09:00:45 <ifat_afek> #startmeeting vitrage
09:00:46 <openstack> Meeting started Wed Jan 13 09:00:45 2016 UTC and is due to finish in 60 minutes.  The chair is ifat_afek. Information about MeetBot at http://wiki.debian.org/MeetBot.
09:00:47 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
09:00:49 <openstack> The meeting name has been set to 'vitrage'
09:00:57 <ifat_afek> Hi everyone :-)
09:01:00 <amir_gur> hi
09:01:02 <alexey_weyl> Hi
09:01:18 <omer_etr_> Hi
09:01:23 <ayaho> hi
09:04:34 <danoffek> Shalom
09:04:49 <elisha_r> Uvracha
09:05:04 <elisha_r> hi all
09:05:26 <danoffek> So we finally have an RPC client server, now we need to integrateit with the Vitrage API, context, ...
09:06:02 <ifat_afek> #topic Current status and progress
09:06:32 <danoffek> We see the graph data is received correctly from the graph, through the rpc server, all the way to the RPC client and the Vitrage API.
09:07:33 <danoffek> The vitrage API though, when returning the data, returns it in HTML format instead of JSON,
09:07:46 <emalin> hi
09:07:49 <ifat_afek> great news!!
09:08:38 <alexey_weyl> So at the moment we have 2 problems with the integration for the demo
09:08:55 <elisha_r> danoffek - what is the reason for the HTML format? do we know?
09:09:08 <alexey_weyl> 1. which dan has reminded, which is quiet small problem
09:09:15 <danoffek> Not yet, just raising a probel
09:09:21 <danoffek> problem
09:09:24 <elisha_r> got it. tnx
09:09:44 <alexey_weyl> 2. integration with the horizon, which we try to understand why it doesn't work on the devstack with all the changed made my Omer
09:09:51 <eyalb> hi
09:10:04 <danoffek> The second part might be a bigger issue
09:11:30 <eyalb> I found the problem why the result of get topology is returned as html
09:11:42 <eyalb> I am working on it
09:11:43 <omer_etrog> We need to  understand how to implement phyton-vitrageclient in Horizon
09:12:25 <danoffek> good newa
09:12:28 <danoffek> newa
09:12:31 <danoffek> news
09:12:34 <ifat_afek> so who is working on it? eyalb, omer_etrog, ...?
09:12:40 <alexey_weyl> Great @eyalb
09:13:13 <alexey_weyl> i am working on it as well
09:13:40 <eyalb> it is solved now
09:13:41 <omer_etrog> we also have issues in AngularJS in Horizon and I try to get answers from the Horizon team
09:13:50 <ifat_afek> cool
09:13:59 <elisha_r> eyalb: cool!
09:14:00 <eyalb> now we need to see why tree representation is not working
09:14:40 <ifat_afek> ok, anything else regarding the integration?
09:15:42 <eyalb> nope
09:17:14 <lhartal_> As result of integration problems, we are going to delay the demo. It will be available on monday in Vitrage website
09:18:04 <lhartal_> #action presenting first Vitrage demo
09:18:18 <ifat_afek> ok, let's move on
09:18:24 <ifat_afek> PinPoint approved the first version of their use cases document
09:18:33 <ifat_afek> #link https://wiki.opnfv.org/_media/pinpoint/use_cases_v2_5.docx
09:19:49 <elisha_r> great!\
09:19:49 <ifat_afek> as the next phase, they plan to do a gap analysis, so they will check if our get topology api answers their use cases
09:20:03 <ifat_afek> Ohad and I plan to go over it and make sure it matches our design. We also plan, together with Eyal, to finish our get topology API definition and see that it matches their requirements.
09:20:34 <ifat_afek> #action review PinPoint use cases and make sure it matches Vitrage implementation
09:20:57 <ifat_afek> I’m almost done writing the documentation of Vitrage graph design and main use cases. Once I’m done, I’ll send it to PinPoint to get their feedback.
09:21:11 <eyalb> tree representation is working now :-)
09:21:20 <ifat_afek> cool!!
09:21:26 <elisha_r> nice :)
09:21:32 <alexey_weyl> great!
09:21:37 <danoffek> great !!
09:21:48 <alexey_weyl> so now we don't have any problems with the html?
09:21:57 <eyalb> yes
09:22:02 <alexey_weyl> great!!! :)
09:22:06 <eyalb> need to check the query
09:22:31 <ifat_afek> eyalb: we will talk with Ohad tomorrow about the query, ok?
09:22:51 <ifat_afek> eyalb: or do you mean there is still a technical problem with the query?
09:23:20 <eyalb> we need to check the api is working with query
09:23:44 <ifat_afek> ok, thanks
09:24:33 <alexey_weyl> but for the demo it's enough to get the full topology
09:24:50 <ifat_afek> right
09:25:32 <ifat_afek> Let's move on
09:25:33 <ifat_afek> Regarding Aodh, I’m still trying to figure out how we are going to integrate with them
09:25:45 <ifat_afek> Our current discussions are about how to get notifications about alarm state change. And there is another open issue of how to trigger vitrage alarms in aodh.
09:25:56 <ifat_afek> #action ifat_afek continue discussions with Aodh
09:26:25 <ifat_afek> We discussed a few days ago Vitrage Consistency design. alexey_weyl, can you update?
09:27:13 <alexey_weyl> yes, we had a discussion and a design meeting about the consistency service.
09:27:53 <alexey_weyl> The consistency service will delete the old deduced alarms on initializing of vitrage
09:28:29 <alexey_weyl> and in addition it will every T time will check what old entities we have and delete them as well.
09:29:25 <alexey_weyl> in addition every T time get_all will be performed for all the entities from the synchronizer and update them in the entity_graph
09:29:46 <alexey_weyl> for more details. you can see the design in our documentation
09:30:48 <ifat_afek> ok, we should also update the blueprint
09:31:28 <ifat_afek> Who can update about the synchronizer status?
09:32:46 <amir_gur> no progress was done this week.
09:32:54 <ifat_afek> ok
09:33:17 <ifat_afek> anyone else has updates?
09:33:57 <ifat_afek> moving on
09:33:58 <ifat_afek> #topic Review action items
09:34:10 <ifat_afek> • alexey_weyl continue with the integration, including zones, hosts and instances
09:34:25 <alexey_weyl> Done :)
09:34:31 <ifat_afek> great
09:34:32 <ifat_afek> • finalize get topology API
09:34:55 <ifat_afek> we plan to work on it this week and document the final API
09:35:03 <ifat_afek> #action finalize get topology API
09:35:16 <ifat_afek> • ifat_afek update the documentation on vitrage main page with our latest design diagrams (of vitrage graph and the synchronizer)
09:35:20 <ifat_afek> almost done
09:35:30 <ifat_afek> #action ifat_afek update the documentation on vitrage main page with our latest design diagrams (of vitrage graph and the synchronizer)
09:35:39 <ifat_afek> •  presenting first Vitrage demo
09:35:51 <ifat_afek> postponed to Monday
09:36:13 <ifat_afek> #topic Next Steps
09:36:35 <ifat_afek> mitaka-2 milestone is at the end of next week. I want to check the status of the blueprints that are scheduled for mitaka-2
09:36:51 <ifat_afek> the following links you can see all blueprints scheduled for mitaka:
09:37:00 <ifat_afek> #link https://blueprints.launchpad.net/vitrage/mitaka/+specs?show=all
09:37:11 <ifat_afek> #link https://blueprints.launchpad.net/python-vitrageclient
09:37:19 <ifat_afek> #link https://blueprints.launchpad.net/vitrage-dashboard/mitaka/+specs?show=all
09:37:37 <ifat_afek> the following blueprints are scheduled for mitaka-2:
09:38:08 <ifat_afek> synchronizer-get-all-instances – nadav_yakar
09:38:26 <ifat_afek> amir_gur: do you know the status of this blueprint?
09:38:54 <amir_gur> moment
09:39:07 <emalin> it's almost done
09:39:26 <ifat_afek> what's missing? can it be ready for next Thursday?
09:40:05 <emalin> We just wait to hear from the transformer people if something is missing
09:40:51 <emalin> I think you may want some timer that run the get all
09:40:58 <lhartal_> noothing is missing :)
09:41:22 <lhartal_> For Mitaka 3 we need to add support for update events
09:41:31 <ifat_afek> emalin: and the timer can be ready for next week?
09:42:07 <ifat_afek> for mitaka-2 our goal was to support get topology for nova entities: zones, hosts and instances. so I'm trying to see if this can be done by next Thursday
09:42:13 <ifat_afek> next blueprint:
09:42:28 <ifat_afek> get-topology-api – eyalb + dany (working on the api handler)
09:43:59 <danoffek> The Standard flow of the "API Handler" (full RPC client server) will be done by next week.
09:45:11 <danoffek> Preparing it for sub processing and standard schedueling on the service side might not be ready by then
09:45:43 <ifat_afek> so I suggest we create another blueprint for the scheduling, and close the basic blueprint for mitaka-2
09:46:08 <ifat_afek> next blueprint:  networkx-graph-driver – idan_hefetz
09:46:45 <danoffek> ok. need approval for that
09:47:07 <ifat_afek> I talked with idan_hefetz, he finished all the basic functionality. what's missing is the sub-graph-matching, which is not needed for mitaka-2. we agreed to split the blueprint
09:47:24 <ifat_afek> next: nova-entity-transformer – lhartal
09:49:28 <emalin> hi again
09:49:31 <lhartal_> I finished to work on nova transformers - it includes transformers for instances, hosts and zones.
09:49:45 <emalin> I was disconnected for some minutes
09:50:13 <emalin> ifat_afek did you ask me something ?
09:50:31 <ifat_afek> emalin: not yet, will do soon :-)
09:50:38 <lhartal_> I'll change the bp status to implemented
09:51:26 <ifat_afek> great. next: synchronizer-get-all-hosts and synchronizer-get-all-zones – inbarsto
09:51:37 <ifat_afek> emalin: do you know inbarsto's status?
09:52:58 <ifat_afek> until email answers:  vitrage-resource-processor – alexey_weyl
09:52:59 <emalin> The gel all stories are done
09:53:07 <emalin> *get all
09:53:13 <ifat_afek> emalin: great, thanks
09:53:22 <emalin> Besides the part of the timer
09:53:45 <ifat_afek> emalin: the timer should be implemented for each plugin separately?
09:54:13 <alexey_weyl> the vitrage-resource-processor is finished a couple of weeks already.
09:54:57 <alexey_weyl> Now after the nova zone and host transformers are done, i have added a test for create entity graph with instances, hosts, zones and node, and it looks great
09:55:10 <alexey_weyl> every thing is passes, and graph is built as expected
09:55:21 <ifat_afek> great, thank
09:55:28 <ifat_afek> next: vitrage-cli - eyalb
09:55:29 <emalin> The timer would be implemented to all plugins
09:55:41 <ifat_afek> emalin: thanks
09:56:08 <eyalb> I am updating it all the time when the get topology api blueprint is updated
09:56:21 <ifat_afek> eyalb: ok, thanks
09:56:23 <alexey_weyl> great, before you push it to gerrit i want to sit with you, ok? (emalin)
09:56:36 <ifat_afek> vitrage-menu – oetrog
09:57:20 <omer_etrog> vitrage-menu is done
09:57:25 <ifat_afek> great
09:57:26 <ifat_afek> ui-system-health-sunburst – alonh
09:58:00 <omer_etrog> sunburest is in progress, we already show the correct data, and labels
09:58:13 <ifat_afek> and it is supposed to be finished by next Thursday?
09:58:44 <omer_etrog> we still have some issues with the AngularJS bindings
09:59:04 <ifat_afek> ok, thanks
09:59:07 <ifat_afek> we are almost out of time. we need to talk about the tempest test...
09:59:17 <danoffek> Do we know of anyone who had similar experience in OpenStack development ?
10:00:20 <ifat_afek> I guess we should meet and talk about the tempest tests
10:01:11 <mkoushnir> we need tempest test plan. the meeting in sunday
10:01:22 <ifat_afek> we have to end the meeting. goodbye everyone
10:01:27 <ifat_afek> thanks mkoushnir
10:01:32 <ifat_afek> #endmeeting