| *** Trozz has joined #openstack-ceilometer | 00:03 | |
| *** matsuhashi has joined #openstack-ceilometer | 00:27 | |
| *** terriyu has quit IRC | 00:58 | |
| *** terriyu has joined #openstack-ceilometer | 01:04 | |
| *** terriyu has quit IRC | 01:04 | |
| *** changbl has joined #openstack-ceilometer | 01:12 | |
| *** matsuhashi has quit IRC | 01:12 | |
| *** matsuhashi has joined #openstack-ceilometer | 01:14 | |
| *** sdake_ has quit IRC | 01:15 | |
| *** sdake has quit IRC | 01:16 | |
| *** kun_huang has joined #openstack-ceilometer | 01:23 | |
| *** liusheng has joined #openstack-ceilometer | 01:23 | |
| *** nosnos has joined #openstack-ceilometer | 01:27 | |
| *** sdake has joined #openstack-ceilometer | 01:35 | |
| *** sdake_ has joined #openstack-ceilometer | 01:35 | |
| *** _cjones_ has quit IRC | 01:37 | |
| *** alexpilotti has quit IRC | 01:47 | |
| *** flwang has quit IRC | 02:00 | |
| *** matsuhashi has quit IRC | 02:00 | |
| *** matsuhashi has joined #openstack-ceilometer | 02:08 | |
| *** anamalagon has joined #openstack-ceilometer | 02:39 | |
| *** flwang has joined #openstack-ceilometer | 02:41 | |
| *** nati_ueno has quit IRC | 02:42 | |
| *** thomasem has joined #openstack-ceilometer | 02:53 | |
| *** anamalagon has quit IRC | 03:05 | |
| *** matsuhashi has quit IRC | 03:15 | |
| *** nosnos has quit IRC | 03:28 | |
| *** thomasem has quit IRC | 03:39 | |
| *** nati_ueno has joined #openstack-ceilometer | 04:02 | |
| *** mihgen has joined #openstack-ceilometer | 04:22 | |
| *** matsuhashi has joined #openstack-ceilometer | 04:25 | |
| *** nosnos has joined #openstack-ceilometer | 04:36 | |
| *** _nadya_ has joined #openstack-ceilometer | 05:02 | |
| *** _nadya_ has quit IRC | 05:03 | |
| *** flwang1 has joined #openstack-ceilometer | 05:27 | |
| *** kun_huang has quit IRC | 05:28 | |
| *** flwang has quit IRC | 05:29 | |
| *** _nadya_ has joined #openstack-ceilometer | 05:30 | |
| *** ildikov_ has quit IRC | 05:52 | |
| openstackgerrit | Piyush Masrani proposed a change to openstack/ceilometer: Documenting hypervisor support for nova meters https://review.openstack.org/79910 | 05:57 |
|---|---|---|
| *** nati_ueno has quit IRC | 05:57 | |
| *** _nadya_ has quit IRC | 06:00 | |
| openstackgerrit | Jenkins proposed a change to openstack/ceilometer: Imported Translations from Transifex https://review.openstack.org/76762 | 06:04 |
| *** sdake has quit IRC | 06:12 | |
| *** sdake_ has quit IRC | 06:12 | |
| *** saju_m has joined #openstack-ceilometer | 06:19 | |
| *** sdake has joined #openstack-ceilometer | 06:24 | |
| *** sdake has quit IRC | 06:24 | |
| *** sdake has joined #openstack-ceilometer | 06:24 | |
| *** sdake_ has joined #openstack-ceilometer | 06:27 | |
| *** matsuhashi has quit IRC | 06:33 | |
| *** matsuhashi has joined #openstack-ceilometer | 06:35 | |
| *** eglynn has joined #openstack-ceilometer | 06:53 | |
| *** liusheng has quit IRC | 06:55 | |
| *** matsuhas_ has joined #openstack-ceilometer | 07:03 | |
| *** flwang1 has quit IRC | 07:03 | |
| *** matsuhas_ has quit IRC | 07:04 | |
| *** flwang has joined #openstack-ceilometer | 07:06 | |
| *** matsuhas_ has joined #openstack-ceilometer | 07:06 | |
| *** matsuhashi has quit IRC | 07:07 | |
| *** ildikov_ has joined #openstack-ceilometer | 07:11 | |
| *** gibi has joined #openstack-ceilometer | 07:11 | |
| *** saju_m has quit IRC | 07:13 | |
| *** liusheng has joined #openstack-ceilometer | 07:28 | |
| *** inc0 has joined #openstack-ceilometer | 07:29 | |
| eglynn | jd__: good morning! | 07:34 |
| jd__ | hi eglynn | 07:34 |
| eglynn | jd__: if you got a spare sec, could you consider restoring your +1 on https://review.openstack.org/82672 ? | 07:34 |
| eglynn | (patch set 2 is just a test improvement, basic fix is unchanged) | 07:34 |
| jd__ | let me check | 07:34 |
| eglynn | jd__: I spoke too soon! :) | 07:35 |
| jd__ | I think so :) | 07:35 |
| eglynn | ildikov_: ta much! | 07:35 |
| jd__ | ildikov_: is fast | 07:35 |
| eglynn | jd__: apols for the noise :) | 07:35 |
| jd__ | np :) | 07:35 |
| ildikov_ | eglynn, jd__: good morning guys :) | 07:37 |
| eglynn | ildikov_: good morning! | 07:37 |
| ildikov_ | eglynn: please check my comment on the second patch set | 07:37 |
| eglynn | ildikov_: looking now | 07:37 |
| eglynn | ildikov_: good point on the volume > 1 being unrealistic for the instance meter | 07:39 |
| eglynn | ildikov_: though in this case, shouldn't impact on what test is seeking to achieve | 07:40 |
| jd__ | FYI rc1 is targetted for Friday | 07:40 |
| eglynn | ildikov_: ... as long as count(sample) > 100 while count(distinct(resource_id)) > 100, the volume won't really matter than much | 07:40 |
| ildikov_ | eglynn: I know that it will not affect the test flow, but after our conversation about testing, I think it is important to use realistic test data (both for positive and negative test cases) | 07:40 |
| eglynn | ildikov_: you're absolutely correct! I'll fix next time my nose is in that code :) | 07:41 |
| ildikov_ | jd__: cool, thanks for the info | 07:41 |
| eglynn | jd__: yeah just reading the backlog for the project/release meeting last night | 07:41 |
| eglynn | jd__: ... that should be doable, or? | 07:42 |
| ildikov_ | eglynn: sorry for arguing so much about tests | 07:43 |
| eglynn | ildikov_: ... nah, you were absolutely correct to point out that issue :) | 07:43 |
| jd__ | eglynn: I hope so | 07:44 |
| eglynn | jd__: I'll line up a ceiloclient release with RC1, there are about 2 more client patches I'd like to see land first | 07:45 |
| eglynn | (... time-constrained alarms and selectable aggregates) | 07:46 |
| jd__ | great | 07:46 |
| ildikov_ | eglynn: I will try to look at those patches today, but I cannot promise anything, I started it yesterday, but I got distracted :( | 07:48 |
| ildikov_ | eglynn: as for the selectable aggregates patch, I haven't seen any test for checking the printed format of the returned data, I assume there should be a test like that, am I correct? | 07:49 |
| *** mihgen has quit IRC | 07:50 | |
| eglynn | ildikov_: for the client tests, there isn't universal coverage of the formatted output of shell commands | 07:53 |
| eglynn | ildikov_: ... usually the pattern is to assert on the returned representation instead IIRC | 07:54 |
| eglynn | ildikov_: ... but in some cases (such as alarm-history) a more complete assertion on the formatted output is done | 07:54 |
| ildikov_ | eglynn: I meant the table format with those ugly assertEqual checks in the test cases | 07:55 |
| eglynn | ildikov_: ... a-ha, ok, the expected-actual ordering issue, do you mean? | 07:55 |
| *** mihgen has joined #openstack-ceilometer | 07:57 | |
| * eglynn is not sure what ugly means in this context ... | 07:58 | |
| ildikov_ | eglynn: I'm talking about the test_shell.py | 07:58 |
| *** matsuhas_ has quit IRC | 07:59 | |
| eglynn | ildikov_: ... a-ha ok, that's kinda what I understood also originally | 07:59 |
| inc0 | ildikov_, true, but well, there is no test for statistics shell call at all | 08:00 |
| ildikov_ | eglynn: there are several query output tests | 08:00 |
| eglynn | ildikov_: so only a representative subset of commands are covered by https://github.com/openstack/python-ceilometerclient/blob/master/ceilometerclient/tests/v2/test_shell.py | 08:00 |
| inc0 | I think thats separate matter, but yes, it is an issue | 08:00 |
| ildikov_ | inc0: it's not a good excuse that there isn't any test currently ;) | 08:01 |
| eglynn | ildikov_: ... but you are correct, we should strive to have more complete coverage there | 08:01 |
| eglynn | ildikov_: ... ugliness notwithstanding ;) | 08:01 |
| inc0 | ildikov_, can't argue with that;) all I'm saying is that should be separate patch imho | 08:01 |
| inc0 | also, we should check if there are any more non-tested methods out there | 08:02 |
| inc0 | I'll run coverage over it | 08:03 |
| *** matsuhashi has joined #openstack-ceilometer | 08:04 | |
| ildikov_ | inc0: I can agree with that it can be a separate patch | 08:04 |
| * eglynn loves the sweet, sweet smell of an emergant consensus ... :) | 08:06 | |
| jd__ | eglynn: https://bugs.launchpad.net/ceilometer/+bug/1297677 wtf? | 08:07 |
| ildikov_ | inc0: it would be good, if you could run a coverage test, I had time only had look at Ceilometer's test coverage report without the client code | 08:07 |
| eglynn | jd__: ... looking | 08:07 |
| inc0 | eglynn, btw, about this trace I got yesteday = AttributeError at https://github.com/openstack/ceilometer/blob/master/ceilometer/compute/util.py#L43 , I think I spotted an bug. Not sure why, but for time to time (every hour to be more exact) I get notification, in which message['payload']['metadata'] == [] | 08:10 |
| inc0 | instead of for example {} | 08:10 |
| eglynn | jd__: ... ickk, that looks very wrong | 08:11 |
| eglynn | jd__: ... I'll need to dig thru the git fossil record to figure out how that happened | 08:11 |
| eglynn | jd__: ... that small fragment of code around alarm ownership/visibility was refactored multiple times IIRC | 08:11 |
| eglynn | inc0: ... only for notifications originating from nova? | 08:12 |
| eglynn | inc0: ... that could be a bug in the nova instance audit logic | 08:12 |
| inc0 | maybe its due to our not-too-fresh nova version | 08:13 |
| jd__ | eglynn: I'm surprised we don't have tests for that | 08:13 |
| eglynn | inc0: ... that generates instance.exists notification every hour, day, week (cadence based on nova config) | 08:13 |
| eglynn | jd__: ... yep, me too | 08:13 |
| *** mihgen has quit IRC | 08:13 | |
| eglynn | jd__: ... I'll confirm it this morning when I get in to the office | 08:13 |
| ildikov_ | eglynn, jd__: there are post alarm test cases only with admin role, if I see right, so it seems that nothing protected that part of the code | 08:13 |
| jd__ | eglynn: cool thanks | 08:13 |
| inc0 | still, I think we should check type before parsing it, bug or no bug in nova | 08:14 |
| ildikov_ | eglynn, jd__: but I will have a deeper look in the test code | 08:14 |
| eglynn | inc0: yes you are correct, we need to tolerant any chaotic variations in notification structure that the other services feel like throwing at us | 08:17 |
| jd__ | inc0: eglynn: that's why there was a plan to schematize these notif :( | 08:17 |
| eglynn | inc0: can you file a couple of bugs, one definite for ceilo and one tentatively for nova? | 08:17 |
| inc0 | I'll report one for ceilo and look if its known issue in nova | 08:18 |
| eglynn | jd__: yeah, but I thought we discussed that and concluded we'd just have to be tolerant of variation anyway? | 08:18 |
| eglynn | inc0: cool, thanks! | 08:18 |
| jd__ | eglynn: no actually there's a bp in oslo.messaging to do that | 08:18 |
| jd__ | just didn't have time to push it more so far | 08:18 |
| eglynn | jd__: a-ha, cool | 08:19 |
| ildikov_ | jd__: which is that BP in oslo.messaging? | 08:19 |
| inc0 | https://bugs.launchpad.net/ceilometer/+bug/1297686 | 08:20 |
| openstackgerrit | Uros Jovanovic proposed a change to openstack/python-ceilometerclient: Adds alarm time constraint support to ceilometer CLI https://review.openstack.org/78356 | 08:25 |
| jd__ | ildikov_: https://blueprints.launchpad.net/oslo.messaging/+spec/notification-structured | 08:26 |
| inc0 | jd__, it doesn't seem to have much activity in it :( | 08:29 |
| ildikov_ | jd__: thanks, I will check it, as I have a bug about inconsitently handled metadata, which would need some centralized handling of this data too | 08:29 |
| openstackgerrit | A change was merged to openstack/ceilometer: Avoid swallowing AssertionError in test skipping logic https://review.openstack.org/82297 | 08:29 |
| openstackgerrit | A change was merged to openstack/ceilometer: Ensure idempotency of cardinality reduction in mongo https://review.openstack.org/82672 | 08:29 |
| jd__ | ildikov_: great | 08:30 |
| ildikov_ | jd__: I had a look at CADF events, but metadata is an unhandled part of the payload in CADF too, I was wondering if we could have some metadata model or something to handle this | 08:32 |
| ildikov_ | jd__: anyway, I will try to find some time in schedule to take a deeper look at, and then we will see, what could be the next step to push this forward | 08:34 |
| openstackgerrit | A change was merged to openstack/ceilometer: spawn multiple workers in services https://review.openstack.org/79962 | 08:38 |
| *** Ruetobas has quit IRC | 08:40 | |
| *** admin0 has joined #openstack-ceilometer | 08:40 | |
| admin0 | hello all | 08:40 |
| admin0 | i was able to write a python daemon that takes the ceilometer stats and sends it to graphite | 08:43 |
| jd__ | ildikov_: well part of the point is not to have metadata anymore, but thigns structured in notifications – so that would be structured data rather than random key/values | 08:43 |
| admin0 | now i am looking for a way to create it as an adapter | 08:43 |
| openstackgerrit | A change was merged to openstack/ceilometer: fix create_or_update logic to avoid rollbacks https://review.openstack.org/81969 | 08:43 |
| ildikov_ | jd__: you mean that there would be no resource/user metadata? | 08:44 |
| ildikov_ | jd__: when I checked the CADF related BPs and wiki pages it did not solve this problem, but if the goal is to move to the metadataless direction and include the data into the notifications in a structured way, I'm not against it | 08:45 |
| *** inc0 has quit IRC | 08:47 | |
| *** Ruetobas has joined #openstack-ceilometer | 08:47 | |
| ildikov_ | jd__: as for the alarm bug, I modified one of the tests to set the role to "tenant" instead of "admin" and I got back a nice Unauthorized error | 08:48 |
| *** inc0 has joined #openstack-ceilometer | 08:49 | |
| *** Ruetobas has quit IRC | 08:51 | |
| *** Ruetobas has joined #openstack-ceilometer | 08:52 | |
| *** mihgen has joined #openstack-ceilometer | 09:06 | |
| eglynn | ildikov_: IIUC I've confirmed the alarm bug in devstack | 09:08 |
| *** yassine has joined #openstack-ceilometer | 09:08 | |
| *** yassine has quit IRC | 09:08 | |
| *** yassine has joined #openstack-ceilometer | 09:08 | |
| eglynn | ildikov_: ... create an alarm with user_id & project_id explicitly set to an admin user_id/project_id that differs from the current non-admin user's identity | 09:09 |
| eglynn | ildikov_: ... in that case the alarm is created with the user_id & project_id of the current user, as opposed to failing with an explicit auth error | 09:11 |
| eglynn | ildikov_, jd__: ^^^ thinking about it some more, it's not outrageously bad behaviour, as there's no leakage of privilege involved | 09:12 |
| eglynn | ... but it does go against the "principle of least astonishment" ;) | 09:13 |
| *** massi_ has joined #openstack-ceilometer | 09:17 | |
| eglynn | ... also inconsistent with the behaviour seen if the alarm query constrains on the admin project_id | 09:18 |
| jd__ | yeah that sounds weird | 09:19 |
| eglynn | (i.e. that case fails as expected with 401 "Not Authorized to access project $ADMIN_TENANT_ID" | 09:19 |
| eglynn | jd__: target to RC1 do you reckon? | 09:20 |
| eglynn | jd__: (... I can prepare a patch in case ZhiQiang Fan doesn't show up with a fix by EoD) | 09:20 |
| admin0 | hi eglynn , i was able to write a python script that formats the messages from udp and push to graphite .. now instead of running another daemon, i want to create a graphite publisher .. .. would adding graphite.py be enough for ceilometer to recognise this publisher which I can use from pipeline ? | 09:21 |
| ildikov_ | eglynn: a-ha, ok, I had a meeting in the meantime, so I did not have time for the specific cases, I just verified the regular behavior :) | 09:21 |
| eglynn | ildikov_: cool, np! | 09:21 |
| eglynn | admin0: I guess that would depend on what's in your graphite.py | 09:22 |
| admin0 | i am copying the udp.py and formatting it send directly to graphite | 09:22 |
| eglynn | admin0: but yeah the general idea is that the publishers are discovered and automatically loaded if the module exists | 09:23 |
| ildikov_ | eglynn: I think it's another point on the side of talking about testing on the summit, do you agree? :) | 09:23 |
| eglynn | ildikov_: ... the lack of coverage for that particular alarm creation usecase? | 09:24 |
| eglynn | ildikov_: ... sure, one of the many gaps ... I guess shows up the limitation of measuring coverage in line-counts | 09:24 |
| ildikov_ | eglynn: yes, I will collect the list of test issues occurred in the past one or two weeks | 09:24 |
| eglynn | ildikov_: nice one! thanks | 09:25 |
| ildikov_ | eglynn: what about sharing a session about this and using your preso and my list? | 09:25 |
| eglynn | ildikov_: a capital idea! :) | 09:26 |
| admin0 | eglynn: so i don't need to declare my class graphite in setup.cfg under celimoeter.publisher , but just leave the graphite.py file insider publishers and it will pick it up automatically ? | 09:26 |
| eglynn | ildikov_: ... do you wanna throw a proposal up on http://summit.openstack.org/ ? | 09:27 |
| ildikov_ | eglynn: with gibi we are also on writing up some gate job proposals for coverage check and some kind of spell checking, so we will have some items to improve on this area | 09:27 |
| ildikov_ | eglynn: sure, I just wanted to ask, if it is ok with you, if I throw up one :) | 09:27 |
| eglynn | ildikov_: cool, thanks! | 09:27 |
| ildikov_ | eglynn: /me thanks :) | 09:27 |
| eglynn | admin0: yeah, sure, that's the way all stevedore extensions are loaded by ceilo | 09:28 |
| ildikov_ | eglynn: I will register the proposal for this today | 09:28 |
| *** Trozz has quit IRC | 09:28 | |
| eglynn | admin0: ... i.e. the entry point is defined so that the extension name (e.g. "udp") can be associated with a concrete class (e.g. "ceilometer.publisher.udp:UDPPublisher") | 09:29 |
| eglynn | admin0: ... what I meant was that your code does not have be present in the ceilometer code tree | 09:29 |
| *** Trozz has joined #openstack-ceilometer | 09:29 | |
| eglynn | admin0: ... or in the ceilometer packages/RPMs that you've installed | 09:29 |
| *** admin0 has quit IRC | 09:29 | |
| eglynn | admin0: ... it just needs to be loadable from somewhere | 09:30 |
| *** Trozz_ has joined #openstack-ceilometer | 09:30 | |
| *** Trozz_ has quit IRC | 09:30 | |
| *** Trozz has quit IRC | 09:30 | |
| eglynn | ... /me scurries off on his commute, laters | 09:31 |
| *** Trozz has joined #openstack-ceilometer | 09:31 | |
| *** eglynn has quit IRC | 09:31 | |
| *** Trozz_ has joined #openstack-ceilometer | 09:31 | |
| *** mich5225 has joined #openstack-ceilometer | 09:32 | |
| *** Trozz has quit IRC | 09:33 | |
| *** Trozz_ has quit IRC | 09:33 | |
| *** mich5225 has quit IRC | 09:33 | |
| *** Trozz has joined #openstack-ceilometer | 09:33 | |
| *** flwang has quit IRC | 09:35 | |
| *** Trozz has quit IRC | 09:39 | |
| *** Trozz has joined #openstack-ceilometer | 09:40 | |
| *** Trozz has quit IRC | 09:40 | |
| *** Trozz has joined #openstack-ceilometer | 09:43 | |
| *** Trozz has quit IRC | 09:45 | |
| openstackgerrit | Ilya Tyaptin proposed a change to openstack/ceilometer: Fix order of arguments in assertEquals https://review.openstack.org/71595 | 09:46 |
| *** admin0 has joined #openstack-ceilometer | 09:49 | |
| admin0 | sorry .. got disconnected .. had a standup | 09:50 |
| *** admin0 has left #openstack-ceilometer | 09:50 | |
| *** admin0 has joined #openstack-ceilometer | 09:50 | |
| *** admin0 has quit IRC | 09:52 | |
| *** admin0 has joined #openstack-ceilometer | 10:06 | |
| *** matsuhashi has quit IRC | 10:09 | |
| admin0 | back again | 10:12 |
| *** eglynn has joined #openstack-ceilometer | 10:19 | |
| *** nacim has joined #openstack-ceilometer | 10:24 | |
| *** eoutin has joined #openstack-ceilometer | 10:24 | |
| *** Trozz has joined #openstack-ceilometer | 10:31 | |
| *** admin0 has quit IRC | 10:33 | |
| *** admin0 has joined #openstack-ceilometer | 10:33 | |
| *** Trozz has quit IRC | 10:34 | |
| *** admin0 has quit IRC | 10:37 | |
| *** matsuhashi has joined #openstack-ceilometer | 10:39 | |
| *** admin0 has joined #openstack-ceilometer | 10:39 | |
| *** Trozz has joined #openstack-ceilometer | 10:41 | |
| *** admin0 has quit IRC | 10:41 | |
| *** Trozz has quit IRC | 10:42 | |
| *** Trozz has joined #openstack-ceilometer | 10:44 | |
| *** Trozz has quit IRC | 10:45 | |
| *** liusheng has quit IRC | 10:46 | |
| openstackgerrit | liusheng proposed a change to openstack/python-ceilometerclient: Display message on HTTPException https://review.openstack.org/83028 | 10:48 |
| openstackgerrit | liusheng proposed a change to openstack/python-ceilometerclient: Display message on HTTPException https://review.openstack.org/83028 | 10:51 |
| *** liusheng has joined #openstack-ceilometer | 10:52 | |
| *** vrovachev has joined #openstack-ceilometer | 10:55 | |
| *** janisg has quit IRC | 10:58 | |
| *** eoutin has quit IRC | 10:58 | |
| openstackgerrit | A change was merged to openstack/ceilometer: Fix updating alarm can specify existing alarm name https://review.openstack.org/78495 | 10:59 |
| *** janisg has joined #openstack-ceilometer | 11:00 | |
| openstackgerrit | Michal Jastrzebski proposed a change to openstack/ceilometer: Fixing problem with inconsistent metadata structure in messages from nova. When message['payload']['metadata'] was not dict ceilometer raised AttributeError. https://review.openstack.org/83034 | 11:10 |
| *** admin0 has joined #openstack-ceilometer | 11:12 | |
| openstackgerrit | Michal Jastrzebski proposed a change to openstack/ceilometer: Metadata in compute.instance.exists fix https://review.openstack.org/83034 | 11:14 |
| *** saju_m has joined #openstack-ceilometer | 11:21 | |
| *** Trozz has joined #openstack-ceilometer | 11:23 | |
| *** Trozz has quit IRC | 11:23 | |
| *** Trozz has joined #openstack-ceilometer | 11:29 | |
| *** Trozz has quit IRC | 11:36 | |
| *** Qlawy has joined #openstack-ceilometer | 11:43 | |
| Qlawy | Hi | 11:43 |
| Qlawy | Can anyone help me with empty list from ceilometer? | 11:43 |
| Qlawy | all lists :( | 11:43 |
| Qlawy | in central.log and compute.log I have 2014-03-26 12:39:46.674 28118 AUDIT ceilometer.publisher.rpc [-] Publishing 1 samples on metering | 11:45 |
| Qlawy | co it looks like sth is working... | 11:45 |
| inc0 | Qlawy, which list? | 11:52 |
| inc0 | what exactly are you doing? | 11:52 |
| *** bogdando has quit IRC | 11:58 | |
| *** Trozz has joined #openstack-ceilometer | 12:02 | |
| Qlawy | inc0: I configured ceilometr (i hope i did it right) | 12:02 |
| *** Trozz has quit IRC | 12:03 | |
| Qlawy | inc0: and then ceilometer resource-list ... ist empty | 12:03 |
| Qlawy | inc0: ceilometer meter-list is empty too | 12:03 |
| inc0 | (btw, are you from Poland by any chance?;)) | 12:03 |
| Qlawy | inc0: tak | 12:03 |
| inc0 | tak sądziłem | 12:03 |
| Qlawy | maj inglish iz soł bed? :P | 12:03 |
| inc0 | anyway, did you install all the agents? | 12:03 |
| Qlawy | nope, just central and compute | 12:03 |
| Qlawy | and ofc api | 12:03 |
| inc0 | confirm that all agents are running | 12:04 |
| inc0 | notification etc | 12:04 |
| Qlawy | I have openstack-ceilometer-api openstack-ceilometer-central and openstack-ceilometer-compute up and running | 12:05 |
| inc0 | you need: ceilometer-agent-central and ceilometer-agent-notification | 12:05 |
| inc0 | on cloud controler | 12:05 |
| inc0 | also running | 12:05 |
| Qlawy | this one is aio | 12:05 |
| Qlawy | I think I have not collector running Oo | 12:05 |
| inc0 | yes, collector is also helpful ;) | 12:06 |
| inc0 | and agent-compute should be running on compute nodes | 12:07 |
| Qlawy | ok, its running | 12:07 |
| inc0 | not cloud controler | 12:07 |
| Qlawy | as I said its all-in-one ;) | 12:07 |
| Qlawy | however I have additional 2 compute nodes where I can run just compute agent | 12:08 |
| inc0 | yup | 12:08 |
| *** flwang has joined #openstack-ceilometer | 12:08 | |
| *** kun_huang has joined #openstack-ceilometer | 12:09 | |
| *** promulo has quit IRC | 12:09 | |
| *** matsuhashi has quit IRC | 12:10 | |
| *** bogdando has joined #openstack-ceilometer | 12:10 | |
| *** matsuhashi has joined #openstack-ceilometer | 12:11 | |
| Qlawy | looks better | 12:11 |
| Qlawy | inc0: now... any idea how to get cpu usage from compute node? Should I change sth in pipeline.yaml? | 12:12 |
| *** matsuhas_ has joined #openstack-ceilometer | 12:13 | |
| *** matsuhashi has quit IRC | 12:13 | |
| inc0 | Qlawy, you mean ticks or what? | 12:14 |
| inc0 | http://docs.openstack.org/developer/ceilometer/measurements.html look here | 12:15 |
| inc0 | you can get eighter usage in % or cpu time | 12:15 |
| Qlawy | inc0: yeah, but I dont see resourceId in ceilometer resurce-list for host | 12:16 |
| Qlawy | inc0: and yeah, % is better for my purposes | 12:17 |
| eglynn | Qlawy: re. "any idea how to get cpu usage from compute node" | 12:18 |
| eglynn | Qlawy: ... note that cpu and cpu_util relate to CPU usage of the *instances* running on the compute node, as opposed to the node itself | 12:19 |
| eglynn | Qlawy: ... for the latter, you'd need to used SNMP monitoring | 12:19 |
| Qlawy | compute_monitors = ComputeDriverCPUMonitor | 12:20 |
| Qlawy | By default, Nova will not collect the following meters related to the host compute node machine. Nova option ‘compute_monitors = ComputeDriverCPUMonitor’ should be set in nova.conf to enable meters. | 12:20 |
| Qlawy | I feel a bit confused :( | 12:21 |
| eglynn | Qlawy: my fault, that's yet another was of getting CPU from the compute | 12:21 |
| eglynn | Qlawy: ... in this case via nova notifications | 12:21 |
| Qlawy | eglynn: thank you anyway ;) | 12:22 |
| Qlawy | but still no compute.node.* in resource list ;/ | 12:23 |
| Qlawy | perhaps I have to configure sth in pipline or somethere | 12:24 |
| *** eoutin has joined #openstack-ceilometer | 12:26 | |
| eglynn | Qlawy: I'm not too familiar with the corresponding nova feature, but IIUC these data should be emmitted as nova notifications and picked up by the catch-all meter_pipeline/meter_source in the ceilometer pipeline.yaml | 12:27 |
| eglynn | Qlawy: ... this BP doesn't seem full complete https://blueprints.launchpad.net/nova/+spec/utilization-aware-scheduling | 12:27 |
| Qlawy | eglynn: yeah... | 12:32 |
| Qlawy | it looks like my openstack "flavor" has no monitor implemented yet | 12:36 |
| Qlawy | I mean this: https://github.com/openstack/nova/tree/8252b4a1dc1b584f82e3d0c52a355b3cb3252ce6/nova/compute/monitors | 12:36 |
| *** promulo has joined #openstack-ceilometer | 12:37 | |
| *** eoutin has quit IRC | 12:39 | |
| *** inc0 has quit IRC | 12:41 | |
| *** jaypipes has quit IRC | 12:51 | |
| *** thomasem has joined #openstack-ceilometer | 12:52 | |
| *** gordc has joined #openstack-ceilometer | 12:54 | |
| *** jaypipes has joined #openstack-ceilometer | 12:56 | |
| *** inc0 has joined #openstack-ceilometer | 13:00 | |
| openstackgerrit | gordon chung proposed a change to openstack/ceilometer: DBDeadlock exception in sql backend https://review.openstack.org/80461 | 13:00 |
| openstackgerrit | gordon chung proposed a change to openstack/ceilometer: improve performance of resource-list in sql https://review.openstack.org/80343 | 13:01 |
| gordc | dhellmann: if you have time, could you take a look at the comment here with your name: https://review.openstack.org/#/c/80343/6/ceilometer/api/controllers/v2.py (tried to find ryanpetrello but he seems to be MIA) | 13:06 |
| *** matsuhas_ has quit IRC | 13:07 | |
| *** julim has joined #openstack-ceilometer | 13:16 | |
| *** mihgen has quit IRC | 13:17 | |
| *** thomasem has quit IRC | 13:19 | |
| openstackgerrit | gordon chung proposed a change to openstack/ceilometer: DBDeadlock exception in sql backend https://review.openstack.org/80461 | 13:21 |
| *** admin0 has joined #openstack-ceilometer | 13:25 | |
| *** pas-ha has joined #openstack-ceilometer | 13:27 | |
| *** yassine_ has joined #openstack-ceilometer | 13:32 | |
| *** nacim has quit IRC | 13:34 | |
| *** nosnos has quit IRC | 13:35 | |
| admin0 | hey | 13:40 |
| *** flwang has quit IRC | 13:49 | |
| openstackgerrit | Eoghan Glynn proposed a change to openstack/ceilometer: Verify user/project ID for alarm created by non-admin user https://review.openstack.org/83072 | 13:49 |
| *** flwang has joined #openstack-ceilometer | 13:50 | |
| *** nacim has joined #openstack-ceilometer | 13:51 | |
| openstackgerrit | Eoghan Glynn proposed a change to openstack/ceilometer: Verify user/project ID for alarm created by non-admin user https://review.openstack.org/83072 | 14:01 |
| *** prad_ has joined #openstack-ceilometer | 14:04 | |
| inc0 | hmm, how long does it take to run tox -e py27 for you? | 14:05 |
| inc0 | because its 5min now and it seems hanged for me... | 14:06 |
| eglynn | inc0: ... /me hopes it wasn't my 200-sample test in https://review.openstack.org/82672 that's slowing it down | 14:07 |
| eglynn | inc0: ... e.g. if sqlite totters under such a "massive" workload | 14:08 |
| eglynn | inc0: ... maybe run tests again after trying: git revert 6426ca9b11bd29dcf9ef5926fcc51763b84f913e | 14:10 |
| eglynn | inc0: ... to see if that's actually the case | 14:10 |
| eglynn | inc0: ... if so, I could look at making that test specific to the mongo scenario | 14:11 |
| inc0 | eglynn, I don't think so. I had some problems with mongo on my osx, so that would be my guess | 14:11 |
| eglynn | inc0: ... k, cool | 14:11 |
| inc0 | I don't trust neighter osx nor mongo | 14:11 |
| eglynn | LOL :) | 14:11 |
| eglynn | ... when /me had a MacBook back in the day, I only ever used it to ssh into a RHEL VM | 14:12 |
| eglynn | ... seemed wrong somehow to try to do real development on what looked like a consumer appliance ;) | 14:13 |
| inc0 | well they asked me if I want mac os or dell, I figured out that I'll give mac a try, I'm still thinking of installing ubuntu here after all | 14:14 |
| *** massi_ has quit IRC | 14:14 | |
| eglynn | you prolly choose well, at least the battery life on the Mac rocks :) | 14:14 |
| inc0 | true, and this magic mouse thingy is greay | 14:15 |
| inc0 | great* | 14:15 |
| inc0 | thats why I'm still enduring mac os | 14:15 |
| eglynn | if only it had a proper gesture for "double click" | 14:16 |
| eglynn | ... tho it prolly does by now | 14:16 |
| eglynn | ... prolly had all along only I was dumb to find it | 14:16 |
| inc0 | well dclick is simply dclick for me | 14:16 |
| inc0 | with few apps its pretty configurable | 14:16 |
| eglynn | ... /me needs to eat, brain power is rapidly declining ;) | 14:17 |
| *** eglynn is now known as eglynn-lunch | 14:17 | |
| admin0 | eglynn, as you said, I created a file called graphite.py and put it inside /usr/lib/python2.6/site-packages/ceilometer/publisher/ .. then in pipeline, i set it to graphite:// .. but i get TRACE ceilometer.pipeline RuntimeError: No 'ceilometer.publisher' driver found, looking for 'graphite .. | 14:28 |
| admin0 | what am i missing | 14:29 |
| openstackgerrit | A change was merged to openstack/python-ceilometerclient: Adds alarm time constraint support to ceilometer CLI https://review.openstack.org/78356 | 14:32 |
| gordc | admin0: you need to create an entry point | 14:33 |
| admin0 | gordc: where do I do that ? | 14:33 |
| gordc | admin0: if you're working against source you should update https://github.com/openstack/ceilometer/blob/master/setup.cfg#L160 | 14:34 |
| openstackgerrit | Michal Jastrzebski proposed a change to openstack/ceilometer: Metadata in compute.instance.exists fix https://review.openstack.org/83034 | 14:34 |
| gordc | admin0: if not, there should be an entry_points.txt file you can edit. | 14:35 |
| *** inc0 has quit IRC | 14:37 | |
| admin0 | gordc: thanks .. i found /usr/lib/python2.6/site-packages/ceilometer-2013.2.2-py2.6.egg-info/entry_points.txt .. and i think i need to add my change under [ceilometer.publisher] and try again | 14:38 |
| gordc | admin0: np, give that a try. | 14:38 |
| *** alexpilotti has joined #openstack-ceilometer | 14:39 | |
| admin0 | pipeline has: - graphite://localhost:5005 , in entry_points, I added: graphite = ceilometer.publisher.graphite:GraphitePublisher and my /usr/lib/python2.6/site-packages/ceilometer/publisher/graphite.py file has class GraphitePublisher(publisher.PublisherBase): and error and code is here: http://pastebin.com/QchMTf03 | 14:44 |
| admin0 | what might I be missing | 14:47 |
| *** jmckind has joined #openstack-ceilometer | 14:48 | |
| *** julim_ has joined #openstack-ceilometer | 15:00 | |
| *** julim has quit IRC | 15:01 | |
| admin0 | gordc: any ideas | 15:02 |
| admin0 | or pointers as where i should look | 15:02 |
| *** mihgen has joined #openstack-ceilometer | 15:07 | |
| gordc | admin0: looks like it found your file... it says it can't process "from ceilometer.publisher import utils" | 15:07 |
| gordc | admin0: i'm not sure why. | 15:07 |
| admin0 | i did a copy paste of the udp.py and just changed the LOG value from udp to graphite .. | 15:08 |
| admin0 | to test | 15:08 |
| admin0 | before I add in my codes | 15:08 |
| admin0 | while udp is working fine, this does not | 15:08 |
| gordc | admin0: does ceilometer.publisher.utils.py file exist? | 15:09 |
| admin0 | not inside publishers .. /usr/lib/python2.6/site-packages/ceilometer/utils.py | 15:10 |
| admin0 | i copied the utils.py from ceilometer to inside publishers .. same error . | 15:12 |
| admin0 | now 2014-03-26 15:11:48.719 20968 ERROR stevedore.extension [-] Could not load 'graphite': no such group: collector | 15:12 |
| admin0 | udp.py works fine | 15:13 |
| *** admin0 has quit IRC | 15:14 | |
| gordc | admin0: are you using havana? | 15:14 |
| gordc | i find it strange none of these files exist...it seems like you're copying master copy of udp.py and trying to use it against havana. | 15:16 |
| *** lsmola_ has quit IRC | 15:19 | |
| *** pas-ha has quit IRC | 15:19 | |
| *** fnaval has joined #openstack-ceilometer | 15:21 | |
| *** lsmola_ has joined #openstack-ceilometer | 15:34 | |
| *** piyushmasrani has quit IRC | 15:34 | |
| *** kun_huang has quit IRC | 15:35 | |
| *** mihgen has quit IRC | 15:36 | |
| *** admin0 has joined #openstack-ceilometer | 15:37 | |
| openstackgerrit | gordon chung proposed a change to openstack/ceilometer: improve performance of resource-list in sql https://review.openstack.org/80343 | 15:39 |
| *** saju_m has quit IRC | 15:43 | |
| *** kun_huang has joined #openstack-ceilometer | 15:44 | |
| *** _cjones_ has joined #openstack-ceilometer | 15:48 | |
| *** sayalilunkad has joined #openstack-ceilometer | 15:52 | |
| *** Ruetobas has quit IRC | 16:01 | |
| *** Ruetobas has joined #openstack-ceilometer | 16:03 | |
| *** lsmola_ has quit IRC | 16:06 | |
| *** Ruetobas has quit IRC | 16:08 | |
| *** eglynn-lunch is now known as eglynn | 16:09 | |
| *** _nadya_ has joined #openstack-ceilometer | 16:10 | |
| *** Ruetobas has joined #openstack-ceilometer | 16:13 | |
| *** lsmola_ has joined #openstack-ceilometer | 16:20 | |
| *** leews has joined #openstack-ceilometer | 16:23 | |
| openstackgerrit | Jenkins proposed a change to openstack/ceilometer: Updated from global requirements https://review.openstack.org/77639 | 16:24 |
| *** boris-42 has quit IRC | 16:24 | |
| *** nacim has quit IRC | 16:30 | |
| *** kun_huang has quit IRC | 16:30 | |
| *** boris-42 has joined #openstack-ceilometer | 16:36 | |
| _nadya_ | guys, it seems to me that we have problems with anotification service again | 16:41 |
| _nadya_ | maybe someone knows anything? I do not see screen with this service in latest Jenkins run | 16:42 |
| _nadya_ | e.g. http://logs.openstack.org/24/83024/1/check/check-tempest-dsvm-full/3fc18a5/logs/ | 16:42 |
| ildikov_ | _nadya_: and also Ceilometer python26, 27 gates are failing continuously for a while now | 16:44 |
| _nadya_ | hmm...let me ask infra-team | 16:45 |
| ildikov_ | _nadya_: as I saw, the gates were failing only in case of Ceilometer, I did not see anything useful after the first quick look | 16:47 |
| _nadya_ | ildikov_: I'll check that | 16:47 |
| ildikov_ | _nadya_: cool, thanks, I did not have time for having a deeper look | 16:49 |
| *** admin0 has quit IRC | 16:52 | |
| openstackgerrit | Eoghan Glynn proposed a change to openstack/ceilometer: Verify user/project ID for alarm created by non-admin user https://review.openstack.org/83072 | 16:52 |
| *** nati_ueno has joined #openstack-ceilometer | 17:00 | |
| *** gpizza has joined #openstack-ceilometer | 17:03 | |
| *** admin0 has joined #openstack-ceilometer | 17:04 | |
| *** kun_huang has joined #openstack-ceilometer | 17:07 | |
| *** anamalagon has joined #openstack-ceilometer | 17:09 | |
| gordc | _nadya_: | 17:09 |
| gordc | _nadya_: whoops too late. i push a fixed here: https://review.openstack.org/#/c/83103/ | 17:09 |
| _nadya_ | gordc: ah, ok :) | 17:10 |
| gordc | _nadya_: want me to close mine. i don't think anyone from infra knows about it yet. | 17:10 |
| _nadya_ | gordc: it's ok, don't mind. Mine is later :) I will abandon | 17:11 |
| gordc | _nadya_: cool cool | 17:11 |
| *** gpizza has quit IRC | 17:13 | |
| openstackgerrit | gordon chung proposed a change to openstack/ceilometer: Sync oslo-incubator code https://review.openstack.org/78132 | 17:13 |
| _nadya_ | gordc: I decided to check that tempest tests are success after several workers commit but they aren't. Hope that after gating fix they will pass :) | 17:17 |
| gordc | _nadya_: there's quite a few issues i'm looking at. | 17:18 |
| gordc | _nadya_: we will need this patch https://review.openstack.org/78132 to get postgres working i believe. | 17:18 |
| _nadya_ | gordc: oh... besides, we had problem without DBDeadlock fix | 17:19 |
| *** anamalagon has quit IRC | 17:22 | |
| admin0 | gordc: i gave up further | 17:24 |
| admin0 | cannot figure it out | 17:24 |
| *** _nadya_ has quit IRC | 17:26 | |
| openstackgerrit | gordon chung proposed a change to openstack/ceilometer: SSL errors thrown with Postgres on multi workers https://review.openstack.org/83140 | 17:43 |
| *** erecio has joined #openstack-ceilometer | 17:44 | |
| *** prad__ has joined #openstack-ceilometer | 17:44 | |
| *** prad_ has quit IRC | 17:45 | |
| *** anamalagon has joined #openstack-ceilometer | 17:48 | |
| *** prad__ has quit IRC | 17:51 | |
| *** admin0 has quit IRC | 17:51 | |
| *** eglynn has quit IRC | 17:52 | |
| *** prad_ has joined #openstack-ceilometer | 17:55 | |
| *** _nadya_ has joined #openstack-ceilometer | 17:55 | |
| *** prad_ has quit IRC | 17:55 | |
| *** saju_m has joined #openstack-ceilometer | 17:57 | |
| *** ruhe has joined #openstack-ceilometer | 18:10 | |
| *** sayalilunkad has quit IRC | 18:12 | |
| *** niteshselkari__ has joined #openstack-ceilometer | 18:15 | |
| *** prad_ has joined #openstack-ceilometer | 18:23 | |
| niteshselkari__ | hi | 18:28 |
| *** prad_ has quit IRC | 18:31 | |
| openstackgerrit | Ildiko Vancsa proposed a change to openstack/ceilometer: Fix ceilometer.conf.sample mismatch https://review.openstack.org/83164 | 18:32 |
| *** prad_ has joined #openstack-ceilometer | 18:34 | |
| *** Alexei_987 has quit IRC | 18:36 | |
| *** prad_ has quit IRC | 18:38 | |
| *** lexholden has joined #openstack-ceilometer | 18:38 | |
| *** flwang has quit IRC | 18:41 | |
| ildikov_ | gordc: hi. are you around? | 18:47 |
| *** nacim has joined #openstack-ceilometer | 18:51 | |
| *** lexholden has quit IRC | 18:52 | |
| *** flwang has joined #openstack-ceilometer | 18:57 | |
| *** troy_psx has joined #openstack-ceilometer | 18:58 | |
| *** lexholden has joined #openstack-ceilometer | 18:59 | |
| *** lexholden has quit IRC | 19:03 | |
| gordc | ildikov_: yep. | 19:04 |
| ildikov_ | gordc: I just wanted to ask you to check this patch, if I'm on a right direction or not: https://review.openstack.org/83164 | 19:05 |
| gordc | ildikov_: lol | 19:06 |
| *** niteshselkari__ has quit IRC | 19:06 | |
| gordc | ildikov_: i was going to just drop the checkuptodate.sh check. | 19:07 |
| gordc | ildikov_: that conf stuff is starting to annoy me with it's random changes. | 19:07 |
| ildikov_ | gordc: there is the same issue in cinder too, as I saw on the gate | 19:07 |
| ildikov_ | gordc: and for me the problem occured only after rebuilding the tox env | 19:08 |
| ildikov_ | gordc: that is why I thought to investigate a bit, I've found a possible responsible keystoneclient change | 19:08 |
| gordc | yeah. it's cause of oslo libraries i think. the gate always pulls in latest version of lib | 19:09 |
| ildikov_ | gordc: https://bugs.launchpad.net/ceilometer/+bug/1297999 | 19:09 |
| ildikov_ | gordc: there was a bug earlier, with a same issue as this after a keystoneclient upgrade: https://bugs.launchpad.net/nova/+bug/1268614 | 19:09 |
| gordc | ildikov_: i know nova pulled out the check a whlie back. don't know if they added it back in again | 19:10 |
| ildikov_ | gordc: a-ha, ok I did not follow these changes | 19:10 |
| gordc | ildikov_: i think we should pull it out for now but we'll see.... it's becoming a hassle guessing which options are right or wrong when we can't generate it locally. | 19:11 |
| *** lexholden has joined #openstack-ceilometer | 19:12 | |
| ildikov_ | gordc: what do you mean here by pull it out? | 19:12 |
| gordc | ildikov_: remove this: https://github.com/openstack/ceilometer/blob/master/tox.ini#L15 | 19:13 |
| gordc | ildikov_: if you can fix it that'd be better but removing it is a lot easier. | 19:13 |
| ildikov_ | gordc: well, it is a bit late now here for making a proper fix I'm afraid :) | 19:15 |
| ildikov_ | gordc: but if you think that it is also a proper solution to remove that line, I definitely can do that :) | 19:16 |
| gordc | ildikov_: i'm lazy. that's the reason i'd do that :) | 19:16 |
| *** erecio has quit IRC | 19:16 | |
| *** lexholden has quit IRC | 19:17 | |
| gordc | ildikov_: i think the issue is the conf.sample generator pulls options from external libraries as well and our local env libraries don't necessarily match up with gate env | 19:18 |
| ildikov_ | gordc: that sound reasonable | 19:19 |
| gordc | ildikov_: it'd be cool if jenkins ran it for us like it does for translations... but that's a dream for me. bring the bug up during meeting (if you are attending) | 19:20 |
| ildikov_ | gordc: what about removing that line as a quick solution to let the gate work, if you do not think that the conf.sample should be changed, and then readd it after having the proper fix? | 19:21 |
| ildikov_ | gordc: that was my plan | 19:21 |
| gordc | ildikov_: i support dropping it for now. | 19:21 |
| *** nacim has quit IRC | 19:22 | |
| ildikov_ | gordc: ok, I will upload a second patch set with dropping that check and then I will bring it up during the meeting | 19:22 |
| gordc | ildikov_: cool cool | 19:23 |
| ildikov_ | gordc: thanks | 19:23 |
| _nadya_ | gordc: could you please take a look on https://review.openstack.org/#/c/82048/3/ceilometer/storage/impl_sqlalchemy.py ? I'm not sure that I'm correct with my suggestion | 19:24 |
| gordc | _nadya_: that is a strange fix... | 19:26 |
| _nadya_ | gordc: it's related to https://bugs.launchpad.net/ceilometer/+bug/1295463 mostly | 19:28 |
| _nadya_ | gordc: very likely my understanding what tsmin and tsmax are is not correct | 19:30 |
| *** anamalagon has quit IRC | 19:32 | |
| _nadya_ | gordc: is it the earliest timestamp within all entries? (tsmin) | 19:33 |
| gordc | tsmin will give you earliest timestamp value of samples based on your filter. | 19:33 |
| _nadya_ | gordc: yep, I meant within a filter. And filter already has start and end. but anyway we may get a situation when tsmin>>filter.start, don't we? | 19:37 |
| *** kun_huang has quit IRC | 19:39 | |
| gordc | _nadya_: you mean have a tsmin value later than filter.start? that makes sense... if tsmin is earlier than filter.start we have a problem. | 19:39 |
| _nadya_ | gordc: let me try to find an example in tests, it will be easier to ask | 19:41 |
| gordc | ok. | 19:41 |
| *** nealph_ has quit IRC | 19:41 | |
| *** nacim has joined #openstack-ceilometer | 19:46 | |
| ildikov_ | gordc: BTW, you were right about Nova, there isn't any checkupdate.sh in tox.ini: https://github.com/openstack/nova/blob/master/tox.ini | 19:50 |
| ildikov_ | gordc: there is a [testenv:genconfig] section instead | 19:51 |
| *** nacim has quit IRC | 19:52 | |
| openstackgerrit | Ildiko Vancsa proposed a change to openstack/ceilometer: Fix ceilometer.conf.sample mismatch https://review.openstack.org/83164 | 19:53 |
| gordc | ildikov_: seems like simplest fix | 19:53 |
| ildikov_ | gordc: I now removed the checkupdate.sh, but I haven't added a genconfig section like the one in Nova's tox.ini | 19:54 |
| *** jergerber has joined #openstack-ceilometer | 19:59 | |
| *** nealph has joined #openstack-ceilometer | 20:00 | |
| ildikov_ | gordc: I will be disconnected a bit, I would like to get home before the meeting, it's not as much fun at 10pm in the office :) | 20:01 |
| *** julim_ has quit IRC | 20:03 | |
| *** ildikov_ has quit IRC | 20:08 | |
| _nadya_ | gordc: looks like I've understood the problem. In case when user sets start, end and period we don't care about tsmin. So e.g. we have 1 Sample 20.01.2013 but user asks from 1.01.2013 till 21.01.2013 with period 1 day. We should return 19 times noting and the last value will not be empty | 20:09 |
| _nadya_ | gordc: but if start is not set we will return only one non-empty value | 20:10 |
| gordc | _nadya_: i don't get it. start is set in example...and period isn't. | 20:11 |
| _nadya_ | gordc: period is 1 day | 20:11 |
| *** saju_m has quit IRC | 20:11 | |
| _nadya_ | gordc: I've reread the bug https://bugs.launchpad.net/ceilometer/+bug/1295463. the get_meter_statistics() method will try to query sample before time of earliest sample - looks like it's correct. We may not have samples for 1 week after 'start' but user asks them and we return smth like [none] (I don't know exactly) | 20:16 |
| gordc | _nadya_: the bug is correct or the code is correct? sorry i only half read the bug. | 20:17 |
| _nadya_ | gordc: as for me, bug is incorrect, code is correct | 20:18 |
| gordc | _nadya_: yeah, i get the same feeling. i'd need to take a look at it completely | 20:18 |
| *** yassine_ has quit IRC | 20:18 | |
| _nadya_ | gordc: I'll try to write a comment to review with example, thanks! sorry for disturbing | 20:20 |
| *** yassine has quit IRC | 20:23 | |
| *** promulo has quit IRC | 20:26 | |
| *** anamalagon has joined #openstack-ceilometer | 20:28 | |
| *** saju_m has joined #openstack-ceilometer | 20:29 | |
| *** _nadya_ has quit IRC | 20:30 | |
| openstackgerrit | gordon chung proposed a change to openstack/ceilometer: fix rollback errors in sql backend https://review.openstack.org/83203 | 20:32 |
| *** anamalagon_ has joined #openstack-ceilometer | 20:32 | |
| *** anamalagon has quit IRC | 20:33 | |
| *** anamalagon_ is now known as anamalagon | 20:33 | |
| openstackgerrit | gordon chung proposed a change to openstack/ceilometer: fix rollback errors in sql backend https://review.openstack.org/83203 | 20:33 |
| *** ildikov_ has joined #openstack-ceilometer | 20:36 | |
| *** saju_m has quit IRC | 20:37 | |
| *** _nadya_ has joined #openstack-ceilometer | 20:41 | |
| *** julim has joined #openstack-ceilometer | 20:45 | |
| *** terriyu has joined #openstack-ceilometer | 20:50 | |
| *** eglynn has joined #openstack-ceilometer | 20:55 | |
| openstackgerrit | gordon chung proposed a change to openstack/ceilometer: enable a single worker by default https://review.openstack.org/83215 | 20:59 |
| *** boris-42 has quit IRC | 20:59 | |
| *** admin0 has joined #openstack-ceilometer | 21:00 | |
| admin0 | hey all | 21:00 |
| *** boris-42 has joined #openstack-ceilometer | 21:16 | |
| *** urulama has joined #openstack-ceilometer | 21:25 | |
| openstackgerrit | gordon chung proposed a change to openstack/ceilometer: enable a single worker by default https://review.openstack.org/83215 | 21:37 |
| *** claudiub_ has joined #openstack-ceilometer | 21:56 | |
| claudiub_ | Hi | 21:58 |
| *** terriyu has quit IRC | 21:59 | |
| claudiub_ | I have some commits that has to be backported. Most of them are small ones. :) | 21:59 |
| claudiub_ | If you have the time, can you take a look at them? :) | 21:59 |
| claudiub_ | https://review.openstack.org/#/c/83023/ | 22:00 |
| claudiub_ | https://review.openstack.org/#/c/83073/ | 22:00 |
| claudiub_ | https://review.openstack.org/#/c/57017/ | 22:00 |
| claudiub_ | disregard the last one. :D | 22:01 |
| *** _nadya_ has quit IRC | 22:02 | |
| claudiub_ | https://review.openstack.org/#/c/83024/ | 22:02 |
| gordc | any core want to approve this: https://review.openstack.org/#/c/83164/ | 22:02 |
| *** dhellmann is now known as dhellmann_ | 22:03 | |
| gordc | claudiub_: i've no approve power on backports. | 22:03 |
| claudiub_ | i see. could you give me some names that can help with the backports? :) | 22:05 |
| *** nikhil__ is now known as nikhil__|afk | 22:12 | |
| eglynn | claudiub_: I'll have a look tmrw | 22:13 |
| claudiub_ | sure thanks. :) | 22:13 |
| *** eglynn has quit IRC | 22:15 | |
| openstackgerrit | gordon chung proposed a change to openstack/ceilometer: fix rollback errors in sql backend https://review.openstack.org/83203 | 22:16 |
| *** troy_psx has quit IRC | 22:23 | |
| openstackgerrit | gordon chung proposed a change to openstack/ceilometer: enable a single worker by default https://review.openstack.org/83215 | 22:27 |
| *** jergerber has quit IRC | 22:28 | |
| *** dhellmann_ is now known as dhellmann | 22:31 | |
| *** jmckind has quit IRC | 22:37 | |
| *** dhellmann is now known as dhellmann_ | 22:41 | |
| *** julim has quit IRC | 23:05 | |
| *** Trozz has joined #openstack-ceilometer | 23:10 | |
| *** nati_ueno has quit IRC | 23:23 | |
| *** admin0 has quit IRC | 23:23 | |
| *** nati_ueno has joined #openstack-ceilometer | 23:23 | |
| openstackgerrit | A change was merged to openstack/ceilometer: Fix ceilometer.conf.sample mismatch https://review.openstack.org/83164 | 23:24 |
| *** urulama has quit IRC | 23:26 | |
| *** admin0 has joined #openstack-ceilometer | 23:26 | |
| *** alexpilotti has quit IRC | 23:28 | |
| openstackgerrit | gordon chung proposed a change to openstack/ceilometer: WIP fix rollback errors in sql backend https://review.openstack.org/83203 | 23:35 |
| *** Trozz has quit IRC | 23:40 | |
| *** admin0 has quit IRC | 23:42 | |
| *** fnaval has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!