Wednesday, 2015-07-15

*** barra204 has quit IRC00:03
openstackgerritOpenStack Proposal Bot proposed openstack/ceilometer: Updated from global requirements  https://review.openstack.org/20065900:09
openstackgerritOpenStack Proposal Bot proposed openstack/ceilometermiddleware: Updated from global requirements  https://review.openstack.org/20133500:09
openstackgerritOpenStack Proposal Bot proposed openstack/python-ceilometerclient: Updated from global requirements  https://review.openstack.org/20183400:14
*** jfluhmann has quit IRC00:16
openstackgerritOpenStack Proposal Bot proposed openstack/ceilometer: Updated from global requirements  https://review.openstack.org/20065900:16
openstackgerritOpenStack Proposal Bot proposed openstack/ceilometermiddleware: Updated from global requirements  https://review.openstack.org/20133500:16
openstackgerritOpenStack Proposal Bot proposed openstack/ceilometer: Updated from global requirements  https://review.openstack.org/20065900:23
openstackgerritOpenStack Proposal Bot proposed openstack/ceilometermiddleware: Updated from global requirements  https://review.openstack.org/20133500:23
*** david-lyle has quit IRC00:25
*** primechuck has quit IRC00:26
*** Kennan2 is now known as Kennan00:29
openstackgerritOpenStack Proposal Bot proposed openstack/python-ceilometerclient: Updated from global requirements  https://review.openstack.org/20183400:29
openstackgerritOpenStack Proposal Bot proposed openstack/ceilometer: Updated from global requirements  https://review.openstack.org/20065900:31
openstackgerritOpenStack Proposal Bot proposed openstack/ceilometermiddleware: Updated from global requirements  https://review.openstack.org/20133500:31
*** ankita_wagh has quit IRC00:32
openstackgerritOpenStack Proposal Bot proposed openstack/python-ceilometerclient: Updated from global requirements  https://review.openstack.org/20183400:36
*** thorst has joined #openstack-ceilometer00:38
*** primechuck has joined #openstack-ceilometer00:41
*** bitblt has quit IRC00:44
*** deepthi has joined #openstack-ceilometer00:58
*** jruano has joined #openstack-ceilometer01:02
*** ankita_wagh has joined #openstack-ceilometer01:10
*** fabian4 has joined #openstack-ceilometer01:21
*** changbl has joined #openstack-ceilometer01:24
*** jfluhmann has joined #openstack-ceilometer01:28
*** primechuck has quit IRC01:28
openstackgerritliusheng proposed openstack/ceilometer: Clean the re-implemented serializers in Ceilometer  https://review.openstack.org/19414801:29
openstackgerritOpenStack Proposal Bot proposed openstack/ceilometer: Updated from global requirements  https://review.openstack.org/20065901:31
openstackgerritOpenStack Proposal Bot proposed openstack/ceilometermiddleware: Updated from global requirements  https://review.openstack.org/20133501:31
*** deepthi has quit IRC01:32
*** jruano has quit IRC01:35
*** thorst has quit IRC01:41
*** liusheng has quit IRC01:41
*** liusheng has joined #openstack-ceilometer01:47
*** boris-42 has quit IRC01:52
*** ankita_wagh has quit IRC01:58
*** Kennan2 has joined #openstack-ceilometer02:02
*** Kennan has quit IRC02:03
*** david-lyle has joined #openstack-ceilometer02:12
*** david-lyle has quit IRC02:35
*** hans_ has joined #openstack-ceilometer02:43
*** Beetnik has quit IRC02:46
openstackgerritgordon chung proposed openstack/gnocchi: fix broken devstack storage logic  https://review.openstack.org/20180502:52
*** ddieterly has quit IRC03:10
*** ankita_wagh has joined #openstack-ceilometer03:29
*** pauldix has joined #openstack-ceilometer03:32
*** david-lyle has joined #openstack-ceilometer03:38
*** _cjones_ has joined #openstack-ceilometer03:41
*** _cjones_ has quit IRC03:41
*** ddieterly has joined #openstack-ceilometer04:11
*** dikonoor has joined #openstack-ceilometer04:14
*** ddieterly has quit IRC04:16
*** deepthi has joined #openstack-ceilometer04:22
openstackgerritSergey Vilgelm proposed openstack/ceilometer: Switch to the oslo_utils.fileutils  https://review.openstack.org/19951504:35
*** yprokule has joined #openstack-ceilometer04:38
*** boris-42 has joined #openstack-ceilometer04:43
*** pauldix has quit IRC04:53
*** rjaiswal has quit IRC05:04
*** ddieterly has joined #openstack-ceilometer05:12
*** exploreshaifali has joined #openstack-ceilometer05:13
*** exploreshaifali has quit IRC05:14
*** ddieterly has quit IRC05:17
*** ankita_wagh has quit IRC05:26
*** ankita_wagh has joined #openstack-ceilometer05:27
*** Longgeek has joined #openstack-ceilometer05:28
*** ankita_wagh has quit IRC05:31
*** deepthi has quit IRC05:33
*** tshtilma has joined #openstack-ceilometer05:43
*** tshtilma has quit IRC05:44
*** tshtilma has joined #openstack-ceilometer05:45
*** _nadya_ has joined #openstack-ceilometer05:45
*** deepthi has joined #openstack-ceilometer05:50
*** fawadkhaliq has joined #openstack-ceilometer05:54
*** _nadya_ has quit IRC05:57
*** _nadya_ has joined #openstack-ceilometer05:58
*** _nadya_ has quit IRC06:02
*** rjaiswal has joined #openstack-ceilometer06:09
*** pauldix has joined #openstack-ceilometer06:11
*** pauldix_ has joined #openstack-ceilometer06:11
*** ddieterly has joined #openstack-ceilometer06:13
*** ddieterly has quit IRC06:18
*** belmoreira has joined #openstack-ceilometer06:21
*** primechuck has joined #openstack-ceilometer06:24
*** afazekas has joined #openstack-ceilometer06:30
*** tshtilma has quit IRC06:31
*** ankita_wagh has joined #openstack-ceilometer06:33
*** fawadkhaliq has quit IRC06:33
*** tshtilma has joined #openstack-ceilometer06:34
*** pauldix_ has quit IRC06:41
*** _nadya_ has joined #openstack-ceilometer06:52
*** ildikov has quit IRC06:53
*** fawadkhaliq has joined #openstack-ceilometer06:54
*** fawadk has joined #openstack-ceilometer06:55
*** fawadkhaliq has quit IRC06:58
*** fawadkhaliq has joined #openstack-ceilometer06:59
*** fawadk has quit IRC07:00
*** shardy has joined #openstack-ceilometer07:10
*** deepthi has quit IRC07:12
*** ddieterly has joined #openstack-ceilometer07:14
*** Ala has joined #openstack-ceilometer07:18
*** shardy_ has joined #openstack-ceilometer07:18
*** shardy has quit IRC07:19
*** ddieterly has quit IRC07:19
*** eglynn has joined #openstack-ceilometer07:20
openstackgerritRyota MIBU proposed openstack/aodh: Add 'event' type and 'event_rule' to alarm API  https://review.openstack.org/20096707:20
*** primechuck has quit IRC07:22
*** shardy_ has quit IRC07:23
*** Longgeek has quit IRC07:23
*** zigo has quit IRC07:24
*** shardy has joined #openstack-ceilometer07:24
*** zigo has joined #openstack-ceilometer07:25
*** ankita_wagh has quit IRC07:25
*** _stowa has quit IRC07:30
*** tshtilma has quit IRC07:41
*** _stowa_ has joined #openstack-ceilometer07:43
openstackgerritRyota MIBU proposed openstack/aodh: Refactor alarm scenario tests (TestAlarmsHistory)  https://review.openstack.org/20196607:44
*** _stowa_ is now known as _stowa07:46
*** ildikov has joined #openstack-ceilometer07:47
*** safchain has joined #openstack-ceilometer07:49
*** eglynn has quit IRC07:49
openstackgerritRyota MIBU proposed openstack/aodh: Refactor alarm scenario tests (TestAlarmsHistory)  https://review.openstack.org/20196607:56
*** Longgeek has joined #openstack-ceilometer08:00
*** cdent has joined #openstack-ceilometer08:01
*** fawadkhaliq has quit IRC08:05
*** boris-42 has quit IRC08:12
*** ddieterly has joined #openstack-ceilometer08:15
*** ddieterly has quit IRC08:19
*** fawadkhaliq has joined #openstack-ceilometer08:22
*** nsaje_ has joined #openstack-ceilometer08:28
openstackgerritAleksey Ripinen proposed openstack/python-ceilometerclient: Complex statistics query in ceilometer client  https://review.openstack.org/20151208:31
*** nsaje has quit IRC08:31
*** jdandrea has quit IRC08:32
*** yassine_ has joined #openstack-ceilometer08:37
*** yatin has joined #openstack-ceilometer08:40
*** _nadya_ has quit IRC08:47
*** _nadya_ has joined #openstack-ceilometer08:47
*** _nadya_ has quit IRC08:52
*** tshtilma has joined #openstack-ceilometer08:52
*** _nadya_ has joined #openstack-ceilometer09:01
*** rjaiswal has quit IRC09:04
*** _nadya_ has quit IRC09:06
*** ddieterly has joined #openstack-ceilometer09:15
*** _nadya_ has joined #openstack-ceilometer09:20
*** ddieterly has quit IRC09:20
*** bdossant has joined #openstack-ceilometer09:34
*** deepthi has joined #openstack-ceilometer09:48
*** jaosorior has joined #openstack-ceilometer09:54
*** Kennan has joined #openstack-ceilometer10:03
*** Kennan2 has quit IRC10:03
*** tshtilma has quit IRC10:15
*** ddieterly has joined #openstack-ceilometer10:16
*** ddieterly has quit IRC10:21
*** tshtilma has joined #openstack-ceilometer10:36
*** _nadya__ has joined #openstack-ceilometer10:48
*** _nadya_ has quit IRC10:50
*** cdent has joined #openstack-ceilometer10:52
*** afazekas has quit IRC11:05
openstackgerritMehdi Abaakouk (sileht) proposed openstack/gnocchi: Remove ceilometer dispatcher  https://review.openstack.org/20040811:14
*** ddieterly has joined #openstack-ceilometer11:17
cdentsileht: are your is liusheng's patches to remove the dispatcher effectively the same?11:21
*** ddieterly has quit IRC11:21
silehtcdent, he just miss an entry in setup.cfg and a useless configuration file11:22
*** fawadkhaliq has quit IRC11:23
openstackgerritMehdi Abaakouk (sileht) proposed openstack/gnocchi: Remove ceilometer dispatcher  https://review.openstack.org/20040811:25
silehtcdent, I just fix the tests of my version11:26
*** openstack has joined #openstack-ceilometer11:36
*** chmouel_ has joined #openstack-ceilometer11:37
*** htruta_ has joined #openstack-ceilometer11:37
*** thorst has joined #openstack-ceilometer11:37
*** fawadkhaliq has joined #openstack-ceilometer11:37
*** openstackstatus has joined #openstack-ceilometer11:37
*** ChanServ sets mode: +v openstackstatus11:37
*** fawadkhaliq has quit IRC11:45
openstackgerritMerged openstack/gnocchi: fix broken devstack storage logic  https://review.openstack.org/20180511:46
*** exploreshaifali has joined #openstack-ceilometer11:55
*** Ala has quit IRC12:05
*** Ala has joined #openstack-ceilometer12:06
*** gordc has joined #openstack-ceilometer12:08
cdentgordc: have you got a speaker bio I can steal as a base?12:17
cdentI'd prefer to say something like "speaker bios suck" but I don't think that will be very productive12:17
*** ddieterly has joined #openstack-ceilometer12:18
*** ddieterly has quit IRC12:23
*** liusheng has quit IRC12:26
gordccdent: is that what you put in vancouver?12:30
cdentno, but I can't find what i said for vancouver and besides I recall it was very boring12:30
gordcpaste doesn't like my bio without new lines... i'll pm you.12:32
gordccdent: i've been told my bio is questionable from a 'professional' pov.12:33
gordcbut that's what i've been using as my intro for everything.12:33
cdentprofessionalism is hogwash12:33
gordcyeah. it's something... all i know is that i lack whatever it is.12:34
*** edmondsw has joined #openstack-ceilometer12:35
cdentIf someone or something is striving for professionalism, it means they don't have any reasl sense of the value of what they are doing and need extenral validation12:35
gordclol12:36
*** liusheng_ has joined #openstack-ceilometer12:36
gordcvery well versed12:36
gordcsome would 'professional'12:37
*** deepthi has quit IRC12:39
cdentI struggle with that12:42
*** yatin has quit IRC12:42
*** afazekas has joined #openstack-ceilometer12:44
*** exploreshaifali has quit IRC12:45
*** deepthi has joined #openstack-ceilometer12:52
*** tshtilma has quit IRC12:53
*** gordc has quit IRC12:54
*** gordc has joined #openstack-ceilometer12:54
*** tshtilma has joined #openstack-ceilometer12:55
cdentliusheng_: do you not regularly tox -r ?12:58
*** yassine_ has quit IRC12:59
liusheng_cdent, sorry, which one you talking about ?13:00
*** julim has joined #openstack-ceilometer13:00
cdentliusheng_: I'm just curious how you ran across a situation where you had something other than the newest gabbi in a testing situation13:00
*** liusheng_ has quit IRC13:07
*** liusheng_phone has joined #openstack-ceilometer13:07
*** yassine_ has joined #openstack-ceilometer13:08
*** liusheng_ has joined #openstack-ceilometer13:11
*** Administrator__ has joined #openstack-ceilometer13:11
liusheng_cdent, sorry, I have a network breakdonw just now, I was also wonder about that :(13:12
*** primechuck has joined #openstack-ceilometer13:22
*** shardy_ has joined #openstack-ceilometer13:29
*** primechuck has quit IRC13:30
*** eglynn has joined #openstack-ceilometer13:31
*** shardy has quit IRC13:31
*** shardy_ has quit IRC13:35
*** shardy has joined #openstack-ceilometer13:35
*** jdandrea has joined #openstack-ceilometer13:37
gordccdent: do you have gnocchi port defined when you launch devstack+gnocchi?13:43
* cdent looks13:43
cdentno, with the latest devstack plugin for gnocchi, the metric endpoint is on port 80 and and /metric/13:44
cdents/and and/and on/13:44
cdentyou can see it configured in /etc/ceilometer/ceilometer.conf13:44
gordcwhat's the horizon endpoint?13:44
cdenthorizon gets / if I recall correctly13:45
cdentbut also horizon? turn that off, it only takes up cpu cycles13:45
gordchmmm this is only if you have mod wsgi enabled?13:45
openstackgerritMerged openstack/gnocchi: Remove ceilometer dispatcher  https://review.openstack.org/20040813:45
cdentgordc: only now that that ^ has merged is there any chance gnocchi+ceilo will work in devstack13:45
cdentgordc: yes13:46
cdentbut mod wsgi is the emerging default13:46
gordcif i have it disabled i need port?13:46
gordci turned it off because it's a bitch to debug.13:46
cdentyes13:46
gordc... is that a bad word? meh.13:46
cdenthow do you do your debugging?13:46
gordcpdb13:46
gordci don't like the remote debugger13:46
*** prad has joined #openstack-ceilometer13:47
gordcit's a hassle to set up.13:47
gordccdent: i'll bring in new patch.13:47
cdentI'm so proud of you gordc13:48
*** liusheng_ has quit IRC13:48
cdentactually use pdb!13:48
cdentit's always seemed _far_ too painful to me13:48
gordci don't like learning new things13:48
*** Administrator__ has quit IRC13:48
cdentI use print and log and sys.stderr.write !13:49
* cdent is a luddite13:49
gordcyeah. i won't lie, once there's multiple workers the pdb solution falls apart.13:49
*** exploreshaifali has joined #openstack-ceilometer13:49
* sileht uses 'raise Exception("boom!")' as breaking point :p13:50
gordc... now we know why the tests say that.13:50
*** ddieterly has joined #openstack-ceilometer13:53
*** pauldix has joined #openstack-ceilometer13:55
*** yprokule has quit IRC13:55
*** yprokule has joined #openstack-ceilometer13:57
gordcsileht: if i use gnocchi+swift+ceilometermiddleware, it will explode right? there's no way to actually have them all running13:58
*** rbak has joined #openstack-ceilometer13:58
silehtgordc, it should work but generates a lot of useless messages on rabbit13:59
silehtgordc, the collector should drop the samples that comes from its (if the project_id is not empty)13:59
gordcsileht: hmmm. maybe it's the new oslo.messaging. i get a lot of 'Duplicate message' errors... even on notification agent side.14:01
silehtgordc, I suspect a race sometimes when the collector start and the gnocchi project/user dones't yet exists when it starts14:01
gordcbasically the notification and collector queues just grow and never get smaller14:02
* sileht stack.sh14:02
openstackgerritMerged openstack/aodh: Optionally create trust for alarm actions  https://review.openstack.org/19958914:03
openstackgerritMerged openstack/aodh: Tolerate alarm actions set to None  https://review.openstack.org/19997414:04
cdentI thought the middleware had been changed?14:04
cdentor is it that it doesn't have access to the right info at boot time?14:05
gordccdent: i added an ignore param... but you need to know project_id of gnocchi to properly ignore those requests.14:05
cdentmeaning? you have to change the config and restart swift?14:06
gordcsileht: i get this: http://paste.openstack.org/show/377642/14:07
gordccdent: basically. or you need to create gnocchi account info before starting swift.14:08
silehtgordc, looks like a messaging bugs :(14:08
gordcsileht: kk. because it seems like it's no clearing the queue at all14:08
cdentgordc: you can't control ordering in that fashion so doing a restart is likely the only option14:14
*** primechuck has joined #openstack-ceilometer14:14
gordccdent: yeah... so not that useful in reality14:15
cdentthere's nothing wrong with doing a restart14:15
*** shardy_ has joined #openstack-ceilometer14:16
cdentthe critical part is making sure the config gets set appropriately14:16
*** shardy has quit IRC14:17
gordccdent: arguably not that useful... that said, it's sort of needed because if you ever want to see your how large a queue can get, i'd suggest you devstack+gnocchi+swift.14:18
gordcyour queue will be in the thousands in seconds.14:18
cdentit needs to work, gordc, because we need to be able to use devstack to do testing, right?14:20
cdentand if there is extraneous stuff in the queue then we are obscuring the testing14:20
*** shardy_ has quit IRC14:21
*** shardy has joined #openstack-ceilometer14:22
gordccdent: it's not exactly extraneous.14:25
gordcit's just that you are now metering gnocchi.14:25
cdentwe've declared that the swiftmiddleware should not send notifications about gnocchi14:25
cdentthat's part of the contract14:25
cdenttherefore we need to test that contract14:25
gordcwhere was that said?14:26
cdentthat's why you wrote that addition to the middleware, yeah?14:26
gordcbasically. because i noticed the load metering gnocchi created.14:27
gordci'm not saying we should be sending notifications about gnocchi. i'm just curious if we formally announced it at some point.14:27
cdentwhat does that even mean? nobody is using gnocchi, really, so how can there be anything formal that needs to be announced?14:28
cdentit's a technical requirement if we actually want stuff to work...14:28
gordcwell not 'formal', but more whether at some point consensus was made on it?14:29
gordcor if it's just everyone's general assumption.14:29
cdentI think it's kind of general assumption14:30
cdentbetter would be just to turn off the ceilometermiddleware as that's what swift wants anyway14:30
gordci think swift wants to turn off openstack.. :)14:30
*** rbrooker has joined #openstack-ceilometer14:31
*** deepthi has quit IRC14:34
*** pauldix has quit IRC14:36
gordcsileht: do you need a bug opened for the paste i sent you?14:44
gordcnot sure if it's just my environment... doesn't seem to be visible in gate.14:45
silehtgordc, I just updated oslo.messaging and got the same error14:46
gordcsileht: :(14:46
gordcsileht: https://review.openstack.org/#/c/20046714:54
gordcfor that, are you going to work on tempest part too or is that it?14:55
gordctrying to see if i need to find resource.14:55
*** _nadya__ has quit IRC15:00
*** _nadya_ has joined #openstack-ceilometer15:01
openstackgerritMerged openstack/ceilometer: monkey_patch thread in tests  https://review.openstack.org/20172315:01
silehtgordc, I'm unsure we need tempest here15:01
silehtgordc, the current tempest tests already exists as gabbit tests in aodh tree15:02
*** pauldix has joined #openstack-ceilometer15:02
*** shardy_ has joined #openstack-ceilometer15:03
*** shardy has quit IRC15:04
*** _nadya_ has quit IRC15:05
openstackgerritRohit Jaiswal proposed openstack/ceilometer: Fixes querying alarm history with severity field  https://review.openstack.org/20023815:08
*** shardy_ has quit IRC15:08
*** shardy has joined #openstack-ceilometer15:09
*** tshtilma has quit IRC15:10
*** jfluhmann has quit IRC15:10
*** tshtilma has joined #openstack-ceilometer15:12
*** lsmola has quit IRC15:13
*** yprokule has quit IRC15:17
gordcsileht: i see. i think we still need to switch over because we'd want heat and other integration tests to use aodh and not ceilometer-alarm.15:17
gordci'll try to find a resource for that.15:17
*** liusheng_phone has quit IRC15:19
*** ddaskal has joined #openstack-ceilometer15:20
*** pauldix has quit IRC15:22
pradgordc, i updated the kv and orch tests to confirm the meters i added are working and generating samples as expected for existing tests15:26
pradgordc, if we go the route of pushing an empty meter.yaml, we can remove those15:27
pradgordc, just some extra validation on top of generic tests i have15:27
pradgordc, i'll address the rest of the comments in a bit and push a new version15:27
*** tshtilma has quit IRC15:29
*** yprokule has joined #openstack-ceilometer15:37
*** ddieterly has quit IRC15:38
*** ig0r__ has quit IRC15:39
*** ig0r_ has joined #openstack-ceilometer15:40
*** pauldix has joined #openstack-ceilometer15:45
openstackgerritChris Dent proposed openstack/ceilometer: [WIP] The start of a grenade plugin for ceilometer  https://review.openstack.org/19644115:48
*** belmoreira has quit IRC15:48
gordcprad: why do we need to validate meter.yaml generates those specific samples?15:49
*** tshtilma has joined #openstack-ceilometer15:50
*** pauldix has quit IRC15:51
pradgordc, for my satisfaction that i'm not breaking the existing workflow :)15:52
pradgordc, those will go away with deprecation task i have anyway15:53
pradgordc, i was thinking we'll get this commit in and then i'll push a subsequent commit to enable it in setup.cfg15:55
pradgordc, in that commit, we can decide whether to remove meters.yaml or deprecate other entrypoints instead15:56
gordcremove meters.yaml?15:57
*** adurbin_ has joined #openstack-ceilometer15:57
gordcprad: i would think that's the only non-option.15:58
pradgordc, crap from meter.yaml.. as in have an empty one15:58
*** pauldix has joined #openstack-ceilometer15:58
pradgordc, and then fill it as we go and deprecate15:58
*** pauldix has quit IRC16:00
*** logan2 has quit IRC16:00
gordcprad: i don't get why'd we commit it full, delete stuff from it, and then fill it in again. i think i'm missing something.lol16:00
*** bdossant_ has joined #openstack-ceilometer16:01
pradgordc, no we dont commit it full.. if i do then i'm not going to remove it for sure16:01
gordcprad: i believe the conclusoin from midcycle was that we'd drop all the meter code/tests as long as it was convered in meter.yaml16:01
*** _nadya_ has joined #openstack-ceilometer16:01
cdentgordc, sileht did the earlier conversations about messaging bugs resolve to anything happy or said?16:01
gordccdent: sad16:02
cdentbummer, what now?16:02
gordcit's a very real issue with 1.17.1 i believe16:02
silehtcdent, I still investigating16:02
gordccdent: work on dream projects.16:02
cdentlet me know if there's anything I can do to test or help out, etc16:02
pradgordc, right hence my point,.. we dont add entry point for this in setup.cfg now, leave the meters.yaml as is and then add entry point as we deprecate16:02
pradgordc, or have an empty one now and then add as we go16:03
*** bdossant has quit IRC16:04
gordcprad: you can check it in with either empty meter.yaml and entry point or non-empty yaml but no entrypoint16:04
gordcbut not non-empty yaml and entrypoint?16:04
pradgordc, i already did " non-empty yaml but no entrypoint" as you see, so lets go with that16:05
pradgordc, i'll add entrypoint as i deprecate code16:05
silehtcdent, gordc that weird I have the callback of a futures.add_done_callback() called multiple times16:05
*** jfluhmann has joined #openstack-ceilometer16:05
pradhas anyone had a clean tox run lately.. i see a bunch of errors spewing16:06
gordcprad: sure... i think i understand... :)16:06
*** pauldix has joined #openstack-ceilometer16:06
gordcprad: which oslo.messaging version?16:06
prad  File "/home/pkilambi/code/openstack/ceilometer/.tox/py27/lib/python2.7/site-packages/eventlet/hubs/epolls.py", line 39, in __init__16:07
pradIOError: [Errno 24] Too many open files16:07
gordcthere were a bunch of release 2 days ago16:07
gordcprad: i think _nadya_ had that yestreday16:07
gordci saw that before too but it was more mongo related.16:07
prad2015-07-15 12:07:43.540 2780 ERROR ceilometer.storage.mongo.utils [req-518b6435-04b1-4c0e-86a0-80e0fe22167d - - - - -] Unable to reconnect to the primary mongodb after 10 retries. Giving up.16:08
pradyea seems mongo related16:08
*** _nadya_ has quit IRC16:09
gordcdo you get the same thing with py-mysql?16:09
*** pauldix has quit IRC16:09
gordci stopped testing with py27 because of the mongo errors.16:10
gordcprad: you can try changing the maxConn setting in setup-test-env-mongodb16:10
gordcit worked on my first dev machine... it didn't work on second.16:11
praddid not try mysql yet16:11
pradcool, i'll try the maxConn and see16:11
*** _cjones_ has joined #openstack-ceilometer16:14
*** ildikov has quit IRC16:14
*** pauldix has joined #openstack-ceilometer16:17
gordcsileht: is pooledexecutor a new executor (alternative to eventlet)?16:18
gordcsileht: never. it's a base.16:18
silehtgordc,yep16:18
silehtgordc, we put the executor logic into a lib called futurist (code come from taskflow)16:19
silehtgordc, the lib allow to retrive statistic on task runned by the executor16:19
silehtit seems the bug is in this lib16:19
silehtgordc, I have tried why the threading executor and everything work16:20
gordcsileht: ah i see.. i haven't looked at the messaging code in a while.16:20
silehtgordc, so this is a bug with futures+eventlet somewhere in fututist16:20
gordcsileht: cool cool. good to have it narrowed down16:20
*** pauldix has quit IRC16:21
gordcsileht: there are a lot of colourful variable names now... i don't know what they do.lol16:22
*** bdossant_ has quit IRC16:22
*** ddaskal has quit IRC16:24
*** afazekas has quit IRC16:27
*** pauldix has joined #openstack-ceilometer16:29
pradgordc, no luck with maxconn tweak, will try mysql16:30
*** Ala has quit IRC16:34
gordcprad: yeah, it didn't work on my current dev machine either.16:35
gordclong live sql16:36
*** [1]tshtilma has joined #openstack-ceilometer16:36
*** tshtilma has quit IRC16:37
*** [1]tshtilma is now known as tshtilma16:37
openstackgerritPradeep Kilambi proposed openstack/ceilometer: Declarative meters support  https://review.openstack.org/19763316:43
*** yassine_ has quit IRC16:45
silehtgordc, cdent that should fix the issue: https://review.openstack.org/#/c/202187/16:47
cdentyay!16:47
silehtI haven't really understand what happend, I will add tests to catch that in oslo.messaging tomorrow16:48
*** ankita_wagh has joined #openstack-ceilometer17:00
openstackgerritChris Dent proposed openstack/gnocchi: Use a settings file for the devstack plugin  https://review.openstack.org/20219117:01
openstackgerritChris Dent proposed openstack/ceilometer: Delay the start of the collector until after apache restart  https://review.openstack.org/20219217:02
*** nsaje__ has joined #openstack-ceilometer17:02
cdentgordc: I'm getting metrics in swifted gnocchi, from the ceilometer dispatcher17:02
*** nsaje_ has quit IRC17:04
*** eglynn has quit IRC17:04
gordccdent: cool cool. i need to patch oslo.messaging on mine.17:10
*** exploreshaifali has quit IRC17:10
*** logan2 has joined #openstack-ceilometer17:10
gordccan't tell how quick it's processing data since the queue never clears.17:10
cdentyeah that part I'm just ignoring at the moment17:11
gordcbbl. grabbing lunch.17:12
*** ityaptin has quit IRC17:15
*** ddieterly has joined #openstack-ceilometer17:18
*** tuvenen has quit IRC17:21
*** yprokule has quit IRC17:23
*** pauldix has quit IRC17:25
*** tuvenen has joined #openstack-ceilometer17:25
*** ityaptin has joined #openstack-ceilometer17:26
openstackgerritRohit Jaiswal proposed openstack/ceilometer: add Trove(DBaaS) events  https://review.openstack.org/20074617:29
openstackgerritChris Dent proposed openstack/ceilometer: [WIP] Attempt to use grenade plugin with devstack plugin  https://review.openstack.org/20220317:29
*** dikonoor has quit IRC17:31
*** mgarza_ has joined #openstack-ceilometer17:32
ityaptingordc: Hi! I have an couple of thoughts in context of py27 tests + mongo failing. Did you run several tox jobs which use setup-test-env-mongodb in same time?17:37
*** _nadya_ has joined #openstack-ceilometer17:37
ityaptingordc: In this case first job may stop mongodb daemon that is used by both before the second job finish tests.17:42
ityaptingordc: It caused by constant port of mongodb tests and fact what tests are not stopped even if no one waited line founded.17:43
ityaptingordc: So, i think "wait_for_line "waiting for connections on port ${MONGO_PORT}" ${MONGO_DATA}/out" does not work as I expected.17:45
gordcityaptin: hmm. you mean did i run more than one tox job at a time or if i ran tox, abort, and re run?17:48
ityaptingordc: more than one tox job at a time17:49
gordci don't think i did.17:49
gordcnot more than one mongo run17:50
*** tshtilma has quit IRC17:51
ityaptingordc: Hmm, then i don't know reason :( . In same time, i think we have a bug in wait_for_line. I think it should stop tests if we haven't got this line before the eof.17:52
gordcityaptin: hmmm so you're saying that it's running tests before mongo has been started? that might be it.17:56
*** nsaje___ has joined #openstack-ceilometer17:56
gordcityaptin: i haven't looked into it to be honest. more focused on sql personally17:57
*** nsaje__ has quit IRC17:59
*** _nadya_ has quit IRC18:00
*** ankita_w_ has joined #openstack-ceilometer18:00
*** _nadya_ has joined #openstack-ceilometer18:00
cdentinfra has taken a dive18:01
*** Kennan has quit IRC18:02
ityaptingordc: The wait_for_line may affect any backend. It's mean that if any error become during the service initialize and we don't find waiting line [1], we run testr nevertheless. So, in this case we run tests on backend that doesn't run. Of course, tests will failed.18:03
*** ankita_wagh has quit IRC18:03
*** Kennan has joined #openstack-ceilometer18:03
ityaptingordc: Same issue in case of starting two same tox jobs in same time18:04
*** _nadya_ has quit IRC18:05
ityaptin1. First job run own service18:05
*** pauldix has joined #openstack-ceilometer18:05
ityaptin2. second job doesn't start service but run tests  nevertheless18:05
gordcityaptin: huh? iiuc correctly, that statement will block until it sees the string in the specified file18:05
ityaptin3. first job stop service18:06
ityaptin4. second job's test fails18:06
gordcah got it.18:06
gordcyeah, i'm not doing anything that fancy.18:06
ityaptingordc: If read line get the eof wait_for_line behaves like it's ok.18:07
gordci see.18:07
*** nsaje____ has joined #openstack-ceilometer18:12
*** nsaje___ has quit IRC18:15
openstackgerritIlya Tyaptin proposed openstack/ceilometer: Stop the tests if backend hasn't started  https://review.openstack.org/20222618:17
cdentityaptin: that's a nice catch18:21
openstackgerritIlya Tyaptin proposed openstack/ceilometer: Stop the tests if backend hasn't started  https://review.openstack.org/20222618:22
*** boris-42 has joined #openstack-ceilometer18:23
*** vishwanathj has joined #openstack-ceilometer18:25
ityaptingordc, cdent: ^^ if you want you could test it with next scenarios: 1) in screen/tmux/other console window run a "sudo mongodb --port 29000" and during it is working run a "tox -e py27"18:25
* cdent nods at ityaptin 18:26
ityaptingordc, cdent: It also affects elasticsearch and postgresql environments18:27
ityaptingordc, cdent: mysql works fine18:28
*** belmoreira has joined #openstack-ceilometer18:28
cdentis doing the desired thing with mongo for me, so seems to work18:28
ityaptincdent: good)18:29
*** ddieterly has quit IRC18:33
*** ddieterly has joined #openstack-ceilometer18:36
*** ddieterly has quit IRC18:36
openstackgerritPradeep Kilambi proposed openstack/ceilometer: Declarative meters support  https://review.openstack.org/19763318:37
*** pauldix has quit IRC18:39
openstackgerritChris Dent proposed openstack/gnocchi: Use a settings file for the devstack plugin  https://review.openstack.org/20219118:41
*** belmoreira has quit IRC18:41
*** _nadya_ has joined #openstack-ceilometer18:46
*** exploreshaifali has joined #openstack-ceilometer18:46
*** annasort has quit IRC18:58
edmondswgordc, you had a comment in https://review.openstack.org/#/c/187593 (patch set 2) about "metering" as the service catalog endpoint type should really be telemetry19:00
edmondsw"this should be telemetry = service/telemetry... i will clean up devstack"19:00
edmondswanything happen on that?19:01
edmondswI've also noticed that "metering" is hardcoded in ceilometerclient, so that would also need to change19:01
edmondswI agree that telemetry is a much better value there19:01
*** Longgeek has quit IRC19:08
*** tshtilma has joined #openstack-ceilometer19:09
gordcedmondsw: it didn't go as easily as i had hoped: https://review.openstack.org/#/c/187992/ and https://review.openstack.org/#/c/190600/19:09
gordcit became low priority for me but i get the feeling that you need to disable tempest, update devstack, update tempest, reenable tempest. just to make the change19:10
*** mgarza_ has quit IRC19:11
*** tongli has joined #openstack-ceilometer19:11
*** mgarza has joined #openstack-ceilometer19:11
gordcedmondsw: if you have someone who wants to do it, they can take it over... but i don't get feeling it's worth the effort unless there's an easier way.19:11
cdentchanging names in the service catalog seems like something that ought to never happen?19:11
cdentunless you are just adding an alias19:11
cdent?19:11
gordccdent: is there an alias?19:11
cdenti dunno, that's why I threw that ? on that at the last minute19:12
gordccdent: the thing is we stopped being 'metering' over 2 years ago.19:12
cdentbollocks19:12
cdentit's just a name19:12
gordcoh. yeah i never looked at alias if that exists or not.19:12
cdentmetering means "we hear stuff"19:12
cdentwe still do that19:12
edmondswit's supposed to be a descriptive name, though19:12
cdenttelemetry means absolutely zero to the human on the street19:13
cdentwhereas metering means every so slightly more than zero?19:13
gordctelemetry was easier to undetsand then metering. (for me)19:13
*** belmoreira has joined #openstack-ceilometer19:13
cdentmy point is it could be "foobar" as long as people knew that foobar was what they were supposed to use19:13
gordci actually never heard anyone use metering until i joined ceilometer.lol19:13
cdenttelemetry has a much more specific meaning to me and thus is artificial, sort of snobbing19:13
edmondswmight be able to duplicate the data in the service catalog, one called telemetry and the other metering, and then remove metering after a suitable deprecation timeframe19:13
cdentedmondsw: that seems like the only likely road if people think that having a new name is a good idea19:14
cdenthowever19:14
cdentas alarming and gnocchi go out to their own apis the umbrella name kind of goes away19:14
edmondswthis is true19:14
cdentso maybe now is not the right time? maybe once things mature?19:14
gordccdent: probably. i only tried it becuase i figured it'd be a one line change19:15
gordcit was not... which is why i just gave up19:15
cdentbro, do you even openstack?19:15
cdentnothing is a one line change19:15
cdent;)19:15
edmondsw:D19:15
gordcyou see those spelling patches!lol19:15
edmondswok, tx guys19:16
*** prad has quit IRC19:19
*** prad has joined #openstack-ceilometer19:22
openstackgerritChris Dent proposed openstack/ceilometer: Set conf.gnocchi_dispatcher.url explicitly in tests  https://review.openstack.org/20224519:28
*** prad has quit IRC19:28
*** tshtilma has quit IRC19:34
cdentgordc: make the world a better place +A this: https://review.openstack.org/#/c/202226/19:35
gordccdent: done19:39
cdentdanke19:39
*** prad has joined #openstack-ceilometer19:40
*** belmoreira has quit IRC19:43
openstackgerritOpenStack Proposal Bot proposed openstack/ceilometer: Updated from global requirements  https://review.openstack.org/20065919:44
*** mgarza has quit IRC19:50
*** mgarza has joined #openstack-ceilometer19:50
*** ankita_w_ has quit IRC19:56
*** ankita_wagh has joined #openstack-ceilometer19:56
*** mgarza has quit IRC20:01
*** mgarza has joined #openstack-ceilometer20:02
*** Longgeek has joined #openstack-ceilometer20:08
openstackgerritChris Dent proposed openstack/gnocchi: Cleanup handling of metricd and its logs in devstack  https://review.openstack.org/20225920:09
*** mgarza has quit IRC20:10
*** mgarza has joined #openstack-ceilometer20:10
*** bdossant has joined #openstack-ceilometer20:11
*** bdossant has quit IRC20:11
*** pauldix has joined #openstack-ceilometer20:12
*** Longgeek has quit IRC20:13
gordcprad: do you know if this was ever picked up https://review.openstack.org/#/c/91474/20:14
pradgordc, i'll need to check, it was suppose to be a juno thing, so i would assume so20:15
pradgordc, neutron did a big refactor and this was suppose to be covered as part of that20:16
pradgordc, i'll poke someone20:16
gordci was going to assume no. ;)20:17
gordcjust asking so i could clean this: https://bugs.launchpad.net/ceilometer/+bug/107001120:17
openstackLaunchpad bug 1070011 in Ceilometer "the notification message payload for delete events should include all metadata" [Medium,Triaged] - Assigned to Nelly (nelly-smitt)20:17
pradfair assumption :)20:17
pradgordc, i'll get back to you after checking20:18
* cdent retires for the evening20:19
* prad waves20:19
gordclates20:19
cdenttomorrow morning I'll go back through the gnocchi+ceilo devstack parade and see what else I can break20:19
cdentnight night20:20
*** cdent has quit IRC20:20
*** ankita_wagh has quit IRC20:23
*** gordc has quit IRC20:27
pradgordc, aodh packaging https://pkilambi.fedorapeople.org/openstack-aodh/20:27
pradthis is based on latest git as we havent released it to pypi yet20:28
*** gordc has joined #openstack-ceilometer20:28
*** htruta_ has quit IRC20:32
*** htruta has joined #openstack-ceilometer20:32
*** jfluhmann has quit IRC20:37
*** ankita_wagh has joined #openstack-ceilometer20:38
*** ankita_wagh has quit IRC20:39
*** ankita_wagh has joined #openstack-ceilometer20:39
openstackgerritOpenStack Proposal Bot proposed openstack/ceilometer: Updated from global requirements  https://review.openstack.org/20065920:39
*** mgarza has quit IRC20:48
*** mgarza has joined #openstack-ceilometer20:49
*** pauldix has quit IRC20:50
*** pauldix has joined #openstack-ceilometer20:50
pradgordc, this should be good to get in i think https://review.openstack.org/197633 .. when you get a chance if you could review20:55
pradgordc, i'm working on subsequent patch to enable entrypoints and deprecate20:55
pradas we discussed20:55
pradgordc, btw i'm working on testing the gnocchi dispatcher work flow, will document something in the docs on configuration after i'm reasonably convinced it works :)20:58
pradiirc we dont have anything in the docs about gnocchi dispatcher20:58
*** ildikov has joined #openstack-ceilometer21:00
*** mgarza has quit IRC21:01
*** mgarza has joined #openstack-ceilometer21:02
*** thorst has quit IRC21:03
gordcprad: yeah docs would be good.21:07
gordcthis is what i'm going by http://docs.openstack.org/developer/gnocchi/devstack.html21:08
gordcare you trying with swift or file?21:08
pradfile21:10
*** julim has quit IRC21:10
pradplanning to try all 3, file, swift, ceph21:10
pradhmm seeing issues with log files missing21:11
*** tongli has quit IRC21:11
*** mgarza has quit IRC21:13
*** mgarza_ has joined #openstack-ceilometer21:13
openstackgerritMerged openstack/ceilometer: Stop the tests if backend hasn't started  https://review.openstack.org/20222621:13
*** fawadkhaliq has joined #openstack-ceilometer21:15
*** safchain has quit IRC21:15
openstackgerritOpenStack Proposal Bot proposed openstack/ceilometer: Updated from global requirements  https://review.openstack.org/20065921:20
*** htruta_ has joined #openstack-ceilometer21:30
*** _nadya_ has quit IRC21:31
*** _nadya_ has joined #openstack-ceilometer21:32
*** _nadya_ has quit IRC21:36
*** pauldix has quit IRC21:37
*** gordc has quit IRC21:43
*** mgarza_ has quit IRC21:43
*** mgarza has joined #openstack-ceilometer21:44
openstackgerritRohit Jaiswal proposed openstack/ceilometer: Fixes querying alarm history with severity field  https://review.openstack.org/20023821:44
openstackgerritRohit Jaiswal proposed openstack/ceilometer: add Trove(DBaaS) events  https://review.openstack.org/20074621:44
*** shardy has quit IRC21:47
*** fawadkhaliq has quit IRC21:55
*** Longgeek has joined #openstack-ceilometer21:57
*** htruta_ has quit IRC22:05
*** edmondsw has quit IRC22:05
*** jruano has joined #openstack-ceilometer22:10
*** mgarza has quit IRC22:11
*** mgarza has joined #openstack-ceilometer22:12
*** nsaje has joined #openstack-ceilometer22:14
*** exploreshaifali has quit IRC22:15
*** Longgeek has quit IRC22:16
*** nsaje____ has quit IRC22:17
*** EmilienM is now known as EmilienM|off22:22
*** jaosorior has quit IRC22:26
*** primechuck has quit IRC22:36
*** EmilienM|off is now known as EmilienM22:37
*** prad has quit IRC22:45
*** htruta_ has joined #openstack-ceilometer22:50
*** rbak has quit IRC23:00
*** EmilienM is now known as EmilienM|off23:03
*** mgarza has quit IRC23:04
*** adurbin_ has quit IRC23:06
*** primechuck has joined #openstack-ceilometer23:12
*** htruta_ has quit IRC23:13
*** ankita_w_ has joined #openstack-ceilometer23:19
*** ankita_wagh has quit IRC23:22
*** rbrooker has quit IRC23:27
*** thorst has joined #openstack-ceilometer23:41
*** thorst has quit IRC23:52

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