Tuesday, 2016-02-09

*** thorst has joined #openstack-telemetry00:29
*** thorst_ has joined #openstack-telemetry00:31
*** thorst has quit IRC00:34
*** thorst_ has quit IRC00:34
*** thorst has joined #openstack-telemetry00:35
*** thorst has quit IRC00:39
*** KrishR has quit IRC00:52
*** thorst has joined #openstack-telemetry01:15
*** rooney has joined #openstack-telemetry01:24
openstackgerritDavanum Srinivas (dims) proposed openstack/ceilometer: [WIP] Trying py27/34 with oslo-master  https://review.openstack.org/27765501:27
*** rooney has quit IRC01:29
*** prashantD has quit IRC02:05
openstackgerritRohit Jaiswal proposed openstack/ceilometer-specs: Propose the removal of WSME from ceilometer and aodh  https://review.openstack.org/21815502:13
*** prashantD has joined #openstack-telemetry02:14
*** achatterjee has quit IRC02:48
*** achatterjee has joined #openstack-telemetry02:50
*** prashantD has quit IRC03:04
*** thorst has quit IRC03:18
*** thorst has joined #openstack-telemetry03:18
*** thorst has quit IRC03:27
*** sanjana has joined #openstack-telemetry03:28
*** rooney has joined #openstack-telemetry03:36
*** rooney has quit IRC03:40
*** links has joined #openstack-telemetry03:41
*** dikonoor has joined #openstack-telemetry03:51
*** rjaiswal has quit IRC04:08
*** thorst has joined #openstack-telemetry04:24
*** thorst has quit IRC04:32
*** dikonoor has quit IRC04:37
*** yprokule has joined #openstack-telemetry05:01
*** thorst has joined #openstack-telemetry05:29
*** rooney has joined #openstack-telemetry05:34
*** thorst has quit IRC05:37
*** nadya has joined #openstack-telemetry05:56
*** nadya is now known as Guest3150705:56
*** ljxiash has joined #openstack-telemetry06:01
*** rooney has quit IRC06:03
*** dikonoor has joined #openstack-telemetry06:09
*** Guest31507 has quit IRC06:26
*** thorst has joined #openstack-telemetry06:35
*** thorst has quit IRC06:42
*** ljxiash has quit IRC06:56
*** ljxiash has joined #openstack-telemetry07:12
openstackgerritMehdi Abaakouk (sileht) proposed openstack/aodh: Allow to configure the evaluator loopback window  https://review.openstack.org/27732607:33
*** thorst has joined #openstack-telemetry07:40
*** belmoreira has joined #openstack-telemetry07:41
*** thorst has quit IRC07:46
openstackgerritMehdi Abaakouk (sileht) proposed openstack/python-gnocchiclient: Add batching to client  https://review.openstack.org/27342908:02
openstackgerritMehdi Abaakouk (sileht) proposed openstack/python-gnocchiclient: Ensure description exists for HTTP error.  https://review.openstack.org/27342808:02
openstackgerritMehdi Abaakouk (sileht) proposed openstack/gnocchi: Extend measures batching to named metrics  https://review.openstack.org/27336808:07
*** guillaume_ has quit IRC08:23
*** shardy has joined #openstack-telemetry08:30
*** thorst has joined #openstack-telemetry08:45
openstackgerritMehdi Abaakouk (sileht) proposed openstack/gnocchi: Fixes routing of aggregation endpoint  https://review.openstack.org/27736408:49
*** thorst has quit IRC08:53
*** nadya has joined #openstack-telemetry08:53
*** nadya is now known as Guest2256908:54
*** safchain has joined #openstack-telemetry09:00
*** dikonoor has quit IRC09:17
*** dikonoor has joined #openstack-telemetry09:18
*** dikonoo has joined #openstack-telemetry09:19
*** dikonoor has quit IRC09:23
*** yassine has joined #openstack-telemetry09:26
*** jwcroppe has joined #openstack-telemetry09:33
*** jwcroppe has quit IRC09:36
*** jwcroppe has joined #openstack-telemetry09:37
*** efoley has joined #openstack-telemetry09:37
*** jwcroppe has quit IRC09:41
*** thorst has joined #openstack-telemetry09:50
*** thorst has quit IRC09:57
openstackgerritMerged openstack/python-ceilometerclient: Enhances client to support unique meter retrieval  https://review.openstack.org/27263310:05
*** r-mibu has quit IRC10:20
*** r-mibu has joined #openstack-telemetry10:21
openstackgerritMerged openstack/ceilometer: sample: remove deprecated option name.  https://review.openstack.org/27610210:37
*** cdent has joined #openstack-telemetry10:41
*** efoley has quit IRC10:52
*** thorst has joined #openstack-telemetry10:54
*** thorst has quit IRC11:02
*** efoley has joined #openstack-telemetry11:02
*** efoley_ has joined #openstack-telemetry11:09
*** efoley has quit IRC11:11
openstackgerritMerged openstack/gnocchi: Don't require trollius on Python 3.4 and newer  https://review.openstack.org/27674211:17
openstackgerritMerged openstack/ceilometer: network: remove deprecated option name  https://review.openstack.org/27610411:25
*** shardy has quit IRC11:34
*** efoley__ has joined #openstack-telemetry11:34
*** shardy has joined #openstack-telemetry11:36
*** efoley_ has quit IRC11:36
*** boris-42 has quit IRC11:43
*** thorst has joined #openstack-telemetry12:00
*** sree has joined #openstack-telemetry12:06
sreehiii, I want proper documentation of how to use ceilometer.12:06
sreeplease someone help12:07
*** thorst has quit IRC12:07
*** efoley__ has quit IRC12:26
*** gordc has joined #openstack-telemetry12:36
*** liamji has joined #openstack-telemetry12:37
*** openstackgerrit_ has joined #openstack-telemetry12:39
*** thorst has joined #openstack-telemetry12:46
*** efoley__ has joined #openstack-telemetry12:46
*** thorst_ has joined #openstack-telemetry12:47
*** thorst has quit IRC12:51
*** cdent has quit IRC13:04
*** cdent has joined #openstack-telemetry13:05
*** ildikov has quit IRC13:06
*** links has quit IRC13:17
*** liamji has quit IRC13:31
openstackgerritMerged openstack/python-gnocchiclient: Ensure description exists for HTTP error.  https://review.openstack.org/27342813:42
*** liamji has joined #openstack-telemetry13:42
openstackgerritMerged openstack/gnocchi: Fixes routing of aggregation endpoint  https://review.openstack.org/27736413:54
*** openstackgerrit_ has quit IRC13:56
*** ljxiash has quit IRC13:57
*** ljxiash has joined #openstack-telemetry13:58
*** liamji has quit IRC13:59
*** ildikov has joined #openstack-telemetry14:02
*** liamji has joined #openstack-telemetry14:03
*** bdossant has joined #openstack-telemetry14:05
openstackgerritZi Lian Ji proposed openstack/ceilometer: Enable the Load Balancer v2 for the Ceilometer(Part One)  https://review.openstack.org/27451814:09
*** cdent has quit IRC14:13
*** links has joined #openstack-telemetry14:16
liamjiIRC上找他一下吧14:16
liamjisorry, ignore the last message I sent14:17
jd__\o/14:18
jd__encoder achieved14:18
*** peristeri has joined #openstack-telemetry14:21
*** efoley__ is now known as efoley14:24
*** yprokule has quit IRC14:33
*** dave-mccowan has joined #openstack-telemetry14:34
*** dikonoo has quit IRC14:35
*** yprokule has joined #openstack-telemetry14:38
*** achatterjee has quit IRC14:38
*** tongli has joined #openstack-telemetry14:40
openstackgerritgordon chung proposed openstack/gnocchi: grab less data when adding measures  https://review.openstack.org/27760414:47
*** dikonoor has joined #openstack-telemetry14:49
*** julim_ has quit IRC14:49
*** cdent has joined #openstack-telemetry14:50
*** pradk has joined #openstack-telemetry14:52
*** sree has quit IRC14:53
*** links has quit IRC14:54
*** peristeri has quit IRC14:55
*** rbak has joined #openstack-telemetry14:56
*** KrishR has joined #openstack-telemetry14:56
openstackgerritgordon chung proposed openstack/gnocchi: minimise swift processing metric size  https://review.openstack.org/27787815:17
*** ildikov has quit IRC15:19
*** ildikov has joined #openstack-telemetry15:19
alextricity25Howdy! I have a question about the 'messaging_urls' directive in ceilometer.conf15:26
alextricity25Is there anyway I can specify urls to multiple rabbitmq vhosts without using more than one 'messaging_urls'15:27
*** shardy has quit IRC15:28
alextricity25In other words, what code is responsible for parsing the value for 'messaging_urls'.15:29
alextricity25I'm very interested in seeing what gives it the behavior specified in the docs15:29
gordcalextricity25: it uses oslo.config15:31
gordcalextricity25: i assume you can list it in one option and use comma as delimiter.15:31
gordcyou'll still need to put full host/vhost details though15:32
*** farid has joined #openstack-telemetry15:34
*** yprokule has quit IRC15:34
*** Guest22569 has quit IRC15:35
*** mragupat has joined #openstack-telemetry15:38
alextricity25gordc: Ah okay. In theory, I should be able to do this , right? "messaging_urls = ["rabbit://glance:glacepassword@172.29.238.115:5671/glance", "rabbit://nova:novapassword@172.29.238.115:5671/nova", etc..]15:38
gordcalextricity25: not entirely sure. you probably don't need the square brackets and quotations... never tried myself.15:40
gordci assume some variation.15:40
*** yprokule has joined #openstack-telemetry15:40
alextricity25gordc: I think the problem is with the way it expects urls to multiple hosts. According to the docs:15:40
alextricity25"(MultiStrOpt) Messaging URLs to listen for notifications. Example: transport://user:pass@host1:port[,hostN:portN]/virtual_host (DEFAULT/transport_url is used if empty)"15:40
alextricity25When I try to list out full rabbitmq connection strings delimited with a comma, it will try to parse it like described above.15:41
gordcoh. err.. yeah, i guess that help message describes it best.15:42
*** liamji has quit IRC15:44
*** ljxiash has quit IRC15:45
*** julim has joined #openstack-telemetry15:45
jd__gordc: https://gist.github.com/jd/d3c23a261bd153d29299#file-bench-py-L11215:46
gordcjd__: nice!15:47
jd__gordc: I've find why LZ4 was slow, it was Pandas doing shit15:47
gordchow'd you get the compression time down?15:47
gordcoh15:47
jd__so LZ4 is constant 0.04s encoding15:47
jd__my XOR encoding is a bit slower15:47
jd__but can get better results sometime15:48
jd__so no… I wonder what to pick15:48
jd__s/no/now/15:48
gordcwhat's "Simple contiguous range"?15:49
jd__gordc: 1 2 3 4 515:50
jd__15:50
jd__https://gist.github.com/jd/d3c23a261bd153d29299#file-bench-py-L1615:50
jd__continuous I guess15:50
jd__:D15:50
gordcah right right. same i guess.15:50
*** shardy has joined #openstack-telemetry15:51
gordcoh... the codes above.lol i probably should've scrolled up15:51
jd__:-)15:51
jd__I'd tend to pick lz4, as it's pretty constant and faster15:52
jd__it's a bit bigger sometimes, but it looks like the best compromise15:53
jd__unless I can speed up my xor implementation?15:53
gordcjd__: i was going to say same thing. reason: faster15:54
gordclooks like xor does the best in first scenario with const increasing value.15:54
*** ljxiash has joined #openstack-telemetry15:55
jd__Python takes 20 % of the time counting 0 in a binary and 10% of its time converting a double to an int15:55
jd__lol.15:55
jd__it's horrible to think my xor is likely faster than LZ4, but since LZ4 is in C, it's faster15:56
jd__damn Python15:56
gordcjd__: that's probably not good considering everything we have is in python...15:57
jd__:)15:57
gordctime to rewrite everything15:57
jd__that's called trade-offs15:58
jd__but well that's easy to improve, my code was pretty naive anyway15:58
*** yprokule has quit IRC15:59
gordcjd__: you going to write a blog on math?16:00
jd__I'll write a blog post on that at least16:00
*** ljxiash has quit IRC16:00
jd__that should be interesting16:00
*** boris-42 has joined #openstack-telemetry16:11
*** belmoreira has quit IRC16:21
jd__omfg16:22
jd__so I read that, but it's actually true16:22
jd__it's faster to count the number of leading and trailing zeroes in Python from a string of "000100101000" than doing it in pure-binary16:22
jd__facepalm16:22
*** guillaume_ has joined #openstack-telemetry16:22
jd__https://gist.github.com/jd/303809d1f6c6425cf5e916:32
jd__version 1 is way faster…16:32
jd__answer from haypo "if you want more speed, write C code"16:32
jd__re-facepalm16:32
*** prashantD has joined #openstack-telemetry16:45
*** dikonoor has quit IRC16:48
*** vishwanathj has joined #openstack-telemetry16:51
*** nadya has joined #openstack-telemetry16:59
*** nadya is now known as Guest7158717:00
*** mragupat has quit IRC17:00
*** bdossant has quit IRC17:00
*** mragupat has joined #openstack-telemetry17:01
*** Guest71587 has quit IRC17:02
openstackgerritJulien Danjou proposed openstack/gnocchi: carbonara: remove carbonara-dump  https://review.openstack.org/27682417:16
openstackgerritJulien Danjou proposed openstack/gnocchi: carbonara: serialize AggregatedTimeSerie using RLE, XOR encoding and LZ4  https://review.openstack.org/27636517:16
openstackgerritJulien Danjou proposed openstack/gnocchi: carbonara: compute sampling at init time  https://review.openstack.org/27797617:16
openstackgerritgordon chung proposed openstack/gnocchi: grab less data when adding measures  https://review.openstack.org/27760417:19
pradkjd__, so the master tag in gnocchi still seems to have 1.3.1.dev* vs 1.3.4 or newer17:21
jd__pradk: yeah, and it's never going to be fixed actually17:21
pradkjd__, we're running into issue with packstack pulling liberty vs mitaka17:21
jd__we can't merge any tag in master17:22
pradkjd__, hmm can we just bump master to a pre release tag? say 2.0.0a0 or something like nova?17:22
pradkcurrently master is 1.3.1 and liberty is 1.3.4 which messes up builds17:23
jd__yeah I know, I'm just not sure how to fix it17:23
jd__tagging a random release is not the best option I think17:23
-openstackstatus- NOTICE: Gerrit is restarting now, to alleviate current performance impact and WebUI errors.17:25
* jd__ thinks17:25
pradkyea understand, in ceilo for example we have 6.0.0.0b3.dev45 .. which always makes master newer, gnocchi needs something like that to advance master to be newer17:28
jd__pradk: that's not true 100% of the time actually17:31
jd__there's a short period of time between 5.0.0 and 6.0.0a1 I think where it can be messed up17:32
jd__which is exactly the problem you're hitting17:32
*** mragupat has quit IRC17:32
*** mragupat has joined #openstack-telemetry17:33
pradkjd__, so when we release gnocchi 2.0 will the master bump to 2.0? or remain at 1.3.117:33
jd__pradk: 2.0.017:33
jd__master is actually 1.3.0 + 264 commits17:33
jd__but pbr says 1.3.1 + 262 commits17:33
jd__(for whatever reason it computes it this way)17:34
*** safchain has quit IRC17:34
pradkso how did we end up with 1.3.1 without the tag getting merged? i'm just wondering if we could do the same that led to 1.3.117:34
jd__it's just pbr computing the version that says 1.3.117:35
pradkhm ok.. so we'll already end up in this situation, even if we bump to 2.0 .. when 2.0.1 is out master tag will be older17:36
pradks/already/always17:36
* jd__ nods17:36
*** mragupat has quit IRC17:36
jd__master is master, there's no "version" until we release17:37
jd__how packstack is failing ?17:37
*** mragupat has joined #openstack-telemetry17:37
pradkyea understand.. the way delorean does it today is by getting the version from python setup.py --version .. which gives us 1.3.1 in trunk .. vs liberty has 1.3.4 so when nvr compare happens, 1.3.4 wins17:38
jd__which… makes sense somehow17:40
jd__you should have a way to force some kind of "dev mode" where it pulls master, whatever the version is17:40
pradktrunk is nightly builds so gets the latest and greatest from master .. but gets messed up with vr17:41
pradkyea i'll experiment and see.. the issue is since gnocchi is in delorean-deps its winning the comapre.. may be we just move it out,17:43
pradkanyhu, when are we shooting for 2.0?17:43
*** KrishR has quit IRC17:46
pradkbrb lunch17:47
*** safchain has joined #openstack-telemetry17:48
*** ntt has quit IRC17:50
*** ildikov has quit IRC17:53
*** shardy has quit IRC18:00
*** vishwanathj has quit IRC18:00
*** cdent has quit IRC18:02
*** nadya has joined #openstack-telemetry18:07
*** nadya is now known as Guest986018:08
*** efoley has quit IRC18:09
*** peristeri has joined #openstack-telemetry18:12
*** safchain has quit IRC18:16
*** llu has quit IRC18:30
*** mragupat_ has joined #openstack-telemetry18:32
*** llu has joined #openstack-telemetry18:32
openstackgerritgordon chung proposed openstack/python-aodhclient: WIP: Add gnocchi alarm rules  https://review.openstack.org/26607318:35
*** mragupat has quit IRC18:36
*** ildikov has joined #openstack-telemetry18:36
*** guillaume_ has quit IRC18:36
*** prashantD has quit IRC18:42
*** cdent has joined #openstack-telemetry18:43
*** prashantD has joined #openstack-telemetry18:44
*** peristeri has quit IRC18:49
*** prashantD has quit IRC18:50
*** prashantD has joined #openstack-telemetry18:51
*** thorst_ is now known as thorst18:57
*** KrishR has joined #openstack-telemetry19:00
*** rooney has joined #openstack-telemetry19:35
cdentjd__: voting has opened: https://www.openstack.org/summit/austin-2016/vote-for-speakers/presentation/696519:39
*** KrishR has quit IRC19:46
*** vishwanathj has joined #openstack-telemetry19:47
jd__cdent: you mean the one that is not used? :)19:49
cdentjd__: yupper19:50
*** rcernin has joined #openstack-telemetry20:08
*** KrishR has joined #openstack-telemetry20:24
openstackgerritDan Travis proposed openstack/ceilometer: Adds timestamp option to Aggregation transformer  https://review.openstack.org/27367220:39
datravisGeneral Gerrit question, how does one get Gerrit to post draft comments?20:49
gordcdatravis: what do you mean 'post draft comments'? i feel like you asking a non-obvious question20:54
*** prashantD has quit IRC20:54
*** safchain has joined #openstack-telemetry20:54
*** prashantD has joined #openstack-telemetry21:00
*** krotscheck is now known as krotscheck_dcm21:03
*** Guest9860 has quit IRC21:06
datravisgordc: I've attempted to respond to a comment in a review.  Gerrit seems to have put that comment in a draft state.  I was having issues getting it to publish that comment21:11
gordcdatravis: for the patch above ^^?21:11
datravisyessir21:11
gordcdatravis: are you commenting to patchset1 or the new patchset?21:11
datravispatchset121:12
datravis.... I see my problem now21:12
gordcyou need to switch view to patchset1 (top right)21:12
datravisthanks21:12
gordcyeah... new gui.21:12
*** datravis has quit IRC21:21
*** datravis has joined #openstack-telemetry21:41
*** rooney has quit IRC21:54
*** jwcroppe has joined #openstack-telemetry22:02
*** yassine has quit IRC22:07
openstackgerritDan Travis proposed openstack/ceilometer: Adds timestamp option to Aggregation transformer  https://review.openstack.org/27367222:10
*** thorst has quit IRC22:14
*** nadya has joined #openstack-telemetry22:15
*** thorst has joined #openstack-telemetry22:15
*** nadya is now known as Guest1739422:15
*** thorst_ has joined #openstack-telemetry22:16
*** thorst has quit IRC22:19
*** Guest17394 has quit IRC22:19
*** thorst_ has quit IRC22:21
*** safchain has quit IRC22:28
*** julim has quit IRC22:32
*** thorst has joined #openstack-telemetry22:39
*** jwcroppe has quit IRC22:42
*** thorst has quit IRC22:43
*** farid has quit IRC22:46
*** pradk has quit IRC22:49
*** cdent has quit IRC22:55
*** tongli has quit IRC22:56
*** rcernin has quit IRC22:58
*** thorst has joined #openstack-telemetry23:00
*** thorst has quit IRC23:09
*** thorst has joined #openstack-telemetry23:09
*** thorst has quit IRC23:14
*** nadya has joined #openstack-telemetry23:16
*** nadya is now known as Guest3193723:17
*** Guest31937 has quit IRC23:21
*** gordc has quit IRC23:39
*** jwcroppe has joined #openstack-telemetry23:40
*** jwcroppe has quit IRC23:44
*** jwcroppe has joined #openstack-telemetry23:45
*** KrishR has quit IRC23:45
*** jwcroppe has quit IRC23:49
*** mragupat_ has quit IRC23:57
*** rbak has quit IRC23:59

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