Thursday, 2016-02-25

« Wednesday, 2016-02-24 Index Friday, 2016-02-26 »
*** diogogmt has quit IRC00:10
*** ljxiash has joined #openstack-telemetry00:18
*** annasort has joined #openstack-telemetry00:18
*** ljxiash has quit IRC00:23
*** ildikov has quit IRC00:48
*** ildikov has joined #openstack-telemetry00:52
*** cheneydc has joined #openstack-telemetry01:08
openstackgerritliusheng proposed openstack/python-aodhclient: WIP: Add composite rule alarm support in aodhclient  https://review.openstack.org/28402201:18
*** julim has joined #openstack-telemetry01:26
*** thorst has joined #openstack-telemetry01:28
*** thorst has quit IRC01:34
*** thorst has joined #openstack-telemetry01:34
openstackgerritliusheng proposed openstack/python-aodhclient: WIP: Add composite rule alarm support in aodhclient  https://review.openstack.org/28402201:35
*** thorst has quit IRC01:39
*** ljxiash has joined #openstack-telemetry01:44
*** Liuqing has joined #openstack-telemetry01:51
r-mibupradk, gordc, hi - i would like to tempest code to migrate code into aodh in this cycle, but let me check the issue first01:57
*** diogogmt has joined #openstack-telemetry02:05
*** liamji has joined #openstack-telemetry02:33
*** Ephur has quit IRC02:36
openstackgerritliusheng proposed openstack/python-aodhclient: WIP: Add composite rule alarm support in aodhclient  https://review.openstack.org/28402202:42
*** achatterjee has joined #openstack-telemetry02:45
*** links has joined #openstack-telemetry02:47
openstackgerritliusheng proposed openstack/aodh: Add releasenote for composite alarm feature  https://review.openstack.org/28450503:03
*** liusheng has left #openstack-telemetry03:04
*** boris-42 has quit IRC03:04
*** liusheng has joined #openstack-telemetry03:04
*** diogogmt has quit IRC03:23
*** diogogmt has joined #openstack-telemetry03:26
*** links has quit IRC03:34
*** Liuqing has quit IRC03:37
*** ljxiash has quit IRC03:38
*** Liuqing has joined #openstack-telemetry03:38
*** ljxiash has joined #openstack-telemetry03:38
*** ljxiash has quit IRC03:43
*** diogogmt has quit IRC03:43
*** chlong_ has quit IRC03:47
*** chlong_ has joined #openstack-telemetry03:49
*** sanjana_ has joined #openstack-telemetry04:07
openstackgerritgordon chung proposed openstack/gnocchi: fix migration from v1.3 to v2.0  https://review.openstack.org/28452804:21
*** Liuqing has quit IRC04:21
*** Liuqing has joined #openstack-telemetry04:22
*** ljxiash has joined #openstack-telemetry04:28
*** links has joined #openstack-telemetry04:29
*** ljxiash has quit IRC04:34
*** ljxiash has joined #openstack-telemetry04:34
*** ljxiash has quit IRC04:39
*** yprokule has joined #openstack-telemetry04:50
*** Liuqing has quit IRC05:08
*** Liuqing has joined #openstack-telemetry05:08
*** belmoreira has joined #openstack-telemetry05:13
*** ljxiash has joined #openstack-telemetry05:14
*** _nadya_ has joined #openstack-telemetry06:20
liamjiliusheng: it seems that the jenkins fails again?06:30
liushengliamji: :(06:31
liushengdid you find any hint ?06:31
liamjihttp://logs.openstack.org/34/277434/13/check/gate-tempest-dsvm-ceilometer-mongodb-full/6ffedbd/console.html06:32
liamji ERROR! Specified hosts options do not match any hosts06:32
*** liamji has quit IRC06:36
*** liamji has joined #openstack-telemetry06:36
liamjiliusheng: according to the log file, it has some  authority problem to create the user.06:42
*** dave-mccowan has quit IRC06:45
liushengliamji: create user? what's the related log06:46
*** ljxiash has quit IRC06:47
*** Liuqing has quit IRC06:53
*** belmoreira has quit IRC06:53
*** ljxiash has joined #openstack-telemetry06:55
*** _nadya_ has quit IRC07:30
liamji:liusheng 2016-02-25 06:54:37.024 | + die_if_not_set 627 swift_tenant_test1 'Failure creating swift_tenant_test1'07:34
liamjihttp://logs.openstack.org/14/280514/10/check/gate-tempest-dsvm-ceilometer-mongodb-full/f6b67d9/logs/devstack-early.txt.gz07:35
*** ljxiash_ has joined #openstack-telemetry07:35
*** ljxiash has quit IRC07:36
openstackgerritMehdi Abaakouk (sileht) proposed openstack/gnocchi: Add some resource types tests  https://review.openstack.org/27041907:36
openstackgerritMehdi Abaakouk (sileht) proposed openstack/gnocchi: Implements resource attribute number  https://review.openstack.org/27009107:36
openstackgerritMehdi Abaakouk (sileht) proposed openstack/gnocchi: Implements resource attribute bool  https://review.openstack.org/27041807:36
openstackgerritMehdi Abaakouk (sileht) proposed openstack/gnocchi: Don't create Ceilometer resource types by default.  https://review.openstack.org/27032207:36
openstackgerritMehdi Abaakouk (sileht) proposed openstack/gnocchi: Move legacy Ceilometer resource into indexer.  https://review.openstack.org/27026607:36
*** ljxiash_ has quit IRC07:49
*** belmoreira has joined #openstack-telemetry07:51
*** ljxiash has joined #openstack-telemetry07:52
*** belmoreira has quit IRC07:58
*** belmoreira has joined #openstack-telemetry08:00
*** belmoreira has quit IRC08:06
*** boris-42 has joined #openstack-telemetry08:51
*** _nadya_ has joined #openstack-telemetry09:08
*** yassine__ has joined #openstack-telemetry09:10
*** ildikov has quit IRC09:15
*** liamji has quit IRC09:38
*** ljxiash has quit IRC09:40
*** ljxiash has joined #openstack-telemetry09:40
*** ljxiash_ has joined #openstack-telemetry09:44
*** efoley has joined #openstack-telemetry09:44
*** ljxiash has quit IRC09:45
openstackgerritliusheng proposed openstack/python-aodhclient: Add composite rule alarm support in aodhclient  https://review.openstack.org/28402209:52
*** r-mibu has quit IRC09:57
*** ildikov has joined #openstack-telemetry10:01
*** cheneydc has quit IRC10:02
*** efoley has quit IRC10:06
*** efoley_ has joined #openstack-telemetry10:06
*** r-mibu has joined #openstack-telemetry10:08
*** ljxiash has joined #openstack-telemetry10:08
*** ljxiash_ has quit IRC10:10
openstackgerritIlya Tyaptin proposed openstack/ceilometer: Add an update interval to compute discovery  https://review.openstack.org/28432210:17
*** ljxiash has quit IRC10:21
*** ljxiash has joined #openstack-telemetry10:27
*** ljxiash has quit IRC10:32
*** pcaruana has joined #openstack-telemetry10:42
*** cdent has joined #openstack-telemetry10:45
*** belmoreira has joined #openstack-telemetry10:49
*** achatterjee has quit IRC10:52
*** efoley__ has joined #openstack-telemetry11:02
*** efoley_ has quit IRC11:04
*** cdent has quit IRC11:16
*** efoley__ has quit IRC11:46
*** cdent has joined #openstack-telemetry11:53
*** thorst has joined #openstack-telemetry12:01
*** Liuqing has joined #openstack-telemetry12:03
*** idegtiarov has joined #openstack-telemetry12:07
*** Liuqing has quit IRC12:16
*** ljxiash has joined #openstack-telemetry12:40
*** diogogmt has joined #openstack-telemetry12:46
*** gordc has joined #openstack-telemetry12:54
*** pcaruana has quit IRC12:55
*** dave-mccowan has joined #openstack-telemetry12:56
* cdent has transcended cattle12:57
*** efoley__ has joined #openstack-telemetry12:57
openstackgerritIlya Tyaptin proposed openstack/ceilometer: Fix the typo in the gnocchiclient exception  https://review.openstack.org/28470412:59
ityaptincdent, sileht, jd__ ^^^12:59
silehtityaptin, done13:00
ityaptinsileht: Thanks!13:01
gordcityaptin: we need to bump requirements. if i remember correctly, that spelling mistake was in gnocchiclient originally.13:02
*** ildikov has quit IRC13:03
*** efoley_ has joined #openstack-telemetry13:08
*** pcaruana has joined #openstack-telemetry13:09
*** diogogmt has quit IRC13:10
*** pradk has joined #openstack-telemetry13:10
*** pradk has quit IRC13:10
*** efoley__ has quit IRC13:11
openstackgerritZhiQiang Fan proposed openstack/ceilometer: Add validation for polling_namespaces option  https://review.openstack.org/28471013:16
ityaptingordc: you think we should to add support of the NamedMetricAreadyExists and NamedMetricAlreadyExists ?13:26
ityaptinOr only change requirements?13:26
*** links has quit IRC13:26
*** liamji has joined #openstack-telemetry13:30
gordcityaptin: we can probably just change requirements. seems easier and more productive13:31
*** nicodemus_ has joined #openstack-telemetry13:31
ityaptingordc: I agree13:35
gordcityaptin: cool cool13:36
openstackgerritIlya Tyaptin proposed openstack/ceilometer: Fix the typo in the gnocchiclient exception  https://review.openstack.org/28470413:40
ityaptingordc: ^^^13:40
ityaptinsileht: ^^^13:40
silehtityaptin, it seems I have broken backward compat with 2.2.013:42
silehtityaptin, perhaps I should put back the mis-spelling exception13:42
openstackgerritIgor Degtiarov proposed openstack/ceilometer: bracketer event transformer  https://review.openstack.org/26648813:44
gordcsileht: *shrugs* i think it's ok. ( because i +2 the change in client.lol)13:45
silehtgordc, lol13:46
*** datravis has joined #openstack-telemetry13:46
silehtgordc, I'm sure I'm the guys that have done the mis-spelling13:46
*** datravis has left #openstack-telemetry13:46
gordcno comment :P13:46
*** yassine__ has quit IRC13:49
openstackgerritIgor Degtiarov proposed openstack/ceilometer: bracketer event transformer  https://review.openstack.org/26648813:49
*** ildikov has joined #openstack-telemetry13:50
ityaptinsileht: Maybe you'll create an additional exception with mis-spelling in the gnocchiclient ?13:56
ityaptinSo, it will support backward compatibility, but allows to use a word with right spelling?13:57
silehtityaptin, yes13:57
*** ddieterly has joined #openstack-telemetry14:00
*** jasonamyers has left #openstack-telemetry14:01
ityaptinsileht: :D I thought that you want to rename current exception to mis-spelled. I was in slowpoke mode)14:01
*** efoley_ has quit IRC14:01
*** efoley_ has joined #openstack-telemetry14:07
*** links has joined #openstack-telemetry14:07
*** yassine__ has joined #openstack-telemetry14:24
*** datravis has joined #openstack-telemetry14:33
*** pradk has joined #openstack-telemetry14:42
*** jwcroppe has joined #openstack-telemetry14:44
liamjigordc: Hi14:52
gordcliamji: hi14:53
gordcwhatsup?14:53
*** rbak has joined #openstack-telemetry14:54
liamjigordc: some UT cases seem to fail. And it leads to the jenkins fails.14:54
liamjihttp://logs.openstack.org/34/277434/13/check/gate-ceilometer-python27/2c16ebf/console.html14:54
gordcliamji: it's related to latest gnocchiclient14:55
gordcit's sileht's fault. :)14:56
silehtalways14:56
liamjigordc: okey. The fix code is ready now?14:56
gordcliamji: https://review.openstack.org/#/c/284704/214:57
gordcthat should fix it14:57
gordcityaptin: we need to drop the requirement for now i think. it won't pass requiremetns checks14:57
liamjigordc: thanks. I will recheck my change set after the fix is merged. :)14:58
gordcliamji: your patch is fine for Mitaka (the polling part at least)14:58
*** diogogmt has joined #openstack-telemetry14:59
*** rbak_ has joined #openstack-telemetry15:00
liamjigordc: okey, thanks :)15:00
gordcshort meetings over at openstack-meeting15:00
ityaptingordc: Did requirements job failed?15:01
gordcityaptin: http://logs.openstack.org/04/284704/2/check/gate-ceilometer-requirements/601ad03/console.html#_2016-02-25_13_57_30_33815:02
*** llu-laptop has joined #openstack-telemetry15:02
*** rbak has quit IRC15:03
*** llu-laptop has quit IRC15:03
ityaptinack. So, I will update my patch. Should I set an update to the global-requirements or somebody else will do it?15:04
*** PsionTheory has joined #openstack-telemetry15:04
*** llu-laptop has joined #openstack-telemetry15:04
gordcityaptin: you can do it if you'd like :)15:05
*** llu-laptop has quit IRC15:05
*** rickyrem has joined #openstack-telemetry15:08
ityaptin:) ok15:08
openstackgerritIlya Tyaptin proposed openstack/ceilometer: Fix the typo in the gnocchiclient exception  https://review.openstack.org/28470415:09
*** llu-laptop has joined #openstack-telemetry15:10
*** links has quit IRC15:12
*** llu-laptop has quit IRC15:13
*** ddieterly has quit IRC15:13
*** liusheng_ has joined #openstack-telemetry15:14
*** davidsha_ has joined #openstack-telemetry15:14
*** KrishR has joined #openstack-telemetry15:17
*** yprokule has quit IRC15:19
*** efoley_ has quit IRC15:24
*** PsionTheory has quit IRC15:26
*** ddieterly has joined #openstack-telemetry15:47
openstackgerritMerged openstack/aodh: Add composite rule alarm API support  https://review.openstack.org/25772215:48
*** ddieterly has quit IRC15:48
*** ddieterl_ has joined #openstack-telemetry15:48
*** liusheng_ has quit IRC15:58
gordcsileht: m-3 is next week16:00
silehtgordc, m-3 == feature freeze ?16:00
gordcyep16:00
silehtgordc, which day ?16:00
gordcwe can keep adding features but we should have a good reason16:00
gordcsileht: any day next week16:01
gordcbut again, assume gate is broken next week and nothing will merge16:01
gordcwe start tagging monday-thursady16:01
gordceveryone usually does wednesday/thursday16:01
silehtgordc, I don't known how will be the code change, so lets see how my works in going-on and what the gate health to take a decision16:02
silehtgordc, on Monday I should have a idea on how the patch will be big16:02
gordcsileht: ack. we can always make batching FFE if we think it's important16:02
gordcbut yeah, let me know16:02
silehtI don't like that16:02
gordcdon't like FFE?16:03
silehtyes16:03
silehtoriginally I have thougth that bachting will be for next release, not mikita16:03
silehtbut we have release Gnocchi 2.0, so we can write the code now :)16:03
gordcsileht: kk. let's assume defer unless you have it ready on monday16:03
silehtthat's sound reasonable16:04
gordcidegtiarov_: we want to keep discussing online? or maybe mailing list?16:04
gordcsileht: cool cool. thanks16:04
*** davidsha_ has quit IRC16:05
*** diogogmt has quit IRC16:08
*** datravis has left #openstack-telemetry16:11
nicodemus_hello gordc16:13
openstackgerritXia Linjuan proposed openstack/ceilometer: Enable the Load Balancer v2 for notification  https://review.openstack.org/28177716:14
gordcnicodemus_: hey, whatsup?16:20
nicodemus_gordc: just wanted to let you know that your patch (https://review.openstack.org/#/c/284528/) worked, I just used it to upgrade our large deployment16:21
gordcnicodemus_: oh nice!16:21
gordcnicodemus_: i wasn't exactly sure, i needed to dig in to code to figure out what we were writing/storing in 1.3.x16:22
gordci'm not entirely sure that's the correct approach still but i think it does handle the migration part16:23
gordcsileht: i don't know if you have any ideas. re: gnocchi storage migration.16:23
silehtgordc, I have though that gnocchi-upgrade already do that16:25
silehtgordc, it does16:25
gordcit doesn't. it was a trick.lol16:25
silehtnicodemus_, did you run gnocchi-upgrade before start metricd after the upgrade16:26
gordchttps://review.openstack.org/#/c/284528/16:26
silehtgordc, https://github.com/openstack/gnocchi/blob/master/gnocchi/storage/_carbonara.py#L31016:27
nicodemus_sileht: yes, I had to run gnocchi-upgrade. So far the API is working without having to hack what I described here https://bugs.launchpad.net/gnocchi/+bug/154836716:27
openstackLaunchpad bug 1548367 in Gnocchi "Gnocchi-upgrade fails when moving to latest master" [High,Triaged]16:27
gordcsileht: right. but if you look at TimeSerieArchive, it uses AggregatedTimeSerie.16:27
*** belmoreira has quit IRC16:27
silehtgordc, we do not support online migration16:27
nicodemus_once i finish upgrading the remaining APIs, I'll upgrade metricd (it's stopped now)16:28
silehthttp://gnocchi.xyz/install.html#upgrading16:28
gordcso basically the test we have is writing a TimeSerieArchive of v2.0 AggregatedTimeSerie and then migrating it to V2.0 AggregatedTimeSerie16:28
silehttouching carbonara.py looks not a good idea16:29
gordcbut in reality, it's a TimeSerieArchive of v1.3 AggregatedTimeSerie16:29
silehtlol very good test !16:29
gordcyeah, and v1.3 -> v2.0 AggregatedTimeSerie has different fields/data16:29
silehtgordc, _check_for_metric_upgrade must be fixed in this case16:30
gordcyeah, i can basically move my hack to replace https://github.com/openstack/gnocchi/blob/master/gnocchi/storage/_carbonara.py#L31016:30
gordcsileht: that should do the same thing.16:30
gordcsileht: i think big concern right now is there is no way to tell what is a v1.3 datapoint vs v2.0.16:32
gordcright now i'm just basing it on fact we're missing 'first_timestamp' attribute16:32
silehtgordc, before all aggregates was in object backend object, now each have an object16:34
silehtin/in one16:34
silehtaggregates/granularity16:35
gordcall in one timeseriearchive you mean16:35
gordcthe aggregatedtimeserie format also changed from v1.3 and that's why i have the check for 'first_timestamp'16:37
gordci probably don't need it but i think we ever change format again, we might need a way to figure out what version it is16:37
silehtgordc,agreed that will help a lot16:38
silehtgordc, at least to bail out when something unexpected is read from the backend16:39
gordcsileht: i'll try moving my hack into the upgrade method for now. i'm not sure it's completely right...16:39
gordcbut it's more right than now.lol16:39
silehtgordc, I'm starting to see the issue16:40
silehtwe expect 'carbonara.TimeSerieArchive.unserialize(data)' to always return the new format16:40
gordcyeah. i had to hack the test to write old format16:42
gordcand then unserialise will explode16:42
gordcmy fix is good for v1.3->v2.0 but if there's another migration, i'm not sure how that will work. unless we tag and keep changing _check_for_metric_upgrade method every time we change format16:44
silehtgordc, I think you just have to move your code in AggregatedTimeSerie.from_dict16:46
silehtgordc, in fact we have two things to upgrade the object format (granularity split for panda measures) and the AggregatedTimeSerie description16:47
silehtgordc, storage/_carbonara.py must handle the first one,16:47
silehtgordc, I think it's ok to handle the seconds one in carbonara.py16:48
nicodemus_one question... after upgrading metricd to 2.0.1.dev1, when I start it it displays all config options and then freezes there, with no error message nor anything and it's not processing). Where should I start looking?16:48
nicodemus_I don't see it attempting to connect to the database (sniffing packets on the net interface)16:49
gordcnicodemus_: um... that's with debugging turned on?16:51
nicodemus_gordc: yes16:51
gordcnicodemus_: from this log, http://logs.openstack.org/54/282454/1/gate/gate-telemetry-dsvm-integration-gnocchi/88688ae/logs/screen-gnocchi-metricd.txt.gz16:53
gordcwhat line does your metricd stop at?16:53
gordcsileht: i'm not entirely sure how to split first and second parts16:54
nicodemus_gordc: this is the entire log output: http://paste.openstack.org/show/488226/16:55
gordchmm... i wonder if that's some driver loading issues16:56
silehtgordc, the old format was having block_size and back_window fields, are they useless to build the new format ?16:57
*** liamji has quit IRC16:58
nicodemus_gordc: on this host I was using 1.3.1.dev280, I pulled master, uninstalled prev version with pip, installed the requirements, updated api-paste.ini and then installed gnocchi16:58
gordcsileht: did we actually write block_size/back_window? i built the v1.3 format based on https://github.com/openstack/gnocchi/blob/stable/1.3/gnocchi/carbonara.py#L41116:59
gordcwhich has https://github.com/openstack/gnocchi/blob/stable/1.3/gnocchi/carbonara.py#L291-L29616:59
silehtgordc, https://github.com/openstack/gnocchi/blob/stable/1.3/gnocchi/carbonara.py#L29116:59
silehthttps://github.com/openstack/gnocchi/blob/stable/1.3/gnocchi/carbonara.py#L21717:00
silehthttps://github.com/openstack/gnocchi/blob/stable/1.3/gnocchi/carbonara.py#L12917:00
silehtgordc, we was using inheritance for that, so each sub class was adding thing into the dict17:00
gordcAggregatedTimeSerie doesn't inherit BoundTimeSerie17:00
gordc(i don't know the diff between both)17:01
silehtoh right sorry17:01
gordcno worries. i'm sort of guessing right now which is why i'm not sure what right fix is.17:02
silehtgordc, so i would said, just move your fix into AggregatedTimeSerie.from_dict17:02
gordcsileht: shoud i keep it in _check_for_metric_upgrade? from_dict i believe is called even when not migrating17:03
*** thorst is now known as thorst_afk17:04
silehtgordc, I think this is not a problem17:04
*** liusheng has quit IRC17:05
*** liusheng has joined #openstack-telemetry17:05
gordcsileht: ack. i'll try that.17:06
*** cdent has quit IRC17:09
*** _nadya_ has quit IRC17:10
*** thorst_afk is now known as thorst17:12
openstackgerritgordon chung proposed openstack/gnocchi: fix migration from v1.3 to v2.0  https://review.openstack.org/28452817:14
gordcnicodemus_: what did you upgrade in api-paste.ini (although it shouldn't matter)17:15
gordcnicodemus_: maybe run ps -ef | grep metricd and kill all workers and try again?17:16
nicodemus_gordc: here are the two files: http://paste.openstack.org/show/488230/17:16
nicodemus_gordc: I tried... changed backend to file just to see what happened, same behavior17:17
silehtgordc, oh I just got why you put it in TimeSerieArchive...17:18
silehtgordc, I think your first version was OK, and you don't have to check 'first_timestamp' it wasn't exists before17:19
silehtsorry for my slow mind17:20
* sileht should stop working today17:20
gordcsileht: :) no worries.17:22
gordci guess they bothe work whether it's in timeseriearchive vs aggregatedtimeserie17:23
gordcit's actually less code in aggregatedtimeserie. https://review.openstack.org/28452817:23
*** KrishR has quit IRC17:27
*** julim has quit IRC17:28
nicodemus_gordc: would gnocchi-metricd 1.3.5.dev6 work if the API and the data were already migrated to 2.0.1?17:28
*** yassine__ has quit IRC17:30
*** yassine__ has joined #openstack-telemetry17:31
*** cdent has joined #openstack-telemetry17:35
*** rbak__ has joined #openstack-telemetry17:35
*** cdent has quit IRC17:36
*** cdent has joined #openstack-telemetry17:36
*** rbak_ has quit IRC17:38
*** jwcroppe has quit IRC17:38
*** jwcroppe has joined #openstack-telemetry17:39
nicodemus_gordc: I got metricd to work, the indexer url was mysql:// instead of mysql+pymysql://. But sadly, metricd is stacktracing for each new measure: http://paste.openstack.org/show/488235/17:42
*** jwcroppe has quit IRC17:43
gordcnicodemus_: and you migrated db?17:44
gordcnicodemus_: i need to drop for a bit, going to eat and head to office. will be back later.17:44
*** gordc has quit IRC17:44
nicodemus_gordc: yes, with the db migrated. Ok, talk to you later!17:45
*** ljxiash has quit IRC17:49
*** jwcroppe has joined #openstack-telemetry17:58
*** rbak__ has quit IRC18:05
*** lsmola has quit IRC18:08
*** julim has joined #openstack-telemetry18:10
*** rickyrem has quit IRC18:13
*** rbak has joined #openstack-telemetry18:15
*** Ephur has joined #openstack-telemetry18:16
*** annasort has quit IRC18:19
*** Ephur has quit IRC18:20
*** rbak has quit IRC18:21
*** rbak has joined #openstack-telemetry18:21
*** AndreiaKumpera has quit IRC18:25
*** KrishR has joined #openstack-telemetry18:43
*** gordc has joined #openstack-telemetry18:47
openstackgerritSteve Lewis proposed openstack/ceilometer: Docs: Configure meters/events dispatch separately  https://review.openstack.org/28328218:51
nicodemus_is it possible that gnocchi-upgrade no longer works with any version prior to 2.0?18:55
gordcnicodemus_: which version?18:56
nicodemus_just cloned 1.3, and gnocchi-upgrade tries to import "upgrade" from gnocchi.cli18:57
gordcnicodemus_: i'm not sure there was any data format changes within 1.x18:57
nicodemus_but upgrade is on cli.py on 2.0, not 1.318:57
nicodemus_I'm trying to have gnocchi back up, so until metricd is able to resume processing metricd I wanted to re-deploy in 1.318:58
gordcgnocchi upgrade doesn't exist in 1.318:58
nicodemus_and use gnocchi-upgrade command to create a new DB18:58
gordcin 1.3 the command is gnocchi-dbsync18:58
gordci believe prior to 2.0, we only migrated indexer(sql) data.18:59
gordcstorage(ceph|swift|file) data didn't change.18:59
nicodemus_if I installed from git, where should I have  gnocchi-dbsync? Because it's nowhere on the filesystem18:59
gordcit should enable gnocchi-dbsync if you install19:00
gordcsame way it enabled gnocchi-upgrade when you installed 2.x19:01
nicodemus_is gnocchi-dbsync it available for download at any location?19:01
*** yassine__ has quit IRC19:01
nicodemus_I only have gnocchi-upgrade :(19:01
nicodemus_with gnocchi (1.3.5.dev6)19:02
gordcdid you pip uninstall gnocchi?19:04
nicodemus_yes19:04
gordcyou shouldn't have gnocchi-upgrade if you actually removed it19:04
nicodemus_I just uninstalled and I still have it, let me delete it by hand19:06
gordci ususally just keep running pip uninstall until it says doesn't exist. i find sometimes i have multiple copies installed19:07
nicodemus_you were right... had to run pip uninstall again O.o19:09
gordcbrute force always works best19:10
*** rbak_ has joined #openstack-telemetry19:20
*** rbak has quit IRC19:20
*** rickyrem has joined #openstack-telemetry19:22
nicodemus_gordc: I promise I won't keep bugging you much more (at least for today :) with the 2.0 deploy w/ migrated db, metricd is still showing the overflowerror19:25
gordcnicodemus_: can you patch the system again. http://paste.openstack.org/show/488255/19:33
*** _nadya_ has joined #openstack-telemetry19:35
openstackgerritMerged openstack/ceilometer: Fix the typo in the gnocchiclient exception  https://review.openstack.org/28470419:35
*** ddieterl_ has quit IRC19:37
*** ddieterly has joined #openstack-telemetry19:39
*** rickyrem has quit IRC19:40
nicodemus_gordc: here's the log extract: http://paste.openstack.org/show/488258/19:42
*** rickyrem has joined #openstack-telemetry19:43
gordccan you take a larger chunk before and after19:43
gordcjust so i can figure out which value is actually throwing error19:43
nicodemus_sure, I'm filtering the previous 10 lines and the 50 lines after like so: tail -f gnocchi-metricd.log | grep -A 50 -B 10 "Error processing new measures"19:44
nicodemus_would 100 before and 100 after be enough?19:44
gordcsure19:44
*** ddieterl_ has joined #openstack-telemetry19:44
nicodemus_gordc: here's the new log: http://paste.openstack.org/show/488260/19:47
*** rickyrem has quit IRC19:47
gordchmm.. i wonder how the hell it's getting DatetimeIndex instead of a timestamp19:49
*** ddieterl_ is now known as ddieterly_19:49
gordcwhat happens if you re-run gnocchi-upgrade19:49
nicodemus_gordc: one thing I didn't mention, I made a change on the archive policy19:50
*** krotscheck is now known as krotscheck_dcm19:50
gordchmm.. i don't think that should be an issue assuming it was done through api19:51
nicodemus_gordc: I changed the first definition on the default policy, from "points: 1440, granularity: 0:01:00, timespan: 1 day" to "points: 288, granularity: 0:05:00, timespan: 1 day"19:51
nicodemus_gordc: but didn't empty the pool or anything19:52
nicodemus_gordc: let me try re-running gnocchi-upgrae19:52
openstackgerritZhiQiang Fan proposed openstack/ceilometer: Add validation for polling_namespaces option  https://review.openstack.org/28471019:53
gordcnicodemus_: i have a meeting with manager. i'll biab. if you can, can you include the above comments in the bug report19:56
nicodemus_sure, thanks a lot!19:56
*** ljxiash has joined #openstack-telemetry20:01
*** ddieterl_ has joined #openstack-telemetry20:03
*** ddieterly has quit IRC20:05
*** ddieterly_ has quit IRC20:05
*** ljxiash has quit IRC20:05
*** annasort has joined #openstack-telemetry20:15
*** thorst is now known as thorst_afk20:38
*** jwcroppe has quit IRC20:40
*** jwcroppe has joined #openstack-telemetry20:40
*** jwcroppe has quit IRC20:45
*** Ephur has joined #openstack-telemetry20:47
*** _nadya_ has quit IRC20:52
*** thorst_afk is now known as thorst20:53
*** ddieterl_ has quit IRC21:02
*** vishwanathj has quit IRC21:04
*** ddieterly has joined #openstack-telemetry21:04
openstackgerritMerged openstack/python-ceilometerclient: Trivial: remove `None` as a redundant argument to dict.get()  https://review.openstack.org/28269921:07
*** ddieterly has quit IRC21:09
*** ildikov has quit IRC21:16
*** annasort has quit IRC21:25
*** annasort has joined #openstack-telemetry21:28
*** thorst is now known as thorst_afk21:39
*** thorst_afk has quit IRC21:43
*** thorst_afk has joined #openstack-telemetry21:43
*** annasort has quit IRC21:44
*** annasort has joined #openstack-telemetry21:50
*** thorst_afk has quit IRC21:52
gordcnicodemus_: i'm not sure i'll get around to looking at your issue today. did gnocchi-upgrade do anything?21:53
*** ddieterly has joined #openstack-telemetry21:53
nicodemus_gordc: not much, at least nothing different :(21:53
gordcnicodemus_: kk. will try to look tomorrow21:54
nicodemus_gordc: np, thanks! :)21:54
*** nicodemus_ has quit IRC21:56
*** julim has quit IRC22:04
*** pradk has quit IRC22:09
*** pradk has joined #openstack-telemetry22:09
*** annasort has quit IRC22:43
*** thorst_afk has joined #openstack-telemetry22:49
*** thorst_afk has quit IRC22:57
*** cdent has quit IRC23:09
gordccool everything is broken. time to go home23:29
*** gordc has quit IRC23:29
*** pradk has quit IRC23:39
*** ddieterly has quit IRC23:47
*** thorst_afk has joined #openstack-telemetry23:55
« Wednesday, 2016-02-24 Index Friday, 2016-02-26 »

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