Wednesday, 2013-10-30

*** malini_afk is now known as malini00:23
*** reed has quit IRC00:33
*** ayoung has quit IRC00:40
*** oz_akan_ has joined #openstack-marconi00:44
*** amitgandhi has joined #openstack-marconi00:46
*** oz_akan_ has quit IRC00:51
*** amitgandhi has quit IRC00:51
*** oz_akan_ has joined #openstack-marconi00:51
*** nosnos has joined #openstack-marconi01:03
*** ayoung has joined #openstack-marconi01:17
*** amitgandhi has joined #openstack-marconi01:18
*** vkmc has quit IRC01:20
*** oz_akan__ has joined #openstack-marconi01:58
*** amit has joined #openstack-marconi02:00
*** amit is now known as Guest6287202:00
*** oz_akan_ has quit IRC02:00
*** amitgandhi has quit IRC02:04
*** Guest62872 has quit IRC03:09
*** oz_akan__ has quit IRC03:13
*** oz_akan_ has joined #openstack-marconi03:14
*** amitgandhi has joined #openstack-marconi03:15
*** tedross has joined #openstack-marconi03:16
*** amitgandhi has quit IRC03:25
*** amitgandhi has joined #openstack-marconi03:26
*** amitgandhi has quit IRC03:30
*** malini is now known as malini_afk03:37
*** oz_akan_ has quit IRC03:44
*** tedross has quit IRC04:15
*** oz_akan_ has joined #openstack-marconi04:15
*** oz_akan_ has quit IRC04:20
*** amitgandhi has joined #openstack-marconi04:26
*** amitgandhi has quit IRC04:30
*** amitgandhi has joined #openstack-marconi04:36
*** amitgandhi has quit IRC04:40
*** oz_akan_ has joined #openstack-marconi05:16
*** oz_akan_ has quit IRC05:21
*** amitgandhi has joined #openstack-marconi05:27
*** amitgandhi has joined #openstack-marconi05:27
*** amitgandhi has quit IRC05:32
*** oz_akan_ has joined #openstack-marconi06:17
*** oz_akan_ has quit IRC06:21
*** amitgandhi has joined #openstack-marconi06:28
*** amitgandhi has quit IRC06:32
*** amitgandhi has joined #openstack-marconi06:38
*** amitgandhi has quit IRC06:42
*** oz_akan_ has joined #openstack-marconi07:18
*** oz_akan_ has quit IRC07:22
*** amitgandhi has joined #openstack-marconi07:28
*** flaper87|afk is now known as flaper8707:44
*** oz_akan_ has joined #openstack-marconi08:18
*** flaper87 is now known as flaper87|afk08:22
*** oz_akan_ has quit IRC08:23
*** amitgandhi has joined #openstack-marconi08:29
*** flaper87|afk is now known as flaper8708:32
*** amitgandhi has quit IRC08:33
*** amitgandhi has joined #openstack-marconi08:39
*** amitgandhi has quit IRC08:44
*** yassine has joined #openstack-marconi09:13
*** amitgandhi has joined #openstack-marconi09:29
*** amitgandhi has joined #openstack-marconi09:30
*** amitgandhi has quit IRC09:34
*** fifieldt has quit IRC10:10
*** amitgandhi has joined #openstack-marconi10:30
*** amitgandhi has quit IRC10:35
*** amitgandhi has joined #openstack-marconi10:40
*** amitgandhi has quit IRC10:45
*** malini_afk is now known as malini10:58
*** flaper87 is now known as flaper87|afk11:24
*** amitgandhi has joined #openstack-marconi11:30
*** tedross has joined #openstack-marconi12:12
*** nosnos has quit IRC12:27
*** amitgandhi has joined #openstack-marconi12:32
*** flaper87|afk is now known as flaper8712:35
*** amitgandhi has quit IRC12:36
*** kgriffs_afk is now known as kgriffs12:39
*** amitgandhi has joined #openstack-marconi12:42
flaper87kgriffs: amitgandhi good morning guys!12:43
*** amitgandhi has quit IRC12:46
*** tedross has quit IRC12:50
*** whenry has quit IRC12:54
kgriffsflaper87: good afternoon!12:55
*** malini is now known as malini_afk13:04
*** oz_akan_ has joined #openstack-marconi13:04
*** oz_akan_ has quit IRC13:05
*** oz_akan_ has joined #openstack-marconi13:05
*** mpanetta has joined #openstack-marconi13:10
*** amitgandhi has joined #openstack-marconi13:23
openstackgerritKurt Griffiths proposed a change to openstack/marconi: chore: Pass cache into drivers  https://review.openstack.org/5444513:38
*** amitgandhi has quit IRC13:47
*** amitgandhi has joined #openstack-marconi13:47
flaper87so much silence in here!14:07
*** jergerber has joined #openstack-marconi14:09
* kgriffs watches a tumbleweed roll by14:10
oz_akan_malini_afk: there?14:10
*** whenry has joined #openstack-marconi14:11
* flaper87 prepared some bamboo slides for the summit!14:14
* ametts wonders how information written on pieces of bamboo will get into the projector.14:15
*** jcru has joined #openstack-marconi14:15
flaper87ametts: mmmh, that's a good question. Isn't there an adapter for that?14:18
flaper87:D14:18
*** malini_afk is now known as malini14:18
kgriffsflaper87: yep, it's called an overhead projector14:19
flaper87LOOOL14:19
amettshttp://www.bhphotovideo.com/c/product/841314-REG/Wacom_ACK40401_Bamboo_Tablet_Wireless_Accessory.html14:20
amettsWireless, even!14:20
malinioz_akan_: ping14:20
flaper87man, I didn't know wacom has a tablet14:20
flaper87aaaaaaanyway!14:21
kgriffsprotip: don't sudo ulimit on a mac14:27
kgriffsflaper87, ametts: I'll be working on my own slides a little later today14:28
flaper87kgriffs: I did some work this morning, I'm keeping everything in the etherpad14:28
flaper87ametts: ^14:28
mpanettakgriffs: Why don't sudo ulimit on a mac?14:31
flaper87kgriffs: LOL @ ulimit! I think I did that once, not sure if it broke anything, though.14:31
*** kgriffs is now known as kgriffs_afk14:44
*** alcabrera has joined #openstack-marconi14:49
alcabreraGood morning!14:49
flaper87alcabrera: GOOOOOD MORNING!14:50
flaper87alcabrera: question, What do you think about proposing marconi-redis-storage as part of OPW?14:50
flaper87I know we kinda talk about this but, I was thinking about it a bit more and it may make sense after all14:51
flaper87plus, it'd be cool to ship an in-memory storage along with marconi14:51
alcabrera+114:52
alcabreraI'm in favor of it, and I'd be happy to serve as mentor on whoever took on the project. :D14:52
alcabreraflaper87: ^14:52
flaper87alcabrera: awesome, thanks a lot!14:53
alcabreraflaper87: How did PyMilano go yesterday? :)14:54
alcabreraflaper87: regarding https://review.openstack.org/#/c/52160/9/marconi/queues/storage/mongodb/catalogue.py (insert vs. update)15:00
alcabreraThe docs are out of date. :P15:00
*** cpallares has joined #openstack-marconi15:00
alcabreraI did mean for an update-on-existing behavior.15:00
flaper87alcabrera: ok, then, I'm cool with it!15:06
flaper87alcabrera: what confused me is that there's an 'update' method!15:07
flaper87why aren't you re-using the insert one instead ?15:07
alcabreraI should probably merge the two at some point. :/15:07
alcabreraProbably soon.15:07
alcabrera:P15:07
alcabreraupdate is leftover from when I didn't realize how this should connect to the transport layer (PUT style)15:08
alcabreraSoooo.'..15:08
alcabreraHmmm15:08
alcabreraI should probably make an upsert method!15:08
flaper87alcabrera: oooooooooooooooooooooooooooooooooooke! :)15:08
alcabreralemme take care of that now. Meanwhile...15:08
alcabrera:D15:08
alcabrera:D15:08
flaper87alcabrera: also, maybe multi=False is not necessary in the update since there's a unique index on PRIMARY_KEY15:09
alcabreraCould you review this other one: https://review.openstack.org/#/c/50998/15:09
alcabrera+115:09
alcabreraflaper87: oh! Just realized something. Hmm....15:11
alcabreraI don't think I want to merge insert/update, since I want to raise a QueueNotMapped if the entry doesn't exist on a PATCH.15:11
alcabrerahmmm...15:12
* alcabrera thinks and open a pymongo console15:12
flaper87hence the `if not res['updatedExisting']`15:13
alcabrerayeah, I think I might be able to use that. :D15:13
flaper87:D15:13
alcabreraI think that's why I want to keep the two separate - upsert creates the entry anyway.15:14
alcabreraI don't want that to happen for a PATCH.15:14
alcabreraSooooo15:14
alcabreraI can reuse a common body for the impl.15:14
alcabreraBut I want the update to continue raising, and to have upsert=False.15:14
malinizyuan: I forgot to mention y'day..But I have the tsung stats test ready15:24
zyuanmalini: thanks! how you use a patch? a github branch, or a gerrit review?15:25
zyuanmalini: review is at https://review.openstack.org/#/c/50969/15:25
malinizyuan: you need to verify the perf before merging , rt?15:25
zyuanyes15:26
*** ayoung has quit IRC15:26
malinihmm..I have a simple standalone cloud server running marconi..Its nothing fancy & doesnt have a replicaset etc. WIll tht be good enuf for testing ur patch ?15:27
zyuani think that's enough15:27
*** whenry has quit IRC15:27
maliniok..I'll give it a try , with & without ur patch15:27
openstackgerritAlejandro Cabrera proposed a change to openstack/marconi: feat: add catalogue storage driver for queues  https://review.openstack.org/5216015:28
alcabreraflaper87: all feedback addressed. ^^15:28
flaper87alcabrera: awesome, thanks a lot!15:40
alcabreranp, thank you! :(15:40
alcabrera...15:40
alcabrera:)15:40
alcabrera^15:40
flaper87alcabrera: LGTM ;)15:43
alcabreraw00t15:43
alcabreracpallares: I found some time to review your 'exceptions -> errors' patch. I left my comments inline. Glad to see you were able to make it past Jenkins! :D15:53
*** vkmc has joined #openstack-marconi15:54
*** vkmc has quit IRC15:54
*** vkmc has joined #openstack-marconi15:54
alcabreraTime to continue working on sharding~15:55
cpallaresalcabrera: thanks!15:56
cpallaresalcabrera: I thought since there's another module from pymongo called errors, there would be name conflicts, so there wouldn't?15:57
alcabreracpallares: hmm...15:58
alcabreraIf the import is done as import pymongo15:58
alcabrerathen you'd have to do 'pymongo.errors' to access that one.15:58
alcabreraThere shouldn't be a name collision in that case, as a result of that.15:58
alcabreraSo there'd be 'marconi.queues.storage' => 'errors' and 'pymongo' => 'pymongo.errors'15:59
cpallaresalcabrera: I get you.15:59
alcabreraOf course, the unit tests could prove me completely wrong. They know better. ;)15:59
openstackgerritCindy Pallares proposed a change to openstack/marconi: Renamed "exceptions" module to "errors"  https://review.openstack.org/5414116:18
cpallaresOkay this one should be okay!16:18
cpallares...I hope.16:19
* alcabrera checks it out16:20
alcabreracpallares: for this file (https://review.openstack.org/#/c/54141/8/marconi/proxy/storage/mongodb/catalogue.py)16:20
alcabrerawas just using 'errors' causing a problem?16:20
cpallaresalcabreara: No, actually it wasn't. I thought I had looked over the files. I'll check them again.16:22
alcabreracpallares: I noticed you removed all the other instances. This one caught my eye while reviewing. :)16:22
alcabrerainstances of 'exceptions -> X_errors'16:22
malinizyuan: I tried with & without the patch & dont see significant improvement in stats16:27
malinihttp://198.61.239.147:8000/log/20131030-1620/report.html - with patch16:27
malinihttp://198.61.239.147:8000/log/20131030-1624/report.html - without patch16:27
cpallaresalcabreara: Mmm... I get a pep8 error redefinition of unused 'errors' from line 2816:32
cpallaresalcabrera: but it passes in py2716:33
alcabrerahmmmmmm16:35
alcabreraahhh16:36
alcabrerayeah16:36
alcabrerathat makes sense!16:36
alcabreraCool16:36
alcabreracpallares: thanks for investigating that. :D16:36
maliniflaper87: ping16:36
alcabreracpallares: +2'd16:36
cpallaresalcabrera: yay :D16:37
*** aniuskad has joined #openstack-marconi16:37
alcabreracpallares: congrats! Once flaper87 and/or kgriffs_afk gets around to reviewing it, it should be merged in shortly. :)16:46
flaper87back17:00
flaper87sorry, I was in a meeting17:00
alcabreraflaper87: wb!17:01
*** reed has joined #openstack-marconi17:02
*** ayoung_ has joined #openstack-marconi17:09
*** kgriffs_afk is now known as kgriffs17:10
*** yassine has quit IRC17:10
zyuanmalini: did you tried interleaving posting and stats?17:10
kgriffsalcabrera: how's your patch coming? I am working on caching but want to wait and base mine on yours17:13
openstackgerritA change was merged to openstack/marconi: feat: add catalogue storage driver for queues  https://review.openstack.org/5216017:13
kgriffsi mean, I am not waiting to implement, just to submit17:14
alcabrerakgriffs: I'm almost through with the implementation. I just finished porting...17:16
alcabrera... the weighted select algorithm from the proxy.17:16
kgriffskewl17:16
alcabreraI also finished putting together a means to convert from a python dict -> oslo.config.cfg.CONF.17:16
alcabreraWhat's left is to unit test this thing! :D17:17
kgriffsrock on17:17
flaper87cpallares: +2 ;) Thanks a lot! great first contribution!17:17
alcabreraflaper87, kgriffs: The last patch left blocking the sharding development in the review queue is: https://review.openstack.org/#/c/50998/17:17
alcabreraThanks for all your help reviewing those!17:18
alcabreracpallares: w00t - keep making marconi awesome! :D17:18
flaper87alcabrera: since that's the last one, kgriffs and I will charge you a fee for the review!17:18
alcabreraflaper87: $3.5017:18
flaper87alcabrera: small one, just a couple of thousand dollars17:18
alcabreraD:17:18
flaper87each17:18
alcabreracocosettes?17:19
alcabreraMaybe?17:19
alcabreraI could do those. ;P17:19
flaper87alcabrera: Man, I got those! :D17:19
alcabreralol17:19
flaper87fvollero: sent me some!17:19
flaper87alcabrera: btw, did you find them there?17:19
kgriffsI prefer bitcoin. Harder for the NSA to track.17:19
alcabrerafvollero: maaan, you broke my bargain chip. :P17:19
* kgriffs looks over shoulder17:19
kgriffsalcabrera: fwi, that last patchset was just git cherry-pick sillyness I guess. I hadn't thought I'd changed anything.17:20
kgriffsflaper87: TBH I haven't been following your cache patch as closely as I would like to lately. How is it looking?17:21
alcabrerakgriffs: weird connection errors solved?17:21
flaper87kgriffs: I addressed the last comments so, hopefully, people will look at it very soon!17:21
kgriffsalcabrera: yes. OS X Mavericks was just being retarded17:22
flaper87kgriffs: no changes to the API, though17:22
kgriffshad a really low ulimit17:22
kgriffsflaper87: cool, gtk17:22
kgriffsI still want to help write backends for it. Things have just been so crazy lately. :p17:22
flaper87I don't think anyone is going to review it before the smmit, TBH!17:23
kgriffsheh. probably not.17:23
alcabreraIt'll be nice to double down on the oslo.cache patch after that.17:24
alcabreraThough I remember seeing a session in the Summit schedule...17:24
alcabreraSomething about better oslo-incubator structuring.17:24
alcabreraThat sounded relevant to the future of oslo.cache17:24
fvolleroflaper87: some of?17:25
fvolleroalcabrera: whaaat ? :)17:25
flaper87fvollero: cocosette chamo, cocosette!17:25
*** notmyname has quit IRC17:25
*** notmyname has joined #openstack-marconi17:26
alcabrerabrb17:32
*** alcabrera is now known as alcabrera|afk17:32
flaper87alcabrera|afk: you come back here, NOW!17:33
*** ayoung_ has quit IRC17:35
fvolleroflaper87: che, eso es un problema, ya la que tenia terminaron :)17:43
*** aniuskad has quit IRC17:44
fvolleroflaper87: some bug i can fix to get again in sync with marconi?17:44
*** whenry has joined #openstack-marconi17:45
flaper87fvollero: erm, lesseeeeeeeeeeeeeeeee17:46
flaper87alcabrera|afk: btw, me saying lessee is your fault!17:46
flaper87:D17:46
flaper87fvollero: https://bugs.launchpad.net/marconi/+bug/124598717:48
flaper87mpanetta: ^17:48
* fvollero checks17:49
flaper87kgriffs: https://review.openstack.org/#/c/50638/17:49
flaper87O.O did I just drop that in here?17:49
flaper87did I really mentioned kgriffs in that message ?17:49
flaper87mention*17:50
kgriffs:D17:54
kgriffsah yes, we need that client for the summit!17:54
kgriffsas soon as I get my slides done and caching for sharding, I'll go into review mode17:54
*** vkmc has quit IRC17:56
openstackgerritCindy Pallares proposed a change to openstack/marconi: Renamed "exceptions" module to "errors"  https://review.openstack.org/5414118:01
*** amitgandhi has quit IRC18:03
flaper87kgriffs: I just asked for a review there, don't pretend it'll work after it's merged!18:04
* flaper87 hides18:04
openstackgerritA change was merged to openstack/python-marconiclient: Implement queue's API methods  https://review.openstack.org/5063818:07
kgriffsflaper87: LOL18:11
kgriffsI will take a look at the other patches ASAP18:12
flaper87kgriffs: cool, thanks a LOT!18:12
kgriffsbtw, here is the inspiration for the t-shirt: https://en.wikipedia.org/wiki/File:Coherer_Rcvr.jpg18:17
openstackgerritCindy Pallares proposed a change to openstack/marconi: Renamed "exceptions" module to "errors"  https://review.openstack.org/5414118:23
*** alcabrera|afk is now known as alcabrera18:23
alcabreralet's see what I missed~18:23
mpanettaflaper87: What ya need? :)18:24
alcabreralol18:25
kgriffsalcabrera: I've got caching ready, just waiting on your patch18:25
alcabreraflaper87: leseeeee is *very* effective in communicating deliberation and ACKs. :D18:25
alcabrerakgriffs: awesome!18:25
alcabreraI'll share a WIP.18:25
kgriffslet's make this real today18:25
flaper87mpanetta: erm money?18:26
flaper87mpanetta: :P18:26
flaper87mpanetta: fvollero is taking a bug you created18:26
mpanettaAh18:26
flaper87he may need more info18:26
mpanettaHmm zyuan looked at that one yesterday18:26
flaper87yup18:26
mpanettaOk18:26
flaper87not sure if you guys decided something18:26
mpanettaI can give him what I gave zyuan18:26
kgriffsflaper87: can you review this one soon? https://review.openstack.org/#/c/50998/18:26
mpanettaIt is a weird bug18:26
flaper87kgriffs: looking18:27
kgriffsbrb (meeting)18:27
kgriffsflaper87: thanks man. Rackspace is really keen on getting sharding in ASAP18:27
kgriffs(albeit with high quality)18:27
kgriffs:D18:27
mpanettaI am going to go get food.18:29
mpanettaI will bbiab18:29
alcabrerakgriffs: about to post a WIP - writing a commit message.18:31
alcabreraThe WIp has the core logic but is missing the unit tests.18:31
alcabrera*WIP18:31
kgriffskk18:31
kgriffsWIP...WIP it good!18:31
*** amitgandhi has joined #openstack-marconi18:32
malinizyuan: ping18:33
zyuanmalini: ?18:33
malinican you join the other channel?18:33
zyuanit'm there18:33
zyuani'm18:34
openstackgerritAlejandro Cabrera proposed a change to openstack/marconi: feat: connect sharding manager to control drivers (WIP)  https://review.openstack.org/5460518:35
openstackgerritAlejandro Cabrera proposed a change to openstack/marconi: feat: integrate shard storage with transport  https://review.openstack.org/5099818:35
alcabrerakgriffs: There's that WIP I promised, with possibly a rebase on the shard storage patch. :P18:35
alcabrerakgriffs: lol about WIP puns. :P18:38
openstackgerritZhihao Yuan proposed a change to openstack/marconi: feat(health): make health API storage-aware  https://review.openstack.org/5345718:38
flaper87alcabrera: I think you need to rebase: https://review.openstack.org/#/c/50998/18:40
flaper87with latest cpallares patch18:41
alcabreraflaper87: I'll do that now! :)18:41
flaper87alcabrera: the patch is being merged18:41
flaper87:P18:41
alcabreraI may have... tripped it up with the WIP sharding patch above.18:41
openstackgerritAlejandro Cabrera proposed a change to openstack/marconi: feat: integrate shard storage with transport  https://review.openstack.org/5099818:42
openstackgerritAlejandro Cabrera proposed a change to openstack/marconi: feat: connect sharding manager to control drivers (WIP)  https://review.openstack.org/5460518:43
alcabreraflaper87: done!18:43
alcabrera:D18:43
flaper87alcabrera: mmh, you sure? it still says it depends on your previous patch18:44
flaper87cpallares' patch hasn't landed yet18:44
flaper87Zuul is stil 'tasting' it18:45
flaper87still18:45
alcabrerahmmm18:45
alcabreraweird18:45
alcabreraflaper87: hmmmm - it *should* depend on the catalogue storage driver. I'm confused. :x18:46
flaper87alcabrera: I meant to say that your patch will conflict with cpallares changes18:48
flaper87s/queues://18:48
alcabreraahh18:48
alcabreraI see.18:48
openstackgerritA change was merged to openstack/marconi: Renamed "exceptions" module to "errors"  https://review.openstack.org/5414118:48
alcabreraThere we go.18:48
alcabreraOkay, I'll rebase now.18:49
flaper87mmh actually, I was wrong but it would conflict anyway18:49
flaper87:D18:49
alcabrera:)18:49
alcabrerayup, super conflict18:49
alcabreraFixing~18:49
flaper87she was working on 2 patches but the one that got merged is the exceptions one18:49
flaper87anyway, we both were confused and right at the same time18:49
flaper87what a strange combination18:49
alcabreracool - only 2 files in conflict18:50
alcabreranot bad18:50
cpallaresflaper87: haha18:50
cpallaresyay my patch merged :D18:50
flaper87cpallares: CONGRATS!18:50
flaper87you're officially a Marconi committer18:51
flaper87ok, gtg now!18:51
cpallares:D18:51
flaper87may be back later!18:51
flaper87take care guys!18:51
cpallareshaha bye!18:51
*** flaper87 is now known as flaper87|afk18:52
alcabreraflaper87|afk: take care! :)18:52
alcabreracpallares: also, congrats! Happy to have you working with us. :)18:54
openstackgerritAlejandro Cabrera proposed a change to openstack/marconi: feat: integrate shard storage with transport  https://review.openstack.org/5099818:55
openstackgerritAlejandro Cabrera proposed a change to openstack/marconi: feat: connect sharding manager to control drivers (WIP)  https://review.openstack.org/5460518:56
alcabreraflaper87|afk, kgriffs: all rebasing is complete. enjoy the patches. ;)18:57
kgriffsthanks!18:57
openstackgerritZhihao Yuan proposed a change to openstack/marconi: WIP: feat(health): make health API storage-aware  https://review.openstack.org/5345719:01
openstackgerritZhihao Yuan proposed a change to openstack/marconi: WIP: feat(health): make health API storage-aware  https://review.openstack.org/5345719:13
*** vkmc has joined #openstack-marconi19:15
*** malini is now known as malini_afk19:34
openstackgerritAlejandro Cabrera proposed a change to openstack/marconi: feat: connect sharding manager to control drivers (WIP)  https://review.openstack.org/5460519:35
alcabreraI need to learn the WIP command on 'git review' :P19:36
zyuanalcabrera: which?19:39
*** ayoung_ has joined #openstack-marconi19:40
alcabrerazyuan: 'git review -D'19:42
alcabreraI forget to do that every time I'm submitting a WIP.19:42
zyuancool19:44
zyuandraft....19:44
alcabrerakgriffs: getting closer to green - I've replaced the SqliteShardedTest from Messages with a TODO(...) and am now working into the MongodbSharded test.19:55
kgriffsw00t19:55
kgriffsI have my patches rebased on the WIP, so should be easy to finalize now19:55
alcabreraawesome!19:56
*** vkmc has quit IRC20:03
alcabreraHeading out. I'll be back in a little bit.20:23
*** alcabrera has quit IRC20:23
*** mpanetta has quit IRC20:24
*** alcabrera has joined #openstack-marconi20:46
*** kgriffs is now known as kgriffs_afk20:51
*** kgriffs_afk is now known as kgriffs20:54
alcabreraHeh, I found a catch-22 in the sharded unit tests.21:04
alcabreraShards need to be registered before queues are created, but queues are created when setUp is called. :P21:04
*** jergerber has quit IRC21:25
alcabrerakgriffs: ping21:31
kgriffso/21:33
alcabreraRequesting assistance around this catch-22. I'm thinking of placing a conditional block in the 'MessageBaseTest' that creates a few shards before queues are created.21:34
alcabreraThoughts?21:34
alcabreraThe condition being - sharded mode is active.21:35
* kgriffs is looking21:36
alcabreratests/unit/queues/marconi/transport/wsgi/test_messages21:37
kgriffsalcabrera: sorry, juggling multiple thought streams21:44
alcabreraI think I've got a way to work it.21:44
alcabrerakgriffs: thanks for looking - I've got some green going. :)21:44
kgriffsok, cool21:44
kgriffsI guess the crux of the issue is the base class doesn't know about it's child classes21:45
kgriffsso it doesn't know whether sharding is enabled21:45
kgriffss/it's/its21:46
kgriffsin the child class you could create the shards before calling super's setup21:47
alcabrerathe issue I ran into there is that neither self.app or self.bootstrap exist at that point.21:48
kgriffsah21:48
alcabreraI resolved this by creating the shards after those two exist, in one of the parent classes, IFF sharding is enabled, with appropriate tearDown if sharding is enabled.21:49
kgriffsok, based on the config INI?21:53
alcabrerayup21:55
alcabreraI've parsed the conf at that point, so I need if sharding is enabled or not, and what the uri of the DB is.21:55
alcabreras/need/know21:55
alcabrerakgriffs: ^21:55
kgriffscool beans21:57
*** oz_akan_ has quit IRC22:12
alcabreraI'm out for the night. Dinner calls~ :)22:15
alcabreraNight guys.22:15
*** alcabrera has quit IRC22:15
*** amitgandhi has quit IRC22:22
*** amitgandhi has joined #openstack-marconi22:23
*** amitgandhi has quit IRC22:27
*** jcru has quit IRC22:58
*** cpallares has quit IRC23:03
*** amitgandhi has joined #openstack-marconi23:23
*** amitgandhi has quit IRC23:30
openstackgerritAlejandro Cabrera proposed a change to openstack/marconi: feat: connect sharding manager to control drivers (WIP)  https://review.openstack.org/5460523:38
*** pquerna has quit IRC23:57
*** pquerna has joined #openstack-marconi23:58

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!