Tuesday, 2014-06-17

*** Obulpathi has joined #openstack-marconi00:01
*** Obulpathi has quit IRC00:15
*** reed has quit IRC00:34
*** rwsu has quit IRC00:52
*** amitgandhi has joined #openstack-marconi00:54
*** haomaiwa_ has joined #openstack-marconi00:54
*** malini2 has joined #openstack-marconi01:05
*** amitgandhi has quit IRC01:05
*** nosnos has joined #openstack-marconi01:26
*** vkmc has quit IRC01:37
*** haomaiwa_ has quit IRC01:43
*** norman has quit IRC01:59
*** norman has joined #openstack-marconi01:59
*** malini2 has quit IRC02:06
*** renlt has joined #openstack-marconi03:01
normanHi all, anyone can help to run the "tox -e py33" in your local env?  I run lots of issue with the *refresh* code (by git stash)03:20
normansuch as   File "./marconi/common/transport/wsgi/helpers.py", line 77, in validate_queue_identification03:25
norman    queue = params['queue_name'].decode('utf-8', 'replace')03:25
normanAttributeError: 'str' object has no attribute 'decode'03:25
*** nosnos has quit IRC03:41
openstackgerritA change was merged to openstack/marconi: Re-enable tests affected by #1321840  https://review.openstack.org/9911704:13
openstackgerritA change was merged to openstack/marconi: remove default=None for config options  https://review.openstack.org/9954504:13
*** norman has quit IRC04:25
*** norman has joined #openstack-marconi04:27
*** norman has quit IRC04:30
*** norman has joined #openstack-marconi04:30
*** nosnos has joined #openstack-marconi04:30
*** haomaiwa_ has joined #openstack-marconi04:39
*** haomaiw__ has joined #openstack-marconi04:54
*** haomaiwa_ has quit IRC04:56
openstackgerritA change was merged to openstack/marconi: Refactor auxiliary methods for MongoDB unit tests  https://review.openstack.org/9891606:04
*** alcabrera|afk is now known as alcabrera06:18
*** alcabrera is now known as alcabrera|afk06:23
*** flaper87|afk is now known as flaper8706:32
*** chandan_kumar has quit IRC06:38
*** ykaplan has quit IRC06:51
*** AAzza_afk is now known as AAzza07:08
AAzzanorman: hi, yes.  There is no problem with your setup. Marconi is not fully py3k compatible, so some tests fails under py33.  I'm currently fixing this.07:23
openstackgerritNataliia Uvarova proposed a change to openstack/marconi: Make storage.utils.keyify py3k compatible  https://review.openstack.org/10008407:30
normanAAzza:  Thanks, then do you mean that I can send out the code review even tox failed?07:32
AAzzanorman: yes, you can if tests fails only for py33. py33 gate is not voting now :)07:38
norman-:)07:38
AAzzanorman: but if it is not difficult can you check that your changes don't introduce new failing tests. simplest way just to look at number of failing one before and after changes.07:39
normanAAzza: yes, I will07:39
AAzzanorman: i suppose current is 28 (if without mongodb driver) or smth like that07:40
normanAAzza : btw, can you help to check the 'tox -e lint2' , why there are tons of errors in my local env ,such as  "instance of "testQueueMisc' has no 'assertEqual' member (no-member)"07:41
AAzzanorman: hmm, not sure about lint2, whether it is used or should be used, sorry) but for submitting review simple check on tox -e pep8 will be enough07:43
normanAAzza: I'd thought I need to run  "tox"07:44
normanAAzza: since it's the first time what I will deliver review in marconi and openstack,  so for another silly question,  do I have to run tempest in devestack before submitting ?  or "tox -e py27", "tox -e pep8"  is Ok now?07:46
AAzzanorman: the py27 and  and pep8 is required minimum. if you can you can (hmm?) run in tempest, but honestly i don't think anyone do it locally. free to submit patch if these two is working and see what gate will tell you07:48
normanAAzza: thanks ,that's helpful07:49
*** renlt has quit IRC07:50
*** nosnos has quit IRC08:16
*** haomaiw__ has quit IRC08:16
*** torgomatic has quit IRC08:16
*** russell_h has quit IRC08:16
*** VeggieMeat has quit IRC08:16
*** mkoderer has quit IRC08:16
*** kgriffs|afk has quit IRC08:16
*** megan_w has quit IRC08:16
*** lvh has quit IRC08:16
*** pquerna has quit IRC08:16
*** tjanczuk_ has quit IRC08:16
*** Ephur has quit IRC08:16
*** alcabrera|afk has quit IRC08:16
*** dragondm has quit IRC08:16
*** boris-42 has quit IRC08:16
*** openstackgerrit has quit IRC08:16
*** fifieldt has quit IRC08:16
*** mwagner_lap has quit IRC08:16
*** norman has quit IRC08:16
*** amalagon has quit IRC08:16
*** AAzza has quit IRC08:16
*** ekarlso has quit IRC08:16
*** tmu_ has quit IRC08:16
*** ChanServ has quit IRC08:16
*** stannie has quit IRC08:16
*** peoplemerge has quit IRC08:16
*** rektide has quit IRC08:16
*** flaper87 has quit IRC08:16
*** ametts has quit IRC08:17
*** malini has quit IRC08:17
*** dmitryme has quit IRC08:17
*** dragondm has joined #openstack-marconi08:22
*** alcabrera|afk has joined #openstack-marconi08:22
*** Ephur has joined #openstack-marconi08:22
*** tjanczuk_ has joined #openstack-marconi08:22
*** pquerna has joined #openstack-marconi08:22
*** lvh has joined #openstack-marconi08:22
*** russell_h has joined #openstack-marconi08:22
*** megan_w has joined #openstack-marconi08:22
*** kgriffs|afk has joined #openstack-marconi08:22
*** mkoderer has joined #openstack-marconi08:22
*** VeggieMeat has joined #openstack-marconi08:22
*** torgomatic has joined #openstack-marconi08:22
*** nosnos has joined #openstack-marconi08:22
*** haomaiw__ has joined #openstack-marconi08:22
*** norman has joined #openstack-marconi08:22
*** ametts has joined #openstack-marconi08:22
*** openstackgerrit has joined #openstack-marconi08:22
*** amalagon has joined #openstack-marconi08:22
*** fifieldt has joined #openstack-marconi08:22
*** AAzza has joined #openstack-marconi08:22
*** mwagner_lap has joined #openstack-marconi08:22
*** boris-42 has joined #openstack-marconi08:22
*** stannie has joined #openstack-marconi08:22
*** tmu_ has joined #openstack-marconi08:22
*** ekarlso has joined #openstack-marconi08:22
*** malini has joined #openstack-marconi08:22
*** peoplemerge has joined #openstack-marconi08:22
*** rektide has joined #openstack-marconi08:22
*** flaper87 has joined #openstack-marconi08:22
*** ChanServ has joined #openstack-marconi08:22
*** dmitryme has joined #openstack-marconi08:22
*** dickson.freenode.net sets mode: +ooo kgriffs|afk flaper87 ChanServ08:22
*** stannie has quit IRC08:22
*** fifieldt has quit IRC08:27
*** fifieldt has joined #openstack-marconi08:28
*** nosnos has quit IRC08:33
*** haomaiw__ has quit IRC08:33
*** torgomatic has quit IRC08:33
*** russell_h has quit IRC08:33
*** VeggieMeat has quit IRC08:33
*** mkoderer has quit IRC08:33
*** kgriffs|afk has quit IRC08:33
*** megan_w has quit IRC08:33
*** lvh has quit IRC08:33
*** pquerna has quit IRC08:33
*** tjanczuk_ has quit IRC08:33
*** Ephur has quit IRC08:33
*** alcabrera|afk has quit IRC08:33
*** dragondm has quit IRC08:33
*** ykaplan has joined #openstack-marconi08:43
*** haomaiw__ has joined #openstack-marconi08:43
*** nosnos has joined #openstack-marconi08:43
*** torgomatic has joined #openstack-marconi08:43
*** lvh has joined #openstack-marconi08:43
*** pquerna has joined #openstack-marconi08:43
*** tjanczuk_ has joined #openstack-marconi08:43
*** Ephur has joined #openstack-marconi08:43
*** alcabrera|afk has joined #openstack-marconi08:43
*** dragondm has joined #openstack-marconi08:43
*** russell_h has joined #openstack-marconi08:43
*** megan_w has joined #openstack-marconi08:43
*** kgriffs|afk has joined #openstack-marconi08:43
*** mkoderer has joined #openstack-marconi08:43
*** VeggieMeat has joined #openstack-marconi08:43
*** dickson.freenode.net sets mode: +o kgriffs|afk08:43
*** ykaplan has quit IRC08:45
*** ykaplan has joined #openstack-marconi08:45
*** mwagner_lap has quit IRC08:48
*** mwagner_lap has joined #openstack-marconi08:48
*** nosnos has quit IRC08:54
*** nosnos has joined #openstack-marconi08:55
*** jamiehannaford has joined #openstack-marconi09:12
*** jamiehannaford has quit IRC09:19
*** ykaplan has quit IRC09:27
*** AAzza is now known as AAzza_afk09:57
*** jamiehannaford has joined #openstack-marconi10:01
*** jamiehannaford has quit IRC10:02
*** AAzza_afk is now known as AAzza10:13
*** AAzza is now known as AAzza_afk10:20
*** ykaplan has joined #openstack-marconi10:34
*** alcabrera|afk is now known as alcabrera10:52
*** mwagner_lap has quit IRC12:06
*** alcabrera is now known as alcabrera|afk12:07
*** vkmc has joined #openstack-marconi12:24
*** vkmc has quit IRC12:24
*** vkmc has joined #openstack-marconi12:24
vkmchi all, good morning :)12:32
*** ykaplan has quit IRC12:35
flaper87vkmc: hey hey12:37
*** sriram has joined #openstack-marconi12:38
vkmcflaper87, hola!12:38
*** sriram has quit IRC12:39
*** sriram has joined #openstack-marconi12:39
*** haomaiw__ has quit IRC12:43
flaper87vkmc: how are you doing?12:43
*** haomaiwang has joined #openstack-marconi12:43
vkmcflaper87, good good, and you?12:43
*** haomaiwang has quit IRC12:46
*** haomaiwang has joined #openstack-marconi12:47
*** ykaplan has joined #openstack-marconi12:49
sriramyo yo yo12:50
*** jmckind has joined #openstack-marconi12:51
*** haomaiw__ has joined #openstack-marconi12:52
*** haomaiwang has quit IRC12:53
*** nosnos has quit IRC12:53
vkmcsriram, yoooooo12:53
sriramwhat's up vkmc? :)12:54
vkmcsriram, not much, and you?12:57
sriramI'm going to sign up to run a 5k.13:03
* sriram must start training13:03
vkmcoh me too13:06
vkmcbut I don't like running... it kills me13:07
* vkmc is the blerch 13:07
*** jmckind has quit IRC13:08
*** abettadapur has joined #openstack-marconi13:16
*** ykaplan has quit IRC13:22
*** malini1 has joined #openstack-marconi13:29
*** Obulpathi has joined #openstack-marconi13:30
*** Obulpathi has quit IRC13:30
*** malini1 has quit IRC13:30
*** malini1 has joined #openstack-marconi13:31
*** Obulpathi has joined #openstack-marconi13:31
*** ykaplan has joined #openstack-marconi13:36
*** alcabrera|afk is now known as alcabrera13:38
alcabreragood morning. :)13:38
flaper87alcabrera: hey hey :)13:39
alcabreraflaper87: how are you? :)13:40
flaper87alcabrera: I'm doing great, you?13:41
alcabreraI am so tired. I made the poor decision of reading so many interesting things last night, and having far too much coffee at 2am13:45
alcabrerasoo13:45
alcabrerahere I am13:45
sriramgood morning! :)13:46
alcabreraawake all day on 2 hours of nap. :P13:46
alcabreraHappy, though!13:46
alcabrerasriram: hey!13:46
sriramalcabrera: ouch, yes I had done that once . good thing is I slept for almost 10 hours the next night :P13:47
alcabrerahaha13:47
*** ykaplan has quit IRC13:48
*** ykaplan has joined #openstack-marconi13:48
openstackgerritOpenStack Proposal Bot proposed a change to openstack/marconi: Updated from global requirements  https://review.openstack.org/9903313:52
*** tonytan4ever has joined #openstack-marconi13:58
*** cpallares has joined #openstack-marconi13:58
*** balajiiyer has joined #openstack-marconi13:59
alcabreraflaper87, kgriffs|afk: https://bugs.launchpad.net/marconi/+bug/1264853 is still relevant14:05
alcabrerasince it affects sqlalchemy14:05
alcabreraI explained the situation on the bug tracker.14:06
flaper87alcabrera: LOL, I just replied :D14:06
flaper87alcabrera: thanks, you did a better job than me14:06
alcabrera:D14:06
* flaper87 should've explained why in the bug tracker too :D14:06
alcabreraI'm helping. ;)14:06
alcabrerasee - we did an *amazing* job of showing the power of implicit concurrency14:07
alcabreraala Haxl14:07
*** amitgandhi has joined #openstack-marconi14:08
*** amitgandhi has quit IRC14:10
*** amitgandhi has joined #openstack-marconi14:10
*** mwagner_lap has joined #openstack-marconi14:17
*** oz_akan_ has joined #openstack-marconi14:20
*** AAzza_afk is now known as AAzza14:20
*** rwsu has joined #openstack-marconi14:24
*** haomaiw__ has quit IRC14:27
*** haomaiwa_ has joined #openstack-marconi14:28
*** tonytan4ever has quit IRC14:37
*** tonytan4ever has joined #openstack-marconi14:38
*** haomai___ has joined #openstack-marconi14:43
*** tonytan4ever has quit IRC14:45
*** reed has joined #openstack-marconi14:46
*** haomaiwa_ has quit IRC14:46
*** tonytan4ever has joined #openstack-marconi14:51
openstackgerritOpenStack Proposal Bot proposed a change to openstack/marconi: Updated from global requirements  https://review.openstack.org/9903314:57
*** ykaplan has quit IRC14:57
*** ametts has quit IRC15:00
malini1meeting15:03
malini1#openstack-meeting-alt15:03
alcabreramalini1: thank you. I'm so tired today I had forgotten about the meeting. @_@15:03
*** chandan_kumar has joined #openstack-marconi15:04
malini1flaper87: meeting15:04
sriramkgriffs|afk: ping, meeting15:04
*** renlt has joined #openstack-marconi15:06
*** kgriffs|afk is now known as kgriffs15:06
*** amitgandhi has quit IRC15:07
*** oz_akan_ has quit IRC15:08
*** oz_akan_ has joined #openstack-marconi15:09
*** tonytan4ever has quit IRC15:11
*** tonytan4ever has joined #openstack-marconi15:17
*** tonytan4_ has joined #openstack-marconi15:26
*** tonytan4ever has quit IRC15:30
*** tonytan4_ has quit IRC15:33
*** tonytan4ever has joined #openstack-marconi15:50
*** tonytan4ever has quit IRC15:53
*** prashanthr_ has joined #openstack-marconi15:54
cpallaresmalini, vkmc regarding reviews, have you looked at http://status.openstack.org/reviews/? (search for marconi) It shows a "prioritized" list for reviews.16:01
vkmccpallares, oh I didn't know about it... fancy!16:01
malini1tht is awesome cpallares!16:02
kgriffscan someone volunteer to post the minutes to the wiki?16:02
malini1kgriffs: I will16:02
kgriffsthanks!16:03
srirambbl lunch16:03
vkmccpallares, I wonder how is the score calculated16:03
*** tonytan4ever has joined #openstack-marconi16:03
vkmcoh just hovering over it I got my answer haha16:04
kgriffsbtw everyone16:04
vkmcimportance of the feature + days in the queue16:04
cpallaresvkmc: yeah I think the patch needs to have some description on gerrit whether it's a bug or a feature16:04
kgriffsif you are looking for something to do, please check out the unassigned bps and bugs here: https://launchpad.net/marconi/+milestone/juno-216:04
*** renlt has quit IRC16:04
malini1vkmc, cpallares: if review continues being a bottle neck, we cud just scrape tht page & have a bot post in IRC16:10
vkmcsounds good16:13
cpallaresmalini1: my thoughts exactly :)16:13
malini1great minds in action :-P16:14
*** tonytan4ever has quit IRC16:15
vkmcalcabrera, one GSoC related question... midterms are next Monday, are they expecting to see some code?16:25
*** malini1 has left #openstack-marconi16:26
alcabreramidterms are next monday! I'd been tracking that.16:26
alcabreraso let's see16:26
* alcabrera pulls up Carol's email16:26
vkmcit says nothing from the students pov16:26
vkmcbut maybe I'm missing something16:27
alcabrerathis looks relevant, vkmc: http://www.google-melange.com/gsoc/document/show/gsoc_program/google/gsoc2014/help_page#9._How_do_evaluations_work16:27
alcabreraoh geez long link. >.>16:27
alcabrerahttp://goo.gl/M0NOgo16:27
alcabreramuch better16:27
vkmcyeah... there it mentions that code should be provided iff the student doesn't agree with the grade16:29
vkmcit's understandable16:30
*** malini1 has joined #openstack-marconi16:31
*** malini1 has quit IRC16:31
*** malini1 has joined #openstack-marconi16:32
*** balajiiyer has quit IRC16:36
*** tonytan4ever has joined #openstack-marconi16:46
*** tonytan4ever has quit IRC16:51
vkmcbrb, out to lunch16:57
*** prashanthr_ has quit IRC16:58
*** AAzza is now known as AAzza_afk16:59
kgriffsbbl - lunch17:13
*** tonytan4ever has joined #openstack-marconi17:16
*** AAzza_afk is now known as AAzza17:16
*** balajiiyer has joined #openstack-marconi17:21
*** kgriffs is now known as kgriffs|afk17:23
flaper87BANG! https://review.openstack.org/#/c/100633/17:42
* flaper87 throws a pop-tart in the channel17:44
* alcabrera replicates the poptart17:44
*** kgriffs|afk is now known as kgriffs17:44
* alcabrera computation diverges, and the channel overflows with poptart17:45
*** tonytan4ever has quit IRC17:53
*** tonytan4ever has joined #openstack-marconi17:53
*** kgriffs is now known as kgriffs|afk17:54
*** tonytan4ever has quit IRC17:57
*** tonytan4ever has joined #openstack-marconi17:58
sriramjust a question though, if we are going to transition to Naav, do we want a Naav specs?18:06
sriramor we'll take care of that later?18:06
sriramflaper87, alcabrera : ^18:06
alcabreraoh, yeah18:07
alcabreragood point, sriram18:07
*** kgriffs|afk is now known as kgriffs18:07
alcabrerakgriffs, flaper87, malini1, flwang: naav-specs?18:07
malini1probably tht should be part of the who;e renaming process?18:08
flaper87it will be renamed as part of the rename process18:08
alcabrerakk18:08
sriramI see, cool :)18:08
alcabrerachecking for a friend -- /cc sriram ;)18:08
*** AAzza is now known as AAzza_afk18:09
sriraminfinite poptarts alcabrera :P18:11
openstackgerritFlavio Percoco proposed a change to openstack/marconi: Rename shards to pool  https://review.openstack.org/9646318:16
flaper87alcabrera: kgriffs malini1 PLEASE ^^^^18:16
flaper87I just went through rebase hell18:16
alcabreraoh ouch18:16
flaper872 different patches against this poor baby18:16
alcabrera+614 lines18:16
alcabrerayikes whata  rebase that must've been18:16
flaper87alcabrera: s/shard/pool/18:17
flaper87I SWEAR I DIDN'T USE SED18:17
* flaper87 silently walks away18:17
*** kgriffs is now known as kgriffs|afk18:17
flaper87... so does kgriffs|afk18:17
alcabreraflaper87: +218:18
alcabreraI paid particular attention to files that had different +/-18:18
alcabreraI saw you expanded comments in one18:18
alcabreraand removed a stale NOTE on the other18:18
alcabreranicely done. :)18:19
flaper87AWESOME! thanks18:19
malini1flaper87: I went through half of tht patch once :-$18:23
malini1But it was too much for my ADD brain :D18:23
flaper87malini1: FWIW, I think we can trust jenkins here :D18:24
flaper87I moved some files around, replaced all shard matches with pool18:24
flaper87and that's it18:24
malini1yeap..I saw tht18:24
malini1I am going to trust jenkins :)18:24
flaper87there are weird terms like: `pooled` or `pooling` :P18:24
flaper87but you know :D18:24
malini1flaper87: I just approved it18:25
*** tonytan4ever has quit IRC18:25
alcabrerayay!18:25
*** tonytan4ever has joined #openstack-marconi18:26
flaper87holy molly18:26
flaper87THANKS!18:26
malini1the bot patch failed pep8 https://review.openstack.org/#/c/99033/ :D18:26
malini1I dont understand tht18:27
alcabrerasilly bot18:28
flaper87malini1: that's because it's syncing the new hacking version which adds new hacking rules18:28
flaper87(and we're obviously not compliant with those rules)18:29
alcabreraah!18:29
alcabrerathat makes sense18:29
*** tonytan4ever has quit IRC18:30
*** tonytan4ever has joined #openstack-marconi18:33
sriramhaha "I swaer I didnt use sed" :P18:36
sriram*swear18:36
alcabrerathat's what flaper87 sed, sriram. ;)18:38
flaper87LOOOOOOOOOOL18:39
sriramhaha :D18:46
*** tonytan4ever has quit IRC18:47
*** tonytan4ever has joined #openstack-marconi18:48
sriramflaper87,malini1: https://review.openstack.org/#/c/91804/ could you guys have a look, please :)18:51
*** balajiiyer1 has joined #openstack-marconi18:53
*** balajiiyer has quit IRC18:53
*** balajiiyer1 has quit IRC19:06
*** balajiiyer has joined #openstack-marconi19:06
*** tonytan4ever has quit IRC19:06
flaper87sriram: NO19:07
* flaper87 ducks19:07
*** kgriffs|afk is now known as kgriffs19:08
*** tonytan4ever has joined #openstack-marconi19:14
*** AAzza_afk is now known as AAzza19:14
*** kgriffs is now known as kgriffs|afk19:17
flaper87http://blog.flaper87.com/post/53a09586d987d23f49c777bf/19:23
*** kgriffs|afk is now known as kgriffs19:24
*** vkmc has quit IRC19:28
sriramis that a message id flaper87? looks like a mongo object id :P19:29
flaper87sriram: it is ;)19:29
sriramawesome!19:29
*** malini1 has quit IRC19:32
openstackgerritA change was merged to openstack/marconi: Implement Lazy Create Queue in v1.1 API  https://review.openstack.org/9180419:33
flaper87sriram: ^19:33
sriramIts an Indian greeting : _/\_19:33
sriramflaper87: ^19:33
flaper87:D19:34
sriramwoot :D19:34
kgriffsmy new desktop wallpaper: https://www.dropbox.com/s/5i17utc7c49y623/naav.jpg19:53
*** tonytan4ever has quit IRC19:53
* kgriffs gets back to doing real work19:53
srirama flying boat, wow!19:53
*** tonytan4ever has joined #openstack-marconi19:54
*** AAzza is now known as AAzza_afk20:00
kgriffsalcabrera: can I trouble you for a quick review? https://github.com/racker/falcon/pull/271/files20:06
* alcabrera clicks20:07
kgriffsalso this one if you have a few minutes this week. much appreciated!20:08
kgriffshttps://github.com/racker/falcon/pull/268/files20:08
*** tonytan4ever has quit IRC20:09
alcabrera#268 is more than I have energy for atm, but I can handle 271, kgriffs. :)20:14
kgriffskk20:14
kgriffsno worries. just looking for some feedback sometime over the next couple days20:15
*** oz_akan_ has quit IRC20:19
*** mwagner_lap has quit IRC20:19
*** torgomatic has quit IRC20:19
*** russell_h has quit IRC20:19
*** balajiiyer has quit IRC20:19
*** VeggieMeat has quit IRC20:19
*** mkoderer has quit IRC20:19
*** kgriffs has quit IRC20:19
*** megan_w has quit IRC20:19
*** lvh has quit IRC20:19
*** pquerna has quit IRC20:19
*** tjanczuk_ has quit IRC20:19
*** Ephur has quit IRC20:19
*** alcabrera has quit IRC20:19
*** dragondm has quit IRC20:19
*** balajiiyer has joined #openstack-marconi20:20
*** oz_akan_ has joined #openstack-marconi20:20
*** mwagner_lap has joined #openstack-marconi20:20
*** torgomatic has joined #openstack-marconi20:20
*** VeggieMeat has joined #openstack-marconi20:20
*** mkoderer has joined #openstack-marconi20:20
*** kgriffs has joined #openstack-marconi20:20
*** megan_w has joined #openstack-marconi20:20
*** russell_h has joined #openstack-marconi20:20
*** dragondm has joined #openstack-marconi20:20
*** alcabrera has joined #openstack-marconi20:20
*** Ephur has joined #openstack-marconi20:20
*** tjanczuk_ has joined #openstack-marconi20:20
*** pquerna has joined #openstack-marconi20:20
*** lvh has joined #openstack-marconi20:20
*** dickson.freenode.net sets mode: +o kgriffs20:20
*** chandan_kumar has quit IRC20:24
*** Obulpathi has quit IRC20:24
*** Obulpathi has joined #openstack-marconi20:31
*** sriram has quit IRC20:32
*** abettadapur has quit IRC20:33
*** chandan_kumar has joined #openstack-marconi20:34
*** cpallares has quit IRC20:34
*** oz_akan_ has quit IRC20:35
*** Obulpathi is now known as 77CAADSV520:36
*** alcabrera is now known as alcabrera|afk20:36
*** Obulpathi has joined #openstack-marconi20:36
*** Obulpathi has quit IRC20:36
*** oz_akan_ has joined #openstack-marconi20:40
*** tonytan4ever has joined #openstack-marconi20:42
openstackgerritFlavio Percoco proposed a change to openstack/marconi: Rename shards to pool  https://review.openstack.org/9646320:44
*** flaper87 is now known as flaper87|afk20:45
*** vkmc has joined #openstack-marconi20:46
*** vkmc has quit IRC20:46
*** vkmc has joined #openstack-marconi20:46
*** tonytan4ever has quit IRC20:47
*** tonytan4ever has joined #openstack-marconi20:48
*** oz_akan__ has joined #openstack-marconi20:58
*** oz_akan__ has quit IRC20:59
*** oz_akan__ has joined #openstack-marconi20:59
openstackgerritOpenStack Proposal Bot proposed a change to openstack/marconi: Updated from global requirements  https://review.openstack.org/9903321:00
*** oz_akan_ has quit IRC21:02
*** oz_akan__ has quit IRC21:04
openstackgerritOpenStack Proposal Bot proposed a change to openstack/python-marconiclient: Updated from global requirements  https://review.openstack.org/9909321:05
vkmchttps://review.openstack.org/#/c/99033 is failing on pep8 due to comments in files21:07
vkmcthat's odd21:07
vkmcin that case, shouldn't it fail for other patch sets as well?21:08
*** 77CAADSV5 has quit IRC21:18
*** Obulpathi has joined #openstack-marconi21:19
*** tonytan4ever has quit IRC21:20
*** Obulpathi has quit IRC21:25
*** mwagner_lap has quit IRC21:30
*** balajiiyer has quit IRC21:34
kgriffsyeah, seems like all should be failing that were gated since hacking changed21:57
kgriffsI am going to try getting a repro and then submit a patch to fix the violations21:59
vkmck :)22:00
flwangkgriffs: ping22:11
kgriffsoh_my_goodness22:11
kgriffsthat is a LOT of hacking violations22:12
* kgriffs sighs22:12
kgriffsflwang: o/22:12
flwangkgriffs: busy? do you have a few minutes as for bp health?22:13
torgomaticthere's definitely something to be said for opting into hacking checks instead of opting out22:13
kgriffssure, I have some time. I am just working on fixing all the hacking violations from the latest update on the rules22:13
vkmckgriffs, I could submit a patch for it in a moment22:13
vkmc(if you are not already working on that)22:14
kgriffsoh, I just started. have you already done the fixes?22:14
torgomaticFWIW, hacking opt-in looks like https://github.com/openstack/swift/blob/master/tox.ini#L6222:14
kgriffstorgomatic: gtk!22:14
vkmcno no22:14
kgriffsok, if you haven't then I don't mind doing it real quick.22:15
vkmck :) thanks kgriffs22:15
* kgriffs flexes his typing muscles22:16
flwangkgriffs: seems I shouldn't disturb you now :)22:18
kgriffsflwang: nah, go ahead. what's up?22:19
flwangkgriffs: thanks, just wanna discuss what the metrics we need to implement against detailed health22:20
flwangkgriffs: below is the list you commented22:20
flwang1. Total number of messages stored for all queues in a given backend partition.22:20
flwang2. average transactions/sec or per-request latency over the last few  minutes (give numbers for each window: 30 seconds, 1 minute, and 5  minutes).22:20
flwang3. Verify full functionality (perform a test that excercises all  endpoints: post a message, list it, claim it, delete it). I guess  operators do this in their own monitoring tools but they would like if  we could do it for them instead.22:20
flwangkgriffs: as for the #1, it's straightforward22:22
kgriffsHow about doing 1 and 3, and putting of #222:22
flwangbut I'm not sure if it's easy to get based on current db schema, but it makes sense and feasible22:23
kgriffsfwiw, I just noticed the word "partition" in #122:23
flwangkgriffs: haha, that's what I want to suggest22:23
kgriffsshould be "pool"22:23
flwangkgriffs: yep, i see22:23
kgriffsflwang: it might also be interesting to see total number of queues per pool22:24
flwangas for #3, that's the one consumer are really interested in, IMHO22:24
flwangkgriffs: given we're moving to no queue22:25
kgriffsflwang: it would be cool to give the exact operation that failed if the functional test doesn't complete successful22:25
flwangkgriffs: so yes, we just need to care about the num of msg from different views22:25
flwangkgriffs: good point22:26
kgriffsoh, right. well, for v1.x we retain the notion of queues22:26
kgriffshmmm22:26
kgriffsI just thought it would be good to see the distribution of queues across the pools22:26
flwangkgriffs: that's what the 'health' mean I think22:26
kgriffsand if pools aren't enabled, just pretend there's only "one" pool22:26
flwangkgriffs: so? we should still consider about the 'queue' for #1 ATM?22:27
kgriffshmmm22:27
kgriffsMaybe let's split this bp22:28
kgriffslet's put #1 and #3 in j-2 if possible22:28
kgriffsand let's define the v1.1 api to have a flexible schema about what /health may actually return.22:28
kgriffsI think as long as you *add* fields, it is ok. you just don't want to remove or rename fields already in the document22:28
kgriffsso,22:29
kgriffsthen we have "low priority" j-3 bp22:29
kgriffsthat contains stuff like #2 and the new one, queues per pool22:29
flwangkgriffs: +122:29
flwangkgriffs:  that means for j-2, we should focus on msg against pool, is it?22:30
kgriffswe then have something to give people to try and they can come back and help us decide what else is useful22:30
kgriffsyeah, just total messages per pool22:30
flwangkgriffs: yes22:30
flwangkgriffs: awesome22:30
kgriffsoh, and also the deep health test thing (#2)22:30
kgriffsthat is definitely important - got a direct ask for that at the ATL summit22:30
* kgriffs wonders how much style fixes he can sneak into this "hacking" patch without anyone realizing they aren't actual hacking rules.22:31
flwangkgriffs: how to define the ^transaction^?22:31
flwanghow to define the 'request'?22:31
kgriffshmm? You mean the deep health check (#3)?22:32
flwang#222:32
kgriffsoh22:32
flwang2. average transactions/sec or per-request latency over the last few  minutes (give numbers for each window: 30 seconds, 1 minute, and 5  minutes).22:32
kgriffswell, I'm trying to decide on this one... this is sort of a new-relic type thing.22:33
kgriffsit may be best left to 3rd-party tools22:33
kgriffsbut if we wanted to track it, I would expect to see22:33
kgriffsfor each operation on each resource, what is the average rate (req/sec)22:34
kgriffslike "post message is x/sec" and "claim message is y/sec"22:34
kgriffsand "list messages is z/sec"22:35
flwangso it's most like for each kind of API calling, we'd better have a statistic, right?22:35
kgriffsyeah, we'd have to used statsd or something22:35
flwangfrom this perspective, I can't see the difference from #2 and #3, am i confused?22:37
kgriffsactually, I think this work should probably be made separate from health22:37
kgriffswe should have a performance stats bp22:37
flwangtrue22:37
kgriffswe could give an operator the ability to turn on timers at different points of the code22:38
flwangIIRC, seems there is a stats bp, is it?22:38
kgriffsit would show off what parts of the stack are slow22:38
kgriffsflwang: possible. would you mind checking and/or creating one?22:38
flwangkgriffs: no problem22:38
kgriffscool22:38
* kgriffs thinks that would be fun to work on22:38
* kgriffs wished he had more time to write fun code22:38
flwangkgriffs: it's really aligning with our current benchmarking work22:39
flwangkgriffs: speed, speed, speed and stable, stable, more stable22:39
kgriffstrue22:40
flwangkgriffs: so, seems we just figured out the #1, as for #2 and #3, seems we're mixing them22:40
kgriffshow are we mixing them?22:41
flwangkgriffs: maybe I'm confused22:41
kgriffsheh22:41
kgriffsok22:41
flwangkgriffs: so for #3, we're planing to move it to a stats bp, right?22:41
kgriffsno22:42
kgriffshmm22:42
flwang(10:37:41) kgriffs: actually, I think this work should probably be made separate from health22:42
flwang(10:37:48) kgriffs: we should have a performance stats bp22:42
kgriffsoh, that was #222:42
openstackgerritA change was merged to openstack/marconi: Fix tests to be Python 3 compatible  https://review.openstack.org/10008722:42
kgriffsso, #1 we want in /health22:42
kgriffsand we also want #3, which is the deep health check in /health22:43
kgriffsthen we move #2, which is really performance monitoring, into the stats blueprint22:43
kgriffsdoes that make sense, or did I just make things more confusing? :p22:44
flwangok, so as for #3, we're more caring about if the end to end function work no not, right?22:44
kgriffsthat is right. We just try out every operation and make sure it succeeds.22:44
flwangkgriffs: we're making it more clear, I think :)22:44
kgriffsdon't need to time them or anything; it's just pass/fail22:45
flwangkgriffs: cool22:45
flwangkgriffs: very clear now, thanks guru22:45
flwanghttps://blueprints.launchpad.net/marconi/+spec/ops-stats22:46
flwanga superman created the stats bp22:46
flwangbefore 1 year ago, what a proactive view :D22:46
kgriffslol. I dream big.22:48
flwangbtw, are we following the bp specs workflow atm?22:49
kgriffsflwang: we are going to try it with new feature proposals. Flavio is going to do the plumbing to make it work this week.22:50
flwangkgriffs: fantastic, just make sure if i should updated the bp content based on current flow22:51
flwangkgriffs: thank you very much for your valuable comments on this bp22:52
kgriffsno problem. I really appreciate your help on the implementation.22:52
flwangkgriffs: dive into coding now :)22:53
flwangcheers22:53
kgriffslet's do this22:53
kgriffs:)22:53
kgriffsjust for the record, I'm pretty sure all of these hacking violations are malini's fault22:58
* kgriffs hides22:58
vkmckgriffs, if you have a moment, I would need your opinion with the AMQP transport driver23:06
vkmcs/need/appreciate23:07
vkmcwell, both23:07
kgriffssure, what's up?23:07
vkmcshould we add it both for v1 and v1.1?23:09
vkmcIMO it worths the effort...23:10
kgriffshmm23:10
kgriffsif I am creating an app and wanting to use AMQP transport, why wouldn't I just want to use the more polished v1.1 API?23:11
kgriffssorry, that is a leading question23:11
kgriffs:p23:11
vkmchaha23:11
vkmcwell... but what happens if you are already deployed with v123:11
kgriffsah, I see23:12
kgriffsthe question is: do we require the operator upgrade to Juno release for the AMQP transport to work?23:12
kgriffspersonally, I would say yes... the more people running the latest code, the better, and this would encourage people to upgrade23:13
kgriffswhat do you think?23:13
vkmcyes that's a fair pov23:14
vkmcand I agree that it would be something that may motivate users to upgrade23:15
vkmcbut since I'm not familiar of how things are done in production23:16
kgriffsah23:16
kgriffswell, I guess I would be more worried about it if we were integrated and had lots of deployments already.23:17
vkmcI don't know if it's usual to upgrade or if it's done as a last resort23:17
vkmcyep23:17
kgriffsvkmc: depends on how painful it is to upgrade23:17
vkmcI see23:18
vkmcwell, I will start with v1.123:18
kgriffsif we make it really easy, AND we make sure each release is high-quality, deployers trust the project and will upgrade more frequently23:18
vkmctrue that :)23:18
kgriffsvkmc: sounds good. We can always backport during K if people ask for it23:18
vkmcgreat!23:19
vkmckgriffs, thanks :D23:19
kgriffshey, no problem. That was a great question and something we need to keep in mind, especially as we get more deployments23:20
kgriffsman, these sqla statements are killing me23:21
kgriffsPython was *not* designed to write SQL queries23:21
kgriffs /marconi/queues/storage/sqlalchemy/messages.py23:22
vkmcI'm not a big fan of SQL >.<23:22
vkmcnope nope nope23:22
kgriffsMaybe LINQ wasn't such a bad idea after all. :p23:22
kgriffsvkmc: check out this monstrosity: http://goo.gl/9YYqNd23:24
vkmclol23:24
kgriffsuuuugly23:25
vkmctwo selects with counts inside a select23:25
vkmcgo nuts23:25
kgriffswoohoo!23:25
kgriffsmuch fun23:25
vkmcyesterday I had to review SQL calls in PHP23:28
vkmcthis is super neat in comparison to that23:28
kgriffsoh yeah?23:28
kgriffsPHP: The Language That Isn't Dead Yet23:29
vkmcyeah... code generated with CakePHP23:29
kgriffsI blame facebook23:29
vkmcugliest thing ever23:29
kgriffs(bless it's heart)23:29
kgriffss/it's/its23:30
kgriffsvkmc: in the south, you can get away with saying just about anything if you follow it up with "bless his/her heart"23:30
vkmcoh :) good to know23:31
vkmcI blame google too23:32
vkmcmost SDKs provided are PHP, Javascript23:32
kgriffsheh23:32
kgriffsanyway, that's what I learned from Henry Cho. Great comedian23:32
vkmcoh I don't know him :)23:33
kgriffsprobably find him on youtube23:33
* vkmc searchs23:33
vkmcwell I have to stop watching haha23:42
vkmche made me remember Seinfeld23:43
kgriffsheh23:45
kgriffsgood stuff23:45

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