Wednesday, 2013-10-16

*** SergeyLukjanov has joined #openstack-metering00:18
*** matsuhashi has joined #openstack-metering00:20
*** SergeyLukjanov has quit IRC00:41
*** ryanpetrello has quit IRC01:07
*** nosnos has joined #openstack-metering01:11
*** ryanpetrello has joined #openstack-metering01:26
*** nati_ueno has quit IRC01:45
*** matsuhashi has quit IRC01:46
*** matsuhashi has joined #openstack-metering01:47
*** matsuhas_ has joined #openstack-metering01:49
*** matsuhashi has quit IRC01:49
*** suo has joined #openstack-metering01:59
*** ryanpetrello has quit IRC02:02
*** fandikurnia01 has joined #openstack-metering02:03
*** fandikurnia01 has quit IRC02:04
*** fandikurnia01 has joined #openstack-metering02:04
*** suo has quit IRC02:14
*** suo has joined #openstack-metering02:15
*** sandywalsh has quit IRC02:20
*** ryanpetrello has joined #openstack-metering02:41
*** ryanpetrello has quit IRC02:45
*** ryanpetrello has joined #openstack-metering02:46
*** ryanpetrello has quit IRC02:47
*** sandywalsh has joined #openstack-metering02:50
*** fandikurnia01 has quit IRC02:56
*** fandikurnia01 has joined #openstack-metering02:56
*** julim has quit IRC03:15
*** ArcTanSusan has joined #openstack-metering03:22
*** ArcTanSusan has quit IRC03:27
*** julim has joined #openstack-metering03:30
*** julim has quit IRC03:36
*** nati_ueno has joined #openstack-metering03:41
*** nati_uen_ has joined #openstack-metering03:50
*** nati_ueno has quit IRC03:54
*** ArcTanSusan has joined #openstack-metering04:10
*** egallen has joined #openstack-metering04:16
fandikurnia01hi all04:21
fandikurnia01may someone share error log like ERROR [ceilometer.openstack.common.threadgroup] Unable to communicate with identity service: {"error": {"message": "Invalid user / password", "code": 401, "title": "Not Authorized"}}. (HTTP 401)04:22
*** evanjfraser has quit IRC04:29
*** nati_uen_ has quit IRC04:36
*** nati_ueno has joined #openstack-metering04:36
*** asalkeld has quit IRC04:59
*** evanjfraser has joined #openstack-metering04:59
*** evanjfraser has quit IRC05:04
*** fandikurnia01 has quit IRC05:16
*** nati_uen_ has joined #openstack-metering05:25
*** nati_ueno has quit IRC05:27
*** ryanpetrello has joined #openstack-metering05:36
*** ryanpetrello has quit IRC05:46
*** Bada has joined #openstack-metering05:58
*** fandikurnia01 has joined #openstack-metering06:00
openstackgerritJenkins proposed a change to openstack/ceilometer: Imported Translations from Transifex  https://review.openstack.org/4968006:03
*** Bada has quit IRC06:05
*** nati_ueno has joined #openstack-metering06:13
*** nati_uen_ has quit IRC06:16
*** ArcTanSusan has quit IRC06:41
*** yfujioka has joined #openstack-metering06:42
*** sandywalsh has quit IRC06:49
fandikurnia01ok i miss some configuration,06:55
fandikurnia01now glance is working06:55
fandikurnia01[root@controller-2 ~]# ceilometer --debug meter-list06:55
fandikurnia01INFO (connectionpool:191) Starting new HTTP connection (1): 192.168.0.306:55
fandikurnia01DEBUG (connectionpool:283) "POST /v2.0/tokens HTTP/1.1" 200 None06:55
fandikurnia01DEBUG (http:109) curl -i -X GET -H 'X-Auth-Token: 2984f952b2354e9e96c439460d26ac93' -H 'Content-Type: application/json' -H 'Accept: application/json' -H 'User-Agent: python-ceilometerclient' http://10.20.0.133:8777//v1/meters06:55
fandikurnia01DEBUG (http:119)06:55
fandikurnia01HTTP/1.0 200 OK06:55
fandikurnia01date: Mon, 14 Oct 2013 12:32:54 GMT06:55
fandikurnia01content-length: 47406:55
fandikurnia01content-type: application/json06:55
fandikurnia01server: WSGIServer/0.1 Python/2.6.606:55
*** eglynn has joined #openstack-metering06:59
*** egallen has quit IRC07:00
*** boris-42 has quit IRC07:01
*** nati_uen_ has joined #openstack-metering07:01
*** sandywalsh has joined #openstack-metering07:04
*** nati_ueno has quit IRC07:05
*** egallen has joined #openstack-metering07:06
*** asalkeld has joined #openstack-metering07:07
*** egallen has quit IRC07:11
*** egallen has joined #openstack-metering07:12
*** suo has quit IRC07:36
*** suo has joined #openstack-metering07:37
*** Bada has joined #openstack-metering07:37
*** shardy_afk is now known as shardy07:37
*** dafter has joined #openstack-metering07:50
*** tvb|afk has joined #openstack-metering07:51
*** tvb|afk has joined #openstack-metering07:51
*** dafter has quit IRC07:55
*** aspiers has joined #openstack-metering07:58
*** boris-42 has joined #openstack-metering08:10
*** nprivalova has joined #openstack-metering08:27
*** nati_uen_ has quit IRC09:09
*** nati_ueno has joined #openstack-metering09:10
*** nati_ueno has quit IRC09:15
*** tvb|afk has quit IRC09:18
*** matsuhas_ has quit IRC09:19
*** matsuhas_ has joined #openstack-metering09:19
*** dafter has joined #openstack-metering09:36
*** dafter has quit IRC09:36
*** dafter has joined #openstack-metering09:36
*** Ruetobas has quit IRC09:43
*** nprivalova has quit IRC09:45
*** Ruetobas has joined #openstack-metering09:47
openstackgerritYuuichi Fujioka proposed a change to openstack/python-ceilometerclient: Fix order of sample list  https://review.openstack.org/4733109:49
*** nprivalova has joined #openstack-metering09:50
*** Ruetobas has quit IRC09:51
*** egallen has quit IRC09:57
*** Ruetobas has joined #openstack-metering09:57
*** yfujioka has quit IRC10:02
*** nprivalova has quit IRC10:17
*** julim has joined #openstack-metering10:30
eglynnjd__, nijaba_: BTW did that etherpad outlining a proposed structure for summit preso make sense?10:41
* eglynn starting to knock up some draft slides on googledocs based on that10:41
eglynn(so feedback would be good if you reckon a different approach/emphasis/whatever would work better ...)10:42
jd__eglynn: let me check that in a few minutes :)10:42
eglynnjd__: sure, no worries :)10:43
jd__eglynn: added a few things, otherwise looks good10:47
eglynnjd__: excellent, thank you sir!10:47
eglynnjd__: cool, I'll continue on that basis to put together some draft slides10:49
eglynn(we can always butcher them again in the review cycle ...)10:49
*** nprivalova has joined #openstack-metering10:50
*** boris-42 has quit IRC10:52
*** egallen has joined #openstack-metering10:53
*** julim has quit IRC10:56
nprivalovahello, guys. I've got the error during hbase testing: http://paste.openstack.org/show/48549/. Is it hbase-specific? Or maybe the same was found for Mongo and has been already fixed?11:03
*** sandywalsh has quit IRC11:26
*** sandywalsh has joined #openstack-metering11:39
*** fandikurnia01 has quit IRC12:04
*** egallen has quit IRC12:07
*** egallen has joined #openstack-metering12:10
*** nprivalova has quit IRC12:12
*** boris-42 has joined #openstack-metering12:22
*** fandikurnia01 has joined #openstack-metering12:31
*** ryanpetrello has joined #openstack-metering12:35
*** nprivalova has joined #openstack-metering12:35
*** terriyu has joined #openstack-metering12:37
*** bogdando has quit IRC12:37
*** bogdando has joined #openstack-metering12:40
*** egallen has quit IRC12:45
*** bpokorny has joined #openstack-metering12:48
*** egallen has joined #openstack-metering12:49
*** nosnos has quit IRC12:55
*** nosnos has joined #openstack-metering12:55
*** gordc has joined #openstack-metering12:59
*** nosnos has quit IRC13:00
*** eglynn is now known as eglynn-lunch13:01
*** thomasm has joined #openstack-metering13:02
*** thomasm is now known as Guest1358213:02
*** suo has quit IRC13:06
*** bingbu has joined #openstack-metering13:07
Guest13582jd__, hey there, ping13:13
*** Guest13582 is now known as thomasem13:13
thomasem? need to fix that13:13
jd__hi Guest13:13
thomasemLol, sorry, it was me, jd__ =]13:14
jd__:-)13:14
thomasemjd__, So, yeah - that optional samples branch does have two points to turn sampling on and off, and I don't like that fact either. I want to think on this a bit more, maybe even write up a BP for the suggestion to push notification -> sample conversion into pipeline.yaml for config.13:16
thomasemjd__, One question I'm left with: Are instance:<flavor name> and ImageCRUD just artifacts of the old way (before one meter one plugin) that just missed the BP implementation?13:16
*** nprivalova is now known as nadya13:25
*** shakayumi has joined #openstack-metering13:32
*** bingbu has quit IRC13:46
*** gordc has quit IRC13:46
*** gordc has joined #openstack-metering13:51
*** Shaan7 has quit IRC13:51
*** Shaan7 has joined #openstack-metering13:51
nijaba_eglynn-lunch: sorry, must have missed the etherpad link13:51
*** eglynn-lunch is now known as eglynn13:52
eglynnnijaba_: np!13:52
eglynnnijaba_: here it is https://etherpad.openstack.org/p/CeiloPlusHeatEqualsAlarming13:52
jd__thomasem: yes I've a patch to remove instance:<flavor name> but the API still limits the ability to remove this one13:54
jd__thomasem: didn't check imageCrud, but likely yes :)13:54
jd__and about notification -> sample in a conf file, we already talked about the idea a couple of time and we'll be happy to move this forward13:55
thomasemjd__, Okay. =] Thanks for handling my continuous questioning. I'm a curious guy. Great, I'll write up a BP for it. =]13:55
jd__I've proposed a session around having more structured notifications for Icehouse that'll help in this regard13:55
thomasemUnless there already is one.13:55
jd__thomasem: no problem :)13:55
thomasemOh, awesome13:55
thomasemVery sad that I don't get to join y'all at the summit. I will at the next one, though, or die trying. =]13:56
*** egallen has quit IRC14:07
*** SergeyLukjanov has joined #openstack-metering14:07
*** egallen has joined #openstack-metering14:08
*** esheffield has quit IRC14:10
nadyahi guys, I decided to ask again about  http://paste.openstack.org/show/48549/ . Have you ever seen this anywhere except HBase?14:18
*** fandikurnia01 has quit IRC14:18
*** ryanpetrello has quit IRC14:20
jd__nadya: I never saw that14:26
*** ryanpetrello has joined #openstack-metering14:34
nadyajd__, we saw it twice on different devstacks. I mean on different deployment of the latest devstack. It is mostly FYI, I will continue to understand what's going on here14:36
*** matsuhas_ has quit IRC14:38
*** matsuhashi has joined #openstack-metering14:39
*** matsuhashi has quit IRC14:39
*** matsuhashi has joined #openstack-metering14:40
*** matsuhashi has quit IRC14:42
*** matsuhashi has joined #openstack-metering14:42
*** SergeyLukjanov is now known as _SergeyLukjanov14:46
*** beav has joined #openstack-metering14:47
*** matsuhashi has quit IRC14:47
*** _SergeyLukjanov is now known as SergeyLukjanov14:50
*** ryanpetrello has quit IRC14:59
*** SergeyLukjanov is now known as _SergeyLukjanov15:03
*** ryanpetrello has joined #openstack-metering15:03
*** _SergeyLukjanov has quit IRC15:03
*** SergeyLukjanov has joined #openstack-metering15:04
*** gordc has quit IRC15:05
*** SergeyLukjanov has quit IRC15:20
*** egallen has quit IRC15:23
*** egallen has joined #openstack-metering15:26
nijaba_eglynn: looks like a lot of content to me for 40 min15:27
nijaba_let me send you what shardy and I used for a 20 min pres15:28
eglynnnijaba_: yeah, I've edited down the ceilometer overview to just a single slide15:28
*** krtaylor has quit IRC15:33
*** egallen has quit IRC15:42
*** SergeyLukjanov has joined #openstack-metering15:53
*** nadya has quit IRC15:56
*** gordc has joined #openstack-metering15:59
sandywalshjd__, gotta sec?16:00
*** aspiers has quit IRC16:00
*** Ruetobas has quit IRC16:01
jd__sandywalsh: just one16:02
*** Ruetobas has joined #openstack-metering16:03
sandywalshjd__, :) k, so, I want to make a metric plugin for nova that will publish samples via udp. I could refactor the udp publisher into oslo, but it's pretty simple code.16:05
sandywalshjd__, Do you think I should just dup the socket code from the publisher or refactor it into oslo and just wrap the call from the metric plugin and the udp publisher?16:05
jd__is it going to publish samples or notifications? and if not notifications, why not notifications?16:06
sandywalshsamples ... these are polled data samples (lots and frequent)16:06
sandywalshit's hook into the periodic_tasks in nova that already does all the HV polling16:07
sandywalsh(ram, disk, network, etc)16:07
sandywalsh*it's a16:07
jd__ok, but you don't mean "Ceilometer samples" right?16:07
sandywalshwhatever the udp publisher sends. I assumed they were CM samples?16:08
*** Ruetobas has quit IRC16:08
jd__the UDP publisher sends samples indeed16:08
jd__but I don't think Nova has code to send samples16:08
jd__s/send/generate/ even16:09
sandywalshname, unit, volume, type16:09
sandywalshit doesn't currently, I have a branch to support it16:09
sandywalshhttps://review.openstack.org/#/c/51249/16:09
jd__ok16:09
* jd__ thinks16:10
*** krtaylor has joined #openstack-metering16:10
sandywalshI'm thinking other services that poll for data will want to do the same thing. So I was thinking oslo.16:11
sandywalshmay not be any need to move/refactor the existing udp publisher16:12
jd__sandywalsh: yeah, but I think your implementation is bit too narrow16:12
jd__I wouldn't like to have this UDP only16:12
sandywalshit's a plugin, it can be anything16:12
sandywalshfirst cut for CM integration will be via udp16:12
jd__sandywalsh: I really think that should send notifications; and then the a notification driver should support UDP16:13
jd__I don't think it's a good idea to have several system to send things out16:13
sandywalshway too many in our environment for that16:13
*** Ruetobas has joined #openstack-metering16:13
jd__sandywalsh: notifications != RPC16:13
*** boris-42 has quit IRC16:14
sandywalshI know that16:14
sandywalshit's still volume in the queue16:14
jd__no16:14
jd__you could have a notification driver that send notifications over UDP16:14
sandywalshoh, you mean to send notification over udp16:15
sandywalshhmm16:15
jd__actually, we even have a notification driver that transforms notifications in samples and send using the multipublisher (and therefore UDP)16:15
jd__it's in ceilometer.notifier I think16:15
sandywalshnot sure ... there are semantics around "notifications" that are assumed16:15
jd__it's really used nor documented, but it was something I had in mind a few months ago so I wrote it as a PoC16:15
sandywalshnot dropping them, being the big one16:15
jd__sandywalsh: well, I think we would need a few more in the notification driver, like some routing16:16
jd__like we do in Ceilometer pipeline sort of16:16
jd__so you can send some notifications on RPC, and others over UDP16:16
sandywalshjd__, routing will be handled in the trigger pipeline16:16
jd__sandywalsh: well by routing I mean, publishing16:16
*** dafter has quit IRC16:16
jd__sandywalsh: point is, I don't think we should yet another interface to send information ouf of Nova or whatever16:17
jd__(dhellmann might have an opinion on that too, HIGHLIGHT! :)16:18
sandywalshI think I see your point. You're implying we should have a notification router and use it to send *some* notifications to rabbit and others to UDP drivers16:18
jd__sandywalsh: yes, that's what I mean16:19
jd__sorry if it wasn't clear :)16:19
sandywalshnp ... this whole game is about terminology and context :)16:19
jd__sandywalsh: which is something that overlaps with what we do with our pipeline in Ceilometer, so there might be way to reconcilie both in Oslo16:19
jd__*reconcile16:20
sandywalshinteresting idea ... lemme stew on it for a little and I'll get back to you.16:20
sandywalshmy immediate worry is a bad configuration flooding the queue, but that might be a false alarm.16:20
sandywalshactually it would be pretty easy to implement without much change to the existing code base. Just a RouterNotifier which is a wrapper on the stock notifiers16:22
jd__stew, stew :)16:22
sandywalshthanks ... stay tuned16:22
sandywalshjd__, stewed. Good idea. I'll get started on that.16:27
jd__thumbs up16:27
jd__feel free to run ideas against me if you need or whatever :)16:27
*** shakayumi has quit IRC16:33
*** julim has joined #openstack-metering17:26
*** ryanpetrello has quit IRC17:28
*** Shaan7 has quit IRC17:28
eglynnjd__, nijaba_: I shared out a very rough first cut of that slidedeck on googledocs17:29
openstackgerritEric Pendergrass proposed a change to openstack/ceilometer: Don't assume localhost for mysql test connection  https://review.openstack.org/5216917:29
openstackgerritEric Pendergrass proposed a change to openstack/ceilometer: Don't assume localhost for mysql test connection  https://review.openstack.org/5216917:32
*** boris-42 has joined #openstack-metering17:38
*** Bada has quit IRC17:47
openstackgerritgordon chung proposed a change to stackforge/pycadf: observer and reporter should be valid Resource  https://review.openstack.org/5217517:51
*** nadya has joined #openstack-metering17:58
thomasemdhellmann, Hey, man. Is there a way for me to alter the return formatting for a particular type in WSME? This is regarding that timestamp precision from the API bug.18:10
thomasemI'm trying to find if there's some declaration for what causes the datetime object to convert to ISO8601 without sub-second (unless sub-second is > 0 in the datetime object), but it looks like the default from the documentation?18:11
openstackgerritgordon chung proposed a change to stackforge/pycadf: observer and reporter should be valid Resource  https://review.openstack.org/5217518:29
openstackgerritDavid Peraza proposed a change to openstack/ceilometer: Changes FakeMemcache to set token to expire on utcnow + 5 mins  https://review.openstack.org/5218218:30
openstackgerritEric Pendergrass proposed a change to openstack/ceilometer: Don't assume localhost for mysql test connection  https://review.openstack.org/5216918:33
openstackgerritEric Pendergrass proposed a change to openstack/ceilometer: Don't assume localhost for mysql test connection  https://review.openstack.org/5216918:34
*** julim has quit IRC18:43
*** julim has joined #openstack-metering19:00
dhellmannthomasem: Good question. If that's the default for datetime, maybe wsme needs to be updated to use a consistent format string.19:00
*** sandywalsh has quit IRC19:02
thomasemdhellmann, WSME follows the default Python DateTime behavior with respect to append sub-second precision, which actually seems like a perfectly fine default. I was just curious if a way to force sub-second existed for WSME. Otherwise I am thinking we could just change those fields to string and format in our controller?19:03
thomasemJust to make it consistent for UX purposes. It is already ISO8601 compliant - but if someone is consuming this without a compliant parser, they're going to have no fun.19:04
thomasemI blame Java. :P19:04
dhellmannthomasem: I blame Java, too, but I can see the argument that WSME should produce the timestamps in the same format all the time.19:05
dhellmannotoh, if it's really compliant than it doesn't seem like a high priority to me19:05
thomasemIt appears to be from my end. I'm not testing according to the spec or anything, but sub-second is optional, and so is timezone - which we address in the documentation by saying it's UTC. =]19:06
thomasemAnywho, the perfect format is here: https://github.com/openstack/ceilometer/blob/master/ceilometer/openstack/common/timeutils.py#L3319:07
thomasemhttps://github.com/openstack/ceilometer/blob/master/ceilometer/openstack/common/timeutils.py#L3119:07
dhellmannthomasem: right19:12
dhellmannI'll leave it up to jd__, but if anything needs to be "fixed" I think the right place for the code change is in WSME and not ceilometer.19:13
dhellmannIMO, it's not a bug if what we are producing meets the standard19:13
thomasemdhellmann, Agreed. I'm thinking a little regex could handle the case just fine anyway from the Java side.19:14
dhellmannthomasem: I find it difficult to believe there is no ISO date time parser for Java19:15
*** sandywalsh has joined #openstack-metering19:15
thomasemdhellmann, Yeah. I couldn't find one quickly. The only date parsers that I saw floating around pretty much had a community "DO NOT USE" pasted across them.19:15
thomasemAnd folks saying just to regex it19:16
dhellmannwow19:17
thomasemLol, it's Java. Whatcha gonna do? The internal calendar object and it's kids are apparently pretty rough around the edges.19:17
thomasemI almost want to take the challenge and write them a Java date parser for 8601.19:17
*** ryanpetrello has joined #openstack-metering19:18
thomasemGonna have to dust off the old JDK.19:18
thomasemIt looks like there's an XML one that has a ISO8601 parse built in. Anyway - not a high priority, methinks.19:21
*** ekarlso has quit IRC19:27
*** nadya has quit IRC19:41
*** ekarlso has joined #openstack-metering19:46
*** evanjfraser has joined #openstack-metering19:47
*** dperaza has joined #openstack-metering19:48
dperazaHello anyone knows what these failures are: http://logs.openstack.org/82/52182/1/check/gate-ceilometer-python27/32e0f38/console.html19:48
dperazaseems unrelated to my patch https://review.openstack.org/#/c/5218219:49
dperazaand other patches get it too19:49
*** julim has quit IRC19:52
thomasemdhellmann, added to the whiteboard here: https://blueprints.launchpad.net/ceilometer/+spec/pipeline-configuration-cleanup19:52
*** julim has joined #openstack-metering19:53
*** lrengan has joined #openstack-metering20:02
*** lrengan has left #openstack-metering20:02
*** SergeyLukjanov is now known as _SergeyLukjanov20:06
*** _SergeyLukjanov has quit IRC20:06
*** SergeyLukjanov has joined #openstack-metering20:08
*** aignatov has quit IRC20:11
*** aignatov has joined #openstack-metering20:11
*** terriyu has quit IRC20:14
*** terriyu has joined #openstack-metering20:16
*** thomasem has quit IRC20:17
*** thomasm has joined #openstack-metering20:18
*** thomasm is now known as Guest1362020:19
*** julim has quit IRC20:26
*** evanjfraser_ has joined #openstack-metering20:28
*** krtaylor has quit IRC20:31
*** boris-42_ has joined #openstack-metering20:36
*** nati_ueno has joined #openstack-metering20:37
*** lexx has joined #openstack-metering20:39
*** boris-42 has quit IRC20:40
*** beav has quit IRC20:51
*** lexx_ has joined #openstack-metering20:52
*** lexx__ has joined #openstack-metering20:54
*** lexx has quit IRC20:55
*** lexx_ has quit IRC20:58
*** lexx has joined #openstack-metering21:00
*** lexx__ has quit IRC21:03
*** julim has joined #openstack-metering21:03
*** lexx has quit IRC21:08
*** fc__ has quit IRC21:13
*** fc__ has joined #openstack-metering21:14
*** eglynn has quit IRC21:19
*** ryanpetrello has quit IRC21:25
*** krtaylor has joined #openstack-metering21:29
*** SergeyLukjanov is now known as _SergeyLukjanov21:35
*** _SergeyLukjanov has quit IRC21:36
*** SergeyLukjanov has joined #openstack-metering21:36
*** asalkeld is now known as asalkeld_afk21:39
*** Guest13620 has quit IRC21:43
*** eglynn has joined #openstack-metering21:52
*** asalkeld_afk is now known as asalkeld21:53
*** bpokorny has quit IRC21:57
*** bpokorny has joined #openstack-metering21:59
*** julim has quit IRC22:10
*** annegentle has quit IRC22:10
*** zigo_ has quit IRC22:11
*** zigo has joined #openstack-metering22:11
*** annegentle has joined #openstack-metering22:13
*** eglynn has quit IRC22:18
*** bpokorny has quit IRC22:34
*** boris-42_ has quit IRC22:41
*** ruhe has joined #openstack-metering22:50
*** ruhe has quit IRC22:51
*** SergeyLukjanov is now known as _SergeyLukjanov22:52
*** _SergeyLukjanov is now known as SergeyLukjanov22:52
*** yfujioka has joined #openstack-metering23:37
*** SergeyLukjanov has quit IRC23:49

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