*** promulo has joined #openstack-ceilometer | 00:04 | |
*** jergerber has joined #openstack-ceilometer | 00:09 | |
*** jergerber has quit IRC | 00:09 | |
*** nati_uen_ has joined #openstack-ceilometer | 00:11 | |
*** nati_ueno has quit IRC | 00:12 | |
*** nati_ueno has joined #openstack-ceilometer | 00:12 | |
*** mgagne has joined #openstack-ceilometer | 00:15 | |
*** nati_uen_ has quit IRC | 00:16 | |
*** ChanServ changes topic to "All of the project infrastructure hosts are being restarted for security updates." | 00:21 | |
*** openstackgerrit has quit IRC | 00:24 | |
*** nosnos has joined #openstack-ceilometer | 00:29 | |
*** _cjones_ has quit IRC | 00:43 | |
*** _cjones_ has joined #openstack-ceilometer | 00:44 | |
*** alexpilotti has quit IRC | 00:44 | |
*** terriyu has joined #openstack-ceilometer | 00:59 | |
*** nosnos has quit IRC | 01:16 | |
*** terriyu has quit IRC | 01:21 | |
*** julim has quit IRC | 01:28 | |
*** _cjones_ has quit IRC | 01:28 | |
*** matsuhashi has joined #openstack-ceilometer | 01:28 | |
*** _cjones_ has joined #openstack-ceilometer | 01:29 | |
*** ChanServ changes topic to "OpenStack Telemetry (Ceilometer) | https://wiki.openstack.org/wiki/Ceilometer" | 01:32 | |
openstackstatus | NOTICE: All services should be back up | 01:32 |
---|---|---|
*** _cjones_ has quit IRC | 01:33 | |
*** openstack has joined #openstack-ceilometer | 01:48 | |
*** changbl has joined #openstack-ceilometer | 02:00 | |
*** flwang has quit IRC | 02:14 | |
*** nati_ueno has quit IRC | 02:19 | |
*** changbl has quit IRC | 02:29 | |
*** flwang has joined #openstack-ceilometer | 02:30 | |
*** changbl has joined #openstack-ceilometer | 03:24 | |
*** matsuhashi has quit IRC | 03:29 | |
*** matsuhashi has joined #openstack-ceilometer | 04:19 | |
*** promulo has quit IRC | 04:48 | |
*** nati_ueno has joined #openstack-ceilometer | 04:53 | |
*** mihgen has joined #openstack-ceilometer | 05:04 | |
*** ildikov_ has quit IRC | 05:08 | |
*** saju_m has joined #openstack-ceilometer | 05:28 | |
*** ildikov_ has joined #openstack-ceilometer | 05:32 | |
*** _nadya_ has joined #openstack-ceilometer | 05:37 | |
*** _nadya_ has quit IRC | 05:42 | |
*** liusheng has joined #openstack-ceilometer | 06:02 | |
openstackgerrit | Jenkins proposed a change to openstack/ceilometer: Imported Translations from Transifex https://review.openstack.org/84359 | 06:04 |
*** _nadya_ has joined #openstack-ceilometer | 06:04 | |
*** AMike has joined #openstack-ceilometer | 06:23 | |
*** AMike has joined #openstack-ceilometer | 06:23 | |
*** _nadya_ has quit IRC | 06:32 | |
*** inc0 has joined #openstack-ceilometer | 06:37 | |
*** saju_m has quit IRC | 07:17 | |
*** eouti has joined #openstack-ceilometer | 07:17 | |
*** saju_m has joined #openstack-ceilometer | 07:25 | |
*** saju_m has quit IRC | 07:32 | |
*** eouti has quit IRC | 07:33 | |
*** matsuhashi has quit IRC | 07:53 | |
*** matsuhashi has joined #openstack-ceilometer | 07:54 | |
*** saju_m has joined #openstack-ceilometer | 07:58 | |
*** saju_m has quit IRC | 08:03 | |
*** mihgen has quit IRC | 08:09 | |
*** nati_ueno has quit IRC | 08:11 | |
*** saju_m has joined #openstack-ceilometer | 08:16 | |
*** nati_ueno has joined #openstack-ceilometer | 08:18 | |
*** nati_ueno has quit IRC | 08:18 | |
*** scroiset_ is now known as scroiset | 08:21 | |
*** admin0 has joined #openstack-ceilometer | 08:22 | |
*** nprivalova has quit IRC | 08:29 | |
*** idegtiarov has joined #openstack-ceilometer | 08:31 | |
*** _nadya_ has joined #openstack-ceilometer | 08:37 | |
*** alexpilotti has joined #openstack-ceilometer | 08:50 | |
*** mihgen has joined #openstack-ceilometer | 09:00 | |
*** saju_m has quit IRC | 09:39 | |
*** matsuhashi has quit IRC | 09:39 | |
*** inc0 has quit IRC | 09:39 | |
*** mgagne has quit IRC | 09:39 | |
*** zul has quit IRC | 09:39 | |
*** bada has quit IRC | 09:39 | |
*** bogdando has quit IRC | 09:39 | |
*** annegentle_ has quit IRC | 09:39 | |
*** Guest21911 has quit IRC | 09:39 | |
*** tasdomas has quit IRC | 09:39 | |
*** EmilienM has quit IRC | 09:39 | |
*** dhellmann has quit IRC | 09:39 | |
*** nijaba has quit IRC | 09:39 | |
*** dragondm has quit IRC | 09:39 | |
*** inc0 has joined #openstack-ceilometer | 09:39 | |
*** bada has joined #openstack-ceilometer | 09:39 | |
*** bogdando has joined #openstack-ceilometer | 09:39 | |
*** matsuhashi has joined #openstack-ceilometer | 09:39 | |
*** EmilienM has joined #openstack-ceilometer | 09:39 | |
*** bada has quit IRC | 09:39 | |
*** bogdando has quit IRC | 09:39 | |
*** idegtiarov has quit IRC | 09:39 | |
*** openstackgerrit has quit IRC | 09:39 | |
*** sdake has quit IRC | 09:39 | |
*** thomasem has quit IRC | 09:39 | |
*** Daviey has quit IRC | 09:39 | |
*** openstack has joined #openstack-ceilometer | 09:49 | |
*** wayneeseguin has joined #openstack-ceilometer | 09:52 | |
*** matsuhashi has quit IRC | 09:52 | |
*** kin has joined #openstack-ceilometer | 09:52 | |
*** matsuhashi has joined #openstack-ceilometer | 09:53 | |
*** nsaje has quit IRC | 09:57 | |
*** nsaje has joined #openstack-ceilometer | 09:57 | |
*** nsaje has quit IRC | 10:03 | |
*** sileht has quit IRC | 10:03 | |
*** zigo has quit IRC | 10:03 | |
*** nsaje has joined #openstack-ceilometer | 10:04 | |
*** zigo has joined #openstack-ceilometer | 10:04 | |
*** sileht has joined #openstack-ceilometer | 10:04 | |
*** yassine has joined #openstack-ceilometer | 10:09 | |
*** annegentle has quit IRC | 10:13 | |
*** Guest99168 has quit IRC | 10:15 | |
*** EmilienM has quit IRC | 10:15 | |
*** mgagne has quit IRC | 10:16 | |
*** _nadya_ has quit IRC | 10:16 | |
*** EmilienM has joined #openstack-ceilometer | 10:22 | |
*** sdake_ has quit IRC | 10:26 | |
*** scroiset has quit IRC | 10:26 | |
*** bogdando has quit IRC | 10:26 | |
*** inc0 has quit IRC | 10:26 | |
*** alexpilotti has quit IRC | 10:26 | |
*** eloib_ has quit IRC | 10:26 | |
*** Wangpan has quit IRC | 10:26 | |
*** yjiang5_1 has quit IRC | 10:26 | |
*** ondergetekende has quit IRC | 10:26 | |
*** scroiset has joined #openstack-ceilometer | 10:27 | |
*** sdake_ has joined #openstack-ceilometer | 10:27 | |
*** sdake_ has joined #openstack-ceilometer | 10:27 | |
*** bogdando has joined #openstack-ceilometer | 10:27 | |
*** Wangpan has joined #openstack-ceilometer | 10:27 | |
*** alexpilotti has joined #openstack-ceilometer | 10:27 | |
*** yjiang5_1 has joined #openstack-ceilometer | 10:27 | |
*** inc0 has joined #openstack-ceilometer | 10:27 | |
*** yjiang5_1 has quit IRC | 10:27 | |
*** yjiang5_1 has joined #openstack-ceilometer | 10:27 | |
*** eloib_ has joined #openstack-ceilometer | 10:28 | |
*** ondergetekende has joined #openstack-ceilometer | 10:31 | |
*** saju_m has quit IRC | 10:34 | |
*** saju_m has joined #openstack-ceilometer | 10:35 | |
*** sdake_ has quit IRC | 10:39 | |
*** sdake has quit IRC | 10:39 | |
*** fc__ has quit IRC | 10:39 | |
*** idegtiarov has quit IRC | 10:39 | |
*** AMike has quit IRC | 10:39 | |
*** liusheng has quit IRC | 10:39 | |
*** lsmola_ has quit IRC | 10:39 | |
*** idegtiarov has joined #openstack-ceilometer | 10:39 | |
*** fc__ has joined #openstack-ceilometer | 10:40 | |
*** sdake_ has joined #openstack-ceilometer | 10:40 | |
*** sdake has joined #openstack-ceilometer | 10:40 | |
*** liusheng has joined #openstack-ceilometer | 10:40 | |
*** sdake has quit IRC | 10:40 | |
*** sdake has joined #openstack-ceilometer | 10:40 | |
*** sdake_ has quit IRC | 10:40 | |
*** sdake_ has joined #openstack-ceilometer | 10:40 | |
*** AMike has joined #openstack-ceilometer | 10:40 | |
*** AMike has quit IRC | 10:40 | |
*** AMike has joined #openstack-ceilometer | 10:40 | |
*** lsmola has joined #openstack-ceilometer | 10:40 | |
*** lsmola has quit IRC | 10:40 | |
*** lsmola has joined #openstack-ceilometer | 10:40 | |
*** matsuhashi has quit IRC | 10:53 | |
*** matsuhashi has joined #openstack-ceilometer | 10:54 | |
*** eglynn has joined #openstack-ceilometer | 10:57 | |
*** sdake has quit IRC | 10:57 | |
*** matsuhashi has quit IRC | 10:58 | |
*** matsuhashi has joined #openstack-ceilometer | 10:59 | |
*** ildikov_ has joined #openstack-ceilometer | 11:02 | |
*** sdake has joined #openstack-ceilometer | 11:31 | |
*** annegentle has joined #openstack-ceilometer | 11:46 | |
*** mgagne has joined #openstack-ceilometer | 11:48 | |
*** lillie has joined #openstack-ceilometer | 11:48 | |
*** lillie is now known as Guest19269 | 11:49 | |
*** Qlawy has quit IRC | 11:51 | |
*** Qlawy has joined #openstack-ceilometer | 11:52 | |
*** Qlawy has joined #openstack-ceilometer | 11:52 | |
sileht | ildikov_, the merge of https://review.openstack.org/#/c/78132/ have failed, I have been forced to rebase it manually, if you can retake a look to it, thx in advance. | 11:54 |
ildikov_ | sileht: sure, I'm already looking | 11:56 |
ildikov_ | sileht: I posted a coment on it, that function call is not needed, as I refactored the way how it works and it was merged an hour ago | 11:59 |
ildikov_ | sileht: sorry for making one more turnaround with this patch, I promise there will not be more, at least not because of me :) | 12:01 |
sileht | ildikov_, good catch ! | 12:08 |
sileht | ildikov_, fixed | 12:08 |
*** sdake has quit IRC | 12:09 | |
*** nprivalova has joined #openstack-ceilometer | 12:14 | |
*** nsaje has quit IRC | 12:16 | |
*** annegentle has quit IRC | 12:17 | |
*** nsaje has joined #openstack-ceilometer | 12:17 | |
*** zigo has quit IRC | 12:17 | |
*** Guest19269 has quit IRC | 12:17 | |
*** matsuhashi has quit IRC | 12:18 | |
*** zigo has joined #openstack-ceilometer | 12:18 | |
*** matsuhashi has joined #openstack-ceilometer | 12:18 | |
*** sileht has quit IRC | 12:19 | |
*** mgagne has quit IRC | 12:19 | |
*** matsuhashi has quit IRC | 12:19 | |
*** matsuhashi has joined #openstack-ceilometer | 12:19 | |
*** mgagne has joined #openstack-ceilometer | 12:21 | |
*** annegentle has joined #openstack-ceilometer | 12:21 | |
*** lillie- has joined #openstack-ceilometer | 12:22 | |
*** eouti has joined #openstack-ceilometer | 12:22 | |
*** sileht has joined #openstack-ceilometer | 12:22 | |
*** _nadya_ has joined #openstack-ceilometer | 12:23 | |
*** erecio has joined #openstack-ceilometer | 12:27 | |
ildikov_ | sileht: thanks for the fix, the patch is on its way to the gate now :) | 12:28 |
*** matsuhashi has quit IRC | 12:30 | |
*** changbl has quit IRC | 12:32 | |
*** erecio has quit IRC | 12:36 | |
*** gordc has joined #openstack-ceilometer | 12:39 | |
*** heyongli has joined #openstack-ceilometer | 12:44 | |
*** jdob has joined #openstack-ceilometer | 12:45 | |
*** lillie- is now known as lillie | 12:49 | |
*** lillie is now known as Guest37432 | 12:50 | |
*** jmckind has joined #openstack-ceilometer | 12:57 | |
*** erecio has joined #openstack-ceilometer | 12:59 | |
*** _nadya_ has quit IRC | 13:00 | |
*** jaypipes has joined #openstack-ceilometer | 13:06 | |
*** _nadya_ has joined #openstack-ceilometer | 13:14 | |
*** openstackgerrit has joined #openstack-ceilometer | 13:23 | |
*** openstackgerrit has quit IRC | 13:23 | |
*** openstackgerrit has joined #openstack-ceilometer | 13:24 | |
*** _nadya_ has quit IRC | 13:28 | |
*** julim has joined #openstack-ceilometer | 13:39 | |
*** thomasem has joined #openstack-ceilometer | 13:41 | |
*** Hao has joined #openstack-ceilometer | 13:42 | |
*** nacim has joined #openstack-ceilometer | 13:42 | |
*** alexpilotti has quit IRC | 13:51 | |
*** rdmcnair has joined #openstack-ceilometer | 13:57 | |
*** prad has joined #openstack-ceilometer | 13:59 | |
eglynn | jd__: can you take another look at this unflattening hackery if you get a chance ... https://review.openstack.org/85990 | 14:02 |
eglynn | (... /me eager to get it landed on master so I can backport to milestone-proposed) | 14:02 |
*** nacim has quit IRC | 14:02 | |
jd__ | done | 14:04 |
eglynn | jd__: thank you sir! | 14:05 |
*** jergerber has joined #openstack-ceilometer | 14:08 | |
nprivalova | eglynn: | 14:10 |
eglynn | nprivalova: | 14:10 |
*** nacim has joined #openstack-ceilometer | 14:10 | |
eglynn | ;) | 14:10 |
nprivalova | eglynn: hi :) about 85990... | 14:11 |
eglynn | nprivalova: shoot! | 14:11 |
nprivalova | eglynn: 1st question: user may post 'complex' metadata now? | 14:12 |
nprivalova | eglynn: I mean dict of dicts | 14:12 |
eglynn | nprivalova: yeah, so the requirement came from a user on the ML who wanted to POST samples for a custom metric and base autoscaling decisions on those data | 14:13 |
eglynn | nprivalova: ... membership of an autoscaling group is represented by user_metadata on the instance | 14:13 |
eglynn | nprivalova: ... which is nested within the resource_metadata | 14:13 |
nprivalova | eglynn: please take a look on https://review.openstack.org/#/c/85990/5/ceilometer/api/controllers/v2.py line 640 | 14:14 |
nprivalova | eglynn: so our validation allows to post only {a:b} and nothing more complex | 14:14 |
eglynn | nprivalova: do you mean: resource_metadata = {wtypes.text: wtypes.text} ? | 14:14 |
nprivalova | eglynn: yep | 14:14 |
eglynn | nprivalova: yeah so that constraint is maintained | 14:14 |
eglynn | nprivalova: the user POSTs the nested data in flattened form | 14:15 |
eglynn | nprivalova: since that's already the "externalized" format for resource metadata' | 14:15 |
eglynn | nprivalova: see for example the test https://review.openstack.org/#/c/85990/5/ceilometer/tests/api/v2/test_post_samples_scenarios.py | 14:15 |
nprivalova | eglynn: ok. yep, the second question is about the test :) | 14:16 |
nprivalova | eglynn: this test is not for dict of dicts too, right? only one-dimention? | 14:17 |
eglynn | nprivalova: sorry distracted with internal chatter | 14:20 |
nprivalova | eglynn: let me try to explain my concerns...it's in https://review.openstack.org/#/c/85990/5/ceilometer/api/controllers/v2.py line 566 | 14:20 |
*** heyongli has quit IRC | 14:20 | |
eglynn | nprivalova: k, the test asserts that a pre-flattened version of a nested dict can be POST'd | 14:20 |
eglynn | nprivalova: is this comment correct? output now: a.b.c=d | 14:22 |
eglynn | nprivalova: (as opposed to "output before: a.b:c=d") | 14:22 |
nprivalova | eglynn: yep, but to keep compatibility we change it back | 14:22 |
eglynn | nprivalova: arggh, really? | 14:22 |
nprivalova | eglynn: yep | 14:22 |
eglynn | sad face | 14:23 |
eglynn | nprivalova: change it back when? | 14:23 |
nprivalova | eglynn: sorry( Gordon and me wanted to change API output but I get -2 | 14:23 |
eglynn | nprivalova: ... before returning the sample with metadata? | 14:23 |
openstackgerrit | A change was merged to openstack/ceilometer: Sync oslo-incubator code https://review.openstack.org/78132 | 14:24 |
nprivalova | eglynn: but! if we are using 1-level deep of dict everything seems to be fine | 14:25 |
eglynn | nprivalova: yeah one-level is all that's needed for the intended use-case | 14:25 |
eglynn | ... /me mulls a follow-up patch | 14:26 |
eglynn | nprivalova: I'm not sure which is worse ... | 14:26 |
eglynn | nprivalova: 1. requiring the caller of POST /v2/meters must know about the special-casing for >1 level of nesting | 14:27 |
eglynn | nprivalova: or: | 14:27 |
eglynn | nprivalova: 2. allowing the caller of POST /v2/meters to nest in the natural way a.b.c and then returning a.b:c when the sample is later queried | 14:28 |
nprivalova | eglynn: about 1: we don't allow to put >1 level of nesting. Let me thing about the case when user send already 'flatten' data like {'a.b.c':d} | 14:30 |
nprivalova | eglynn: {'a.b.c': 'd'} | 14:30 |
nprivalova | eglynn: *think | 14:30 |
eglynn | nprivalova: "we don't allow to put >1 level of nesting" ==> where is that asserted? | 14:35 |
nprivalova | eglynn: in validation resource_metadata = {wtypes.text: wtypes.text} | 14:35 |
nprivalova | eglynn: we will get ValidationError if resource_metadata field in POST will be '{ 'a' : {'b': 'c'} }' | 14:37 |
nprivalova | *is | 14:37 |
eglynn | nprivalova: but we don't POST { 'a' : {'b': 'c'} } | 14:38 |
eglynn | nprivalova: instead we POST a flattened form | 14:38 |
eglynn | nprivalova: ... resource_metadata = {'a.b': 'c'} | 14:38 |
eglynn | nprivalova: ... which passes validation | 14:38 |
nprivalova | eglynn: yep, agreed... | 14:39 |
nprivalova | eglynn: let me write one test to check something | 14:40 |
eglynn | nprivalova: ... so are you proposing that we accept resource_metadata = {'a.b': 'foo'} but not resource_metadata = {'a.b.c': 'foobar'} ? | 14:40 |
*** saju_m has quit IRC | 14:41 | |
nprivalova | eglynn: no. I'm still finding solution | 14:42 |
eglynn | nprivalova: k | 14:42 |
eglynn | nprivalova: ... is there an existing test that asserts the a.b:c=d versus a.b.c=d logic in the response from 'GET /v2/meters/foo'? | 14:44 |
nprivalova | eglynn: yep...Let me 1 sec | 14:44 |
eglynn | nprivalova: cool enough | 14:44 |
*** dhellmann has quit IRC | 14:45 | |
*** jd__ has quit IRC | 14:45 | |
*** tasdomas has quit IRC | 14:45 | |
*** apmelton has quit IRC | 14:45 | |
*** mihgen has quit IRC | 14:45 | |
*** SergeyLukjanov has quit IRC | 14:45 | |
*** shadower has quit IRC | 14:45 | |
*** nikhil__ has quit IRC | 14:45 | |
*** shadower has joined #openstack-ceilometer | 14:48 | |
*** shadower has quit IRC | 14:49 | |
*** shadower has joined #openstack-ceilometer | 14:49 | |
*** apmelton has joined #openstack-ceilometer | 14:49 | |
*** apmelton has quit IRC | 14:49 | |
*** SergeyLukjanov has joined #openstack-ceilometer | 14:49 | |
*** apmelton has joined #openstack-ceilometer | 14:49 | |
*** dhellmann has joined #openstack-ceilometer | 14:49 | |
*** nikhil__ has joined #openstack-ceilometer | 14:50 | |
*** jd__ has joined #openstack-ceilometer | 14:50 | |
nprivalova | eglynn: e.g. tests.api.v2.test_list_meters_scenarios.TestListMeters.test_list_samples_with_dict_metadata | 14:50 |
*** tasdomas has joined #openstack-ceilometer | 14:52 | |
eglynn | nprivalova: so we already have a pretty bad inconsistency there ... | 14:54 |
eglynn | nprivalova: ... in the sense that the metadata query must be expressed with the nested property identified as metadata.properties.prop_2.sub_prop_1 | 14:54 |
nprivalova | eglynn: yep, it was my first bug in Ceilometer :) | 14:54 |
eglynn | nprivalova: ... i.e. no colon, periods all the way thru' in the query.field | 14:55 |
eglynn | nprivalova: ... ah, nostalgia! :) | 14:55 |
eglynn | nprivalova: whereas in the response to the metadata query, we returned "properties.prop_2:sub_prop_1" in the actual sample | 14:57 |
nprivalova | eglynn: I can suggest two solutions... 1. remove https://review.openstack.org/#/c/85990/5/ceilometer/api/controllers/v2.py line 578 about replace. But in this case we change API output. All tests should be changed | 14:57 |
eglynn | nprivalova: so a matches b, where a != b | 14:57 |
nprivalova | eglynn: yep | 14:57 |
eglynn | nprivalova: so given that currently when the user needs to specify nested metadata in an API call, they do so with periods through-out | 14:58 |
eglynn | nprivalova: (i.e. that case arises when the user is making a metadata query) | 14:58 |
eglynn | nprivalova: ... then I think we should allow them to use the same form to specify nested metadata when POSTing samples | 14:59 |
eglynn | nprivalova: ... otherwise we'd have to require that incoming nested metadata was expressed in two different ways by the API caller | 14:59 |
eglynn | nprivalova: ... does my reasoning above make sense? | 15:00 |
nprivalova | eglynn: 2. Add one more hack like before if k not in metadata: k.replace('.', ':').replace(':', '.', 1) before line 578 . Let me read you suggestions above :) | 15:01 |
*** Hao has quit IRC | 15:05 | |
nprivalova | eglynn: I don't think we should make user think about our 'broken' format... We should make only one thing now: if user provides flattened structure we should retun it just in the same way. So if POST '{a.b.c: d}' => '{a.b.c: d}' after GET, if '{a.b$whatever_separator$c: d}' the same should be returned | 15:08 |
nprivalova | eglynn: so in 578 we should NOT do replacing in key if metadata dict contains this key. what do you think? | 15:09 |
prad | nprivalova: what you think of my suggestion https://review.openstack.org/#/c/84438/3 .. is that an acceptable balance? basically keep the None checks for image ref (As its needed) and remove it for instance name and status | 15:09 |
*** mihgen has joined #openstack-ceilometer | 15:09 | |
eglynn | nprivalova: hmmm, are you suggesting that we support BOTH {'a.b.c': 'foo'} AND {'a.b:c': 'foo'} to represent {'a': {'b': {'c': 'foo'}}} in POST'd samples? | 15:11 |
nprivalova | prad, I will take a look now | 15:12 |
nprivalova | eglynn: no. In the case you described we return only {'a.b:c': 'foo'}. But in the case of POST {'a.b.c': 'foo'} we return {'a.b.c': 'foo'} and in case of {'a.b:c': 'foo'} we return {'a.b:c': 'foo'} | 15:13 |
*** Hao has joined #openstack-ceilometer | 15:14 | |
nprivalova | eglynn: now in case of POST {'a.b.c': 'foo'} we return POST {'a.b:c': 'foo'}. it's wrong | 15:14 |
eglynn | nprivalova: "in the case of POST {'a.b.c': 'foo'} we return {'a.b.c': 'foo'}" ==> are you talking about the immediate response from the POST? | 15:14 |
eglynn | nprivalova: ... or the response to a subsequent GET? | 15:15 |
nprivalova | eglynn: after GET of course | 15:15 |
eglynn | nprivalova: I confused as to how that could ever work | 15:16 |
eglynn | nprivalova: ... as we don't maintain the distinction in the stored representation | 15:16 |
eglynn | nprivalova: ... the nested metadata is stored as {'a': {'b': {'c': 'foo'}}} regardless of how it was POST'd | 15:16 |
eglynn | nprivalova: ... or? | 15:17 |
eglynn | nprivalova: ... i.e. how would be remember that the original POST'd form was {'a.b:c': 'foo'} as opposed to {'a.b.c': 'foo'} | 15:17 |
eglynn | how would *we remember | 15:18 |
nprivalova | eglynn: I believe that if user POSTs {'a.b.c': 'foo'} we store it it the same way. don't convert to {'a': {'b': {'c': 'foo'}}} | 15:19 |
eglynn | nprivalova: ... no, the whole point of that patch is to unwind POST {'a.b.c': 'foo'} to {'a': {'b': {'c': 'foo'}}} prior to storage | 15:19 |
eglynn | nprivalova: ... otherwise the 'a.b.c' fails to be persisted in mongo | 15:20 |
openstackgerrit | A change was merged to openstack/ceilometer: Allowed nested resource metadata in POST'd samples https://review.openstack.org/85990 | 15:20 |
eglynn | nprivalova: ... (because of the embedded '.' in the key) | 15:21 |
eglynn | nprivalova: ... were you referring to the situation *prior* to the patch just landed? | 15:21 |
nprivalova | eglynn: prior patch. Now I understand your question "are you suggesting that we support BOTH {'a.b.c': 'foo'} AND {'a.b:c': 'foo'} to represent {'a': {'b': {'c': 'foo'}}} in POST'd samples" | 15:22 |
*** changbl has joined #openstack-ceilometer | 15:23 | |
eglynn | nprivalova: but prior to the patch, *nothing* is persisted in mongo when the user POSTs nested metadata | 15:23 |
jd__ | eglynn: I think you can push the backport now | 15:24 |
eglynn | jd__: yep I'm on it | 15:24 |
nprivalova | eglynn: let me try to provide specific example from your patch | 15:25 |
eglynn | jd__: ... https://review.openstack.org/86061 | 15:26 |
*** eouti has quit IRC | 15:29 | |
*** nacim has quit IRC | 15:36 | |
openstackgerrit | Pradeep Kilambi proposed a change to openstack/ceilometer: Include instance state in metadata https://review.openstack.org/84438 | 15:46 |
nprivalova | eglynn: cannot quick create a test with both post_json and get_json in one test. But I think I've understood your change better. So looks like we should provide additional info in docs... I think you are right about format in POST. We may require that nested structures should be separated by '.'. But from the other hand we always returns in a.b:c format after GET :( And it should be in docs too | 15:51 |
eglynn | nprivalova: at least now there's some kind of consistency ... | 15:52 |
eglynn | nprivalova: 1. user-specified flattening ==> periods throughout | 15:52 |
eglynn | nprivalova: 2. API-returned flattening ==>periods for initial levels, colon for last level if > 1 | 15:53 |
nprivalova | eglynn: yep. are we going to have APIv3 :)? | 15:53 |
eglynn | nprivalova: I think the weight of the accreted cruft in v2 is gonna force that upon us sooner rather than later | 15:54 |
*** idegtiarov has quit IRC | 15:57 | |
*** _cjones_ has joined #openstack-ceilometer | 15:57 | |
nprivalova | eglynn: thanks for discussion, sorry about this metadata stuff but it's really nightmare and I wanted to be sure that it is not becoming even worst | 15:58 |
eglynn | nprivalova: well thank you! ... you've opened my eyes to some of the madness that lurks beneath the surface | 15:59 |
*** Ruetobas has quit IRC | 16:01 | |
*** Ruetobas has joined #openstack-ceilometer | 16:03 | |
*** nacim has joined #openstack-ceilometer | 16:04 | |
*** julim has quit IRC | 16:05 | |
*** Ruetobas has quit IRC | 16:08 | |
*** sdake has joined #openstack-ceilometer | 16:11 | |
*** mihgen has quit IRC | 16:13 | |
*** mihgen has joined #openstack-ceilometer | 16:14 | |
*** Ruetobas has joined #openstack-ceilometer | 16:14 | |
*** kun_huang has joined #openstack-ceilometer | 16:16 | |
*** eglynn has quit IRC | 16:22 | |
openstackgerrit | Pradeep Kilambi proposed a change to openstack/ceilometer: Include instance state in metadata https://review.openstack.org/84438 | 16:22 |
openstackgerrit | Pradeep Kilambi proposed a change to openstack/ceilometer: Include instance state in metadata https://review.openstack.org/84438 | 16:24 |
*** saju_m has joined #openstack-ceilometer | 16:35 | |
openstackgerrit | Pradeep Kilambi proposed a change to openstack/ceilometer: Ensure instance.image is not None https://review.openstack.org/86085 | 16:38 |
*** ildikov_ has quit IRC | 16:41 | |
*** yassine has quit IRC | 16:42 | |
*** promulo has joined #openstack-ceilometer | 16:48 | |
*** nacim has quit IRC | 16:51 | |
*** gordc has quit IRC | 16:54 | |
*** gordc has joined #openstack-ceilometer | 16:58 | |
*** admin0 has quit IRC | 16:58 | |
*** julim has joined #openstack-ceilometer | 17:02 | |
*** _nadya_ has joined #openstack-ceilometer | 17:03 | |
*** gordc has quit IRC | 17:07 | |
*** gordc has joined #openstack-ceilometer | 17:09 | |
*** Hao has quit IRC | 17:11 | |
*** Hao has joined #openstack-ceilometer | 17:11 | |
*** jdob_ has joined #openstack-ceilometer | 17:13 | |
*** Hao has quit IRC | 17:15 | |
*** Hao has joined #openstack-ceilometer | 17:19 | |
*** _nadya_ has quit IRC | 17:22 | |
*** Hao has quit IRC | 17:23 | |
*** jdob_ has quit IRC | 17:25 | |
*** cdent has joined #openstack-ceilometer | 17:36 | |
openstackgerrit | Pradeep Kilambi proposed a change to openstack/ceilometer: Set appropriate metadata image value https://review.openstack.org/86085 | 17:39 |
*** Hao has joined #openstack-ceilometer | 17:42 | |
*** Hao has quit IRC | 17:46 | |
*** gordc has quit IRC | 17:47 | |
*** _nadya_ has joined #openstack-ceilometer | 17:48 | |
*** cdent has quit IRC | 17:52 | |
*** cdent has joined #openstack-ceilometer | 17:54 | |
*** anteaya has quit IRC | 17:56 | |
*** terriyu has joined #openstack-ceilometer | 17:56 | |
*** _cjones_ has quit IRC | 18:04 | |
*** _cjones_ has joined #openstack-ceilometer | 18:05 | |
*** _cjones_ has quit IRC | 18:09 | |
*** cosmin_ has joined #openstack-ceilometer | 18:09 | |
*** terriyu has quit IRC | 18:10 | |
cosmin_ | part | 18:11 |
*** cosmin_ has left #openstack-ceilometer | 18:11 | |
*** gordc has joined #openstack-ceilometer | 18:11 | |
*** cosmin_ has joined #openstack-ceilometer | 18:11 | |
*** cosmin_ has quit IRC | 18:12 | |
*** kun_huang has quit IRC | 18:12 | |
*** Hao has joined #openstack-ceilometer | 18:22 | |
*** Hao has quit IRC | 18:26 | |
*** _nadya_ has quit IRC | 18:26 | |
*** mihgen has quit IRC | 18:32 | |
*** mihgen has joined #openstack-ceilometer | 18:33 | |
*** mihgen has quit IRC | 18:34 | |
*** _cjones_ has joined #openstack-ceilometer | 18:35 | |
*** eglynn has joined #openstack-ceilometer | 18:41 | |
*** saju_m has quit IRC | 18:44 | |
*** ildikov_ has joined #openstack-ceilometer | 18:51 | |
*** inc0 has quit IRC | 18:55 | |
*** cdent has quit IRC | 19:17 | |
*** ildikov has joined #openstack-ceilometer | 19:18 | |
*** Hao has joined #openstack-ceilometer | 19:22 | |
*** mihgen has joined #openstack-ceilometer | 19:26 | |
*** Hao has quit IRC | 19:26 | |
*** mihgen has quit IRC | 19:50 | |
openstackgerrit | A change was merged to openstack/ceilometer: Include instance state in metadata https://review.openstack.org/84438 | 20:01 |
*** Vetalll has joined #openstack-ceilometer | 20:10 | |
Vetalll | Hi guys. I am new in ceilometer, and would like to work with you | 20:13 |
Vetalll | I started with a little patch, and i would like to ask you - could you check it https://review.openstack.org/#/c/85560/ ? | 20:14 |
Vetalll | Also i have a question. I think we can use custom assertion for all objects which have length | 20:15 |
Vetalll | I need your opinion about this. | 20:16 |
*** erecio has quit IRC | 20:19 | |
*** mihgen has joined #openstack-ceilometer | 20:19 | |
*** Hao has joined #openstack-ceilometer | 20:23 | |
*** gordc has quit IRC | 20:25 | |
*** Hao has quit IRC | 20:27 | |
openstackgerrit | ZhiQiang Fan proposed a change to openstack/ceilometer: Exit collector service if no dispatcher available https://review.openstack.org/86155 | 20:52 |
*** kin is now known as nikhil___ | 20:53 | |
*** janisg has quit IRC | 21:05 | |
*** janisg has joined #openstack-ceilometer | 21:06 | |
*** jdob has quit IRC | 21:07 | |
*** Vetalll has quit IRC | 21:14 | |
*** jmckind has quit IRC | 21:15 | |
*** Hao has joined #openstack-ceilometer | 21:24 | |
*** eglynn has quit IRC | 21:26 | |
*** Hao has quit IRC | 21:28 | |
*** promulo has quit IRC | 21:30 | |
*** thomasem has quit IRC | 21:45 | |
*** Hao has joined #openstack-ceilometer | 22:21 | |
*** Hao has quit IRC | 22:23 | |
*** Hao has joined #openstack-ceilometer | 22:23 | |
*** Hao has quit IRC | 22:28 | |
*** jergerber has quit IRC | 22:30 | |
openstackgerrit | ZhiQiang Fan proposed a change to openstack/ceilometer: Disable specifying alarm itself in combination rule https://review.openstack.org/86181 | 22:31 |
*** prad has quit IRC | 22:37 | |
*** nikhil__ is now known as nikhil__|afk | 22:51 | |
*** changbl has quit IRC | 23:00 | |
*** Hao has joined #openstack-ceilometer | 23:24 | |
*** fnaval has joined #openstack-ceilometer | 23:25 | |
*** fnaval has quit IRC | 23:25 | |
*** Hao has quit IRC | 23:28 | |
*** fnaval has joined #openstack-ceilometer | 23:28 | |
*** fnaval has quit IRC | 23:28 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!