Monday, 2015-08-10

*** thumpba has quit IRC00:03
*** kbyrne has quit IRC00:17
*** thumpba has joined #openstack-ceilometer00:20
*** kbyrne has joined #openstack-ceilometer00:20
*** thumpba has quit IRC00:38
openstackgerritOpenStack Proposal Bot proposed openstack/ceilometer: Updated from global requirements  https://review.openstack.org/21085701:04
openstackgerritOpenStack Proposal Bot proposed openstack/ceilometermiddleware: Updated from global requirements  https://review.openstack.org/21088301:05
*** haigang has joined #openstack-ceilometer01:09
*** haigang has quit IRC01:26
*** fabian1 has joined #openstack-ceilometer01:54
*** alejandrito has joined #openstack-ceilometer02:08
*** thumpba has joined #openstack-ceilometer02:08
*** ankita_wagh has quit IRC02:14
*** ankita_wagh has joined #openstack-ceilometer02:14
*** thumpba has quit IRC02:15
*** alejandrito has quit IRC02:37
*** ankita_wagh has quit IRC02:38
*** thumpba has joined #openstack-ceilometer03:12
*** thumpba has quit IRC03:17
*** ddieterly has quit IRC03:51
*** yprokule has joined #openstack-ceilometer04:23
*** vishwanathj has quit IRC04:36
*** ddieterly has joined #openstack-ceilometer04:52
*** ankita_wagh has joined #openstack-ceilometer04:53
*** ankita_wagh has quit IRC04:56
*** ankita_wagh has joined #openstack-ceilometer04:56
*** ddieterly has quit IRC04:57
*** thumpba has joined #openstack-ceilometer05:14
*** thumpba has quit IRC05:18
*** ddieterly has joined #openstack-ceilometer05:53
*** ddieterly has quit IRC05:57
*** thumpba has joined #openstack-ceilometer06:15
*** thumpba has quit IRC06:19
*** nadya has joined #openstack-ceilometer06:22
*** afazekas_ has joined #openstack-ceilometer06:24
*** ildikov has quit IRC06:30
*** jkraj has joined #openstack-ceilometer06:43
*** ddieterly has joined #openstack-ceilometer06:53
*** ankita_wagh has quit IRC06:54
openstackgerritliusheng proposed openstack/aodh: Avoid re-registering transport configure options  https://review.openstack.org/21100206:56
*** ildikov has joined #openstack-ceilometer06:58
*** ddieterly has quit IRC06:59
*** belmoreira has joined #openstack-ceilometer07:02
*** thumpba has joined #openstack-ceilometer07:15
*** thumpba has quit IRC07:20
*** ankita_wagh has joined #openstack-ceilometer07:45
*** ddieterly has joined #openstack-ceilometer07:54
*** ddieterly has quit IRC07:59
jd__sileht: around?08:05
*** yassine has joined #openstack-ceilometer08:06
openstackgerritJulien Danjou proposed openstack/aodh: Stop registering oslo.messaging option  https://review.openstack.org/21102508:08
jd__sileht: opinion welcom on ^08:08
openstackgerritJulien Danjou proposed openstack/aodh: storage: re-add and deprecate alarm_connection  https://review.openstack.org/21028608:24
openstackgerritJulien Danjou proposed openstack/aodh: storage: re-add and deprecate alarm_connection  https://review.openstack.org/21028608:25
*** haigang has joined #openstack-ceilometer08:40
*** ankita_wagh has quit IRC08:43
openstackgerritChris Dent proposed openstack/ceilometer: Correct thread handling in TranslationHook  https://review.openstack.org/21050308:52
*** ddieterly has joined #openstack-ceilometer08:55
openstackgerritChris Dent proposed openstack/ceilometer: Make ConnectionRetryTest more reliable  https://review.openstack.org/21054208:58
*** ddieterly has quit IRC09:00
*** afazekas_ is now known as afazekas09:04
*** thumpba has joined #openstack-ceilometer09:17
*** belmoreira has quit IRC09:18
*** belmoreira has joined #openstack-ceilometer09:19
*** thumpba has quit IRC09:22
*** alexpilotti has joined #openstack-ceilometer09:25
*** cdent has joined #openstack-ceilometer09:45
openstackgerritJulien Danjou proposed openstack/gnocchi: sqlalchemy: use DBReferenceError to generate the correct exception  https://review.openstack.org/21106909:46
openstackgerritJulien Danjou proposed openstack/gnocchi: indexer: remove wrong FK catch  https://review.openstack.org/21107009:46
cdentjd__: your topic name fits pretty well: db-improv: keep trying stuff until the audience laughs09:47
jd__cdent: hmmm not sure I follow your joke here :)09:53
cdentyou mean improvement, right? but it sort of reads like "improvisation"09:53
jd__haha right I didn't realize09:53
jd__excellent :))09:54
*** ddieterly has joined #openstack-ceilometer09:56
* cdent is inspired to listen to some jazz09:57
*** ddieterly has quit IRC10:00
jd__hum today is gonna be Gnocchi 1.1.0rc110:02
jd__fasten your seat belt10:02
cdentI seem to have left mine with the other computer10:02
openstackgerritJulien Danjou proposed openstack/gnocchi: indexer: remove wrong FK catch  https://review.openstack.org/21107010:15
openstackgerritJulien Danjou proposed openstack/gnocchi: sqlalchemy: use DBReferenceError to generate the correct exception  https://review.openstack.org/21106910:15
jd__cdent: what's your thinking on the InfluxDB driver?10:15
cdentjd__: I haven't had a chance to actually use it, just make the tests go10:15
jd__ok10:16
cdentso I haven't had a chance to develop a string opinion10:16
cdentstrong10:16
cdentI was going to look into devstacking it last friday but fell into a giant hole of random bugs and never came out10:16
jd__ack10:17
jd__I guess it'll wait for 1.210:17
cdentI've been falling into a lot of holes lately10:18
cdentjd__: did you see the notes I wrote up on the file backend? nothing too formal, just some poking: https://tank.peermore.com/tanks/cdent-rhat/LongRunningGnocchi10:19
jd__cdent: I don't think I did, did you send them somewhere?10:19
* jd__ reading10:19
cdentthat was from the last productization meeting10:19
cdent(I think)10:19
jd__"Short version: Behavior is better than might have been expected" ok I stop reading then.10:19
jd__:>10:19
jd__ah ok, I missed it :(10:19
cdent:)10:20
jd__cdent: this is uber interesting, thank you for doing it10:23
cdentI had a more detailed version that I managed to accidently trashed, so that the filtered down version10:24
cdentwhich parts stand out as interesting10:24
jd__cdent: you have an idea on how many measures per second that represents?10:24
cdentthat's one of the numbers I lost :)10:24
jd__ok :)10:25
cdentI can make a vague guess: I had a watch ls |wc going on in the measures directory10:25
* cdent dredges his memory10:25
jd__but indeed the conclusion is that metricd works pretty well finally10:26
jd__I have a few ideas on how to improve things but it's a bit of work so I've postponed until someone really complains I guess10:27
*** fabian1 has quit IRC10:27
cdentthe number of dirs in measures oscillated between high 50s and low 70s10:27
cdentthat doesn't really tell us that much about how many measures/sec it was10:27
cdentbut it does give a sense that sometimes metricd was able to get ahead of the flow10:28
jd__nice10:28
jd__cdent: re: sharing file backend, I guess you can go with NFS but well, perf might not be terrific because well… NFS :)10:29
cdentI'm fairly convinced that if I were running my own little gnocchi setup I'd use file, not swift, as it seems like the http overhead would be a bear10:29
cdentthe way around the nfs ick would be to write to local disk and make it read only somewhere else10:29
cdentthat doesn't have any ha10:29
cdentbut meh10:30
cdentbut it doesn't seem kind of fun to have the notion of a node which is just doing gnocchi-metricd, spinning away wildly10:30
*** shardy has joined #openstack-ceilometer10:41
openstackgerritMerged openstack/gnocchi: Drop downgrade field in alembic script.py.mako  https://review.openstack.org/20930810:47
*** ddieterly has joined #openstack-ceilometer10:57
openstackgerritChris Dent proposed openstack/gnocchi: Some clarification of install and config procedures  https://review.openstack.org/21110310:59
*** ddieterly has quit IRC11:01
*** fawadkhaliq has joined #openstack-ceilometer11:11
*** liusheng has joined #openstack-ceilometer11:30
*** exploreshaifali has joined #openstack-ceilometer11:32
*** haigang has quit IRC11:33
openstackgerritMerged openstack/gnocchi: Don't update resource if nothing has changed  https://review.openstack.org/20752211:33
*** exploreshaifali has quit IRC11:34
*** haigang has joined #openstack-ceilometer11:35
*** exploreshaifali has joined #openstack-ceilometer11:36
*** exploreshaifali has quit IRC11:37
*** jwcroppe has quit IRC11:39
*** jwcroppe has joined #openstack-ceilometer11:40
*** haigang has quit IRC11:40
*** thorst has joined #openstack-ceilometer11:40
*** haigang has joined #openstack-ceilometer11:42
*** jwcroppe has quit IRC11:44
*** julim has joined #openstack-ceilometer11:48
*** ddieterly has joined #openstack-ceilometer11:57
*** ddieterly has quit IRC12:02
*** htruta has joined #openstack-ceilometer12:04
*** ddieterly has joined #openstack-ceilometer12:09
*** yassine has quit IRC12:11
*** yassine has joined #openstack-ceilometer12:12
*** gordc has joined #openstack-ceilometer12:13
openstackgerritgordon chung proposed openstack/ceilometer: support multiple-meter payloads  https://review.openstack.org/21051212:23
*** ddieterly has quit IRC12:30
nadyagordc: hi Gordon! I have the same question about event processing but for Juno... There is no event-pipeline yet there so the messages may be converted into samples or events but not to both, right? There is only one set of listeners12:32
gordcnadya: it'll convert to both still. it still uses same endpoint mechanism.12:34
gordcnadya: only difference is in juno it wrote straight to db12:34
nadyagordc: aha, all messages are processed by all endpoints, rigt12:35
nadyagordc: I'm describing the differences between Juno and Kilo for events...12:37
*** edmondsw has joined #openstack-ceilometer12:38
openstackgerritMerged openstack/aodh: storage: re-add and deprecate alarm_connection  https://review.openstack.org/21028612:45
gordcnadya: sorry. was reading another screen. the main difference between juno/kilo for events, is that it writes to db in juno and gets sent to pipeline in kilo... they are still processed by listeners the same way.12:50
nadyagordc: yep, I've checked all the bugs and bps. Also, you have added elasticsearch support, several improvements for sql and keeping 'raw' events12:52
*** julim has quit IRC12:56
*** tongli has joined #openstack-ceilometer12:58
*** Kennan has quit IRC13:03
*** lan__ has quit IRC13:04
*** haigang has quit IRC13:09
openstackgerritMerged openstack/gnocchi: Some clarification of install and config procedures  https://review.openstack.org/21110313:11
*** lan__ has joined #openstack-ceilometer13:16
*** prad_ has joined #openstack-ceilometer13:18
*** rbak has quit IRC13:18
*** thumpba has joined #openstack-ceilometer13:19
*** Kennan has joined #openstack-ceilometer13:20
openstackgerritJulien Danjou proposed openstack/gnocchi: Add to influx command's $PATH  https://review.openstack.org/20956113:21
openstackgerritJulien Danjou proposed openstack/gnocchi: InfluxDB storage driver  https://review.openstack.org/16540713:21
*** julim has joined #openstack-ceilometer13:22
*** jwcroppe has joined #openstack-ceilometer13:24
*** jwcroppe_ has joined #openstack-ceilometer13:27
*** jwcropp__ has joined #openstack-ceilometer13:28
*** jwcroppe has quit IRC13:29
*** haigang has joined #openstack-ceilometer13:31
*** jwcroppe_ has quit IRC13:31
*** changbl has quit IRC13:34
*** prad_ is now known as pradk13:36
*** haigang has quit IRC13:47
*** alexpilotti has quit IRC13:55
*** rbak has joined #openstack-ceilometer13:59
gordccould a core review this patch so it can be included in next release14:00
gordchttps://review.openstack.org/#/c/195515/14:00
*** jdandrea has joined #openstack-ceilometer14:00
gordcnadya: ^14:01
gordcsince you were around :)14:01
pradkgordc, i pushed a patch to support the math volume stuff.. i still marked it as a wip as the resource id part of cpu is kinda weird.. it is a combination string14:02
pradkgordc, i tried something, but i'm not very happy with that14:03
*** boris-42 has joined #openstack-ceilometer14:03
gordcpradk: yeah... it's not the cleanest... i'm not sure if we should not just revisit that.14:03
gordci complained about crazy ids from other services and then we have stuff like that in ours.14:04
pradkyea14:05
gordcpradk: which meters were they? the disk meters?14:05
pradkgordc, i thought of just leaving it as nodename but wasnt sure if it would break assumption people whould have made in their queries14:05
pradkcompute.node.cpu.*14:06
pradkresource_id is  "payload.host"_"payload.nodename"14:06
gordcoh. is that why you have two resource_ids listed in the .yaml14:06
pradkway i did it is, i made it a list in the yaml and joined14:06
pradkyea14:06
*** lan_ has joined #openstack-ceilometer14:07
gordcyeah, that's weird14:07
pradkyea agree14:07
nadyagordc: will do in 10-15 minutes14:08
gordcnadya: thanks!14:08
*** Kennan2 has joined #openstack-ceilometer14:09
*** Kennan has quit IRC14:09
*** lan__ has quit IRC14:11
*** liusheng has quit IRC14:18
gordcildikov: thanks!14:19
*** r-daneel has joined #openstack-ceilometer14:19
ildikovgordc: sure, np!14:19
pradkgordc, what you mean by "what if vol isnt the first value" ?14:24
pradkgordc, first value where? we only call this if the field is a volume14:25
pradkgordc, regarding multiplication yea i'm just defaulting to that as thats the only pattern we have so far ..if its not, just define it as an expr14:25
pradkgordc, we can go a bit more fancy and define an operator and use that14:26
pradkgordc, basically if the expression is a string we eval it, else we just scale the volume with whatever number we set14:27
pradki'm enhancing this logic a bit today to be a bit more generic, this is something i just whipped up so far still wip14:28
openstackgerritMerged openstack/ceilometermiddleware: Updated from global requirements  https://review.openstack.org/21088314:28
*** afazekas has quit IRC14:31
tongli@gordc, hi gordon, do we have a UI for Ceilometer on Horizon?14:31
gordcpradk: right, so you're assumption is that it's always some sort of scaling mechanism... but that's not always the case. https://github.com/openstack/ceilometer/blob/master/ceilometer/dns/notifications.py14:32
pradkgordc, right in this case it would be an expression which would be a string and matches the eval case14:33
pradkgordc, so the assumption is not always a scaling mechanism.. the asumption is if its not a string, its probably scaling14:35
gordc... hmm, it seems slightly strange but we'll see how it goes as you implement14:37
pradkcool cool..14:37
pradkyea i havent tested every possible case yet14:37
gordctongli: i don't have horizon enabled, but yes, there's a ceilometer tab in Horizon.14:37
gordcit's under admin/resource info i believe14:37
tongli@gordc, so if ceilometer is installed on devstack, it will be there by default, no special configurations to enable the UI, right?14:38
gordcyes, it's there by default.14:38
tongli@gordc, cool. thanks.14:39
gordci can't remember when it was added specifically.. i'm going to guess around havana/icehouse time frame14:39
tongli@gordc, cool. I was assigned to give a Ceilometer introduction talk. Do you have any good material handy?14:40
gordctongli: a technical talk?14:40
tongli@gordc, an introduction to people in IBM who has not had much OpenStack experiences yet.14:41
tonglinot very technical.14:41
*** exploreshaifali has joined #openstack-ceilometer14:44
gordctongli: some old slides http://www.slideshare.net/enovance/ceilometer-presentation-ods-havana-final-published14:46
gordcsome more recent slides with architecture/deployment scenarios: http://www.slideshare.net/praddevel/stabilizing-the-jenga-tower-scaling-out-ceilometer14:46
tongli@gordc, cool. thanks.14:46
*** fawadkhaliq has quit IRC15:06
*** jasonamyers has joined #openstack-ceilometer15:17
openstackgerritMerged openstack/python-ceilometerclient: Add ceilometerclient support for api-no-pipeline  https://review.openstack.org/19551515:20
*** ddieterly has joined #openstack-ceilometer15:21
*** ddieterly has quit IRC15:22
*** lan_ has quit IRC15:26
*** lan_ has joined #openstack-ceilometer15:26
*** shakamunyi has joined #openstack-ceilometer15:27
*** jruano has joined #openstack-ceilometer15:29
*** thumpba has quit IRC15:30
*** thumpba has joined #openstack-ceilometer15:32
*** thumpba has quit IRC15:37
*** thumpba has joined #openstack-ceilometer15:37
*** belmoreira has quit IRC15:38
*** thumpba has quit IRC15:38
*** thumpba has joined #openstack-ceilometer15:38
*** changbl has joined #openstack-ceilometer15:39
*** thumpba has quit IRC15:40
*** thumpba has joined #openstack-ceilometer15:41
*** exploreshaifali has quit IRC15:43
*** thumpba has quit IRC15:44
*** thumpba has joined #openstack-ceilometer15:45
*** _cjones_ has joined #openstack-ceilometer15:55
*** nadya has quit IRC16:01
*** pauldix has joined #openstack-ceilometer16:02
*** yprokule has quit IRC16:08
*** exploreshaifali has joined #openstack-ceilometer16:12
*** ankita_wagh has joined #openstack-ceilometer16:22
*** shardy has quit IRC16:25
*** jkraj has quit IRC16:29
*** fawadkhaliq has joined #openstack-ceilometer16:34
*** jruano has quit IRC16:38
*** yassine has quit IRC16:47
*** jasonamyers has quit IRC16:48
*** alexpilotti_ has joined #openstack-ceilometer16:57
*** ildikov has quit IRC16:57
*** ankita_wagh has quit IRC16:57
*** ankita_wagh has joined #openstack-ceilometer16:58
*** alexpilotti_ has quit IRC17:01
*** fawadkhaliq has quit IRC17:01
*** ankita_wagh has quit IRC17:02
*** exploreshaifali has quit IRC17:04
*** jruano has joined #openstack-ceilometer17:05
*** shardy has joined #openstack-ceilometer17:07
openstackgerritJulien Danjou proposed openstack/aodh: Stop registering oslo.messaging option  https://review.openstack.org/21102517:10
*** shardy has quit IRC17:12
openstackgerritChris Dent proposed openstack/aodh: Correct thread handling in TranslationHook  https://review.openstack.org/21041817:13
openstackgerritChris Dent proposed openstack/aodh: Correct database functional tests  https://review.openstack.org/20784017:13
*** ankita_wagh has joined #openstack-ceilometer17:14
*** nadya has joined #openstack-ceilometer17:17
*** ankita_w_ has joined #openstack-ceilometer17:24
*** ankita_wagh has quit IRC17:27
*** jruano has quit IRC17:39
*** jruano has joined #openstack-ceilometer17:43
*** nadya has quit IRC17:58
openstackgerritMerged openstack/gnocchi: storage/carbonara: add timestamp as measure suffix  https://review.openstack.org/21075018:00
openstackgerritMerged openstack/gnocchi: storage/carbonara: add timestamp as measure suffix  https://review.openstack.org/21075018:00
*** ildikov has joined #openstack-ceilometer18:12
openstackgerritMerged openstack/aodh: Correct thread handling in TranslationHook  https://review.openstack.org/21041818:17
cdentgordc: I changed the ordering so that with that ^ merge, this is ready to go and passing okay: https://review.openstack.org/#/c/207840/18:18
cdentso it would be super keen if we can get that in and start running the experimental checks all of aodh and then make them real18:18
gordccdent: ah cool. yeah i saw it this morning and noticed it was failing. will take a look again18:20
*** ankita_wagh has joined #openstack-ceilometer18:30
*** ankita_w_ has quit IRC18:32
cdentdanke gordc18:35
openstackgerritMerged openstack/aodh: Correct database functional tests  https://review.openstack.org/20784018:37
gordcdamn. that merged quickly18:38
*** rjaiswal has joined #openstack-ceilometer18:39
rjaiswalgordc: just curious to know why this was abandoned: https://review.openstack.org/#/c/30124/18:40
*** ddieterly has joined #openstack-ceilometer18:40
gordcrjaiswal: i don't want to name names... but i was told to go away. :)18:42
gordcrjaiswal: it actually exists in keystonemiddleware now18:43
gordchttps://github.com/openstack/keystonemiddleware/blob/master/keystonemiddleware/audit.py18:43
gordcand is processed as an event: https://github.com/openstack/ceilometer/blob/master/etc/ceilometer/event_definitions.yaml#L332-L36818:44
*** jruano has quit IRC18:44
rjaiswalgordc: ok18:45
rjaiswalthanks gordc18:46
gordcnp18:46
*** jruano has joined #openstack-ceilometer18:52
*** ddieterly has quit IRC18:58
*** ddieterly has joined #openstack-ceilometer19:03
*** exploreshaifali has joined #openstack-ceilometer19:11
*** ddieterly has quit IRC19:11
*** ankita_w_ has joined #openstack-ceilometer19:12
*** ankita_wagh has quit IRC19:15
*** fawadkhaliq has joined #openstack-ceilometer19:18
openstackgerritgordon chung proposed openstack/ceilometer: start rpc deprecation  https://review.openstack.org/21130419:32
*** belmoreira has joined #openstack-ceilometer19:34
*** nadya has joined #openstack-ceilometer19:36
*** ankita_w_ has quit IRC19:38
*** exploreshaifali has quit IRC19:40
*** nadya has quit IRC19:52
*** jruano has quit IRC19:55
*** ankita_wagh has joined #openstack-ceilometer19:57
openstackgerritgordon chung proposed openstack/ceilometer: add test to validate jsonpath  https://review.openstack.org/21131420:03
*** jasonamyers has joined #openstack-ceilometer20:04
*** Ephur has joined #openstack-ceilometer20:09
*** fawadkhaliq has quit IRC20:29
*** jasonamyers has quit IRC20:45
*** jasonamyers has joined #openstack-ceilometer20:51
*** changbl has quit IRC20:51
*** Saba has joined #openstack-ceilometer20:51
*** Saba is now known as Guest2093020:52
Guest20930Hi , how do I query from CLI something like ceilometer -m cpu_util -q "resource_id=[xxxxxx, yyyyyyy]"   i.e. fetching cpu_util values for multiple instances20:53
*** cdent has quit IRC20:53
gordcGuest20930: you'll need to use complex queries21:00
gordcGuest20930: https://blog.berendt.io/querying-ceilometer/21:00
Guest20930Thanks21:02
*** thorst has quit IRC21:02
*** tongli has quit IRC21:03
*** belmoreira has quit IRC21:03
gordcnp21:07
*** jasonamyers has quit IRC21:10
*** thumpba has quit IRC21:12
*** pradk has quit IRC21:14
openstackgerritMerged openstack/ceilometer: Make ConnectionRetryTest more reliable  https://review.openstack.org/21054221:17
*** julim has quit IRC21:22
*** ankita_w_ has joined #openstack-ceilometer21:27
*** ankita___ has joined #openstack-ceilometer21:27
*** ankita___ has quit IRC21:28
*** ankita___ has joined #openstack-ceilometer21:29
*** ankita_wagh has quit IRC21:30
*** ankita_w_ has quit IRC21:31
*** ankita_wagh has joined #openstack-ceilometer21:34
*** ankita___ has quit IRC21:34
*** thorst has joined #openstack-ceilometer21:34
openstackgerritgordon chung proposed openstack/ceilometer: initialise opencontrail client in tests  https://review.openstack.org/21133521:34
openstackgerritgordon chung proposed openstack/ceilometer: start rpc deprecation  https://review.openstack.org/21130421:35
*** nadya has joined #openstack-ceilometer21:52
*** nadya has quit IRC21:57
*** gordc has quit IRC22:01
*** Guest20930 has quit IRC22:07
*** rbak has quit IRC22:13
*** edmondsw has quit IRC22:25
*** ankita_w_ has joined #openstack-ceilometer22:36
*** ankita_wagh has quit IRC22:37
*** r-daneel has quit IRC22:39
*** julim has joined #openstack-ceilometer22:43
*** ddieterly has joined #openstack-ceilometer23:01
*** ddieterly has quit IRC23:05
*** rjaiswal has quit IRC23:10
*** julim has quit IRC23:21
*** pauldix has quit IRC23:23
*** titilambert has quit IRC23:31
*** titilambert has joined #openstack-ceilometer23:31
*** changbl has joined #openstack-ceilometer23:38

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