| *** SergeyLukjanov is now known as SergeyLukjanov_ | 00:00 | |
| *** lsmola has quit IRC | 00:01 | |
| *** jhurlbert has quit IRC | 00:35 | |
| *** mst89 has joined #openstack-ceilometer | 00:37 | |
| *** xmltok has quit IRC | 00:41 | |
| *** mst89 has quit IRC | 01:03 | |
| *** luisg has quit IRC | 01:48 | |
| *** luisg has joined #openstack-ceilometer | 01:49 | |
| *** ildikov_ has quit IRC | 04:39 | |
| *** tongli has quit IRC | 04:49 | |
| *** julienvey_ has quit IRC | 04:50 | |
| openstackgerrit | A change was merged to openstack/ceilometer: Fix some typos in architecture doc https://review.openstack.org/68714 | 05:50 |
|---|---|---|
| *** rwsu has quit IRC | 05:59 | |
| openstackgerrit | Jenkins proposed a change to openstack/ceilometer: Imported Translations from Transifex https://review.openstack.org/62808 | 06:03 |
| *** doug-fish has quit IRC | 06:19 | |
| *** sayali_ has joined #openstack-ceilometer | 06:58 | |
| *** sayali has joined #openstack-ceilometer | 06:58 | |
| *** ildikov_ has joined #openstack-ceilometer | 07:19 | |
| openstackgerrit | Mehdi Abaakouk proposed a change to openstack/ceilometer: Replace oslo.rpc by oslo.messaging https://review.openstack.org/57457 | 08:17 |
| *** sileht has quit IRC | 08:42 | |
| *** sileht has joined #openstack-ceilometer | 08:47 | |
| *** julienvey_ has joined #openstack-ceilometer | 08:54 | |
| *** Alexei_987 has joined #openstack-ceilometer | 08:55 | |
| *** eglynn has joined #openstack-ceilometer | 09:04 | |
| openstackgerrit | Mehdi Abaakouk proposed a change to openstack/ceilometer: Replace oslo.rpc by oslo.messaging https://review.openstack.org/57457 | 09:07 |
| *** yassine has joined #openstack-ceilometer | 09:12 | |
| openstackgerrit | Julien Danjou proposed a change to openstack/ceilometer: storage: store recording timestamp https://review.openstack.org/70166 | 09:12 |
| openstackgerrit | Julien Danjou proposed a change to openstack/ceilometer: Update oslo https://review.openstack.org/69903 | 09:14 |
| openstackgerrit | Julien Danjou proposed a change to openstack/ceilometer: Update oslo https://review.openstack.org/69903 | 09:18 |
| sileht | jd__, I just realize the database dispatcher pass the sample (named meter...) from the rpc directly to the database backend without convert it into storage.models.Sample | 09:25 |
| jd__ | not surprised indeed | 09:25 |
| jd__ | I think it'd be correct to do so OTOH it kind of add overhead | 09:31 |
| * jd__ going to debug ceilometer eventlet loop, good bye friends | 09:31 | |
| sileht | jd__, if you need some help... | 09:34 |
| ekarlso | jd__: hey, is there a simple way of adding support to PaaS alike services like Designate in Ceilo ? | 09:34 |
| jd__ | ekarlso: depends on what you mean by simple | 09:40 |
| jd__ | but yes we can | 09:41 |
| *** SergeyLukjanov_ is now known as SergeyLukjanov | 09:43 | |
| jd__ | sileht: this looks bad | 09:44 |
| jd__ | sileht: the collector get stucks in the impl_kombu consume() method from the RPC since it blocks on socket.recv() now | 09:46 |
| jd__ | I wonder if oslo.messaging is going to help us on that | 09:46 |
| ekarlso | jd__: any pointers ? | 09:46 |
| jd__ | or if it's as badly 'designed' | 09:46 |
| sileht | jd__, if ceilometer use oslo.messaging with blocking mode, we can build our Service class that handle only thread (ceilometer use only one thread) and use normal timer signal for timed tasks | 09:46 |
| jd__ | ekarlso: 0xd3adb33f | 09:47 |
| sileht | lol | 09:47 |
| ekarlso | jd__: HAH! :p | 09:47 |
| jd__ | sileht: I'm not sure a timer signal would work correctly if you mean SIGALRM but we can build an event loop with some other lib | 09:48 |
| sileht | jd__, agreed over eventloop will be better | 09:48 |
| jd__ | the thing is I don't know what to do | 09:48 |
| sileht | jd__, but what about python2.6/2.7/3.3 | 09:49 |
| jd__ | sileht: http://paste.openstack.org/show/62226/ FWIW | 09:49 |
| sileht | jd__, I have speak to quickly collector use two green thread, one for rpc, one for udp | 09:49 |
| jd__ | ah right, that tooo | 09:49 |
| jd__ | this is getting me nuts | 09:50 |
| *** sayali__ has joined #openstack-ceilometer | 09:55 | |
| *** sayali_ has quit IRC | 09:58 | |
| jd__ | sileht: you know if in oslo.messaging there's a polling function? | 10:13 |
| sileht | jd__, no evernever thing is think async | 10:16 |
| sileht | evernever/everything | 10:16 |
| openstackgerrit | Mehdi Abaakouk proposed a change to openstack/ceilometer: Replace oslo.rpc by oslo.messaging https://review.openstack.org/57457 | 10:17 |
| *** SergeyLukjanov is now known as SergeyLukjanov_ | 10:20 | |
| jd__ | sileht: ok so you have to call something to consume() and it blocks I imagine | 10:21 |
| jd__ | sounds like the same crap | 10:21 |
| *** yassine has quit IRC | 10:22 | |
| *** yassine has joined #openstack-ceilometer | 10:22 | |
| sileht | jd__, no you need to register callback to get the notification/rpc message | 10:23 |
| sileht | jd__, the oslo.messaging executor, must fit in your eventloop | 10:23 |
| jd__ | sileht: how the read() handled then? | 10:23 |
| jd__ | ah ok | 10:23 |
| jd__ | that's part of the executor too, I though it was just for request | 10:23 |
| sileht | jd__, blocking block on read, eventlet spawn a greenlet, | 10:23 |
| jd__ | sileht: ah I see, perfect | 10:24 |
| jd__ | so that would solve our problem in this case | 10:24 |
| sileht | jd__, but (because their always have a but) | 10:25 |
| jd__ | :( | 10:25 |
| sileht | jd__, if we create a asyncio executor for example | 10:25 |
| sileht | the executor can retreive the event loop and add the read/write task, but , kombu is a blocking library, so how to deal with taht | 10:26 |
| sileht | eventlet monkeypatch, but what for asyncio ? | 10:26 |
| jd__ | sileht: well if you do a write()/read() operation it'll block yeah | 10:27 |
| jd__ | e.g. if komby sends a request and wait for the reply | 10:27 |
| jd__ | that's the main problem with library not designed in a async mannter | 10:27 |
| jd__ | that's also why I keep explaining that async and event loop are different things | 10:28 |
| jd__ | so the executor is about integrating in the event loop, not doing async | 10:28 |
| sileht | jd__, yep | 10:28 |
| jd__ | in the end I don't care if kombu blocks, I prefer to have something safe that something with concurrency where it has not been designed for | 10:28 |
| sileht | jd__, latest kombu version have an exprimental async API but the oslo.messaging don't use it | 10:29 |
| jd__ | in the end I'm pretty sure we could replace the current eventlet greenthread by real threads + a big lock to be sure only one thread run at the same time | 10:29 |
| jd__ | sileht: yeah oslo.messaging has not been designed in async manner so anyway we're good to keep working on that | 10:29 |
| jd__ | (unless I'm wrong and I'd be happy to be :)) | 10:30 |
| sileht | jd__, asyncio can deal a bit with block io, by starting a new (real) thread for block IO | 10:33 |
| sileht | jd__, I think the oslo.messaging executor can use that, but I need to PoC the thing | 10:33 |
| openstackgerrit | A change was merged to openstack/ceilometer: Update oslo https://review.openstack.org/69903 | 10:38 |
| Alexei_987 | sileht: > new (real) thread for block IO This won't work | 10:42 |
| Alexei_987 | it depends on libraries that do blocking IO | 10:42 |
| Alexei_987 | sileht: not all of the release GIL | 10:42 |
| Alexei_987 | sileht: I guess it's better to try something like nova approach - spawn some number of worker processes | 10:43 |
| Alexei_987 | s/of the/of them/ | 10:43 |
| *** SergeyLukjanov_ is now known as SergeyLukjanov | 10:50 | |
| openstackgerrit | Mehdi Abaakouk proposed a change to openstack/ceilometer: Replace oslo.rpc by oslo.messaging https://review.openstack.org/57457 | 10:52 |
| *** SergeyLukjanov is now known as SergeyLukjanov_a | 10:53 | |
| *** SergeyLukjanov_a is now known as SergeyLukjanov_ | 10:54 | |
| *** SergeyLukjanov_ is now known as SergeyLukjanov | 10:59 | |
| openstackgerrit | A change was merged to openstack/ceilometer: Correct a misuse of RestController in the Event API https://review.openstack.org/69885 | 11:08 |
| *** SergeyLukjanov is now known as SergeyLukjanov_ | 11:11 | |
| jd__ | sileht: I don't think it'll help, since if you call a function like requests.get() that block because it does write()/read(), there's no way having a thread here is going to going to help since you expect the result as a return value :) | 11:27 |
| sileht | jd__, in oslo.messaging you don't want a return value, but passing the value into a callback (ran in the main thread) | 11:33 |
| jd__ | sileht: which is exactly the same actually | 11:34 |
| jd__ | I mean it's going to block | 11:34 |
| jd__ | unless you spawn a thread, and now you program is multithreaded | 11:34 |
| jd__ | which is NOT WHAT I WANT | 11:34 |
| jd__ | :) | 11:35 |
| sileht | jd__, ack, so I have no clue | 11:35 |
| jd__ | sileht: oh I'm happy with your answer about oslo.messaging, so you have a clue | 11:35 |
| jd__ | I mean the impl_blocking is exactly what I'd want | 11:35 |
| jd__ | my main concern is that we can't use oslo.messaging yet | 11:36 |
| jd__ | I'm trying to find a solution but it's like playing with explosives | 11:36 |
| openstackgerrit | Mehdi Abaakouk proposed a change to openstack/ceilometer: Replace oslo.rpc by oslo.messaging https://review.openstack.org/57457 | 11:43 |
| eglynn | http://en.wikipedia.org/wiki/File:HLposterUSA2.jpg | 11:43 |
| sileht | :) | 11:44 |
| openstackgerrit | Mehdi Abaakouk proposed a change to openstack/ceilometer: Replace oslo.rpc by oslo.messaging https://review.openstack.org/57457 | 11:50 |
| *** llu has quit IRC | 11:58 | |
| *** tian has quit IRC | 11:58 | |
| *** llu has joined #openstack-ceilometer | 11:58 | |
| *** tian has joined #openstack-ceilometer | 11:58 | |
| openstackgerrit | Mehdi Abaakouk proposed a change to openstack/ceilometer: Replace oslo.rpc by oslo.messaging https://review.openstack.org/57457 | 11:58 |
| openstackgerrit | Mehdi Abaakouk proposed a change to openstack/ceilometer: Replace oslo.rpc by oslo.messaging https://review.openstack.org/57457 | 12:04 |
| eglynn | super-dumb question on the conversion logic from DB model -> API representation | 12:17 |
| eglynn | https://github.com/openstack/ceilometer/blob/master/ceilometer/api/controllers/v2.py#L160 | 12:17 |
| *** vrovachev has joined #openstack-ceilometer | 12:17 | |
| eglynn | surely using wsme.Unset as the sentinel requires that the storage layer knows about WSME in order to supress an attribute? | 12:17 |
| eglynn | ... that smells wrong to me, why not just use None for the same purpose? | 12:18 |
| eglynn | ... maybe /me just hasn't had enough coffee yet this morning | 12:18 |
| sileht | eglynn, I think wsme.Unset means user haven't set it through API, None means user have set the an attribut to None | 12:20 |
| eglynn | sileht: yeah, still seems odd to leak wsme.Unset into the storage layer | 12:21 |
| sileht | so if the attribut is None we pass it to the storage, if it is wsme.Unset we don't pass it to the storage | 12:21 |
| eglynn | sileht: I'm talking about the conversion on the return path, i.e. the opposite direction (storage->API) | 12:22 |
| sileht | eglynn, ok ok I see | 12:22 |
| sileht | eglynn, this is because as_dict() is used for both direction | 12:23 |
| sileht | eglynn, but wsme.Unset is only valid for one direction | 12:23 |
| eglynn | sileht: yeah, maybe just need to make the DB model constructors more flexible | 12:24 |
| eglynn | ... so as to allow some attributes to remain unset (i.e. as opposed to being explicitly set to None) | 12:24 |
| * sileht loose between as_dict() from the storage model and as_dict wsme base type | 12:26 | |
| sileht | eglynn, I confirm the method as_dict() L160 is for "API -> storage" (that removes wsme.Unset attributes), opposite direction is done by the storage model object as_dict() method | 12:30 |
| openstackgerrit | Mehdi Abaakouk proposed a change to openstack/ceilometer: Replace oslo.rpc by oslo.messaging https://review.openstack.org/57457 | 12:35 |
| eglynn | sileht: yeah, looking at the code again, you seem to be correct | 12:36 |
| eglynn | sileht: though weirdly, I found that setting the unwanted attributes to wsme.Unset within the response storage model did have the result I expected originally | 12:37 |
| eglynn | (i.e. those attributes were dropped from the API representation in the response) | 12:37 |
| eglynn | ... meh, /me has gone off-piste somewhere | 12:37 |
| andreykurilin | Hi everyone, I'm working on bp db-sql-tests-migrations-refactoring and and I have question about it. | 12:43 |
| andreykurilin | When all the unit tests are run at a real MySQL and a real PostgreSQL servers, MySQL creates Indexes for each Foreign key because it can't create fk without index, PosgreSQL can create Foreign keys without Index, so it doesn't create indexies. Therefore, we have different db structure on different backends. | 12:43 |
| andreykurilin | One of the ways to avoid failing test due to difference between DBs on backends is to add validation of relation Index to Foreign key in MySQL, other way is to add Indexes to PostgreSQL via migrations. | 12:43 |
| andreykurilin | what are your thoughts on this? | 12:44 |
| vrovachev | hi all | 12:44 |
| *** ildikov_ has quit IRC | 12:44 | |
| vrovachev | help me, i have this problem with ceilo: http://paste.openstack.org/show/62235/ | 12:45 |
| jd__ | andreykurilin: I think we should simply ignore things we didn't create, no? | 12:47 |
| andreykurilin | jd__: I think so too. but I do not like the fact that we will have different databases in MySQL and PostgreSQL | 12:50 |
| *** SergeyLukjanov_ is now known as SergeyLukjanov | 12:54 | |
| *** SergeyLukjanov is now known as SergeyLukjanov_a | 12:54 | |
| *** SergeyLukjanov_a is now known as SergeyLukjanov | 12:54 | |
| jd__ | andreykurilin: you will always have different things because they are different engine | 12:55 |
| jd__ | with different types, etc | 12:55 |
| *** SergeyLukjanov is now known as SergeyLukjanov_a | 12:55 | |
| jd__ | learning to deal with it is better than layering everything to the bottom | 12:55 |
| andreykurilin | ok, thanks | 12:55 |
| openstackgerrit | Julien Danjou proposed a change to openstack/ceilometer: Revert "Revert "Ensure we are not exhausting the sqlalchemy pool"" https://review.openstack.org/70337 | 12:56 |
| openstackgerrit | Julien Danjou proposed a change to openstack/ceilometer: service: re-enable eventlet just for sockets https://review.openstack.org/70338 | 12:56 |
| *** SergeyLukjanov_a is now known as SergeyLukjanov | 12:56 | |
| jd__ | sileht: if you have an opinion ^ | 12:56 |
| sileht | jd__, socket only ? CTRL+C work with that ? you don't need to patch threading too ? | 12:56 |
| jd__ | sileht: no, C-c is not fixed by that | 12:57 |
| jd__ | I think we need to revert my patch in oslo | 12:57 |
| sileht | jd__, or patch threading too | 12:57 |
| jd__ | sileht: I don't want to patch threading because it breaks libraries using threading | 12:57 |
| sileht | jd__, yep, I agree, the entire service.py and deps must be rewritte with an other event loop | 12:58 |
| * sileht is sad to see again its sqlalchemy pool hack :( | 13:00 | |
| *** SergeyLukjanov is now known as SergeyLukjanov_ | 13:00 | |
| jd__ | me too | 13:00 |
| *** AMike has quit IRC | 13:01 | |
| *** SergeyLukjanov_ is now known as SergeyLukjanov | 13:03 | |
| *** ruhe_ has joined #openstack-ceilometer | 13:09 | |
| *** ildikov_ has joined #openstack-ceilometer | 13:10 | |
| jd__ | sileht: https://review.openstack.org/#/c/70339/ | 13:16 |
| sileht | jd__, for the first I think if we patch socket we can patch everything, that don't change so much :( | 13:19 |
| jd__ | sileht: everything except threading? | 13:20 |
| sileht | I'm talking about https://review.openstack.org/70338 | 13:20 |
| jd__ | sileht: I feel better knowing that's the only place infected by greenlet | 13:20 |
| *** SergeyLukjanov is now known as SergeyLukjanov_ | 13:28 | |
| *** jdob has joined #openstack-ceilometer | 13:38 | |
| jd__ | sileht: if you got 5 minutes I'd be happy to have your opinion on https://review.openstack.org/#/c/70166/ failure wrt PostgreSQL, as I think you wrote the PreciseTimestamp in the first place | 13:43 |
| openstackgerrit | Julien Danjou proposed a change to openstack/ceilometer: storage: store recording timestamp https://review.openstack.org/70166 | 13:46 |
| sileht | jd__, DATETIME is a alias of Datetime in sqlalchemy | 13:48 |
| jd__ | lol | 13:50 |
| jd__ | dunnow why I though it was different | 13:50 |
| jd__ | though that's not the problem sileht, that's my trying to fix the failure | 13:50 |
| openstackgerrit | Julien Danjou proposed a change to openstack/ceilometer: storage: store recording timestamp https://review.openstack.org/70166 | 13:51 |
| sileht | jd__, lol I have looked into the code they are a small difference, the case ! | 13:53 |
| jd__ | sileht: http://logs.openstack.org/66/70166/1/check/check-tempest-dsvm-postgres-full/7bd1c1e/console.html | 13:54 |
| sileht | jd__, all other PreciseTimestamp are created with Datetime and then converted in PreciseTimestamp for mysql only | 13:56 |
| openstackgerrit | Julien Danjou proposed a change to openstack/ceilometer: service: re-enable eventlet just for sockets https://review.openstack.org/70338 | 13:56 |
| jd__ | sileht: except that 'datetime' does not exists in pg | 13:56 |
| sileht | it should be convert in timestamp in postgres | 13:57 |
| vrovachev | Guys, please help me with this problem: http://paste.openstack.org/show/62235/ | 13:57 |
| *** eglynn is now known as eglynn-afk | 13:58 | |
| vrovachev | I installed devstack of master branch with this localrc: http://paste.openstack.org/show/62239/ | 13:59 |
| *** SergeyLukjanov_ is now known as SergeyLukjanov | 14:00 | |
| jd__ | sileht: you're not helping ;-) | 14:01 |
| sileht | jd__, sqlalchemy but I think: https://github.com/zzzeek/sqlalchemy/blob/master/lib/sqlalchemy/dialects/postgresql/base.py#L1267 datetime is convert but not DATETIME | 14:02 |
| sileht | but/bug/ | 14:02 |
| jd__ | not sure it's a bug | 14:02 |
| sileht | jd__, in all case, you have right with your first tought, we must use Datetime and not DATETIME | 14:04 |
| jd__ | ack | 14:04 |
| jd__ | let's try that | 14:05 |
| openstackgerrit | Julien Danjou proposed a change to openstack/ceilometer: storage: store recording timestamp https://review.openstack.org/70166 | 14:05 |
| *** jmckind has joined #openstack-ceilometer | 14:11 | |
| *** SergeyLukjanov is now known as SergeyLukjanov_a | 14:26 | |
| *** SergeyLukjanov_a is now known as SergeyLukjanov_ | 14:26 | |
| *** tian has quit IRC | 14:36 | |
| *** llu has quit IRC | 14:37 | |
| *** ruhe_ has quit IRC | 14:38 | |
| *** ruhe has quit IRC | 14:41 | |
| openstackgerrit | Julien Danjou proposed a change to openstack/ceilometer: storage: store recording timestamp https://review.openstack.org/70166 | 14:43 |
| *** SergeyLukjanov_ is now known as SergeyLukjanov | 14:45 | |
| *** tian has joined #openstack-ceilometer | 14:52 | |
| *** llu has joined #openstack-ceilometer | 14:52 | |
| *** SergeyLukjanov is now known as SergeyLukjanov_ | 15:04 | |
| eglynn-afk | sileht: d'oh, I see my confusion now ... apols for the noise earlier | 15:12 |
| *** eglynn-afk is now known as eglynn | 15:13 | |
| eglynn | turns out Statistics is a special case, we don't even use the v2._Base.from_db_model() to create it | 15:13 |
| eglynn | (unlike the other API response representations) | 15:13 |
| eglynn | so the Unset is just passed straight thru and then dropped by WSME when marshalling up the response | 15:13 |
| *** rwsu has joined #openstack-ceilometer | 15:13 | |
| eglynn | ... anyhoo, alls that's needed is to make the Statistics model ctor a tad more flexible in terms of missing args | 15:14 |
| *** SergeyLukjanov_ is now known as SergeyLukjanov | 15:14 | |
| *** SergeyLukjanov is now known as SergeyLukjanov_ | 15:15 | |
| *** kwhitney has joined #openstack-ceilometer | 15:17 | |
| *** SergeyLukjanov_ is now known as SergeyLukjanov | 15:20 | |
| *** sayali__ has quit IRC | 15:24 | |
| *** SergeyLukjanov is now known as SergeyLukjanov_ | 15:29 | |
| *** prad_ has joined #openstack-ceilometer | 15:29 | |
| *** ruhe has joined #openstack-ceilometer | 15:31 | |
| *** ruhe is now known as ruhe_ | 15:31 | |
| *** SergeyLukjanov_ is now known as SergeyLukjanov | 15:34 | |
| *** ruhe_ has quit IRC | 15:36 | |
| *** SergeyLukjanov is now known as SergeyLukjanov_a | 15:43 | |
| *** SergeyLukjanov_a is now known as SergeyLukjanov_ | 15:44 | |
| sileht | jd__, what do you think about https://review.openstack.org/#/c/68939/ | 15:47 |
| *** SergeyLukjanov_ is now known as SergeyLukjanov | 15:54 | |
| *** SergeyLukjanov is now known as SergeyLukjanov_ | 15:56 | |
| *** SergeyLukjanov_ is now known as SergeyLukjanov | 16:04 | |
| *** SergeyLukjanov is now known as SergeyLukjanov_a | 16:19 | |
| *** SergeyLukjanov_a is now known as SergeyLukjanov_ | 16:20 | |
| openstackgerrit | Mehdi Abaakouk proposed a change to openstack/python-ceilometerclient: Ensure url sent to proxy don't have redondant / https://review.openstack.org/70368 | 16:21 |
| *** ildikov_ has quit IRC | 16:21 | |
| *** SergeyLukjanov_ is now known as SergeyLukjanov | 16:24 | |
| openstackgerrit | Andreas Jaeger proposed a change to openstack/python-ceilometerclient: Improve help strings https://review.openstack.org/70370 | 16:28 |
| *** SergeyLukjanov is now known as SergeyLukjanov_a | 16:39 | |
| *** SergeyLukjanov_a is now known as SergeyLukjanov_ | 16:40 | |
| *** SergeyLukjanov_ is now known as SergeyLukjanov | 16:41 | |
| *** kwhitney has left #openstack-ceilometer | 16:41 | |
| *** viktors has left #openstack-ceilometer | 16:45 | |
| *** SergeyLukjanov is now known as SergeyLukjanov_ | 16:51 | |
| *** julienvey_ has quit IRC | 16:53 | |
| *** SergeyLukjanov_ is now known as SergeyLukjanov | 16:54 | |
| *** doug-fish has joined #openstack-ceilometer | 16:54 | |
| *** SergeyLukjanov is now known as SergeyLukjanov_ | 16:55 | |
| *** xmltok has joined #openstack-ceilometer | 16:56 | |
| *** vrovachev has left #openstack-ceilometer | 17:04 | |
| *** ildikov_ has joined #openstack-ceilometer | 17:05 | |
| *** yassine has quit IRC | 17:13 | |
| *** yassine has joined #openstack-ceilometer | 17:14 | |
| *** yassine has quit IRC | 17:16 | |
| *** yassine has joined #openstack-ceilometer | 17:16 | |
| *** yassine has quit IRC | 17:16 | |
| *** yassine has joined #openstack-ceilometer | 17:17 | |
| *** yassine has quit IRC | 17:18 | |
| *** yassine has joined #openstack-ceilometer | 17:19 | |
| *** yassine has quit IRC | 17:21 | |
| *** yassine has joined #openstack-ceilometer | 17:22 | |
| *** sayali__ has joined #openstack-ceilometer | 17:25 | |
| *** yassine has quit IRC | 17:27 | |
| *** yassine has joined #openstack-ceilometer | 18:27 | |
| *** KresiusMengg has joined #openstack-ceilometer | 18:55 | |
| *** KresiusMengg has left #openstack-ceilometer | 18:55 | |
| *** jaypipes has quit IRC | 18:59 | |
| *** eglynn has quit IRC | 19:04 | |
| *** jaypipes has joined #openstack-ceilometer | 19:09 | |
| *** eglynn has joined #openstack-ceilometer | 19:23 | |
| *** julienvey_ has joined #openstack-ceilometer | 19:53 | |
| *** julienvey_ has quit IRC | 19:53 | |
| *** yassine has quit IRC | 20:12 | |
| *** yassine has joined #openstack-ceilometer | 20:12 | |
| dhellmann | jd__: ping? | 20:32 |
| jd__ | dhellmann: ? | 20:32 |
| dhellmann | jd__: do you still need that pbr release? I'm trying to catch mordred to see if he has any objections | 20:33 |
| jd__ | yes | 20:33 |
| dhellmann | jd__: the gate runs from source, so where do you need the release? it's late enough in the afternoon on a friday that I'd like to hold off until monday if that's not a problem | 20:37 |
| jd__ | I don't understand the question but I'm ok to wait monday | 20:38 |
| dhellmann | jd__: why do you need a release? | 20:38 |
| jd__ | dhellmann: to depends on it and add the support for python 3 tests in oslo-incubator | 20:38 |
| dhellmann | ok, right, the tests not in the gate don't run from source | 20:39 |
| dhellmann | I was thinking about the fact that devstack gate uses the git version of pbr, not a pypi package | 20:39 |
| dhellmann | ignore me | 20:39 |
| jd__ | got it :) | 20:39 |
| jd__ | yeah I do need a release, but no hurry, I can prepare a patch without that for monday anyway | 20:39 |
| dhellmann | jd__: will email monty about doing a release, and go ahead and tag it monday morning | 20:39 |
| *** sayali__ has quit IRC | 20:40 | |
| jd__ | thumbs up | 20:40 |
| *** thomasem has joined #openstack-ceilometer | 20:50 | |
| *** thomasem has quit IRC | 21:39 | |
| *** jdob has quit IRC | 21:40 | |
| *** openstackgerrit has quit IRC | 21:53 | |
| *** openstackgerrit has joined #openstack-ceilometer | 21:53 | |
| openstackgerrit | Eoghan Glynn proposed a change to openstack/ceilometer: Selectable aggregation functions for statistics [WIP] https://review.openstack.org/70454 | 22:06 |
| *** doug-fish has quit IRC | 22:17 | |
| *** yassine has quit IRC | 22:26 | |
| *** jmckind has quit IRC | 23:04 | |
| *** julienvey_ has joined #openstack-ceilometer | 23:21 | |
| *** julienvey_ has quit IRC | 23:25 | |
| *** julienvey_ has joined #openstack-ceilometer | 23:26 | |
| *** prad_ has quit IRC | 23:31 | |
| *** julienvey_ has quit IRC | 23:33 | |
| *** eglynn has quit IRC | 23:37 | |
| *** Alexei_987 has quit IRC | 23:41 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
