Wednesday, 2014-03-05

*** fnaval has quit IRC00:11
*** jmckind has quit IRC00:11
*** raymondr_ has quit IRC00:14
*** mihgen has joined #openstack-ceilometer00:18
*** mihgen has quit IRC00:22
*** changbl has quit IRC00:24
*** shakayumi has joined #openstack-ceilometer00:24
*** jdob has quit IRC00:25
*** gordc has quit IRC00:27
*** mihgen has joined #openstack-ceilometer00:29
*** nati_ueno has quit IRC00:30
*** Yathi has quit IRC00:31
*** mihgen has quit IRC00:32
*** nati_ueno has joined #openstack-ceilometer00:41
*** _cjones_ has quit IRC00:55
*** _cjones_ has joined #openstack-ceilometer00:56
*** prad has joined #openstack-ceilometer01:08
*** rwsu has quit IRC01:29
openstackgerritLuis A. Garcia proposed a change to openstack/ceilometer: Allow Messages in NotImplementedErrors  https://review.openstack.org/7806301:49
openstackgerritLuis A. Garcia proposed a change to openstack/ceilometer: Allow individual lazy _() imports  https://review.openstack.org/7082401:49
*** tianst20 has joined #openstack-ceilometer01:52
*** flwang has quit IRC01:54
*** nosnos has joined #openstack-ceilometer01:56
*** marcoemorais has quit IRC01:57
*** xianghui has joined #openstack-ceilometer01:59
openstackgerritLuis A. Garcia proposed a change to openstack/ceilometer: Allow Messages in NotImplementedErrors  https://review.openstack.org/7806302:06
openstackgerritLuis A. Garcia proposed a change to openstack/ceilometer: Allow individual lazy _() imports  https://review.openstack.org/7082402:06
*** prad has quit IRC02:17
*** terriyu has quit IRC02:25
*** _cjones_ has quit IRC02:29
*** terriyu has joined #openstack-ceilometer02:36
*** prad has joined #openstack-ceilometer02:40
*** prad has quit IRC02:40
*** nati_ueno has quit IRC02:43
openstackgerritAdrian Turjak proposed a change to openstack/ceilometer: Adds a state metric to ceilometer  https://review.openstack.org/7519302:46
*** yfujioka has joined #openstack-ceilometer02:46
*** changbl has joined #openstack-ceilometer02:50
*** thomasem has joined #openstack-ceilometer02:59
*** gordc has joined #openstack-ceilometer03:17
*** flwang has joined #openstack-ceilometer03:19
openstackgerritZhiQiang Fan proposed a change to openstack/ceilometer: [WIP] Skip central agent interval_task when keystone fails  https://review.openstack.org/7807903:28
*** raymondr has joined #openstack-ceilometer03:30
*** fnaval has joined #openstack-ceilometer03:48
openstackgerritPiyush Masrani proposed a change to openstack/ceilometer: Changes for networking metrics support for vSphere  https://review.openstack.org/7735504:03
openstackgerritRob Raymond proposed a change to openstack/ceilometer: Replace hard coded WSGI application creation  https://review.openstack.org/7795704:59
*** raymondr has quit IRC05:00
openstackgerritLuis A. Garcia proposed a change to openstack/ceilometer: Allow Messages in NotImplementedErrors  https://review.openstack.org/7806305:01
openstackgerritLuis A. Garcia proposed a change to openstack/ceilometer: Allow individual lazy _() imports  https://review.openstack.org/7082405:01
*** marcoemorais has joined #openstack-ceilometer05:07
*** marcoemorais has quit IRC05:09
*** gordc has quit IRC05:23
openstackgerritJenkins proposed a change to openstack/ceilometer: Imported Translations from Transifex  https://review.openstack.org/7676206:03
*** mihgen has joined #openstack-ceilometer06:13
*** terriyu has quit IRC06:19
*** Yathi has joined #openstack-ceilometer06:25
*** mihgen has quit IRC06:32
openstackgerritAkhil Hingane proposed a change to openstack/ceilometer: VMware vSphere support: Disk rates  https://review.openstack.org/7657106:47
*** Yathi has quit IRC07:01
*** saju_m has joined #openstack-ceilometer07:02
*** _nadya_ has joined #openstack-ceilometer07:03
*** lsmola has joined #openstack-ceilometer07:04
*** promulo_ has joined #openstack-ceilometer07:07
*** promulo has quit IRC07:08
*** taplax has joined #openstack-ceilometer07:15
*** eglynn has joined #openstack-ceilometer07:21
*** ildikov_ has joined #openstack-ceilometer07:21
*** ildikov_ has quit IRC07:21
*** ildikov_ has joined #openstack-ceilometer07:22
eglynnjd__: morning07:23
eglynnjd__: could you take a peek at my responses on https://review.openstack.org/75970 ?07:23
eglynnjd__: ... would be great to sneak that last one over the line before ttx pulls the trigger on the i3 milestone-proposed branch07:23
eglynnjd__: (... presumably this bugfix https://review.openstack.org/77959 also i3-targetted, would be candidate for a master->milestone-proposed backport)07:25
*** _nadya_ has quit IRC07:39
*** nati_ueno has joined #openstack-ceilometer07:47
*** eglynn has quit IRC08:11
*** nosnos has quit IRC08:11
*** AMike has quit IRC08:11
*** openstackgerrit has quit IRC08:11
*** Alexei_987 has quit IRC08:11
*** nprivalova has quit IRC08:11
*** ilyashakhat has quit IRC08:11
*** saju_m has quit IRC08:11
*** llu has quit IRC08:11
*** shakayumi has quit IRC08:11
*** bogdando has quit IRC08:11
*** esheffie1d has quit IRC08:11
*** tasdomas has quit IRC08:11
*** nikhil___ has quit IRC08:11
*** dhellmann_ has quit IRC08:11
*** Kim-Chi-San has quit IRC08:11
*** sdake has quit IRC08:11
*** nati_ueno has quit IRC08:11
*** tianst20 has quit IRC08:11
*** parkerr_ has quit IRC08:11
*** krtaylor has quit IRC08:11
*** sileht has quit IRC08:11
*** ondergetekende has quit IRC08:11
*** sbadia has quit IRC08:11
*** gibi_ has quit IRC08:11
*** taplax has quit IRC08:11
*** promulo_ has quit IRC08:11
*** lsmola has quit IRC08:11
*** annegentle has quit IRC08:11
*** dragondm has quit IRC08:11
*** yjiang5 has quit IRC08:11
*** janisg has quit IRC08:11
*** scroiset has quit IRC08:11
*** jd__ has quit IRC08:11
*** ashestakov has quit IRC08:11
*** nijaba has quit IRC08:11
*** jaypipes has quit IRC08:11
*** Ruetobas has quit IRC08:11
*** lakspace has quit IRC08:11
*** SergeyLukjanov has quit IRC08:11
*** ruhe has quit IRC08:11
*** sdake_ has quit IRC08:11
*** jburkhart has quit IRC08:11
*** kin has quit IRC08:11
*** fc__ has quit IRC08:11
*** hhamalai has quit IRC08:11
*** gpocentek has quit IRC08:11
*** Daviey has quit IRC08:11
*** boris-42 has quit IRC08:11
*** nsaje_ has quit IRC08:11
*** EmilienM has quit IRC08:11
*** zigo has quit IRC08:11
*** Guest34377 has quit IRC08:11
*** ildikov_ has quit IRC08:11
*** flwang has quit IRC08:11
*** ekarlso has quit IRC08:11
*** anteaya has quit IRC08:11
*** apmelton1 has quit IRC08:11
*** eglynn has joined #openstack-ceilometer08:28
*** nosnos has joined #openstack-ceilometer08:28
*** eglynn has quit IRC08:29
*** nosnos has quit IRC08:29
*** nprivalova has joined #openstack-ceilometer09:01
*** AMike has joined #openstack-ceilometer09:01
*** openstackgerrit has joined #openstack-ceilometer09:01
*** nosnos_ has joined #openstack-ceilometer09:01
*** ilyashakhat_ has joined #openstack-ceilometer09:01
*** nati_ueno has joined #openstack-ceilometer09:01
*** ildikov_ has joined #openstack-ceilometer09:01
*** taplax has joined #openstack-ceilometer09:01
*** promulo_ has joined #openstack-ceilometer09:01
*** lsmola has joined #openstack-ceilometer09:01
*** saju_m has joined #openstack-ceilometer09:01
*** flwang has joined #openstack-ceilometer09:01
*** tianst20 has joined #openstack-ceilometer09:01
*** jaypipes has joined #openstack-ceilometer09:01
*** parkerr_ has joined #openstack-ceilometer09:01
*** ondergetekende has joined #openstack-ceilometer09:01
*** Ruetobas has joined #openstack-ceilometer09:01
*** annegentle has joined #openstack-ceilometer09:01
*** krtaylor has joined #openstack-ceilometer09:01
*** ekarlso has joined #openstack-ceilometer09:01
*** bogdando has joined #openstack-ceilometer09:01
*** lakspace has joined #openstack-ceilometer09:01
*** llu has joined #openstack-ceilometer09:01
*** dragondm has joined #openstack-ceilometer09:01
*** jd__ has joined #openstack-ceilometer09:01
*** sileht has joined #openstack-ceilometer09:01
*** boris-42 has joined #openstack-ceilometer09:01
*** SergeyLukjanov has joined #openstack-ceilometer09:01
*** ruhe has joined #openstack-ceilometer09:01
*** sdake_ has joined #openstack-ceilometer09:01
*** jburkhart has joined #openstack-ceilometer09:01
*** nsaje_ has joined #openstack-ceilometer09:01
*** yjiang5 has joined #openstack-ceilometer09:01
*** janisg has joined #openstack-ceilometer09:01
*** scroiset has joined #openstack-ceilometer09:01
*** esheffie1d has joined #openstack-ceilometer09:01
*** tasdomas has joined #openstack-ceilometer09:01
*** nikhil___ has joined #openstack-ceilometer09:01
*** dhellmann_ has joined #openstack-ceilometer09:01
*** sbadia has joined #openstack-ceilometer09:01
*** gibi_ has joined #openstack-ceilometer09:01
*** Kim-Chi-San has joined #openstack-ceilometer09:01
*** ashestakov has joined #openstack-ceilometer09:01
*** sdake has joined #openstack-ceilometer09:01
*** nijaba has joined #openstack-ceilometer09:01
*** EmilienM has joined #openstack-ceilometer09:01
*** kin has joined #openstack-ceilometer09:01
*** fc__ has joined #openstack-ceilometer09:01
*** anteaya has joined #openstack-ceilometer09:01
*** hhamalai has joined #openstack-ceilometer09:01
*** gpocentek has joined #openstack-ceilometer09:01
*** Daviey has joined #openstack-ceilometer09:01
*** zigo has joined #openstack-ceilometer09:01
*** Guest34377 has joined #openstack-ceilometer09:01
*** apmelton1 has joined #openstack-ceilometer09:01
*** raymondr has joined #openstack-ceilometer09:12
*** sayalilunkad has joined #openstack-ceilometer09:14
*** eglynn has joined #openstack-ceilometer09:14
sayalilunkadeglynn: hey!09:14
eglynnsayalilunkad: good morning!09:14
sayalilunkadMorning eglynn !09:14
eglynnsayalilunkad: how goes it with sparklines?09:14
eglynnsayalilunkad: that cardinality support we spoke about before has finally landed09:14
eglynnsayalilunkad: (for sqlalchemy and mongo)09:15
*** flwang has quit IRC09:15
sayalilunkadeglynn: Going fine, want to use sparklines to show the cpu_util meter in ceilometer  but somehow my meters are not seen09:15
sayalilunkadeglynn: Oh has it. Can you send me a link to that patch?09:15
*** yassine has joined #openstack-ceilometer09:16
eglynnsayalilunkad: https://review.openstack.org/#/c/76278/ & https://review.openstack.org/7687909:16
sayalilunkadeglynn: Initially I had the compute meters so I could show sparklines for the instance meter but I recloned and now I can see no meters from the ceilometer cli and in the dashboard I see glance meters09:17
sayalilunkadeglynn: thanks I'll look at it :)09:18
eglynnsayalilunkad: obvious things to check ... if the compute agent running on every comute node?09:18
eglynnsayalilunkad: ... are the logs for those compute agents free of errors?09:18
eglynnsayalilunkad: ... is the notification agent running?09:18
eglynnsayalilunkad: ... when you say you "recloned", do you mean unstack.sh && stack.sh ?09:19
sayalilunkadeglynn: yes notification agent is running09:19
eglynnsayalilunkad: ... if so, unstack will ave blown away your prior instances09:19
*** saju_m has quit IRC09:19
eglynnsayalilunkad: ... do the expected instances show up in "nova list"09:19
eglynnsayalilunkad: ... another common error is a mismatch between the hostname nova associates with instances and the hostname used to contrain the query on the nova-api issued by the ceilo compute agent09:21
eglynnsayalilunkad: ... e.g. one a FQDN, the other a bare hostname09:21
eglynnsayalilunkad: ... the default host should be used in both cases, or overrideen in both cases09:21
openstackgerritRob Raymond proposed a change to openstack/ceilometer: Replace hard coded WSGI application creation  https://review.openstack.org/7795709:21
eglynnsayalilunkad: ... both cases == /etc/nova/nova.conf && /etc/ceilometer/ceilometer.conf09:22
*** sayalilunkad has quit IRC09:22
*** raymondr has quit IRC09:23
eglynnsayalilunkad: ... ^^^ do the above things to check all make sense to you?09:24
*** sayalilunkad has joined #openstack-ceilometer09:24
sayalilunkadeglynn: sorry got disconnected09:25
eglynnsayalilunkad: I see you've dropped off and rejoined IRC09:25
eglynnsayalilunkad: see the tail on http://eavesdrop.openstack.org/irclogs/%23openstack-ceilometer/%23openstack-ceilometer.2014-03-05.log09:25
sayalilunkadAnd by reclone I don't mean unstack. I recloned the repo in a new VM09:26
sayalilunkadeglynn: thanks for the logs09:26
sayalilunkadeglynn: and nope nova list also gives me an empty output09:27
eglynnsayalilunkad: well that would seem to suggest that there are no instances ;)09:27
eglynnsayalilunkad: ... did you run unstack.sh when recloning?09:27
eglynnsayalilunkad: ... that will remove all running instances09:28
sayalilunkadeglynn: haha no I can the instances on the dashboard right09:28
sayalilunkadI cloned then followed the normal procedure and then launched an instance09:29
eglynnsayalilunkad: ... surely the dashboard and nova list are hitting the *same* nova API? unless you're running nova list with the wrong credentials09:29
eglynnsayalilunkad: ... i.e. no admin user where the instances are owned by another user09:29
eglynnsayalilunkad: ... what is "normal procedure"?09:30
eglynnsayalilunkad: ... are you using devstack?09:30
eglynnsayalilunkad: ... if so, how did you set up the OS_* env vars?09:30
eglynnsayalilunkad: ... $ . devstack/openrc admin admin09:30
eglynnor:09:30
eglynn$ . devstack/openrc09:30
*** xianghui has quit IRC09:31
eglynnsayalilunkad: ^^^ the latter will set up creds for a non-admin user (demo)09:31
sayalilunkadeglynn: yes it's devstack. So clone then stack.sh09:31
eglynnsayalilunkad: ... then devstack/openrc?09:32
sayalilunkadi did devstack/openrc09:32
sayalilunkadAnd I have an instance for both default users- admin and demo09:33
openstackgerritMehdi Abaakouk proposed a change to openstack/ceilometer: Consume notif. from multiple message bus  https://review.openstack.org/7761209:33
openstackgerritMehdi Abaakouk proposed a change to openstack/ceilometer: Use NotificationPlugin as an oslo.msg endpoint  https://review.openstack.org/7761109:33
openstackgerritMehdi Abaakouk proposed a change to openstack/ceilometer: Use oslo.msg retry API in rpc publisher  https://review.openstack.org/7784509:33
openstackgerritMehdi Abaakouk proposed a change to openstack/ceilometer: Replace oslo.rpc by oslo.messaging  https://review.openstack.org/5745709:33
eglynnsayalilunkad: ... ok, how do you know that if "nova list" gives empty output?09:34
eglynnsayalilunkad: (... just on the basis of what the dashboard is telling you?)09:34
sayalilunkadeglynn: It shows me empty tables on the console09:35
*** saju_m has joined #openstack-ceilometer09:35
eglynnsayalilunkad: what is the state of those phantom instances reported by the dashboard?09:35
eglynnsayalilunkad: ... active or error?09:35
sayalilunkadeglynn: active09:36
sayalilunkad:(09:36
eglynnsayalilunkad: what user/tenant have you used to log into the dashboard?09:37
eglynnsayalilunkad: are you sure the dashboard is pointing to the *same* devstack?09:37
sayalilunkadeglynn: admin to login and yes there is only one devstack so it is definitely to that only09:38
eglynnsayalilunkad: to check quickly ... on devstack node: screen -r ... naviagte to nova-api screen ... ctrl+c ... refresh dashboard09:39
eglynnsayalilunkad: you should see a failure in the dashboard now that nova-api is no longer running09:40
eglynnsayalilunkad: ... do you see a failure?09:40
sayalilunkadeglynn: just a sec09:42
*** Alexei_987 has joined #openstack-ceilometer09:43
sayalilunkadeglynn: no error that I can see09:50
eglynnsayalilunkad: no error on the dashboard when you refresh the instances page even though the nova-api service is not running?09:50
sayalilunkadeglynn: ok my system crashed09:52
sayalilunkadeglynn: I have to start a new session09:52
eglynnsayalilunkad: what do you mean by your "session"?09:54
eglynnsayalilunkad: ... dashboard session or entire VM running devstack?09:55
sayalilunkadeglynn: dashboard session09:55
eglynnsayalilunkad: ... so have you restarted it? ... are the phantom instances still visible? ... or is the restarted dashboard reporting an issue?09:56
sayalilunkadeglynn: I could not rejoin the session for some reason so I had to unstack and now running stack.sh09:57
sayalilunkadeglynn: so will have to create the instances again09:58
*** sayalilunkad has quit IRC10:19
*** admin0 has joined #openstack-ceilometer10:22
admin0Hey all10:23
*** admin0 has quit IRC10:25
*** admin0 has joined #openstack-ceilometer11:18
*** admin0 has quit IRC11:53
*** gibi has joined #openstack-ceilometer11:54
*** sbadia_ has joined #openstack-ceilometer11:55
*** nati_ueno has quit IRC11:56
*** sbadia has quit IRC11:56
*** gibi_ has quit IRC11:56
*** ondergetekende has quit IRC11:57
*** ondergetekende has joined #openstack-ceilometer11:57
*** sbadia_ is now known as sbadia11:57
*** admin0 has joined #openstack-ceilometer12:09
openstackgerritYuuichi Fujioka proposed a change to openstack/ceilometer: Implements monitoring-network-from-opendaylight  https://review.openstack.org/6389012:14
nprivalovaeglynn, sileht, jd__: hi! do you know anything about mysql-specific in ceilometer? I'm trying to make tests in tempest work, it is ok for postgress and fail for mysql. Do you know anything about error "This application has not enabled MySQL traditional mode, which means silent data corruption may occur"?12:19
eglynnnprivalova: have you seen this thread http://www.mail-archive.com/openstack-dev@lists.openstack.org/msg18019.html12:22
nprivalovaeglynn: just fount it, looking12:22
nprivalovaeglynn: is anyone going to sync oslo.db? Actually this WARN is the last idea why notifications may be lost :( cannot reproduce this in my env12:28
eglynnnprivalova: I don't think a sync with oslo.db is on the cards for icehouse, but for juno yes12:28
*** admin0 has quit IRC12:32
*** admin0 has joined #openstack-ceilometer12:32
admin0Hi eglynn12:42
eglynnadmin0: 'sup?12:43
admin0i have been looking into the codes ( also new in python )  .. but was not able to find the exact function that is getting the data from the compute :)12:44
eglynnadmin0: here's the interaction with the hypervisor12:46
eglynn... https://github.com/openstack/ceilometer/blob/master/ceilometer/compute/virt/libvirt/inspector.py12:46
*** promulo has joined #openstack-ceilometer12:52
*** promulo_ has quit IRC12:53
openstackgerritYuuichi Fujioka proposed a change to openstack/ceilometer: Implements monitoring-network-from-opendaylight  https://review.openstack.org/6389013:03
*** tongli has joined #openstack-ceilometer13:37
tongli@eglynn, ping.13:37
eglynntongli: just heading out the door, is it urgent?13:38
tongli@eglynn, just that patch we were talking about.13:38
eglynntongli: new version?13:38
tongli@eglynn, made some changes, and hope you give it another look.13:38
eglynntongli: ... I though that had already been deferred to Juno?13:38
eglynn... /me is prioritize icehouse-targetted work13:39
eglynn... but will try to fit it in later on today13:39
tongli@eglynn, please.13:39
*** nosnos_ has quit IRC13:45
*** eglynn is now known as eglynn-lunch13:45
*** krtaylor has quit IRC13:46
*** nosnos has joined #openstack-ceilometer13:48
*** nosnos has quit IRC13:49
*** nosnos has joined #openstack-ceilometer13:50
*** julim has joined #openstack-ceilometer13:52
*** nosnos has quit IRC13:54
*** eglynn-lunch has quit IRC13:54
*** flwang has joined #openstack-ceilometer14:07
*** mihgen has joined #openstack-ceilometer14:10
*** dhellmann_ is now known as dhellmann14:14
*** saju_m has quit IRC14:15
*** stzilli has joined #openstack-ceilometer14:17
openstackgerritMehdi Abaakouk proposed a change to openstack/ceilometer: Consume notif. from multiple message bus  https://review.openstack.org/7761214:21
openstackgerritMehdi Abaakouk proposed a change to openstack/ceilometer: Use NotificationPlugin as an oslo.msg endpoint  https://review.openstack.org/7761114:21
openstackgerritMehdi Abaakouk proposed a change to openstack/ceilometer: Use oslo.msg retry API in rpc publisher  https://review.openstack.org/7784514:21
openstackgerritMehdi Abaakouk proposed a change to openstack/ceilometer: Replace oslo.rpc by oslo.messaging  https://review.openstack.org/5745714:21
*** parkerr_ has quit IRC14:28
*** ildikov_ has quit IRC14:38
*** raymondr has joined #openstack-ceilometer14:39
*** admin0 has quit IRC14:41
*** parkerr_ has joined #openstack-ceilometer14:42
openstackgerritMehdi Abaakouk proposed a change to openstack/ceilometer: Consume notif. from multiple message bus  https://review.openstack.org/7761214:46
openstackgerritMehdi Abaakouk proposed a change to openstack/ceilometer: Use NotificationPlugin as an oslo.msg endpoint  https://review.openstack.org/7761114:46
openstackgerritMehdi Abaakouk proposed a change to openstack/ceilometer: Use oslo.msg retry API in rpc publisher  https://review.openstack.org/7784514:46
openstackgerritMehdi Abaakouk proposed a change to openstack/ceilometer: Replace oslo.rpc by oslo.messaging  https://review.openstack.org/5745714:46
*** eglynn has joined #openstack-ceilometer14:49
*** thomasem has joined #openstack-ceilometer14:52
openstackgerritLuis A. Garcia proposed a change to openstack/ceilometer: Don't override the lazy _()  https://review.openstack.org/7082414:53
*** admin0 has joined #openstack-ceilometer14:53
*** gordc has joined #openstack-ceilometer14:54
*** ildikov_ has joined #openstack-ceilometer14:54
openstackgerritNadya Privalova proposed a change to openstack/ceilometer: Make Resource able to store several meters  https://review.openstack.org/7824414:54
*** julienvey_ has joined #openstack-ceilometer15:01
*** julienvey_ has quit IRC15:01
*** yfujioka has quit IRC15:02
*** raymondr has quit IRC15:03
*** raymondr_ has joined #openstack-ceilometer15:05
*** raymondr_ has quit IRC15:10
*** fnaval has quit IRC15:11
*** raymondr has joined #openstack-ceilometer15:37
*** fnaval has joined #openstack-ceilometer15:38
*** raymondr has quit IRC15:41
*** ryanpetrello has joined #openstack-ceilometer15:43
*** taplax has quit IRC15:43
*** Ruetobas has quit IRC16:01
*** Ruetobas has joined #openstack-ceilometer16:03
*** jaypipes has quit IRC16:03
*** jergerber has joined #openstack-ceilometer16:04
*** Ruetobas has quit IRC16:09
*** gordc1 has joined #openstack-ceilometer16:09
*** Alexei_987 has quit IRC16:10
*** gordc has quit IRC16:10
*** Ruetobas has joined #openstack-ceilometer16:14
*** gordc1 is now known as gordc16:14
*** jmckind has joined #openstack-ceilometer16:15
*** stzilli has quit IRC16:19
*** promulo_ has joined #openstack-ceilometer16:20
*** promulo has quit IRC16:21
openstackgerritEoghan Glynn proposed a change to openstack/ceilometer: Ensure user metadata mapped for instance notifications  https://review.openstack.org/7795916:25
eglynnsileht, ildikov_: could you take another look at that instance user_metadata in notificatons bugfix ^^^16:26
ildikov_eglynn: sure, I was just about to do that :)16:26
eglynnildikov_: thanking you! :)16:27
nprivalovagordc: hi! are you here?16:27
gordcnprivalova: yep.16:27
nprivalovagordc: I'd like to discuss https://review.openstack.org/#/c/6567116:28
nprivalovagordc: it's hbase part16:28
gordcnprivalova: sure.16:28
nprivalovagordc: I see that we remove meters from resources at all. but take a look on get_meter16:29
nprivalovagordc: *get_meters16:29
nprivalovagordc:  if m.startswith('f:m_'):16:29
gordcnprivalova: you mean how we're building meters from resource_table?16:30
nprivalovagordc: we scan resource table in get_meters. And I do not see how f:m_ may appear in resource. It cannot16:30
gordcthat is weird we get meters from resource when we have a meter table...16:31
nprivalovagordc: I agree16:33
nprivalovagordc: get samples from meters and meters from resource16:34
nprivalovagordc: but it's too late to fix this behavior now. For HBase it means to rewrite it again. I've just finish the first iteration :)16:34
gordcnprivalova: :) i keep forgetting the reason for that.16:35
gordcnprivalova: do you mean you rewrote entire hbase impl for Juno?16:35
ildikov_eglynn: sorry, I found two typos :)16:36
nprivalovagordc: in alarms patch I did a lot of refactoring16:36
gordcnprivalova: yeah. i saw. our backends could use some refactoring.16:36
ildikov_eglynn: that is what I referred that I saw earlier, I just was not 100% sure, that this is still the actual format of that notification, thanks for investigating it16:36
gordcnprivalova: i see '# store meters with prefix "m_"' comment in record_metering_data function16:36
*** admin0 has quit IRC16:37
eglynnildikov_: thanks for the eagle eyes :)16:37
eglynnildikov_: ... /me should have gone to specsavers16:37
eglynnildikov_: http://www.youtube.com/watch?v=jq3H1QZlct016:37
nprivalovagordc: I don't understand why tests passed successful after removing meter from resource in your prev patch. Do you agree with me that we cannot get_meters after removing?16:38
gordcnprivalova: you mean the patch that removed it? i think our tests are testing the model and not the data in backend specifically.16:39
gordcnprivalova: from what i understand, it's storing meter data but it's never being used.16:39
ildikov_eglynn: maybe the situation is not that bad ;)16:40
gordceglynn: is that an irish commerical?16:41
*** lsmola has quit IRC16:41
eglynngordc: Scotish I think16:41
nprivalovagordc: yes, I meant 65671. We have api tests that are rather ok to see such problems. mater info if Resource is used in get_meters16:41
*** mihgen has quit IRC16:41
nprivalovagordc: so I'm trying to say that we cannot remove meter data from Resource in HBase16:42
nprivalovagordc: current impl doesn't allow it16:42
gordceglynn: oh should've read the comments... :) don't have specsaver on this side of the atlantic.16:44
ildikov_nprivalova, gordc: it seemed to me that resource_table.put() does not check the data against the Resource model16:44
nprivalovaildikov_: sure. in HBase we don't care about 'model'16:45
ildikov_nprivalova, gordc: and then get_meters uses this resource_table16:45
ildikov_nprivalova: so I assume it is the explanation of the tests are working16:46
gordcnprivalova: do we have bad tests or a test gap (actually, let me read your bug)16:46
eglynngordc: yeah there's a whole series of those funny adverts poking fun as us visually-challenged folks ;)16:47
eglynn... here's another http://www.funnyplace.org/stream/specsavers-opticians-diamond-ring-19347/16:47
nprivalovagordc: I'll checkout your patch and will take a look on tests16:47
eglynn... it's kinda become one of those pop-cultural memes on this side of the atlantic16:48
openstackgerritEoghan Glynn proposed a change to openstack/ceilometer: Ensure user metadata mapped for instance notifications  https://review.openstack.org/7795916:48
gordceglynn: lol... our eyewear commericals aren't as clever... just people dancing around with glasses on.16:48
gordcnprivalova: it's not my patch ;)16:49
nprivalovagordc: oh, sorry:)16:49
nprivalovagordc: it's eglynn's creation :)16:50
gordcnprivalova: :) i did approve it so if it's broken my bad.lol16:50
nprivalovagordc: I've approved too lol16:51
eglynnnprivalova: sorry, did I break something again? d'oh!16:51
*** _cjones_ has joined #openstack-ceilometer16:52
ildikov_eglynn: why haven't you said that you've copied that comment with Retrived, which should be Retrieved in line 58 too??? ;)16:52
eglynnildikov_: right I'll get my coat ... defo time for a visit to specsavers ;)16:53
*** raymondr has joined #openstack-ceilometer16:54
ildikov_eglynn: mine too, I've recognized it for the second time :)16:54
ildikov_eglynn: and I've already have glasses...16:55
nprivalovaeglynn: actually I made a bug in hbase refactoring and gordc found that the code that I'm fixing should not exist because of your patch :)16:55
openstackgerritEoghan Glynn proposed a change to openstack/ceilometer: Ensure user metadata mapped for instance notifications  https://review.openstack.org/7795916:56
nprivalovaeglynn: but It cannot be removed. But you managed with it :) I'm wondering about tests, looking into it16:56
openstackgerritRob Raymond proposed a change to openstack/ceilometer: Replace hard coded WSGI application creation  https://review.openstack.org/7795716:58
*** jmckind has quit IRC16:58
nprivalovaeglynn: gordc: May I ask you about notifications?17:00
eglynnsure17:01
ildikov_eglynn: maybe I'm too tired, but in the test in line 735 it seems that there is no source of that 'c'17:02
nprivalovaeglynn: gordc: we have big problems on tempest. Notification tests passed on postgress-full but fails on dsvm-full. And we cannot repro it. Today we even deployed gating-like env17:02
*** rwsu has joined #openstack-ceilometer17:03
eglynnnprivalova: does it smell like a timing issue?17:05
eglynnnprivalova: ... e.g. do we wait a sufficiently long time for samples derived from notifications to appear?17:05
nprivalovaeglynn: gordc: I've looked through all the logs and do not see any important differences. I'm going to send email to all nova, tempest guys because cannot see any futher steps17:05
openstackgerritEoghan Glynn proposed a change to openstack/ceilometer: Ensure user metadata mapped for instance notifications  https://review.openstack.org/7795917:06
nprivalovaeglynn: no. we wait untill instance becomes active and do not see notificcations17:06
nprivalovaeglynn: I tried to wait for 200 sec but unsuccessful17:06
eglynnnprivalova: that is very strange ... so clearly AMQP is not backed up if the run_instance RPC messages are getting thru to cause the instance to be spun up17:07
nprivalovaeglynn: for postgress I saw even 2 sec notification. The longest one is 40 sec17:07
gordcnprivalova: i'm not familiar with what differences in postgres-full and dsvm-full are.17:07
eglynnnprivalova: in the ceilo notification agent logs, are you seeing the events appearing?17:07
*** jdob has joined #openstack-ceilometer17:08
nprivalovagordc: only backends are different. And nova-metadata server is up for postgress-case17:08
eglynnnprivalova: i.e. from this line https://github.com/openstack/ceilometer/blob/master/ceilometer/notification.py#12917:08
eglynn(asuming debug=True in ceilometer.conf)17:09
eglynnnprivalova: ... I wondering if we're possibly seeing the notifications but not generating samples from them17:09
nprivalovaeglynn: it's AUDIT. and I see only ceilometer.pipeline [-] Pipeline meter_pipeline: Published samples17:10
eglynnnprivalova: ... we had an issue a while back with incoming notification failing the message signature verification for spurious reasons17:10
eglynnnprivalova: ... in that case the message is discarded17:10
eglynnnprivalova: ... so never show up via the API service17:11
eglynnnprivalova: ... in that case it only applied to nova scheduler notifications17:11
*** raymondr has quit IRC17:11
nprivalovaeglynn: yes, I cannot find out where is the problem: nova or ceilo. Because in nova logs there is no message 'notification is sent', no info about it. Ceilo logs are not informative17:12
*** raymondr has joined #openstack-ceilometer17:12
eglynnnprivalova: ... in that case I mentioned, ceilo was at fault https://bugs.launchpad.net/ceilometer/+bug/126225517:12
nprivalovaeglynn: maybe change the level. IT should be commit to devstack I guess17:12
eglynnnprivalova: yep, either that or just change that individual log message17:13
eglynnnprivalova: ... to avoid a deluge of debug logging17:13
nprivalovaeglynn: but the fix is released. do you think that issue still may occur?17:14
eglynnnprivalova: ... we could temporarily change the severity of  https://github.com/openstack/ceilometer/blob/master/ceilometer/notification.py#129 to audit17:14
eglynnnprivalova: ... no I was just thinking that a similar issue might have occured, perhaps the original fix wasn't complete17:15
gordceglynn: that makes sense. that or submit a devstack patch to switch debug on with no intention of actually merging... and let the gates run to see logs17:16
nprivalovaeglynn: the problem that it's tempest env. I cannot repro it. And I cannot just go to conf and change anything without commit17:16
*** raymondr has quit IRC17:16
*** raymondr has joined #openstack-ceilometer17:16
nprivalovagordc: eglynn: ok, I will come back to you after I get debug logs for ceilo :)17:17
eglynngordc: yeah that would work17:17
gordcnprivalova: if you submit patch that changes log level and say do no merge, could you not take a look at logs that way? (you'll have to wait a long time to get the results though)17:18
eglynnnprivalova: actually scratch the idea of a message signature verification failure17:18
eglynnnprivalova: ... as you would have seen a warning in the logs from https://github.com/openstack/ceilometer/blob/master/ceilometer/dispatcher/database.py#L7217:18
nprivalovagordc: no. because I need to run 2 patches again. With changed logging and new notification test17:19
gordcnprivalova: ah i see. two separate items.17:20
nprivalovaeglynn: in screen-ceilometer-anotification ?17:20
eglynnnprivalova: nope in the collector logs or screen17:20
eglynnnprivalova: ... as the verification is done just before the metering message is persisted away to the DB17:21
nprivalovaeglynn: I see "dropping sample with no predecessor: (<ceilometer.sample.Sample object at 0x35b2c90>,)". why I missed that :(?17:21
eglynnnprivalova: that one is OK is you only see one of them per instance17:23
eglynnnprivalova: ... it basically just an artifact of the way we derive cpu_util etc. from cpu17:23
nprivalovaeglynn: I'll check that. yes, postgress' logs contains the same17:23
eglynnnprivalova: ... need at to have seen at least two samples for the instance before we can calculate a rate17:23
eglynnnprivalova: ... and its related to pollster-driver samples in any case17:24
nprivalovaeglynn: that's sad17:24
eglynn:(17:24
nprivalovaeglynn: btw collector log is full of "BaseException.message has been deprecated as of Python 2.6"17:25
nprivalovaeglynn: and nothing more17:25
gordcnprivalova: that's a oslo.db code issue.17:25
gordcnprivalova: https://github.com/openstack/oslo-incubator/blob/master/openstack/common/db/sqlalchemy/session.py#L37817:26
*** HT_Sergio has joined #openstack-ceilometer17:27
HT_Sergiohey all17:27
HT_Sergiowhen we get samples from ceilometer, can we always expect them to be in chronological order?17:27
nprivalovaok. I'll try to find friends in infra and commit DEBUG lvl for ceilometer :)17:27
eglynnnprivalova: cool17:27
nprivalovaeglynn: gordc: thanks for discussion!17:28
eglynnnprivalova: np!17:28
HT_Sergiosorry for intruding :(17:28
gordcHT_Sergio: get them using the client?17:28
HT_Sergiogordc: yes17:28
HT_Sergioceilometer.samples.list(...) will always return samples sorted by timestamp?17:29
gordcHT_Sergio: are you asking because you're not sure what you're seeing is correct or not? :)17:30
gordcHT_Sergio: i'd have to check... i vaguely remember there being some sort logic... not sure if it's on timestamp though.17:30
HT_Sergiogordc: It seems to be sorted, but I'm fairly certain I got samples out of order before17:31
HT_Sergiogordc: Wanted to ask before I make assumptions17:31
*** openstackgerrit has quit IRC17:32
HT_SergioI'm not very familiar with ceilometer code yet :p17:33
gordcHT_Sergio: not familiar myself either. :)17:35
HT_Sergiogordc: any chance you can point me in the right direction?17:36
gordcHT_Sergio: based on code, it looks like it's sorting by first element (so resource_id)17:36
HT_Sergiogordc: where do you see this?17:36
gordchttps://github.com/openstack/python-ceilometerclient/blob/master/ceilometerclient/v2/shell.py#L9117:36
*** mihgen has joined #openstack-ceilometer17:37
HT_Sergiogordc: Oh I'm not using the CLI lol17:38
gordcHT_Sergio: err.. actually it might be sorting on nothing17:38
HT_SergioI was asking if the API server sorted them before sending17:38
gordcHT_Sergio: oh. nm17:38
HT_Sergiogordc: lol sorry I should have been more descript17:38
gordcHT_Sergio: i'm pretty sure there's no ordering by default. you could use complexquery to order if you're working against master17:41
HT_Sergiogordc: Thanks!17:41
gordcHT_Sergio: np, the api code is here: https://github.com/openstack/ceilometer/blob/master/ceilometer/api/controllers/v2.py#L76017:42
gordcHT_Sergio: https://github.com/openstack/ceilometer/blob/master/ceilometer/storage/impl_sqlalchemy.py#L61917:42
HT_Sergiogordc: ah, this is what I was looking for. Thank you!17:44
gordcnp17:44
*** marcoemorais has joined #openstack-ceilometer17:58
*** eglynn has quit IRC18:00
*** nati_ueno has joined #openstack-ceilometer18:02
*** jmckind has joined #openstack-ceilometer18:03
*** fnaval has quit IRC18:06
*** raymondr has quit IRC18:09
*** fnaval has joined #openstack-ceilometer18:10
*** marcoemorais has quit IRC18:12
*** raymondr has joined #openstack-ceilometer18:23
*** openstackgerrit has joined #openstack-ceilometer18:26
*** raymondr has quit IRC18:27
*** tong_ has joined #openstack-ceilometer18:31
*** tongli has quit IRC18:33
*** ildikov_ has quit IRC18:38
*** raymondr has joined #openstack-ceilometer18:42
*** raymondr_ has joined #openstack-ceilometer18:44
*** raymondr has quit IRC18:46
*** raymondr_ has quit IRC18:48
openstackgerritUros Jovanovic proposed a change to openstack/python-ceilometerclient: Adds alarm time constraint support to ceilometer CLI  https://review.openstack.org/7835618:50
*** raymondr has joined #openstack-ceilometer18:53
*** raymondr has quit IRC18:58
*** admin0 has joined #openstack-ceilometer18:58
*** raymondr has joined #openstack-ceilometer19:02
*** openstackgerrit has quit IRC19:02
*** nati_uen_ has joined #openstack-ceilometer19:03
admin0hey all .. again :)19:05
*** nati_ueno has quit IRC19:06
*** raymondr has quit IRC19:07
*** packet has joined #openstack-ceilometer19:09
*** raymondr has joined #openstack-ceilometer19:12
admin0hi all .. if I want to try ceilometer .. what is the best option to get messages from cloudstack .. does a vagrant based or soothing like devstack work ?19:12
*** _nadya_ has joined #openstack-ceilometer19:13
*** admin0 has quit IRC19:16
*** raymondr has quit IRC19:16
*** raymondr has joined #openstack-ceilometer19:21
*** raymondr_ has joined #openstack-ceilometer19:24
*** raymondr has quit IRC19:26
*** _nadya_ has quit IRC19:36
*** ildikov_ has joined #openstack-ceilometer19:40
*** admin0 has joined #openstack-ceilometer19:41
*** jmckind has quit IRC19:44
*** mihgen has quit IRC19:49
*** _nadya_ has joined #openstack-ceilometer19:51
*** janisg has quit IRC19:53
*** dmsimard has joined #openstack-ceilometer19:58
dmsimardHi guys19:58
dmsimardCeilometer with swift question.. I have ceilometer in my proxy-server pipeline and I get meters for bytes in, bytes out and API calls. There seems to be more meters than that - something to do with a pollster ? Am I missing something ? Ref: http://docs.openstack.org/developer/ceilometer/measurements.html#object-storage-swift19:59
*** janisg has joined #openstack-ceilometer19:59
*** _nadya_ has quit IRC20:05
*** admin0 has quit IRC20:12
*** ddutta has joined #openstack-ceilometer20:14
*** promulo_ has quit IRC20:24
*** dmsimard has quit IRC20:26
*** promulo has joined #openstack-ceilometer20:28
*** mihgen has joined #openstack-ceilometer20:31
*** admin0 has joined #openstack-ceilometer20:38
*** admin0 has quit IRC20:40
*** admin0 has joined #openstack-ceilometer20:40
*** admin0 has quit IRC20:44
*** admin0 has joined #openstack-ceilometer20:48
*** Yathi has joined #openstack-ceilometer20:51
*** admin0 has quit IRC20:51
*** yassine has quit IRC21:02
*** yassine has joined #openstack-ceilometer21:03
*** thomasem_ has joined #openstack-ceilometer21:11
*** admin0 has joined #openstack-ceilometer21:12
*** thomasem has quit IRC21:12
*** admin0 has quit IRC21:15
*** mikitenko has joined #openstack-ceilometer21:16
*** yassine has quit IRC21:17
*** yassine has joined #openstack-ceilometer21:18
*** admin0 has joined #openstack-ceilometer21:18
*** yassine has quit IRC21:24
*** admin0 has quit IRC21:24
*** admin0 has joined #openstack-ceilometer21:25
*** yassine has joined #openstack-ceilometer21:25
*** admin0 has quit IRC21:28
*** admin0 has joined #openstack-ceilometer21:33
*** admin0 has quit IRC21:37
*** admin0 has joined #openstack-ceilometer21:38
*** thomasem_ has quit IRC21:55
*** admin0 has quit IRC22:01
*** mihgen has quit IRC22:01
*** admin0 has joined #openstack-ceilometer22:03
*** thomasem_ has joined #openstack-ceilometer22:04
*** thomasm_ has joined #openstack-ceilometer22:06
*** thomasem_ has quit IRC22:09
*** tong_ has quit IRC22:30
*** mihgen has joined #openstack-ceilometer22:36
*** thomasem has joined #openstack-ceilometer22:46
*** thomasem has quit IRC22:49
*** thomasm_ has quit IRC22:50
*** Yathi has quit IRC22:51
*** admin0 has quit IRC22:55
*** ddutta has quit IRC22:58
*** gordc has quit IRC23:00
*** Yathi has joined #openstack-ceilometer23:02
*** openstack has joined #openstack-ceilometer23:03
*** mikitenko has quit IRC23:04
*** raymondr_ has quit IRC23:08
*** ryanpetrello has quit IRC23:11
*** julim has quit IRC23:13
*** ryanpetrello has joined #openstack-ceilometer23:15
*** ryanpetrello has quit IRC23:15
*** jergerber has quit IRC23:24
*** ddutta has joined #openstack-ceilometer23:35
*** yfujioka has joined #openstack-ceilometer23:38
*** ddutta has quit IRC23:50
*** ddutta has joined #openstack-ceilometer23:58

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