| *** pauldix has joined #openstack-ceilometer | 00:08 | |
| *** pauldix has quit IRC | 00:11 | |
| *** thorst has quit IRC | 00:20 | |
| *** pauldix has joined #openstack-ceilometer | 00:20 | |
| *** prad has quit IRC | 00:43 | |
| *** _cjones_ has quit IRC | 00:52 | |
| *** changbl has quit IRC | 00:59 | |
| *** jdandrea has quit IRC | 01:02 | |
| *** alexpilotti has quit IRC | 01:03 | |
| *** liusheng has quit IRC | 01:04 | |
| *** david-lyle has joined #openstack-ceilometer | 01:19 | |
| *** changbl has joined #openstack-ceilometer | 01:28 | |
| *** jruano has quit IRC | 01:30 | |
| openstackgerrit | liusheng proposed openstack/gnocchi: Drop downgrade field in alembic script.py.mako https://review.openstack.org/209308 | 01:44 |
|---|---|---|
| *** jruano has joined #openstack-ceilometer | 01:44 | |
| openstackgerrit | liusheng proposed openstack/aodh: Drop downgrade field in alembic script.py.mako https://review.openstack.org/209310 | 01:46 |
| *** liusheng has joined #openstack-ceilometer | 01:46 | |
| *** julim has joined #openstack-ceilometer | 01:51 | |
| *** fabian1 has joined #openstack-ceilometer | 01:54 | |
| *** fabian1 has quit IRC | 02:00 | |
| *** ddieterly has joined #openstack-ceilometer | 02:01 | |
| *** fabian1 has joined #openstack-ceilometer | 02:20 | |
| *** pauldix has quit IRC | 02:30 | |
| *** dikonoor has joined #openstack-ceilometer | 02:45 | |
| *** cmyster has quit IRC | 02:54 | |
| *** cmyster has joined #openstack-ceilometer | 02:58 | |
| *** cmyster has joined #openstack-ceilometer | 02:58 | |
| *** rjaiswal has quit IRC | 03:10 | |
| openstackgerrit | liusheng proposed openstack/ceilometer: Use cpu amount as the default workers https://review.openstack.org/209333 | 03:40 |
| *** jruano has quit IRC | 03:55 | |
| *** jruano has joined #openstack-ceilometer | 03:57 | |
| *** ddieterly has quit IRC | 03:59 | |
| openstackgerrit | liusheng proposed openstack/ceilometer: Use cpu amount as the default workers https://review.openstack.org/209333 | 04:06 |
| *** jwcroppe has joined #openstack-ceilometer | 04:23 | |
| *** dikonoor has quit IRC | 04:23 | |
| *** yprokule has joined #openstack-ceilometer | 04:25 | |
| *** fawadkhaliq has joined #openstack-ceilometer | 04:25 | |
| *** pauldix has joined #openstack-ceilometer | 04:30 | |
| *** pauldix has quit IRC | 04:35 | |
| *** julim has quit IRC | 04:38 | |
| *** jruano has quit IRC | 04:46 | |
| *** ildikov has quit IRC | 04:54 | |
| *** fawadkhaliq has quit IRC | 04:57 | |
| *** fawadkhaliq has joined #openstack-ceilometer | 04:58 | |
| *** pcaruana has quit IRC | 04:59 | |
| *** ddieterly has joined #openstack-ceilometer | 05:00 | |
| *** ddieterly has quit IRC | 05:05 | |
| *** ildikov has joined #openstack-ceilometer | 05:42 | |
| *** nadya has joined #openstack-ceilometer | 05:48 | |
| *** nadya has quit IRC | 05:49 | |
| *** neelabh_ has joined #openstack-ceilometer | 05:51 | |
| *** neelabh_ has quit IRC | 05:51 | |
| *** ddieterly has joined #openstack-ceilometer | 06:01 | |
| *** ddieterly has quit IRC | 06:05 | |
| *** neelabh has joined #openstack-ceilometer | 06:09 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/ceilometer: Imported Translations from Transifex https://review.openstack.org/208833 | 06:14 |
| *** neelabh has quit IRC | 06:17 | |
| *** neelabh has joined #openstack-ceilometer | 06:33 | |
| *** nadya has joined #openstack-ceilometer | 06:34 | |
| *** rjaiswal has joined #openstack-ceilometer | 06:36 | |
| *** pcaruana has joined #openstack-ceilometer | 06:36 | |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/aodh: gnocchi: percent_of_overlap=0 for agg. alarms https://review.openstack.org/209109 | 06:39 |
| *** neelabh has quit IRC | 06:48 | |
| *** ddieterly has joined #openstack-ceilometer | 07:02 | |
| *** afazekas has joined #openstack-ceilometer | 07:02 | |
| *** neelabh has joined #openstack-ceilometer | 07:04 | |
| *** ddieterly has quit IRC | 07:06 | |
| *** nadya has quit IRC | 07:08 | |
| *** neelabh has quit IRC | 07:11 | |
| *** Ala has joined #openstack-ceilometer | 07:12 | |
| *** safchain has joined #openstack-ceilometer | 07:16 | |
| *** nadya has joined #openstack-ceilometer | 07:23 | |
| *** neelabh has joined #openstack-ceilometer | 07:27 | |
| *** nadya has quit IRC | 07:27 | |
| *** lsmola has joined #openstack-ceilometer | 07:29 | |
| *** neelabh has quit IRC | 07:30 | |
| *** nadya has joined #openstack-ceilometer | 07:34 | |
| *** fabian1 has quit IRC | 07:40 | |
| nadya | sileht: Mehdi, hi! I'm trying to finish "Initial separating unit and functional tests" and looks like it's ready, but I'm not 100% sure. Could you please take a look if you have time? As I understand, success criteria is gate-ceilometer-dsvm-functional-mongodb job, isn't it? | 07:46 |
| *** neelabh has joined #openstack-ceilometer | 07:47 | |
| *** neelabh has quit IRC | 07:47 | |
| sileht | nadya, hi I will take a look, if mongo db is covered again, that should good to land :) | 07:48 |
| sileht | nadya, do you have the infra patch to pass the job as voting ? | 07:49 |
| nadya | sileht: not yet | 07:49 |
| nadya | sileht: should it be done before landing? | 07:49 |
| sileht | nadya, after, but we need to reduce the period without gating on mongo | 07:50 |
| sileht | nadya, a Approved infra patch with DependsOn: set will be awesome | 07:50 |
| nadya | sileht: ok, I'll prepare the one and will mark it as 'in progress' | 07:50 |
| nadya | sileht: aha, ok | 07:51 |
| sileht | nadya, so when we merged the ceilometer side, the infra side will be merged directly | 07:51 |
| sileht | and mongo will be always tested | 07:51 |
| nadya | sileht: and one more doubt... the job should run on mongo backend but test result contains all possible variants | 07:52 |
| sileht | nadya, that's good sqlite is always tested in functionnal tests | 07:54 |
| sileht | nadya, and db2 code is tested against a mongod server | 07:54 |
| nadya | sileht: but hbase :) | 07:55 |
| sileht | nadya, most of hbase tests as skipped | 07:55 |
| sileht | I guess In the future we can replace the scenario stuffs with the same kind of setup we have in gnocchi | 07:56 |
| *** fabian1 has joined #openstack-ceilometer | 07:56 | |
| nadya | sileht: ok, then I'm starting infra patch | 07:56 |
| sileht | ie: running only tests against the storage url and not generate tests for all backends and skip them when the backend is not present | 07:57 |
| *** yassine has joined #openstack-ceilometer | 08:02 | |
| *** ddieterly has joined #openstack-ceilometer | 08:03 | |
| *** neelabh has joined #openstack-ceilometer | 08:05 | |
| *** fawadkhaliq has quit IRC | 08:06 | |
| *** ddieterly has quit IRC | 08:07 | |
| *** neelabh has quit IRC | 08:09 | |
| *** belmoreira has joined #openstack-ceilometer | 08:23 | |
| *** barra204 has quit IRC | 08:25 | |
| *** neelabh has joined #openstack-ceilometer | 08:25 | |
| *** fawadkhaliq has joined #openstack-ceilometer | 08:28 | |
| *** jwcroppe has quit IRC | 08:29 | |
| *** boris-42 has quit IRC | 08:30 | |
| *** jwcroppe has joined #openstack-ceilometer | 08:35 | |
| *** neelabh has quit IRC | 08:42 | |
| openstackgerrit | Julien Danjou proposed openstack/ceilometer: Stop using openstack.common from keystoneclient https://review.openstack.org/209402 | 08:44 |
| *** nadya_ has joined #openstack-ceilometer | 08:56 | |
| *** nadya has quit IRC | 08:59 | |
| *** neelabh has joined #openstack-ceilometer | 09:00 | |
| *** ddieterly has joined #openstack-ceilometer | 09:04 | |
| openstackgerrit | Thomas Herve proposed openstack/ceilometer: Handle None actions in Alarm.delete_actions https://review.openstack.org/209410 | 09:04 |
| therve | aodh possible patch ^^^ | 09:05 |
| sileht | therve, this is already fixed in aodh | 09:06 |
| sileht | therve, we should break heat less often ;) : https://review.openstack.org/#/c/207298/ | 09:08 |
| *** ddieterly has quit IRC | 09:08 | |
| sileht | the aodh fixe: https://github.com/openstack/aodh/commit/b40b4513e61cd76a11e031d5b77bffbda1d012f6 | 09:09 |
| *** neelabh has quit IRC | 09:11 | |
| therve | Ah :) | 09:14 |
| therve | sileht, Interestingly, it doesn't seem to break test, just return a 500 that we deal with somehow | 09:15 |
| sileht | therve, it breaks my tests because the stack deletion fail because of the error 500 | 09:15 |
| *** neelabh has joined #openstack-ceilometer | 09:28 | |
| *** rjaiswal has quit IRC | 09:30 | |
| *** neelabh has quit IRC | 09:32 | |
| *** jwcroppe has quit IRC | 09:38 | |
| *** jwcroppe has joined #openstack-ceilometer | 09:39 | |
| *** jwcroppe has quit IRC | 09:44 | |
| *** cdent has joined #openstack-ceilometer | 09:45 | |
| *** neelabh has joined #openstack-ceilometer | 09:49 | |
| therve | jd__, https://review.openstack.org/#/c/207140/ | 09:50 |
| jd__ | therve: si? | 09:50 |
| therve | jd__, The fix is already in aodh | 09:50 |
| jd__ | therve: ah, then you should just assign #1479274 to ceilometer too and propose the backport as a fix for that I guess | 09:51 |
| jd__ | I didn't realize it was the same | 09:51 |
| therve | Well, me neither | 09:51 |
| jd__ | hihi | 09:51 |
| jd__ | I'll reorganize things around | 09:51 |
| jd__ | therve: that's because you're still using Ceilometer and not Aodh? | 09:52 |
| therve | jd__, Whatever my devstack does by default | 09:52 |
| jd__ | therve: :( | 09:52 |
| jd__ | we're removing Ceilometer from devstack anywya | 09:52 |
| jd__ | therve: marked as duplicate | 09:53 |
| therve | I did it already | 09:53 |
| therve | sileht, Can you do the backport? I'll abandon my patch | 09:54 |
| sileht | therve, I will try | 09:54 |
| *** neelabh has quit IRC | 09:54 | |
| therve | Coolio | 09:54 |
| *** nadya_ has quit IRC | 09:58 | |
| cdent | is this pymongo failure in the py34 tests something we're already aware of: http://logs.openstack.org/41/196441/11/check/gate-ceilometer-python34/3f7f531/console.html#_2015-08-04_23_03_34_627 | 10:04 |
| *** nadya has joined #openstack-ceilometer | 10:04 | |
| *** ddieterly has joined #openstack-ceilometer | 10:04 | |
| * sileht just see postgresql and mysql doesn't return same elements order with order by on a nullable column and when some items are null | 10:07 | |
| * sileht cries | 10:07 | |
| cdent | oh dear | 10:07 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: WIP: Implements list_resources limit/ordering https://review.openstack.org/209438 | 10:07 |
| *** Qlawy has quit IRC | 10:08 | |
| *** ddieterly has quit IRC | 10:09 | |
| *** Qlawy has joined #openstack-ceilometer | 10:09 | |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Implements list_resources limit/ordering https://review.openstack.org/209438 | 10:11 |
| *** neelabh has joined #openstack-ceilometer | 10:11 | |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Implements list_resources limit/ordering https://review.openstack.org/209438 | 10:12 |
| *** neelabh has quit IRC | 10:13 | |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/ceilometer: api: fix alarm deletion and update https://review.openstack.org/209441 | 10:18 |
| sileht | therve, jd__ https://review.openstack.org/209441 | 10:18 |
| sileht | jd__, I'm writting the pagination stuffs for 'resources' listing/searching, a good practive is to provide links | 10:19 |
| sileht | rel=self/next/previous | 10:19 |
| sileht | but the returned json is just a list of resources | 10:20 |
| sileht | (not encapsuled in {'resources': [...]}) | 10:20 |
| sileht | so I can't add links in payload without breaking API | 10:20 |
| sileht | The alternative can be put links into headers like 'https://tools.ietf.org/html/draft-nottingham-link-hint-00#page-11' proposed | 10:21 |
| sileht | I haven't found any implementation examples of that, but I don't see other alternative without breaking gnocchi v1 API | 10:22 |
| sileht | cdent, jd__ ^ | 10:22 |
| sileht | wdyt ? I put them into header or just add a TODO for next API version | 10:23 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Implements list_resources limit/ordering https://review.openstack.org/209438 | 10:24 |
| cdent | sileht: I've never been too keen on rel links in the headers but in the absence of of an encapsulation of the data it is rather hard to do | 10:26 |
| cdent | I guess at least a TODO and maybe header links may be the right way to go? | 10:27 |
| cdent | unless the client uses the header links they don't much matter? | 10:27 |
| sileht | cdent, many other openstack API use link into the JSON payload | 10:27 |
| sileht | cdent,I don't like to do something different for this | 10:27 |
| *** neelabh has joined #openstack-ceilometer | 10:29 | |
| cdent | I think links in the JSON makes relatively good sense, its just the headers that are not great | 10:30 |
| cdent | they feel invisible | 10:30 |
| *** pauldix has joined #openstack-ceilometer | 10:32 | |
| *** fabian1 has quit IRC | 10:32 | |
| cdent | sileht: the encapsulation that you've described is probably something we're going to need to make happen at some point as its the only reliable way to make room for expansion | 10:35 |
| sileht | cdent, yes | 10:35 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Implements list_resources limit/ordering https://review.openstack.org/209438 | 10:37 |
| *** exploreshaifali has joined #openstack-ceilometer | 10:42 | |
| *** jwcroppe has joined #openstack-ceilometer | 10:45 | |
| *** nadya has quit IRC | 10:47 | |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: InfluxDB storage driver https://review.openstack.org/165407 | 10:48 |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: indexer: always eagerly load archive_policy https://review.openstack.org/209208 | 10:48 |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: rest: validate timestamp in metric value search query https://review.openstack.org/209450 | 10:48 |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: rest: directly pass metric to search_value() https://review.openstack.org/209451 | 10:48 |
| *** jruano has joined #openstack-ceilometer | 10:50 | |
| *** nadya has joined #openstack-ceilometer | 10:50 | |
| *** fabian1 has joined #openstack-ceilometer | 10:53 | |
| openstackgerrit | Merged openstack/ceilometer: gnocchi: retry with a new token on 401 https://review.openstack.org/206560 | 10:55 |
| *** h00327910__ has quit IRC | 10:58 | |
| *** fabian1 has left #openstack-ceilometer | 11:00 | |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: InfluxDB storage driver https://review.openstack.org/165407 | 11:01 |
| *** exploreshaifali has quit IRC | 11:04 | |
| *** ddieterly has joined #openstack-ceilometer | 11:05 | |
| *** ityaptin has quit IRC | 11:09 | |
| *** ityaptin has joined #openstack-ceilometer | 11:10 | |
| *** ddieterly has quit IRC | 11:10 | |
| therve | sileht, You rock once again | 11:13 |
| cdent | he's been rocking a lot lately | 11:14 |
| cdent | might be time for a music video | 11:15 |
| *** alexpilotti has joined #openstack-ceilometer | 11:26 | |
| *** gordc has joined #openstack-ceilometer | 11:28 | |
| cdent | gordc: do you have the power to delete this: https://pypi.python.org/pypi/ceilometer | 11:29 |
| cdent | or maybe dhellmann can do it? | 11:29 |
| gordc | cdent: i probably don't. i didn't create it and the powers above started removing release abilities from everyone. | 11:31 |
| gordc | cdent: about the bug? | 11:31 |
| cdent | yeah, I'm just responding to the poor bastard now | 11:31 |
| gordc | it's pretty funny it got the other packages installed... it's some mashup of different 2-3 year old builds. | 11:32 |
| *** jwcroppe has quit IRC | 11:33 | |
| cdent | I made friendly and hopefully helpful comment and marked it invalid | 11:34 |
| nadya | cdent: gordc: sileht: folks, let me rebase it the last time and we merge it? Infra patch is ready, functional job is successful | 11:35 |
| * cdent is happy | 11:35 | |
| gordc | cdent: friendly? well done. i'll lend you my suit one day. | 11:35 |
| nadya | I forget to mention the patch.... | 11:35 |
| gordc | nadya: :) yay | 11:36 |
| nadya | https://review.openstack.org/#/c/204128/ ! | 11:36 |
| gordc | why does ask.openstack.org tell me to "please start posting anonymously" but then it tells me to signin to post. | 11:37 |
| gordc | nadya: is it difficult to get it gating on sql as well? | 11:38 |
| gordc | not telling you to do it. just asking. | 11:38 |
| nadya | gordc: tbh, sql is tested as well | 11:38 |
| nadya | gordc: and I don't understand why :) | 11:39 |
| nadya | gordc: let me find the latest test results... | 11:39 |
| gordc | nadya: tested on sqlite or mysql/postgres? | 11:40 |
| nadya | gordc: sqlite | 11:40 |
| gordc | oh. yeah. i think that's because of the scenario stuff. | 11:41 |
| gordc | to get mysql/postgres it needs to be configured i believe. | 11:41 |
| nadya | yep, I believe it's needed to change CEILOMETER_TEST_BACKEND | 11:41 |
| nadya | gordc: ok, let me rebase it, run experimental job and then I ping you. The problem is that the patch outdates very often | 11:43 |
| nadya | gordc: the latest successful run http://logs.openstack.org/28/204128/13/experimental/gate-ceilometer-dsvm-functional-mongodb/33b7b10/ | 11:44 |
| sileht | nadya,I have added a comment on the infra patch | 11:44 |
| sileht | bbl | 11:44 |
| *** pauldix has quit IRC | 11:45 | |
| *** thorst has joined #openstack-ceilometer | 11:45 | |
| *** afazekas_ has joined #openstack-ceilometer | 11:46 | |
| nadya | sileht: I should add smth like - name: gate-ceilometer-dsvm-functional-mongodb branch: ^master$ ? | 11:47 |
| openstackgerrit | Nadya Shakhat proposed openstack/ceilometer: Initial separating unit and functional tests https://review.openstack.org/204128 | 11:48 |
| *** afazekas has quit IRC | 11:48 | |
| *** julim has joined #openstack-ceilometer | 11:49 | |
| *** bdossant has joined #openstack-ceilometer | 11:51 | |
| openstackgerrit | gordon chung proposed openstack/ceilometer: add poll history to avoid duplicate samples https://review.openstack.org/209160 | 11:51 |
| *** yassine has quit IRC | 11:51 | |
| gordc | nadya: i think adding "branch: ^(?!stable/juno/kilo).*$" should work | 11:53 |
| nadya | gordc: ok, adding now... | 11:53 |
| gordc | hmmm... seems like there's a different regex for everything in that layout file. | 11:57 |
| *** bdossant_ has joined #openstack-ceilometer | 12:00 | |
| *** bdossant has quit IRC | 12:02 | |
| *** ddieterly has joined #openstack-ceilometer | 12:06 | |
| *** yassine has joined #openstack-ceilometer | 12:07 | |
| *** ddieterly has quit IRC | 12:07 | |
| *** ddieterly has joined #openstack-ceilometer | 12:08 | |
| *** jwcroppe has joined #openstack-ceilometer | 12:12 | |
| *** ddieterly has quit IRC | 12:12 | |
| nadya | gordc: indeed | 12:17 |
| openstackgerrit | Chris Dent proposed openstack/ceilometer: Grenade plugin using devstack plugin for ceilometer https://review.openstack.org/196441 | 12:19 |
| gordc | nadya: opps i gave you wrong regex. | 12:19 |
| *** alejandrito has joined #openstack-ceilometer | 12:19 | |
| openstackgerrit | Chris Dent proposed openstack/ceilometer: Grenade plugin using devstack plugin for ceilometer https://review.openstack.org/196441 | 12:19 |
| gordc | "branch: ^(?!stable/(juno|kilo)).*$" | 12:20 |
| nadya | gordc: yep, I have too many parallel tasks today....btw, I wanted to discuss performance testing with you today | 12:21 |
| gordc | nadya: sure. i have a meeting at 1400UTC but should be ok rest of the time. | 12:22 |
| nadya | gordc: ok | 12:22 |
| *** ddieterly has joined #openstack-ceilometer | 12:24 | |
| cdent | jd__, gordc, liusheng I feel pretty alone on this issue, but I really think numcpu settings for all services is not a great idea as a default: https://bugs.launchpad.net/ceilometer/+bug/1481254 | 12:29 |
| openstack | Launchpad bug 1481254 in Ceilometer "ceilometer default multiple workers option is unrelated with cpu amount" [Medium,In progress] - Assigned to Liusheng (liusheng) | 12:29 |
| *** pauldix has joined #openstack-ceilometer | 12:30 | |
| *** edmondsw has joined #openstack-ceilometer | 12:34 | |
| *** fawadkhaliq has quit IRC | 12:38 | |
| gordc | cdent: just commented. | 12:39 |
| gordc | i agreed | 12:39 |
| *** jkraj has joined #openstack-ceilometer | 12:43 | |
| *** jwcroppe has quit IRC | 12:43 | |
| * cdent needs a proofreader | 12:49 | |
| *** prad has joined #openstack-ceilometer | 12:52 | |
| gordc | nadya: what happens if you make ceilometer patch depend on infra patch? | 12:58 |
| gordc | cdent: is it for a murder mystery story? | 12:58 |
| nadya | gordc: http://lists.openstack.org/pipermail/openstack-dev/2015-February/056515.html | 12:59 |
| cdent | gordc, no its so when I go back and read comments I leave on bugs I don't get embarassed by the strange typos and word rearrangments and drops and such that happen | 12:59 |
| nadya | gordc: I need to read this as well, because I've read only first paragraph | 13:00 |
| gordc | nadya: oh. i meant can you switch the depends on tag? i'd like to check in infra patch first and then merge the ceilometer patch. | 13:00 |
| gordc | nadya: not sure if this is possibel. | 13:00 |
| gordc | possible* | 13:00 |
| *** thumpba has joined #openstack-ceilometer | 13:01 | |
| gordc | cdent: oh. i type garbage 75% of time. which is why you see me always double responding like i'm talking to myself... or maybe i am. | 13:01 |
| nadya | gordc: I may add 'depends-on' in commit message and we will see | 13:01 |
| gordc | nadya: cool cool. | 13:02 |
| openstackgerrit | Nadya Shakhat proposed openstack/ceilometer: Initial separating unit and functional tests https://review.openstack.org/204128 | 13:03 |
| *** thumpba has quit IRC | 13:04 | |
| *** edmondsw has quit IRC | 13:05 | |
| cdent | pissing down rain here | 13:06 |
| * cdent wants summer back | 13:07 | |
| gordc | you live in uk. | 13:08 |
| gordc | isn't that par for the course? | 13:08 |
| cdent | yes, but I've never been one for living by the rules | 13:09 |
| nadya | gordc: hmm, I got This change depends on a change that failed to merge. | 13:09 |
| *** afazekas_ has quit IRC | 13:10 | |
| *** alejandrito has quit IRC | 13:10 | |
| nadya | jd__: wrong channel :) | 13:11 |
| jd__ | nadya: I thought so :) | 13:12 |
| nadya | jd__: https://review.openstack.org/#/c/165407/ failed | 13:13 |
| gordc | nadya: drop the depends on the infra patch. you can't have both depend on each other :) | 13:13 |
| jd__ | nadya: ah my bad I didn't test with py34, I'll fix it | 13:13 |
| gordc | ... is influx almost working :) | 13:13 |
| jd__ | gordc: yes! | 13:14 |
| gordc | sweet! | 13:14 |
| gordc | will get ready to test. | 13:14 |
| nadya | gordc: I thought It was that you wanted :) ok, will do | 13:14 |
| nadya | jd__: could you please add Ilya to co-authors? I don't know why he didn't do it by himself... | 13:15 |
| gordc | nadya: oh. yeah i should have clarified. i want ceilometer->depends on->infra... not infra->depends on->ceilometer | 13:15 |
| jd__ | nadya: yup I will | 13:16 |
| *** primechuck has joined #openstack-ceilometer | 13:16 | |
| jd__ | cdent: is there any plan to enhance $REPONSE in Gabbi to have all the older reponses available? | 13:18 |
| cdent | jd__: you mean to go back the entire stack of tests? no. did you have a particular use case in mind? | 13:19 |
| openstackgerrit | liusheng proposed openstack/ceilometer: Use cpu amount as the default workers https://review.openstack.org/209333 | 13:19 |
| *** bdossant_ has quit IRC | 13:20 | |
| jd__ | cdent: create 2 metrics in Gnocchi and then search in both so you need both ids | 13:20 |
| *** alexpilotti has quit IRC | 13:21 | |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: InfluxDB storage driver https://review.openstack.org/165407 | 13:21 |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: indexer: always eagerly load archive_policy https://review.openstack.org/209208 | 13:21 |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: rest: validate timestamp in metric value search query https://review.openstack.org/209450 | 13:21 |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: rest: directly pass metric to search_value() https://review.openstack.org/209451 | 13:21 |
| jd__ | cdent: I might work on that then | 13:21 |
| cdent | I might not accept it | 13:21 |
| * cdent is super hard core about complexity in gabbi | 13:21 | |
| cdent | just ask my friend FND about it, he hates me | 13:21 |
| jd__ | :((( | 13:22 |
| cdent | I'm not saying no, I'm just saying that changes to the test format are the hardest thing to get over the acceptance bar | 13:22 |
| *** exploreshaifali has joined #openstack-ceilometer | 13:22 | |
| nadya | gordc: I was trying to apply "Multi-way CRD allow for bidirectional links between changes" from the letter :) | 13:22 |
| nadya | gordc: infra is updated | 13:22 |
| nadya | jd__: thank you! | 13:23 |
| *** jmank has joined #openstack-ceilometer | 13:23 | |
| cdent | jd__: if you've got an idea for the syntax join me over in #gabbi to talk about it | 13:23 |
| *** alexpilotti has joined #openstack-ceilometer | 13:23 | |
| * sileht grrr oslo.db still use mox | 13:23 | |
| gordc | lol | 13:24 |
| *** bdossant has joined #openstack-ceilometer | 13:26 | |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/ceilometer: api: fix alarm deletion and update https://review.openstack.org/209441 | 13:31 |
| *** neelabh has quit IRC | 13:31 | |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Implements list_resources limit/ordering https://review.openstack.org/209438 | 13:32 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Implements list_resources limit/ordering https://review.openstack.org/209438 | 13:34 |
| *** jd__` has joined #openstack-ceilometer | 13:34 | |
| *** jd__- has joined #openstack-ceilometer | 13:35 | |
| * sileht remembers one of the reason we move to mock, reported errors are just not helpfull :( | 13:35 | |
| sileht | mox3.mox.UnexpectedMethodCallError: Unexpected method call function.__call__(<mock.MagicMock object at 0x7f02238cf1d0>) -> None | 13:36 |
| *** jdandrea has joined #openstack-ceilometer | 13:39 | |
| *** jd__` has quit IRC | 13:45 | |
| *** jd__- has quit IRC | 13:45 | |
| *** jd__ has quit IRC | 13:45 | |
| *** jd__ has joined #openstack-ceilometer | 13:45 | |
| gordc | any know the reason for defining db connection in hook? | 13:45 |
| *** jd__ has quit IRC | 13:46 | |
| *** jd__ has joined #openstack-ceilometer | 13:46 | |
| sileht | gordc, we just want to not use global I guess | 13:46 |
| *** exploreshaifali has quit IRC | 13:46 | |
| gordc | sileht: reason i ask is the versionedobjects patches stop using it: ie https://review.openstack.org/#/c/190521/ | 13:47 |
| gordc | i can't tell if that's a bad thing or not. | 13:47 |
| *** rbak has joined #openstack-ceilometer | 13:51 | |
| *** dobson has quit IRC | 13:52 | |
| *** briancline has quit IRC | 13:52 | |
| *** dobson has joined #openstack-ceilometer | 13:52 | |
| * sileht sad to see people continue using global for things that really don't need to | 13:53 | |
| sileht | gordc, VersionnedObject creates 1 connections per object types ... | 13:54 |
| sileht | I hope I'm wrong | 13:54 |
| *** briancline has joined #openstack-ceilometer | 13:54 | |
| sileht | I have added this comment to the review | 13:55 |
| gordc | sileht: yeah that's what i understand too... which seems crazy | 13:56 |
| *** rbak has quit IRC | 13:59 | |
| *** edmondsw has joined #openstack-ceilometer | 14:02 | |
| *** rbak has joined #openstack-ceilometer | 14:05 | |
| *** zigo has quit IRC | 14:05 | |
| *** zigo has joined #openstack-ceilometer | 14:06 | |
| *** thumpba has joined #openstack-ceilometer | 14:07 | |
| *** afazekas_ has joined #openstack-ceilometer | 14:12 | |
| *** bdossant has quit IRC | 14:19 | |
| *** bdossant has joined #openstack-ceilometer | 14:19 | |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: InfluxDB storage driver https://review.openstack.org/165407 | 14:19 |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: indexer: always eagerly load archive_policy https://review.openstack.org/209208 | 14:19 |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: rest: directly pass metric to search_value() https://review.openstack.org/209451 | 14:19 |
| *** bdossant has quit IRC | 14:21 | |
| *** bdossant has joined #openstack-ceilometer | 14:21 | |
| *** fawadkhaliq has joined #openstack-ceilometer | 14:22 | |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/gnocchi: Implements list_resources limit/ordering https://review.openstack.org/209438 | 14:23 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/aodh: gnocchi: percent_of_overlap=0 for agg. alarms https://review.openstack.org/209109 | 14:31 |
| openstackgerrit | Chris Dent proposed openstack/ceilometer: Grenade plugin using devstack plugin for ceilometer https://review.openstack.org/196441 | 14:40 |
| cdent | one more try | 14:41 |
| *** afazekas_ has quit IRC | 14:43 | |
| *** afazekas_ has joined #openstack-ceilometer | 14:43 | |
| *** nadya has quit IRC | 14:45 | |
| *** primechuck has quit IRC | 14:46 | |
| *** jruano has quit IRC | 14:49 | |
| *** afazekas_ has quit IRC | 14:54 | |
| *** nadya has joined #openstack-ceilometer | 14:55 | |
| cdent | gordc: "wasn't aware of namespace" me neither. It blew a hole out the side of my head. | 15:02 |
| gordc | cdent: i've no idea what it does. i'm just going to accept it for onw | 15:02 |
| gordc | now* | 15:02 |
| cdent | basically turns a nested dict into a nest object so that you can get stuff by attribute, which works better in the quoted environment required in eval and pipeline.yaml | 15:03 |
| gordc | hmmm... maybe prad can use it for meter definitions. | 15:06 |
| gordc | cdent: is it to avoid using eval() or to encourage using eval() | 15:07 |
| gordc | or neither | 15:07 |
| cdent | it allows the use of eval in a way that doesn't require escaping quotes | 15:08 |
| prad | that sounds useful, i was researching something called safe_eval yesterday | 15:09 |
| prad | which is eval with better safety checks | 15:10 |
| cdent | someday there will be safe_safe_eval | 15:10 |
| gordc | imo, i don't really care. if the person hsa access to pipeline already, you're already screwed. | 15:14 |
| cdent | tru | 15:14 |
| gordc | but if it makes people feel better, we can avoid eval. | 15:14 |
| *** afazekas_ has joined #openstack-ceilometer | 15:15 | |
| *** nadya has quit IRC | 15:16 | |
| *** rbak has quit IRC | 15:16 | |
| *** Ephur has quit IRC | 15:16 | |
| gordc | prad: http://stackoverflow.com/questions/2371436/evaluating-a-mathematical-expression-in-a-string seems like pyparsing is already in requirements | 15:18 |
| gordc | don't know if you played with it before | 15:18 |
| *** rjaiswal has joined #openstack-ceilometer | 15:18 | |
| prad | gordc, yea I ran into that too. This seems definitely better than direct eval | 15:20 |
| gordc | prad: what do we have left? just port everything over, and support math in volume field? | 15:22 |
| prad | gordc, yea pretty much | 15:22 |
| gordc | cool cool. | 15:23 |
| *** Ephur has joined #openstack-ceilometer | 15:26 | |
| *** jasonamyers has joined #openstack-ceilometer | 15:27 | |
| openstackgerrit | Chris Dent proposed openstack/gnocchi: Add to influx command's $PATH https://review.openstack.org/209561 | 15:31 |
| gordc | cdent: sileht: i feel like i should tell you, we merged part of my code but before my spec: https://review.openstack.org/#/c/199331/ | 15:32 |
| cdent | no soup for you | 15:32 |
| sileht | gordc, I will revert everything, sorry | 15:33 |
| gordc | already got the soup. | 15:33 |
| gordc | hahah | 15:33 |
| sileht | ;) | 15:33 |
| gordc | no! | 15:33 |
| *** afazekas_ has quit IRC | 15:33 | |
| *** rbak has joined #openstack-ceilometer | 15:34 | |
| *** shakamunyi has joined #openstack-ceilometer | 15:35 | |
| *** _cjones_ has joined #openstack-ceilometer | 15:36 | |
| cdent | jd__, anyone: Are there already discussions/plans in place related to influxdb and devstack? | 15:37 |
| gordc | haven't heard anything | 15:38 |
| gordc | i assume it's the same as elasticsearch... just put it up? | 15:38 |
| sileht | cdent, devstack == ubuntu should package it, so I don't think this will happen soon | 15:38 |
| cdent | I think there are sourcing-of-packages issues | 15:38 |
| *** belmoreira has quit IRC | 15:39 | |
| *** bdossant has quit IRC | 15:39 | |
| gordc | i think it needs to be packaged in ubuntu/fedora to gat on... but you can still include it in devstack. or that was the criteria when i did elasticsearch (elasticsearch isn't packaged either) | 15:40 |
| gordc | (packaged in LTS) | 15:40 |
| *** drjones has joined #openstack-ceilometer | 15:40 | |
| openstackgerrit | Chris Dent proposed openstack/gnocchi: Add to influx command's $PATH https://review.openstack.org/209561 | 15:41 |
| *** safchain has quit IRC | 15:41 | |
| *** drjones has quit IRC | 15:41 | |
| *** drjones has joined #openstack-ceilometer | 15:43 | |
| *** _cjones_ has quit IRC | 15:44 | |
| * cdent looks at the elasticsearch stuff | 15:47 | |
| *** drjones has quit IRC | 15:47 | |
| cdent | gordc: are their other projects use es? | 15:48 |
| gordc | glance was suppose to be using it. | 15:49 |
| gordc | but if i recalled the merged stuff when the spec was still being reviewed... i don't know if it ever did | 15:49 |
| * cdent makes a bug to not forget | 15:51 | |
| *** adurbin_ has joined #openstack-ceilometer | 15:54 | |
| *** Ala has quit IRC | 15:57 | |
| openstackgerrit | Merged openstack/ceilometer: Corrected unit of snmp based harware disk and memory meters https://review.openstack.org/207199 | 16:00 |
| *** ildikov has quit IRC | 16:01 | |
| *** alexpilotti has quit IRC | 16:06 | |
| *** ddieterly has quit IRC | 16:09 | |
| *** jasonamyers has quit IRC | 16:12 | |
| *** exploreshaifali has joined #openstack-ceilometer | 16:16 | |
| *** drjones has joined #openstack-ceilometer | 16:17 | |
| *** yprokule has quit IRC | 16:21 | |
| gordc | anyone know what kiloeyes is? | 16:22 |
| gordc | https://review.openstack.org/#/c/203259/ | 16:22 |
| openstackgerrit | Chris Dent proposed openstack/ceilometer: Grenade plugin using devstack plugin for ceilometer https://review.openstack.org/196441 | 16:22 |
| cdent | gordc: sounds like it is probably what is says on the tin: cloud monitoring with monasca metrics api and elasticsearch | 16:24 |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: InfluxDB storage driver https://review.openstack.org/165407 | 16:25 |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: indexer: always eagerly load archive_policy https://review.openstack.org/209208 | 16:25 |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: rest: validate timestamp in metric value search query https://review.openstack.org/209450 | 16:25 |
| openstackgerrit | Julien Danjou proposed openstack/gnocchi: rest: directly pass metric to search_value() https://review.openstack.org/209451 | 16:25 |
| *** pcaruana has quit IRC | 16:26 | |
| gordc | cdent: you know what the metrics api part is? | 16:27 |
| cdent | gordc: no, but I know there's been plans to replicate the monasca api or apis in python for a while. I guess this is it. | 16:29 |
| *** openstackgerrit_ has joined #openstack-ceilometer | 16:30 | |
| gordc | i'm not sure what the functionality you're describing is... | 16:31 |
| cdent | nor me, but we do know that a thing called monasca has existed for quite some time, in java form | 16:32 |
| *** drjones has quit IRC | 16:33 | |
| gordc | ... right. i'm just going to take it as a bit tent thing... seems like the catch all term. | 16:34 |
| *** thorst has quit IRC | 16:34 | |
| *** ddieterly has joined #openstack-ceilometer | 16:35 | |
| cdent | gordc: are your curious or concerned? | 16:39 |
| gordc | curious. i don't know what it is to be concerned. | 16:39 |
| openstackgerrit | Ryota MIBU proposed openstack/aodh: Refactor alarm scenario tests (RuleCombination) https://review.openstack.org/204336 | 16:40 |
| * cdent wonders if you'll get an answer to your comment | 16:40 | |
| gordc | no clue... seems like it'd make sense to. | 16:41 |
| *** ildikov has joined #openstack-ceilometer | 16:43 | |
| *** _cjones_ has joined #openstack-ceilometer | 16:47 | |
| *** _cjones_ has quit IRC | 16:49 | |
| gordc | cool influx patch seems to be passing for me. | 16:49 |
| *** thorst has joined #openstack-ceilometer | 16:51 | |
| *** yassine has quit IRC | 16:56 | |
| *** _cjones_ has joined #openstack-ceilometer | 17:01 | |
| *** jwcroppe has joined #openstack-ceilometer | 17:03 | |
| *** openstackgerrit_ has quit IRC | 17:05 | |
| openstackgerrit | Ryota MIBU proposed openstack/aodh: Add 'event' type and 'event_rule' to alarm API https://review.openstack.org/200967 | 17:08 |
| *** nadya has joined #openstack-ceilometer | 17:10 | |
| *** tongli has joined #openstack-ceilometer | 17:12 | |
| *** nadya has quit IRC | 17:13 | |
| *** openstackgerrit_ has joined #openstack-ceilometer | 17:24 | |
| *** ddieterly has quit IRC | 17:30 | |
| rjaiswal | gordc: regarding https://bugs.launchpad.net/ceilometer/+bug/1475441, PollsterPermanentError is thrown in context of multiple tenants/resources | 17:31 |
| openstack | Launchpad bug 1475441 in Ceilometer "rgw still throwing errors" [Medium,Triaged] - Assigned to Rohit Jaiswal (rohit-jaiswal-3) | 17:31 |
| rjaiswal | https://github.com/openstack/ceilometer/blob/master/ceilometer/objectstore/rgw.py#L97 | 17:32 |
| *** nadya has joined #openstack-ceilometer | 17:33 | |
| rjaiswal | also, looks like the log message: Prevent pollster rgw.objects for polling source meter_source anymore! is in the except block https://github.com/openstack/ceilometer/blob/master/ceilometer/agent/base.py#L200 | 17:34 |
| *** openstackgerrit_ has quit IRC | 17:34 | |
| *** nadya has quit IRC | 17:38 | |
| *** jwcroppe has quit IRC | 17:41 | |
| *** lsmola has quit IRC | 17:42 | |
| *** jwcroppe has joined #openstack-ceilometer | 17:42 | |
| gordc | rjaiswal: right. the exception should be returning the list of resources which it would add to blacklist in agent/base so it would avoid polling them again. | 17:42 |
| rjaiswal | gordc: ah, ok | 17:42 |
| gordc | right now it's passing back the error instead of resources so it's blacklisting the error. | 17:42 |
| gordc | that said, i think you need to adjust the exception to all passing back a list of resources rather than one | 17:43 |
| gordc | to allow* | 17:43 |
| rjaiswal | yeah | 17:45 |
| rjaiswal | was thinking the same | 17:45 |
| *** fawadkhaliq has quit IRC | 17:45 | |
| rjaiswal | cool, will have a patch up today | 17:45 |
| gordc | rjaiswal: cool cool | 17:45 |
| rjaiswal | thanks gordc | 17:45 |
| gordc | rjaiswal: awesome! thanks for the help. | 17:45 |
| rjaiswal | gordc: yw | 17:46 |
| *** nadya has joined #openstack-ceilometer | 17:46 | |
| *** jwcroppe has quit IRC | 17:47 | |
| openstackgerrit | Merged openstack/aodh: Refactor alarm scenario tests (RuleCombination) https://review.openstack.org/204336 | 17:51 |
| *** rbak has quit IRC | 17:53 | |
| *** nadya has quit IRC | 18:02 | |
| cdent | jd__, liusheng: this is fun: https://bugs.launchpad.net/aodh/+bug/1481875 | 18:10 |
| openstack | Launchpad bug 1481875 in Aodh "test_migrations test does not do what it is supposed to do" [Undecided,New] | 18:10 |
| *** rbak has joined #openstack-ceilometer | 18:14 | |
| *** exploreshaifali has quit IRC | 18:20 | |
| *** rbak has quit IRC | 18:21 | |
| *** exploreshaifali has joined #openstack-ceilometer | 18:21 | |
| *** nadya has joined #openstack-ceilometer | 18:23 | |
| *** nadya has quit IRC | 18:28 | |
| openstackgerrit | Merged openstack/aodh: Add 'event' type and 'event_rule' to alarm API https://review.openstack.org/200967 | 18:33 |
| jd__ | cdent: meh | 18:40 |
| cdent | meh indeed | 18:40 |
| cdent | skip is dangerous | 18:40 |
| *** rbak has joined #openstack-ceilometer | 18:41 | |
| gordc | cdent: if i run strace and it just loops the following non-stop, what's the deal? http://paste.openstack.org/show/409210/ | 18:43 |
| * cdent looks | 18:43 | |
| cdent | gordc, sileht tells me that is the expected behavior of something that is using oslo.messaging | 18:44 |
| cdent | basically it is waiting for activity on file descriptor 3 | 18:44 |
| cdent | now, the normal way that epoll_wait works is that it, uh, waits, until the kernel has something | 18:45 |
| gordc | is it a listener thing? because i don't see it on collector ... i get something else | 18:45 |
| cdent | but you can set the timeout really low | 18:45 |
| cdent | I think it it is a listener thing yes | 18:45 |
| cdent | I haven't got a clue _why_ this is the way it works as it seems fundamentally broken to me | 18:45 |
| cdent | it is the anti-epoll | 18:46 |
| gordc | hmmm... my collector just spews: | 18:46 |
| gordc | wait4(0, 0x7ffeef378dc4, WNOHANG, NULL) = 0 | 18:46 |
| gordc | select(0, NULL, NULL, NULL, {0, 9983}) = 0 (Timeout) | 18:46 |
| *** openstackgerrit has quit IRC | 18:46 | |
| *** openstackgerrit has joined #openstack-ceilometer | 18:46 | |
| cdent | yeah, so that's the same concept | 18:46 |
| cdent | but done with an older system call | 18:46 |
| cdent | is dinner time, I'll be back a bit later | 18:47 |
| gordc | kk. so confused why notification agent has cpu load but not collector. | 18:47 |
| *** kfox1111 has joined #openstack-ceilometer | 18:54 | |
| kfox1111 | how do you change how frequently openstack-ceilometer-compute gathers? | 18:55 |
| gordc | kfox1111: you'll need to edit pipeline.yaml... probably in /etc/ceilometer/pipeline.yaml | 18:59 |
| gordc | by default it does everything at 600s... you'll need to list meters explicitly if you want different intervals for different meters | 19:00 |
| kfox1111 | ah. I see it. thanks. | 19:00 |
| gordc | np | 19:00 |
| *** nadya has joined #openstack-ceilometer | 19:03 | |
| *** rbak has quit IRC | 19:05 | |
| nadya | gordc: should I mark https://review.openstack.org/209398 as 'in progress' ? | 19:07 |
| nadya | gordc: ok if it's merged? | 19:07 |
| gordc | nadya: i thought you said it was passing? | 19:10 |
| nadya | gordc: yep, it is. but we should merge our patch as well, right? | 19:10 |
| gordc | nadya: sure. i don't have any problems. merging it. | 19:11 |
| nadya | gordc: great :) | 19:12 |
| cdent | gordc: I couldn't figure it out for the life of me either, but it seems like sileht will probably know? | 19:12 |
| cdent | or at least be able to figure out? | 19:12 |
| cdent | I assume there must be something different about the kind of listener each is | 19:12 |
| *** jkraj has quit IRC | 19:13 | |
| cdent | anyway, I've been at this far too long today, need to stop thinking openstack for a few hours | 19:13 |
| *** cdent has quit IRC | 19:13 | |
| gordc | damn... cdent logs off very quickly.lol | 19:13 |
| openstackgerrit | Merged openstack/gnocchi: rest: validate timestamp in metric value search query https://review.openstack.org/209450 | 19:16 |
| *** r-daneel has joined #openstack-ceilometer | 19:20 | |
| openstackgerrit | Pradeep Kilambi proposed openstack/ceilometer: Convert instance, bandwidth and SwiftMiddleware meters https://review.openstack.org/209637 | 19:20 |
| *** ddaskal has joined #openstack-ceilometer | 19:21 | |
| *** rbak has joined #openstack-ceilometer | 19:26 | |
| gordc | prad: is there a reason your patches cover parts of different notifications?lol | 19:27 |
| prad | gordc, the latest one? | 19:27 |
| prad | gordc, that a WIP.. i'm wrapping up all meters that dont need special processing | 19:28 |
| prad | gordc, there are some meter where we put the yield through a loop like swiftmiddlewaremeters etc | 19:28 |
| prad | which i need to handle differently and ofcourse the math one which is just the cpu | 19:29 |
| gordc | you could make each definition an endpoint... there'd be no reason to loop | 19:31 |
| gordc | although i don't understand the looping part... which meter are you talking about? | 19:33 |
| prad | well we have two kinds, one is something like https://github.com/openstack/ceilometer/blob/master/ceilometer/objectstore/notifications.py#L72 | 19:34 |
| *** rbak has quit IRC | 19:36 | |
| prad | gordc, other is when multiple resources are returned like in network notifications https://github.com/openstack/ceilometer/blob/master/ceilometer/network/notifications.py#L77 | 19:36 |
| gordc | for the first one i don't think you need anything special -- you just don't know what is in measurements. | 19:38 |
| prad | gordc, so these would be weird to handle declarative unless we know exactly whats in that payload | 19:38 |
| gordc | what happens when you meter definitino and field doesn't exist? | 19:38 |
| prad | if definition doesnt exist obviously we wont generate any samples | 19:39 |
| gordc | prad: no. i meant if the definition exists but the field(s) don't exist in payload | 19:40 |
| *** ddaskal has quit IRC | 19:40 | |
| prad | for the fields, we expect certain fields to be present, else we raise a handled exception with field name.. for fields like unit/resourece etc we default | 19:40 |
| prad | to None is all fails | 19:40 |
| prad | we look first in the toplevel message else we look in payload else None | 19:40 |
| prad | fields like name, event, volume, unit etc are required | 19:41 |
| gordc | the ones with multiple meters in payload i think you need a conditional to check... that or just ignore meters with missing fields | 19:41 |
| prad | <gordc> for the first one i don't think you need anything special - what ya mean? | 19:42 |
| prad | dont convert it? | 19:42 |
| gordc | this is payload for swift notifications: https://github.com/openstack/ceilometer/blob/master/ceilometer/tests/objectstore/test_notifications.py#L21-L89 | 19:42 |
| prad | yea the problem is the measurements can be any # | 19:43 |
| gordc | is the issue that you can't tell the path because of list | 19:43 |
| prad | the example has 2 metrics, we could have more when swift returns | 19:44 |
| *** rbak has joined #openstack-ceilometer | 19:44 | |
| *** rbak has quit IRC | 19:46 | |
| gordc | two is the max... but i take it it's because you don't know ordering. | 19:47 |
| *** rbak has joined #openstack-ceilometer | 19:51 | |
| prad | gordc, do you know some projects send us bulk payload vs others just 1? this inconsistency makes things complicated | 19:52 |
| prad | standardize notifications ... ugh wish it happens someday | 19:54 |
| gordc | prad: well there are a lot of different payloads which have more than one metric. every definition that act on same event_type has multiple payloads in it | 19:54 |
| gordc | prad: the swift notification is the only one with a standard.lol | 19:55 |
| *** thumpba has quit IRC | 19:58 | |
| *** rbak has quit IRC | 20:01 | |
| gordc | i think i can see your problem with list. biab. need to step away | 20:01 |
| *** boris-42 has joined #openstack-ceilometer | 20:12 | |
| *** rbak has joined #openstack-ceilometer | 20:15 | |
| *** rbak has quit IRC | 20:25 | |
| *** rbak has joined #openstack-ceilometer | 20:30 | |
| *** changbl has quit IRC | 20:39 | |
| *** rbak has quit IRC | 20:40 | |
| *** changbl has joined #openstack-ceilometer | 20:40 | |
| *** wayneeseguin has quit IRC | 20:41 | |
| *** wayneeseguin has joined #openstack-ceilometer | 20:43 | |
| *** rbak has joined #openstack-ceilometer | 20:44 | |
| *** rbak has quit IRC | 20:59 | |
| *** gordc has quit IRC | 20:59 | |
| *** rbak has joined #openstack-ceilometer | 21:00 | |
| *** jwcroppe has joined #openstack-ceilometer | 21:02 | |
| *** rbak has quit IRC | 21:03 | |
| -openstackstatus- NOTICE: Zuul has been restarted to resolve a reconfiguration failure: previously running jobs have been reenqueued but change events between 19:50-20:54 UTC have been lost and will need to be rechecked or their approvals reapplied to trigger testing. | 21:04 | |
| -openstackstatus- NOTICE: Correction: change events between 20:50-20:54 UTC (during the restart only) have been lost and will need to be rechecked or their approvals reapplied to trigger testing. | 21:09 | |
| *** julim has quit IRC | 21:10 | |
| *** exploreshaifali has quit IRC | 21:15 | |
| *** rbak has joined #openstack-ceilometer | 21:18 | |
| *** exploreshaifali has joined #openstack-ceilometer | 21:20 | |
| *** rbak has quit IRC | 21:20 | |
| *** rbak has joined #openstack-ceilometer | 21:21 | |
| *** rbak has quit IRC | 21:23 | |
| *** nadya has quit IRC | 21:25 | |
| *** jwcroppe has quit IRC | 21:25 | |
| *** jwcroppe has joined #openstack-ceilometer | 21:26 | |
| *** rbak has joined #openstack-ceilometer | 21:30 | |
| *** jwcroppe has quit IRC | 21:32 | |
| *** exploreshaifali has quit IRC | 21:36 | |
| *** rbak has quit IRC | 21:36 | |
| *** pauldix_ has joined #openstack-ceilometer | 21:38 | |
| *** pauldix has quit IRC | 21:39 | |
| *** prad has quit IRC | 21:48 | |
| *** edmondsw has quit IRC | 22:00 | |
| openstackgerrit | Merged openstack/ceilometer: Stop using openstack.common from keystoneclient https://review.openstack.org/209402 | 22:06 |
| *** rbak has joined #openstack-ceilometer | 22:18 | |
| *** tongli has quit IRC | 22:28 | |
| *** thorst has quit IRC | 22:32 | |
| *** prad has joined #openstack-ceilometer | 22:38 | |
| *** pauldix_ has quit IRC | 22:56 | |
| *** pauldix has joined #openstack-ceilometer | 22:59 | |
| *** pauldix has quit IRC | 22:59 | |
| *** rbak has quit IRC | 23:01 | |
| *** julim has joined #openstack-ceilometer | 23:50 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!