| *** angdraug has quit IRC | 00:35 | |
| *** dshulyak has joined #openstack-solar | 05:44 | |
| *** salmon_ has joined #openstack-solar | 07:32 | |
| openstackgerrit | Merged openstack/solar: Use WeakValueDictionary instead of WeakSet in DBModelProxy https://review.openstack.org/295675 | 08:41 |
|---|---|---|
| openstackgerrit | Dmitry Shulyak proposed openstack/solar: Refactoring of graph.py and usage of it in scheduler https://review.openstack.org/294605 | 08:58 |
| *** openstackgerrit has quit IRC | 09:03 | |
| *** openstackgerrit has joined #openstack-solar | 09:04 | |
| openstackgerrit | Dmitry Shulyak proposed openstack/solar: Implement traversal based on number of childs https://review.openstack.org/295348 | 09:18 |
| mkwiek | hi guys, what is the preferred way to add transport to resource via code? | 09:32 |
| pigmej | mkwiek: the only one way | 09:33 |
| pigmej | add it :) | 09:33 |
| *** zefciu has joined #openstack-solar | 09:33 | |
| pigmej | mkwiek: check torrent example | 09:33 |
| pigmej | basically you need to connect transport resource with any other resource | 09:33 |
| dshulyak | salmon_: have you noticed how long it takes to deploy 3 controllers with current solar? maybe you have history? | 09:33 |
| pigmej | BUT because transports are kinda 'special' (ugly hack around transports_id) | 09:34 |
| pigmej | yuou need create transports container first | 09:34 |
| pigmej | mkwiek: genrally, transprots are let's say lazy evaluated | 09:34 |
| mkwiek | ok, thanks :) | 09:34 |
| pigmej | each resource have transports_id, and using this transports_id, we search for actual transports resource data, to *not* trigger updates of all resources when you change ssh_user | 09:34 |
| mkwiek | I see, makes sense | 09:35 |
| pigmej | so basically, you need a transports_container stuff (resoruces/transports) | 09:35 |
| pigmej | then you connect these transports to *any* resource as you want | 09:35 |
| pigmej | then you connect your transport to this transports_container | 09:35 |
| pigmej | UX <3 | 09:35 |
| mkwiek | indeed :) | 09:36 |
| pigmej | torrent/example.py is your manual :) | 09:36 |
| pigmej | but well... why you need it ? | 09:36 |
| pigmej | because I'm making currently tests of transports / hanlders | 09:37 |
| salmon_ | dshulyak: too long | 09:41 |
| salmon_ | dshulyak: I';m deploying 2 controllers now | 09:41 |
| pigmej | salmon_: int(too long) is invalid :( | 09:41 |
| dshulyak | salmon_: 2 and 3 shouldnt be any different, there is probably a bug in f2s parser or elsewhere | 09:43 |
| salmon_ | dshulyak: what kind of bug? | 09:43 |
| dshulyak | i dont know :) but total time of deployment for 2 and 3 controllers should be ~ equal | 09:44 |
| dshulyak | or if you dont have enough cpu than for 3 vms than it wont be | 09:45 |
| mkwiek | pigmej: I am just making tests of my changes | 09:45 |
| pigmej | k | 09:45 |
| pigmej | you could mock BAT function though | 09:46 |
| mkwiek | pigmej: but this is what I want to test :) | 09:52 |
| *** dshulyak has quit IRC | 10:25 | |
| *** dshulyak has joined #openstack-solar | 10:45 | |
| openstackgerrit | Dmitry Shulyak proposed openstack/solar: Refactoring of graph.py and usage of it in scheduler https://review.openstack.org/294605 | 11:08 |
| openstackgerrit | Dmitry Shulyak proposed openstack/solar: Implement traversal based on number of childs https://review.openstack.org/295348 | 11:08 |
| salmon_ | dshulyak: do you want results from 2 controllers deploy? | 11:34 |
| dshulyak | salmon_: yep, why not :) | 11:35 |
| salmon_ | dshulyak: https://bpaste.net/show/39b9bc274d61 | 11:36 |
| dshulyak | hm | 11:37 |
| dshulyak | 6890 ? | 11:38 |
| dshulyak | it is about 2 hours | 11:38 |
| pigmej | ;/ | 11:39 |
| pigmej | salmon_: why it's that slow ? | 11:39 |
| pigmej | you said me that you was out of ram for a moment (at least), right? | 11:39 |
| dshulyak | it looks like something just stuck, i will add delta for each task in to report | 11:41 |
| dshulyak | pigmej: can you take a look at this patch again https://review.openstack.org/#/c/294605/? i had to add save_all_lazy before lock is released, otherwise there was concurrency issues with riak backend | 11:42 |
| pigmej | how could it be ? | 11:42 |
| dshulyak | what? | 11:43 |
| pigmej | I mean, why this save_all_lazy fixes the concurrency issue? | 11:43 |
| dshulyak | because lock is released before session_end | 11:43 |
| dshulyak | i think it might be connected to smart waiter, but i am not sure | 11:43 |
| pigmej | so we need to find a way to release it after session_end :) | 11:43 |
| pigmej | smart waiter just sends signals, and it then uses "normal" logic, so if smart waiter causes error then lock is not working properly :/ | 11:44 |
| salmon_ | I said it's slow :P | 11:44 |
| dshulyak | in fuel deployment of 3 controllers ~ 30-40 mins with provisioning | 11:44 |
| dshulyak | so it is either lack of ram/cpu of bugs in solar | 11:45 |
| dshulyak | of/or | 11:45 |
| pigmej | or salmon environment | 11:46 |
| pigmej | P: | 11:46 |
| pigmej | dshulyak: is there any always failing case for this lock? | 11:46 |
| pigmej | beceause if solution should be "unlock after session is ended" I would prefer to have it this way, because I can imagine easily situation that we still may have similar problem in future | 11:48 |
| dshulyak | pigmej: u can checkout my patch and remove save_all_lazy in scheduler.py, you wont have 100% reproducibility - but i have 1-2 functional tests failures every 2-3 run | 11:48 |
| dshulyak | actually previously we were saving all records in graph before releasing lock | 11:49 |
| dshulyak | so it is not regression :) | 11:49 |
| salmon_ | I think it's ram | 11:50 |
| pigmej | dshulyak: yaeh but my question is 'Should it be the fix, or we rather should change when lock is released' | 11:50 |
| pigmej | we could easily add "callbacks" for session_end | 11:50 |
| dshulyak | i think save_all_lazy explicitly is good enough for now, also maybe it would be better to get rid from lock at all | 11:52 |
| pigmej | dshulyak: Could you please make note in these lines ? | 11:53 |
| pigmej | because well, this sounds "hacky" to me :D | 11:53 |
| dshulyak | ok, will do in a moment’ | 12:06 |
| dshulyak | salmon_: do you have script to bootstrap solar on fuel master? | 12:06 |
| salmon_ | dshulyak: nope. I'm doing it manually | 12:07 |
| pigmej | dshulyak: just take parts from centos provisioning, should be fine | 12:08 |
| pigmej | :) | 12:08 |
| mkwiek | pigmej: do we have any fixtures for tests with resources with transports connected to them? | 13:37 |
| pigmej | nope | 13:39 |
| pigmej | I just have some composer files | 13:39 |
| pigmej | https://bpaste.net/show/a3d28f084a6e | 13:39 |
| pigmej | like that | 13:40 |
| mkwiek | can I add them to fixtures? | 13:41 |
| pigmej | what ? | 13:41 |
| pigmej | what fixtures ? | 13:41 |
| mkwiek | does it even make sense to do it? to add this kind of file to let's say, solar/test/resource_fixtures ? | 13:52 |
| pigmej | well, but what's the point ? I mean, this is for 'bat' tests ? | 13:53 |
| pigmej | if so, then adding resources and loading them will test in fact "resources transports passing" | 13:53 |
| pigmej | and not BAT 'load' stuff | 13:53 |
| mkwiek | when I have resources with transports attached to them, I can test the BAT behaviour. At least that's my understanding | 13:55 |
| pigmej | why not just mock transports() from resource? | 13:58 |
| mkwiek | because I still don't know what format is returned from there :( | 13:59 |
| pigmej | mkwiek: just load cli with ANY resource | 13:59 |
| pigmej | and call resource.transports() | 13:59 |
| pigmej | r = resource.load('node1') | 13:59 |
| pigmej | print r.transports() | 14:00 |
| pigmej | :) | 14:00 |
| mkwiek | crap, it makes sense, it seems I had a brainfart... | 14:00 |
| pigmej | ;D | 14:00 |
| *** openstack has joined #openstack-solar | 14:22 | |
| *** spyzalski has quit IRC | 14:58 | |
| *** openstackstatus has joined #openstack-solar | 15:12 | |
| *** ChanServ sets mode: +v openstackstatus | 15:12 | |
| *** dshulyak has quit IRC | 15:31 | |
| *** dshulyak has joined #openstack-solar | 16:20 | |
| *** dshulyak has quit IRC | 17:30 | |
| *** angdraug has joined #openstack-solar | 17:54 | |
| *** angdraug has quit IRC | 18:33 | |
| *** angdraug has joined #openstack-solar | 18:49 | |
| *** angdraug has quit IRC | 18:50 | |
| *** openstack has joined #openstack-solar | 19:08 | |
| *** openstack has joined #openstack-solar | 19:21 | |
| *** pigmej has joined #openstack-solar | 19:22 | |
| *** salmon_ has joined #openstack-solar | 19:23 | |
| *** openstackstatus has joined #openstack-solar | 19:24 | |
| *** ChanServ sets mode: +v openstackstatus | 19:24 | |
| *** openstack has joined #openstack-solar | 20:31 | |
| *** angdraug has joined #openstack-solar | 20:50 | |
| *** angdraug has quit IRC | 23:02 | |
| *** openstack has joined #openstack-solar | 23:23 | |
| *** openstackstatus has joined #openstack-solar | 23:24 | |
| *** ChanServ sets mode: +v openstackstatus | 23:24 | |
| *** angdraug has joined #openstack-solar | 23:45 | |
| *** salmon_ has quit IRC | 23:53 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!