Friday, 2016-02-26

« Thursday, 2016-02-25 Index Saturday, 2016-02-27 »
*** thorst_afk has quit IRC00:01
*** hughhalf has quit IRC00:02
*** chlong_ has quit IRC00:03
*** ddieterly has joined #openstack-telemetry00:54
*** rbak_ has quit IRC00:57
*** ddieterl_ has joined #openstack-telemetry00:58
*** ddieterly has quit IRC00:58
*** thorst_afk has joined #openstack-telemetry00:59
*** thorst_afk has quit IRC01:05
*** thorst_afk has joined #openstack-telemetry01:05
*** thorst_afk has quit IRC01:06
*** thorst_afk has joined #openstack-telemetry01:07
*** thorst_afk has quit IRC01:12
*** ddieterly has joined #openstack-telemetry01:16
*** ddieterl_ has quit IRC01:19
*** rbak has joined #openstack-telemetry01:20
openstackgerritMerged openstack/ceilometer: Enable the Load Balancer v2 for the Ceilometer(Part Two)  https://review.openstack.org/27743401:22
openstackgerritliusheng proposed openstack/python-aodhclient: Add composite rule alarm support in aodhclient  https://review.openstack.org/28402201:39
*** rbak has quit IRC01:47
*** cheneydc has joined #openstack-telemetry01:48
*** chlong_ has joined #openstack-telemetry01:49
*** ljxiash has joined #openstack-telemetry01:49
*** annasort has joined #openstack-telemetry01:56
*** rbak has joined #openstack-telemetry01:56
*** thorst_afk has joined #openstack-telemetry01:57
*** rbak has quit IRC02:01
*** rbak has joined #openstack-telemetry02:13
*** rbak has quit IRC02:15
*** ljxiash has quit IRC02:16
*** ljxiash has joined #openstack-telemetry02:18
*** jwcroppe has joined #openstack-telemetry02:31
*** achatterjee has joined #openstack-telemetry02:45
*** rbak has joined #openstack-telemetry03:00
*** rbak has quit IRC03:08
*** thorst_afk has quit IRC03:09
*** thorst_afk has joined #openstack-telemetry03:09
*** rbak has joined #openstack-telemetry03:11
*** rbak has quit IRC03:11
*** thorst_afk has quit IRC03:18
*** boris-42 has quit IRC03:24
*** ljxiash_ has joined #openstack-telemetry03:25
*** ljxiash_ has quit IRC03:26
*** links has joined #openstack-telemetry03:26
*** ljxiash_ has joined #openstack-telemetry03:26
*** ljxiash has quit IRC03:27
*** ddieterly has quit IRC03:30
*** ljxiash_ has quit IRC03:31
*** jwcroppe has quit IRC03:37
*** jwcroppe has joined #openstack-telemetry03:37
*** jwcroppe has quit IRC03:42
*** thorst_afk has joined #openstack-telemetry04:15
*** thorst_afk has quit IRC04:23
*** jwcroppe has joined #openstack-telemetry04:39
openstackgerritRyota MIBU proposed openstack/ceilometer: tempest: use ceilometer client manager instead of tempest one  https://review.openstack.org/25571104:48
openstackgerritRyota MIBU proposed openstack/ceilometer: WIP: tempest: migrate api/test_check_nova_notification  https://review.openstack.org/26805404:48
*** yprokule has joined #openstack-telemetry04:59
*** ddieterly has joined #openstack-telemetry05:01
*** ddieterly has quit IRC05:05
*** jwcroppe has quit IRC05:12
*** ljxiash has joined #openstack-telemetry05:13
*** ljxiash has quit IRC05:17
*** ljxiash has joined #openstack-telemetry05:17
*** thorst_afk has joined #openstack-telemetry05:20
*** ljxiash has quit IRC05:22
*** thorst_afk has quit IRC05:27
*** ljxiash has joined #openstack-telemetry05:41
openstackgerritRyota MIBU proposed openstack/aodh: fix: option name duplication  https://review.openstack.org/28511705:42
openstackgerritRyota MIBU proposed openstack/aodh: fix: option name duplication  https://review.openstack.org/28511705:43
*** cheneydc has quit IRC05:54
openstackgerritliusheng proposed openstack/python-aodhclient: Use cli.base.execute of tempest-lib  https://review.openstack.org/28511805:55
*** ljxiash has quit IRC05:57
*** ljxiash has joined #openstack-telemetry05:58
openstackgerritliusheng proposed openstack/python-aodhclient: Add composite rule alarm support in aodhclient  https://review.openstack.org/28402205:58
*** ljxiash_ has joined #openstack-telemetry05:59
*** ljxiash__ has joined #openstack-telemetry06:01
*** ljxiash has quit IRC06:02
*** ddieterly has joined #openstack-telemetry06:02
*** ljxiash_ has quit IRC06:05
*** ddieterly has quit IRC06:07
*** jwcroppe has joined #openstack-telemetry06:19
*** thorst_afk has joined #openstack-telemetry06:25
*** links has quit IRC06:26
*** ljxiash__ has quit IRC06:27
openstackgerritliusheng proposed openstack/python-aodhclient: Use cli.base.execute of tempest-lib  https://review.openstack.org/28511806:32
*** thorst_afk has quit IRC06:32
*** ljxiash has joined #openstack-telemetry06:33
*** ljxiash has quit IRC06:34
*** ljxiash has joined #openstack-telemetry06:34
*** ljxiash has quit IRC06:39
*** lsmola has joined #openstack-telemetry06:47
*** ljxiash has joined #openstack-telemetry06:50
*** jwcroppe has quit IRC06:52
*** _nadya_ has joined #openstack-telemetry07:01
*** ljxiash has quit IRC07:01
*** ljxiash has joined #openstack-telemetry07:02
*** ddieterly has joined #openstack-telemetry07:03
*** rcernin has joined #openstack-telemetry07:05
*** ddieterly has quit IRC07:08
*** Qiming has joined #openstack-telemetry07:13
*** ljxiash has quit IRC07:20
*** chlong_ has quit IRC07:26
*** thorst_afk has joined #openstack-telemetry07:30
*** ljxiash has joined #openstack-telemetry07:32
*** thorst_afk has quit IRC07:39
*** ildikov has joined #openstack-telemetry07:46
*** jwcroppe has joined #openstack-telemetry07:49
*** hparekh has quit IRC07:55
*** cheneydc has joined #openstack-telemetry07:56
*** ddieterly has joined #openstack-telemetry08:04
*** belmoreira has joined #openstack-telemetry08:08
*** ddieterly has quit IRC08:09
openstackgerritliusheng proposed openstack/aodh: Fix py34 error of indexing 'dict_keys' object  https://review.openstack.org/28515308:11
*** ljxiash has quit IRC08:16
*** jwcroppe has quit IRC08:22
*** larainema has quit IRC08:25
*** larainema has joined #openstack-telemetry08:25
*** ljxiash has joined #openstack-telemetry08:31
*** thorst_afk has joined #openstack-telemetry08:36
*** thorst_afk has quit IRC08:43
*** Liuqing has joined #openstack-telemetry09:00
*** ddieterly has joined #openstack-telemetry09:05
*** ddieterly has quit IRC09:11
*** Liuqing has quit IRC09:13
*** rcernin has quit IRC09:14
*** jwcroppe has joined #openstack-telemetry09:19
*** yassine__ has joined #openstack-telemetry09:27
*** rcernin has joined #openstack-telemetry09:28
*** rpodolyaka_ has joined #openstack-telemetry09:33
*** jwcroppe_ has joined #openstack-telemetry09:34
*** rpodolyaka has quit IRC09:35
*** jwcroppe has quit IRC09:35
*** rpodolyaka_ is now known as rpodolyaka09:35
*** thorst_afk has joined #openstack-telemetry09:41
*** thorst_afk has quit IRC09:48
*** jwcroppe_ has quit IRC09:53
*** cheneydc has quit IRC10:01
*** ddieterly has joined #openstack-telemetry10:07
*** r-mibu has quit IRC10:07
*** r-mibu has joined #openstack-telemetry10:08
*** ddieterly has quit IRC10:12
*** achatterjee has quit IRC10:13
*** ljxiash has quit IRC10:14
*** Qiming has quit IRC10:16
openstackgerritZi Lian Ji proposed openstack/ceilometer: Ceilometer: fix MeterDefinitionException for py3  https://review.openstack.org/28519610:16
*** ljxiash has joined #openstack-telemetry10:23
*** thorst_afk has joined #openstack-telemetry10:45
openstackgerritXia Linjuan proposed openstack/ceilometer: Enable the Load Balancer v2 for notification  https://review.openstack.org/28177710:49
*** jwcroppe has joined #openstack-telemetry10:50
*** thorst_afk has quit IRC10:53
*** ljxiash has quit IRC10:57
*** ddieterly has joined #openstack-telemetry11:08
*** ddieterly has quit IRC11:12
*** Qiming has joined #openstack-telemetry11:22
*** jwcroppe has quit IRC11:23
*** sanjana_ has quit IRC11:29
*** cdent has joined #openstack-telemetry11:44
openstackgerritIlya Tyaptin proposed openstack/ceilometer: Add an update interval to compute discovery  https://review.openstack.org/28432211:44
*** thorst_afk has joined #openstack-telemetry11:50
*** thorst_afk has quit IRC11:57
* sileht reads the nova case insensitive ML threads, then takes mysql and throws it away through the window12:00
cdentsileht++12:02
*** ddieterly has joined #openstack-telemetry12:09
*** krotscheck_dcm is now known as krotscheck12:13
*** ddieterly has quit IRC12:13
*** jwcroppe has joined #openstack-telemetry12:20
*** samstudio8 has quit IRC12:25
*** samstudio8 has joined #openstack-telemetry12:25
*** julim has joined #openstack-telemetry12:45
*** thorst_afk has joined #openstack-telemetry12:50
*** jwcroppe has quit IRC12:53
*** gordc has joined #openstack-telemetry13:00
*** ddieterly has joined #openstack-telemetry13:09
*** ddieterly has quit IRC13:14
*** thorst_afk is now known as thorst13:18
*** pradk has joined #openstack-telemetry13:31
*** pradk has quit IRC13:31
*** nicodemus_ has joined #openstack-telemetry13:37
*** pradk has joined #openstack-telemetry13:43
*** AndreiaKumpera has joined #openstack-telemetry13:48
*** jwcroppe has joined #openstack-telemetry13:50
*** ljxiash has joined #openstack-telemetry13:57
*** jwcroppe has quit IRC13:58
*** ljxiash has quit IRC14:01
openstackgerritMerged openstack/python-aodhclient: Make the alarm list output more concise  https://review.openstack.org/28391014:10
*** ddieterly has joined #openstack-telemetry14:10
ityaptinFolks, do anyone catch failings of gate-ceilometer-dsvm-functional-[mongodb, mysql, pastgresql] jobs  for the fresh patches?14:14
*** ddieterly has quit IRC14:15
*** therve has left #openstack-telemetry14:16
gordcityaptin: not sure why, seems like sql gates are timing out14:18
openstackgerritMehdi Abaakouk (sileht) proposed openstack/gnocchi: ceph: fix help string  https://review.openstack.org/28532414:23
ityaptinMongoDB gate fails with response status on the this test: https://github.com/openstack/ceilometer/blob/master/ceilometer/tests/functional/gabbi/gabbits/clean-samples.yaml#L4114:23
openstackgerritMerged openstack/aodh: Add releasenote for composite alarm feature  https://review.openstack.org/28450514:24
gordcsileht: these notifications tests take a very long time https://github.com/openstack/ceilometer/blob/master/ceilometer/tests/functional/test_notification.py#L329 after we dropped eventlet. is there anyway to force progress like we did with eventlet.sleep?14:24
silehtgordc, time.sleep(0) ?14:24
ityaptincdent: Hi!14:25
gordcsileht: is that a question?lol i thought about that, but i guess that becuase it looks the same14:25
cdenthi ityaptin14:25
gordcsileht: i'll give it a try locally.14:25
ityaptincdent: Is gabbi expects an status:200 in this test by default?   https://github.com/openstack/ceilometer/blob/master/ceilometer/tests/functional/gabbi/gabbits/clean-samples.yaml#L4114:25
silehtgordc, before we was using eventlet.sleep() to schedule another greenlet14:26
*** rbak has joined #openstack-telemetry14:26
silehtgordc, now we use threads, we can use time.sleep() to schedule other thread14:26
cdentityaptin: yeah, the default for status is 20014:26
gordcsileht: oh cool. ok. let's do that.14:26
silehtgordc, with eventlet you have to use tricks to go back to the main greenlet, with cpython you have to use tricks to release GIL14:27
ityaptincdent: but it's a post request and it returns an 201 status, doesn't it?14:27
ityaptincdent: http://paste.openstack.org/show/r9iXZjpObrFE8JHGutWH/14:28
ityaptinIn same time it doesn't affect gate failing, because I caught this trace in my devstack and all test passed nevertheless14:29
cdentityaptin: look at lines 42 and 43, because it is marked xfail it won't catch that problem14:30
cdentthe xfail needs to be removed and the test updated14:31
cdentor the xfail needs to stay there because the bug is still present14:31
cdentdepends on what the truth is14:31
ityaptincdent: ack, thanks)14:33
ityaptinso, how I understand,  MongoDB gate fails with an timeout too (14:34
*** ljxiash has joined #openstack-telemetry14:37
gordcityaptin: probably just gate exploding because everyone is merging now14:40
ityaptingordc: You are right, I think14:41
*** ddieterly has joined #openstack-telemetry14:49
*** jwcroppe has joined #openstack-telemetry14:54
*** ddieterly has quit IRC14:59
*** ddieterly has joined #openstack-telemetry15:00
gordcityaptin: seems like it's hanging after it finishes all the tests for some reason.15:01
gordcif you look at logs for good check, it takes same amount of time to run all tests,15:01
gordchttp://logs.openstack.org/14/285014/1/check/gate-ceilometer-dsvm-functional-mongodb/f75ab3c/console.html#_2016-02-26_11_13_23_98715:01
gordcnot sure why there's 1.5 hrs between last test and next action15:02
*** rickyrem has joined #openstack-telemetry15:05
*** Liuqing has joined #openstack-telemetry15:07
*** ddieterl_ has joined #openstack-telemetry15:08
ityaptinI found out same exit points for the several patch: http://logs.openstack.org/33/284433/1/check/gate-ceilometer-dsvm-functional-mongodb/52ed212/console.html#_2016-02-26_11_49_21_15815:09
*** ddieterl_ is now known as ddieterly_15:09
ityaptinhttp://logs.openstack.org/33/284433/1/check/gate-ceilometer-dsvm-functional-mongodb/52ed212/console.html#_2016-02-26_11_49_21_15815:10
*** ddieterly has quit IRC15:10
*** jwcroppe has quit IRC15:11
*** liamji has joined #openstack-telemetry15:11
gordcyeah i'm not sure what's happening there15:13
*** Liuqing has quit IRC15:27
*** belmoreira has quit IRC15:43
*** ddieterly_ is now known as ddieterly15:49
rickyremI have Ceilometer writing samples MongoDB. I have a YUM repo being stored in Swift and the small sized requests aren't showing up in MongoDB. Could anyone explain why that would be the case?15:53
rickyremI have debug logging on for Ceilometer and I see that the collector is pulling out of the metering.sample queue in RabbitMQ.15:53
rickyremThe 50 byte requests show up in the logs, but again not in MongoDB.15:54
gordcwhich logs?15:54
rickyremceilometer/collector.log15:55
gordcwhich exact meter are you trying to resolve?15:55
rickyremoh yeah i didn't say that: storage.objects,outgoing.bytes15:55
gordcrickyrem: do you see storage.api.request meter?15:57
gordcthat meter comes from ceilometermiddleware. did you enable it in swift/proxy-server.conf?15:58
rickyremgordc: Yes, I see in the collector.log15:58
rickyremThe the ceilometermiddleware is enabled.15:59
rickyremI created a duplicate queue in RabbitMQ to see what was going to Ceilometer. I can verify that every API request to Swift is turning into a message in RabbitMQ and being consumed by Ceilometer.16:00
gordcweird, you see that but not storage.objects.outgoing.bytes16:00
rickyremSorry, i meant I see storage.objects.outgoing.bytes in the log of the collector, but not every entry makes it into mongodb.16:01
gordcdo you see any measurements in those message payloads?16:01
gordcoh16:01
gordcand you don't get any errors?16:01
gordcin collector.log?16:01
rickyremmetering data storage.objects.outgoing.bytes for fde0a5f… @ 2016-02-26 15:34:50.494699: 46 record_metering_data /usr/lib/python2.7/site-packages/ceilometer/dispatcher/database.py:8216:01
rickyremNo errors, lines just like that one ^16:02
gordcso only some of them are written to mongodb or none of them?16:02
rickyremSome of them.16:03
gordchow do you know only some of them are? the count wrong when you run statistics?16:04
rickyremPretty much no requests of small size (under a kilobyte) make it to MongoDB. A pretty good amount of the requests for a 500MB file are written to MongoDB.16:04
rickyremI have been running statistics to check, and it showed what I stated as being the case. I also manually queried MongoDB to double check.16:06
rickyremThe counter_volume for the meter is correct, when it does show up.16:07
*** Qiming has quit IRC16:09
gordcthe data is being stored as bytes right?16:10
rickyremCorrect16:10
gordcweird, i don't really see why it would store 500 000 000 and not 50016:11
gordcif you see the 'metering data ...' log, there is no filtering done after that point to storage (that would raise a warning/error)16:14
gordcthe only filtering after that is TTL.16:15
rickyremWe have HA controllers so there are 3 nodes. One one of them we sometimes get a log entry like this: oslo_messaging._drivers.amqpdriver [-] received message msg_id: None reply to None __call__ /usr/lib/python2.7/site-packages/oslo_messaging/_drivers/amqpdriver.py:19616:16
gordcsileht: ^ that's just normal logging right?16:17
rickyremoh DEBUG level16:17
silehtgordc, for notification yes16:17
silehtgordc, for rpc.cast too I think16:17
rickyremhmm16:17
*** rcernin has quit IRC16:31
*** thorst is now known as thorst_afk16:33
rickyremgordc: Thank you for your help16:34
rickyremBut i'm starting to think it's MongoDB that is the problem16:34
gordcrickyrem: np. sorry cna't really debug it well without visible errors16:35
rickyremI think that's why there aren't any errors.16:35
rickyremI'm pretty sure MongoDB is getting the insert call and not doing it.16:35
gordcrickyrem: you can always try another backend, ie gnocchi, or writing to a custom tsdb16:35
gordcrickyrem: tbh, it seems like people are moving away from mongodb16:36
rickyremgordc: Sounds about right16:36
rickyremgordc: is there a way we can have Ceilometer duplicate what it does to MongoDB to Gnocchi.16:37
rickyrem?16:37
gordcrickyrem: if you are using mitaka, you can set meter_dispatchers option to write to both: 'database' and 'gnocchi'16:38
rickyremgordc: thanks, unfortunately i'm on liberty16:38
gordcif you are using liberty, you'll need to change pipeline.yaml so the sinks publisher to notifier:// and notifier://?topic=<custom>16:39
gordcand then have one collector as it is now which will listen to default queue16:39
gordcand have another collector listen to <custom> topic queue and be configured with dispatcher='gnocchi' in conf16:40
gordcyou'll may have some write performance issues in liberty since the gnocchi dispatcher wasn't really optimised until mitaka16:40
gordcyou might want to try with gnocchi2.0 to minimise migrations (if you plan on keeping data)16:41
rickyremgordc: thanks for the detailed answer16:43
rickyremI'm going to see if I can apply a short term fix to MongoDB to get things back to normal.16:43
rickyremMy long term goal would be to swap to Gnocchi though.16:44
gordcrickyrem: sounds good. if you do figure it out, feel free to patch it upstgream as well16:44
gordcupstream*16:44
rickyremgordc: Absolutely16:45
pradkgordc, can you approve this backport https://review.openstack.org/#/c/284135/16:45
gordcpradk: done. let's see if it passes... then maybe there is something up with master gate16:47
pradkgordc, thx, ah ok16:48
*** rickyrem has quit IRC16:51
*** yprokule has quit IRC16:56
openstackgerritIlya Tyaptin proposed openstack/ceilometer: Cache getters for the decalarative definitions  https://review.openstack.org/28543116:57
*** rickyrem has joined #openstack-telemetry16:59
openstackgerritgordon chung proposed openstack/ceilometer: test  https://review.openstack.org/28543717:04
openstackgerritIlya Tyaptin proposed openstack/ceilometer: Cache getters for the decalarative definitions  https://review.openstack.org/28543117:07
*** rbak_ has joined #openstack-telemetry17:11
*** elemoine- has joined #openstack-telemetry17:13
*** ljxiash_ has joined #openstack-telemetry17:13
*** j_king_ has joined #openstack-telemetry17:15
*** gzhai1 has joined #openstack-telemetry17:15
*** tonyb_ has joined #openstack-telemetry17:16
*** kfox1111_ has joined #openstack-telemetry17:16
*** r-mibu_ has joined #openstack-telemetry17:18
*** ljxiash has quit IRC17:20
*** rbak has quit IRC17:20
*** r-mibu has quit IRC17:20
*** yassine__ has quit IRC17:20
*** lsmola has quit IRC17:20
*** ekarlso- has quit IRC17:20
*** krotscheck has quit IRC17:20
*** j_king has quit IRC17:20
*** tonyb has quit IRC17:20
*** gzhai has quit IRC17:20
*** afazekas has quit IRC17:20
*** smurke has quit IRC17:20
*** bapalm has quit IRC17:20
*** elemoine has quit IRC17:20
*** kfox1111 has quit IRC17:20
*** krotscheck has joined #openstack-telemetry17:20
*** afazekas has joined #openstack-telemetry17:21
*** bapalm has joined #openstack-telemetry17:21
ityaptingordc: I tried to reduce time to the ceilometer.tests.functional.test_notification. For my devstack it's a old: Sum of execute time for each test: 367.8496 sec, new: new: Sum of execute time for each test: 15.1279 sec.17:22
gordcityaptin: how did you reduce it?17:25
nicodemus_hello gordc17:25
gordci don't think it's the test runtime that's failing gate but if it helps that's good regardless17:25
gordcnicodemus_: hey, whatsup?17:25
ityaptingordc: I reduce count of running ExtentedJsonPathParser.parse(fields).find in declarative.Definitions17:25
ityaptingordc: I'll start to cache all repeated fields getter in the class field.17:26
*** yassine__ has joined #openstack-telemetry17:27
*** ekarlso- has joined #openstack-telemetry17:27
nicodemus_gordc: trying to backport gnocchi's groupby in stable/1.3... were you able to look into bug #1548909 ?17:28
openstackbug 1548909 in Gnocchi "gnocchi-metricd cannot process new metrics on stable/2.0" [Undecided,Confirmed] https://launchpad.net/bugs/154890917:28
gordcnicodemus_: no sorry, our gate started acting up.17:28
ityaptingordc: it's also reduce time to start notification agent, because it irritates me very well.17:28
gordctook a quick look... but couldn't figure out exactly who/what was passing in the wrong vale17:28
gordcvalue*17:28
*** lsmola has joined #openstack-telemetry17:29
nicodemus_gordc: no problem. Do you think it could be worth to try 2.0 with a new, clean ceph pool ?17:31
gordcityaptin: hmm.. it seems to make sense? it does take a long time to build definitions, i agree17:31
gordcnicodemus_: yeah, that could work. do you have the broken ceph pool as well? would be nice to figure out what's wrong with upgrade17:32
nicodemus_yes, the broken pool is still there17:33
nicodemus_I have three API hosts now, one with the broken pool, one with 1.3 and a third one for 2.0 w/ the new pool17:34
gordcnicodemus_: nice. i guess we have all our scenarios covered17:34
ityaptingordc: For me it makes sense. According to the profiling files it's difference between 17.5 seconds without it and 4.7 with it for 4 workers.17:35
ityaptinstarting of the ceilometer-agent-notification17:35
*** smurke has joined #openstack-telemetry17:41
gordcityaptin: i think it looks safe. we can merge it during rc period or right before m-3 if needed.17:42
gordcbiab. lunchtime17:42
ityaptingordc: Let's wait a Jenkins results. If it help we can merge it before m-317:43
*** _nadya_ has quit IRC17:44
*** liamji has quit IRC17:52
*** thorst_afk is now known as thorst_18:23
*** agireud has quit IRC18:55
*** agireud has joined #openstack-telemetry18:57
*** jwcroppe has joined #openstack-telemetry19:11
*** gzhai2 has joined #openstack-telemetry19:20
*** datravis has joined #openstack-telemetry19:21
*** gzhai1 has quit IRC19:23
*** rickyrem has quit IRC19:29
*** ddieterly has quit IRC19:37
*** rickyrem has joined #openstack-telemetry19:41
*** ddieterly has joined #openstack-telemetry19:41
*** _nadya_ has joined #openstack-telemetry19:47
*** rickyrem has quit IRC19:47
*** gordc has quit IRC19:51
*** lsmola has quit IRC19:53
*** jwcroppe has quit IRC19:54
*** safchain has joined #openstack-telemetry20:10
*** alejandrito has joined #openstack-telemetry20:20
nicodemus_gordc, I'm having a new issue with stable/2.0: http://paste.openstack.org/show/488416/20:21
nicodemus_this is a clean deploy with no data, new database, etc. gnocchi-upgrade run correctly without errors20:21
nicodemus_but when ceilometer is trying to create each resource, gnocchi-api throws that db error20:22
*** yarkot_ has joined #openstack-telemetry20:22
*** gordc has joined #openstack-telemetry20:23
*** yarkot_ has quit IRC20:26
*** cdent has quit IRC20:34
alejandritogordc, nicodemus_ can you help us ?20:35
gordcalejandrito: possibly?20:36
alejandritogordc, the weird thing is that is a fresh install20:36
gordcalejandrito: was there a message before "the weird thing is that is a fresh install"?20:38
alejandritogordc, nope, when we upgraded from 1.3 to 2.0, there was no such error20:39
gordcalejandrito: ... yeah i think i missed a message. all i got was 'can you help us' and 'fresh install'... i assume there's a message where you say what the issue is?20:40
alejandritogordc, sure ... no problem, the error with gnocchi 2.0 fresh install is this20:41
alejandritogordc,  http://paste.openstack.org/show/488416/20:41
nicodemus_that error was with pymysql 0.7.2, we've just downgraded to pymysql 0.6.6 and there's a different error: http://paste.openstack.org/show/488426/20:42
alejandritogordc, that's what i meant with ... that error didnt happen when we upgraded from 1.3 to 2.020:44
gordci'll need to check. i don't actually use mysql20:44
alejandritogordc, thats because we are using mysql+pymysql20:45
alejandritogordc, on the connection string20:46
gordcthat's correct. i'm just saying i dont use mysql usually... i use postgresql for most of testing20:46
alejandritogordc, we took the +pymysql from the connection string and worked20:47
*** _nadya_ has quit IRC20:49
gordcit shouldn't matter. they both resolve to same code20:49
gordchttps://github.com/openstack/gnocchi/blob/master/setup.cfg#L125-L12620:49
gordcwhat version of oslo.db are you using. i'm not even sure what line it's failing in gnocchi20:51
*** annasort has quit IRC20:54
*** dave-mccowan has quit IRC20:56
nicodemus_gordc: gnocchi stable/2.0 is broken.20:56
nicodemus_I have just deployed a new gnocchi, with a new database, a new pool, etc20:57
nicodemus_no upgrade whatsoever20:57
nicodemus_and metricd can't process a single measure, with the overflowerror20:58
gordcsomething seems off. we are gating on master ~= 2.021:05
alejandritogordc, good ... there has to be something weird there21:07
*** rbak__ has joined #openstack-telemetry21:08
gordcwhat oslo.db are you using?21:08
*** rbak_ has quit IRC21:10
nicodemus_gordc: oslo.db 4.6.021:10
nicodemus_gordc: keep in mind that now the API is working fine21:11
nicodemus_the issue is the same as before, gnocchi-metricd can't process any measure21:11
gordcwhat's the pymysql comments from before?21:11
nicodemus_gordc: http://paste.openstack.org/show/488260/21:12
alejandritogordc, nevermind, the bug that is hitting us right now is ^^21:12
alejandritothat one from nicodemus_21:12
nicodemus_gordc: that I fixed simply by removing pymysql from the indexer url21:13
*** yarkot_ has joined #openstack-telemetry21:13
*** alejandrito has quit IRC21:16
gordcnicodemus_: you shouldn't have to remove it. i feel like there is an issue with new oslo.db for that issue21:23
gordctry older oslo.db21:23
*** yarkot_ has quit IRC21:29
openstackgerritgordon chung proposed openstack/ceilometer: close services in test  https://review.openstack.org/28554721:32
*** ddieterly has quit IRC21:39
*** agireud has quit IRC21:41
*** AndreiaKumpera has quit IRC21:42
*** agireud has joined #openstack-telemetry21:43
gordcnicodemus_: ok. i'll try to look now. i think i resolved gate issue.21:45
*** nicodemus_ has quit IRC22:00
*** safchain has quit IRC22:05
*** thorst_ has quit IRC22:26
*** liusheng has quit IRC22:37
*** liusheng has joined #openstack-telemetry22:38
*** thorst has joined #openstack-telemetry22:42
*** thorst has quit IRC22:46
*** gordc has quit IRC22:56
*** pradk has quit IRC22:56
*** yassine__ has quit IRC23:07
*** rbak__ has quit IRC23:48
*** Qiming has joined #openstack-telemetry23:57
« Thursday, 2016-02-25 Index Saturday, 2016-02-27 »

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