doctor impelmentation (will update about it soon), and hopefully eyalb will finish the collectd integration
08:05:15 <ifat_afek> I’m doing a fast-fowrard update here ;-) maybe we should ask everyone to update about their own status
08:05:37 <alexey_weyl> Hi Guys :)
08:05:41 <danoffek> Hi. Working on Vitrage ID.
08:05:55 <yujunz> I will continue working on static data source and I think it will catch up with ocata
08:05:59 <ifat_afek> So I’ll update about Doctor: I’m done writing a Doctor datasource for Vitrage, that will implement Doctor SB API. Waiting for your code reviews.
08:06:08 <nadav> hi
08:06:17 <ifat_afek> #link https://review.openstack.org/#/c/414742/
08:06:19 <ifat_afek> #link https://review.openstack.org/414975
08:06:45 <ifat_afek> I suggest we go one by one, and then have a topic about Ocata. Makes sense?
08:06:52 <yujunz> Agree
08:07:10 <ifat_afek> So who has other updates?
08:07:53 <yujunz> I will
08:08:08 <yujunz> static datasource is pending for a few weeks and now resumed
08:08:27 <yujunz> I'll finish the transformer part as soon as possible
08:08:44 <yujunz> But I have a question on the recent refactoring on transformer
08:08:53 <yujunz> Will it affect static datasource?
08:09:09 <yujunz> #link https://review.openstack.org/#/c/415684/
08:09:21 <yujunz> ping alexey_weyl
08:10:04 <yujunz> OK, I saw a reply on gerrit
08:10:11 <alexey_weyl> yes, it will affect you a bit, but don't use it yet, because the Doctor datasource should be pushed first, and then we will approve the refactoring
08:10:17 <yujunz> Is there any background for this refactoring?
08:10:25 <yujunz> What is changed in summary?
08:10:43 <alexey_weyl> after that you can pull it from the master and use it, but don't worry it's not such a big change
08:11:09 <yujunz> OK
08:12:11 <alexey_weyl> the reason fro the refactoring is that there are 2 new datasources that are being added to Vitrage at the moment, and they saw that those methods of the create_placeholder and the create_neighbor are basically the same and should sit at the base where everyone can use it
08:13:11 <yujunz> I see. Maybe adding these notes into the commit message will help reviewers to understand :-)
08:13:19 <ifat_afek> alexey_weyl: you just moved a function to the base class, right?
08:13:43 <alexey_weyl> yujunz: no problem :)
08:14:22 <yujunz> OK thanks. That's all from my side
08:14:30 <alexey_weyl> ifat_afek: kind of, I moved them, and adjusted them so they will be much more robust for usage
08:14:46 <ifat_afek> alexey_weyl: ok, thanks. I’ll review it soon
08:14:55 <alexey_weyl> Thanks
08:15:51 <ifat_afek> I forgot another update on my side: I plan to start writing another blueprint, for implementing Doctor SB API in Vitrage. We should have an API for “send event” or something similar, that will send the event to the message bus. Then, Doctor datasource will read it from there
08:16:18 <ifat_afek> Anyone else wants to update?
08:16:47 <eyalb> I am still working on the collectd datasource
08:17:50 <usman> Hello everyone
08:18:09 <alexey_weyl> I am waiting for feedback from aodh on the commits that I have pushed
08:18:16 <ifat_afek> Hi usman
08:18:45 <alexey_weyl> you can see the changes here:
08:18:46 <alexey_weyl> https://review.openstack.org/#/c/413008/
08:18:49 <ifat_afek> alexey_weyl: did you push all of your code, or is it just the first phase?
08:18:55 <alexey_weyl> https://review.openstack.org/#/c/413594/
08:19:13 <alexey_weyl> not all the code yet, but anyfeedback would be great
08:19:55 <ifat_afek> alexey_weyl: Great. I’ll try to review, although I’m not familiar with Aodh code
08:20:26 <ifat_afek> Any other update anyone?
08:21:11 <ifat_afek> #topic Ocata Release
08:21:25 <ifat_afek> So let’s see what we expect to have ready for Ocata
08:21:40 <ifat_afek> Doctor Datasource + SB API should be ready
08:22:02 <yujunz> low risk on static datasource evolution
08:22:03 <danoffek> Vitrage Id
08:22:50 <ifat_afek> dwj: Are you here? you finished the Aodh notifications for Vitrage, right? I see that the blueprint is still open
08:23:01 <dwj> yes
08:23:02 <ifat_afek> https://blueprints.launchpad.net/vitrage/+spec/aodh-message-bus-notifications
08:23:14 <ifat_afek> BTW, you can see all Ocata blueprints here:
08:23:15 <dwj> It's already finished
08:23:23 <ifat_afek> #link https://blueprints.launchpad.net/vitrage/+spec/aodh-message-bus-notifications
08:23:29 <ifat_afek> #link https://blueprints.launchpad.net/vitrage/ocata
08:23:33 <dwj> I'll change the status later
08:23:35 <ifat_afek> dwj: great, thanks
08:23:54 <ifat_afek> eyalb: what about collectd and red hat?
08:24:11 <eyalb> I hope so  :-)
08:24:33 <ifat_afek> great
08:24:50 <ifat_afek> alexey_weyl: I guess Aodh is in risk because you got no reviews yet?
08:25:09 <yujunz> What is red hat?
08:25:09 <yujunz> apex installer?
08:25:32 <alexey_weyl> yes, it is a risk, because there are a lot of stuff yet to do there, and I need to get feedbacks and responses from the Aodh team
08:25:45 <ifat_afek> yujunz: We would like to support Apex installer, and base it on TripleO installation. Currently we are waiting for RDO
08:26:26 <eyalb> I pushed the initial spec file of vitrage to rdo and I am waiting for there response
08:26:27 <ifat_afek> So the apex installer is also in risk
08:26:41 <yujunz> So red hat = https://blueprints.launchpad.net/vitrage/+spec/support-tripleo-instllation
08:26:43 <yujunz> right?
08:26:52 <yujunz> And apex is another story
08:27:00 <ifat_afek> (But the apex installer not a pure Vitrage task, it’s an OPNFV one)
08:27:10 <ifat_afek> yujunz: right
08:27:14 <ifat_afek> Both are in risk
08:28:11 <yujunz> I see. Thanks
08:28:16 <ifat_afek> eyalb: I’ll mark collectd blueprint for Ocata as well
08:28:23 <ifat_afek> #link https://blueprints.launchpad.net/vitrage/+spec/collectd-datasource
08:28:36 <eyalb> ok
08:29:09 <ifat_afek> Anything that I forgot?…
08:30:07 <yujunz> How is neo4j going?
08:30:12 <yujunz> Any chance to get it ready?
08:30:18 <ifat_afek> BTW, yujunz: eyalb is working on collectd datasource, and collectd DPDK plugin raises alarms on the bridges. This is the reason he wants to use the new static datasource
08:31:10 <yujunz> Thanks for reminding ifat_afek. I'll try to speed up
08:32:33 <ifat_afek> yujunz: no problme
08:33:09 <yujunz> How about neo4j?
08:33:15 <alexey_weyl> yujunz: at the moment we are thinking about our steps regarding the neo4j because in the community edition it doesn't support HA, and thus it is a problem.
08:33:50 <ifat_afek> yujunz: and in any case, I doubt if have enough time left to finish it for Ocata
08:34:17 <yujunz> Is there a big gap to migrate community edition to commercial neo4j?
08:34:41 <ifat_afek> yujunz: there are two issues with the commerical Neo4J
08:34:42 <yujunz> I mean integrate community version in openstack/vitrage and leave it to commercial user the migration tasks
08:35:24 <alexey_weyl> The persistent graph DB is very important, but we need to make sure that we are using the correct one that gives us the most tools as a DB.
08:35:36 <ifat_afek> You can either pay for the license, or use GPL3 license which is “contaminating” (meaning your entire code must become open source as well)
08:35:58 <yujunz> OK, so the community version is licensed under GPL3
08:36:09 <yujunz> That's really something
08:36:33 <ifat_afek> Your suggestion is to have non-HA version in openstack, and a possibility to use the commercial version in another product (Nokia/ZTE)?
08:36:52 <ifat_afek> the community version is apache license, no problem with that. The commerical version is GPL3
08:36:53 <yujunz> Yes, things like that
08:37:32 <yujunz> Alright, I kind of understand the issue now
08:38:55 <ifat_afek> I already checked the license issue. Neo4j does not exist in openstack at the moment, but there is no problem to add it to the global requirements (I believe the package name is py2neo)
08:39:37 <ifat_afek> But we did not check the issue of using the community version vs. the commerical version in different distributions
08:40:21 <yujunz> I guess we need a lawyer for this issue :-)
08:40:30 <ifat_afek> Right :-)
08:40:39 <yujunz> Is there any alternative solution?
08:40:48 <yujunz> I remember we mentioned another in summit
08:40:54 <ifat_afek> We did not investigate it yet
08:41:09 <yujunz> But forgot the name
08:41:23 <ifat_afek> I know there are a few graph databases, but I don’t know of anyone from Vitrage who did a thorough evaluation
08:41:31 <ifat_afek> I don’t remember either
08:44:31 <ifat_afek> Anything else, for Ocata or on a different topic?
08:46:57 <ifat_afek> See you next week!
08:47:04 <nbloom> bye :)
08:47:07 <eyalb> bye
08:47:09 <yujunz> See
08:47:12 <alexey_weyl> bye bye
08:47:14 <dwj> bye
08:47:39 <usman> bye
08:47:54 <ifat_afek> #endmeeting