| *** claudiub has quit IRC | 00:04 | |
| *** ddieterly has quit IRC | 00:04 | |
| *** eglynn has joined #openstack-ceilometer | 00:12 | |
| *** boris-42 has quit IRC | 00:12 | |
| *** sbfox has quit IRC | 00:27 | |
| *** nellysmitt has joined #openstack-ceilometer | 00:39 | |
| *** eglynn has quit IRC | 00:40 | |
| *** nellysmitt has quit IRC | 00:43 | |
| *** jaypipes has quit IRC | 00:55 | |
| *** amalagon has quit IRC | 00:57 | |
| *** itisit has quit IRC | 01:02 | |
| *** _cjones_ has quit IRC | 01:08 | |
| openstackgerrit | liusheng proposed openstack/ceilometer-specs: Support pagination for ceilometer APIs https://review.openstack.org/105343 | 01:14 |
|---|---|---|
| openstackgerrit | liusheng proposed openstack/ceilometer-specs: Add alarm severity https://review.openstack.org/100752 | 01:17 |
| *** promulo__ has joined #openstack-ceilometer | 01:18 | |
| *** promulo has quit IRC | 01:22 | |
| *** shakamunyi has quit IRC | 01:28 | |
| *** alexpilotti has joined #openstack-ceilometer | 01:43 | |
| *** ryanpetrello has joined #openstack-ceilometer | 01:45 | |
| *** lhcheng has quit IRC | 01:47 | |
| *** Yanyanhu has joined #openstack-ceilometer | 02:01 | |
| *** alexpilotti has quit IRC | 02:06 | |
| *** ryanpetrello has quit IRC | 02:10 | |
| *** ddieterly has joined #openstack-ceilometer | 02:11 | |
| openstackgerrit | liusheng proposed openstack/python-ceilometerclient: Fix Bad Request responses from Ceilometer APIs https://review.openstack.org/140205 | 02:20 |
| *** julim has quit IRC | 02:27 | |
| *** fnaval has quit IRC | 02:30 | |
| *** fabiog has quit IRC | 02:34 | |
| *** nellysmitt has joined #openstack-ceilometer | 02:40 | |
| *** nellysmitt has quit IRC | 02:44 | |
| *** fnaval has joined #openstack-ceilometer | 02:46 | |
| *** david-lyle is now known as david-lyle_afk | 03:24 | |
| *** ryanpetrello has joined #openstack-ceilometer | 03:27 | |
| *** jaypipes has joined #openstack-ceilometer | 03:32 | |
| *** ViswaV has quit IRC | 03:34 | |
| *** thomasem has joined #openstack-ceilometer | 03:35 | |
| *** ViswaV has joined #openstack-ceilometer | 03:40 | |
| *** deepthi has joined #openstack-ceilometer | 03:41 | |
| *** ryanpetrello has quit IRC | 04:02 | |
| *** thomasem has quit IRC | 04:07 | |
| *** ddieterly has quit IRC | 04:14 | |
| *** swamireddy has joined #openstack-ceilometer | 04:21 | |
| *** nellysmitt has joined #openstack-ceilometer | 04:40 | |
| *** sbfox has joined #openstack-ceilometer | 04:42 | |
| *** ddieterly has joined #openstack-ceilometer | 04:45 | |
| *** nellysmitt has quit IRC | 04:45 | |
| *** ddieterly has quit IRC | 04:49 | |
| *** ajayaa has joined #openstack-ceilometer | 05:05 | |
| *** yatin has joined #openstack-ceilometer | 05:09 | |
| *** ddieterly has joined #openstack-ceilometer | 05:11 | |
| openstackgerrit | liusheng proposed openstack/ceilometer: Add support for dynamic meters https://review.openstack.org/140232 | 05:15 |
| *** ddieterly has quit IRC | 05:16 | |
| *** harlowja is now known as harlowja_away | 05:16 | |
| *** sbfox has quit IRC | 05:20 | |
| *** ViswaV_ has joined #openstack-ceilometer | 05:25 | |
| *** ViswaV has quit IRC | 05:26 | |
| *** sbfox has joined #openstack-ceilometer | 05:38 | |
| *** promulo__ has quit IRC | 05:45 | |
| *** deepthi has quit IRC | 05:49 | |
| *** fnaval has quit IRC | 05:55 | |
| *** fnaval has joined #openstack-ceilometer | 05:56 | |
| *** fnaval has quit IRC | 06:00 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/ceilometer: Imported Translations from Transifex https://review.openstack.org/139527 | 06:09 |
| *** ajayaa has quit IRC | 06:10 | |
| *** yatin has quit IRC | 06:11 | |
| *** ddieterly has joined #openstack-ceilometer | 06:11 | |
| *** Ephur has quit IRC | 06:13 | |
| *** Ephur has joined #openstack-ceilometer | 06:13 | |
| *** ddieterly has quit IRC | 06:16 | |
| openstackgerrit | liusheng proposed openstack/ceilometer: Add support for dynamic meters https://review.openstack.org/140232 | 06:21 |
| openstackgerrit | liusheng proposed openstack/ceilometer-specs: Add alarm severity https://review.openstack.org/100752 | 06:21 |
| *** ildikov_ is now known as ildikov_afk | 06:28 | |
| *** cmyster has quit IRC | 06:29 | |
| *** cmyster has joined #openstack-ceilometer | 06:31 | |
| *** cmyster has joined #openstack-ceilometer | 06:31 | |
| *** yatin has joined #openstack-ceilometer | 06:33 | |
| *** nellysmitt has joined #openstack-ceilometer | 06:42 | |
| *** russellb has quit IRC | 06:43 | |
| *** russellb has joined #openstack-ceilometer | 06:44 | |
| *** nellysmitt has quit IRC | 06:46 | |
| *** ViswaV_ has quit IRC | 06:48 | |
| *** ViswaV has joined #openstack-ceilometer | 06:48 | |
| *** sudipto has joined #openstack-ceilometer | 06:51 | |
| *** eglynn has joined #openstack-ceilometer | 06:53 | |
| *** Longgeek has joined #openstack-ceilometer | 06:57 | |
| *** eglynn has quit IRC | 06:59 | |
| *** ddieterly has joined #openstack-ceilometer | 07:11 | |
| *** k4n0 has joined #openstack-ceilometer | 07:15 | |
| *** ddieterly has quit IRC | 07:16 | |
| *** drjones has joined #openstack-ceilometer | 07:21 | |
| *** nellysmitt has joined #openstack-ceilometer | 07:23 | |
| *** drjones has quit IRC | 07:26 | |
| *** nellysmitt has quit IRC | 07:28 | |
| *** NellyK has joined #openstack-ceilometer | 07:30 | |
| openstackgerrit | Lena Novokshonova proposed openstack/ceilometer: Add docs about volume/snapshot measurements https://review.openstack.org/139969 | 07:33 |
| *** NellyK is now known as nellysmitt | 07:36 | |
| *** eglynn has joined #openstack-ceilometer | 07:45 | |
| *** nellysmitt has quit IRC | 07:46 | |
| *** eglynn has quit IRC | 07:52 | |
| *** eglynn has joined #openstack-ceilometer | 07:59 | |
| *** cmyster has quit IRC | 08:01 | |
| *** ddieterly has joined #openstack-ceilometer | 08:11 | |
| *** ddieterly has quit IRC | 08:14 | |
| *** NellyK has joined #openstack-ceilometer | 08:14 | |
| *** ddieterly has joined #openstack-ceilometer | 08:15 | |
| *** ddieterly has quit IRC | 08:19 | |
| *** cmyster has joined #openstack-ceilometer | 08:22 | |
| *** cmyster has joined #openstack-ceilometer | 08:22 | |
| *** NellyK has quit IRC | 08:23 | |
| *** nellysmitt has joined #openstack-ceilometer | 08:24 | |
| *** cmyster has quit IRC | 08:35 | |
| *** deepthi has joined #openstack-ceilometer | 08:48 | |
| *** ala_ has joined #openstack-ceilometer | 08:51 | |
| *** claudiub has joined #openstack-ceilometer | 09:04 | |
| *** IvanBerezovskiy1 has joined #openstack-ceilometer | 09:12 | |
| *** Flcn has joined #openstack-ceilometer | 09:15 | |
| *** safchain has joined #openstack-ceilometer | 09:15 | |
| *** Yanyanhu has quit IRC | 09:26 | |
| *** Longgeek has quit IRC | 09:32 | |
| *** Longgeek has joined #openstack-ceilometer | 09:35 | |
| *** sbfox has quit IRC | 09:39 | |
| *** isviridov_away is now known as isviridov | 09:41 | |
| *** sudipto has quit IRC | 09:48 | |
| *** _nadya_ has joined #openstack-ceilometer | 09:48 | |
| *** cmyster has joined #openstack-ceilometer | 09:48 | |
| *** cmyster has joined #openstack-ceilometer | 09:48 | |
| *** sudipto has joined #openstack-ceilometer | 09:55 | |
| sudipto | Hi, is there a way to create an alarm in ceilometer for a group of resources and then when the alarm is generated - we should be able to identify the resource on which it was raised? | 09:57 |
| DinaBelova | sudipto, probably eglynn can help you here | 10:00 |
| sudipto | eglynn: Could you please help? | 10:00 |
| sudipto | Thanks DinaBelova! Does he work during this time? | 10:01 |
| sudipto | *he/she | 10:01 |
| DinaBelova | sudipto, he's UTC+0 now I believe | 10:01 |
| eglynn | sudipto: the alarm would be triggered on the average statistic over the resource group? | 10:01 |
| DinaBelova | eglynn, good morning sir! | 10:02 |
| sudipto | eglynn: No, if it is based on a min/max of a value. Let's say cpu_util... | 10:03 |
| eglynn | sudipto: the alarm evaluator uses the ceilometer statistics API which doesn't identify which resource provided the max/min value | 10:04 |
| eglynn | sudipto: so as things currently stand, it would be problematic to include that info in the alarm notificiatoin | 10:04 |
| sudipto | eglynn: Could you please let me know the reason for this? | 10:05 |
| eglynn | sudipto: the reason the statistics API was designed that way? | 10:05 |
| sudipto | eglynn: Ok, then this is a limitation by which we cannot identify the resource for which the alarm was raised? We would have to individually define the alarm on individual resources? | 10:06 |
| *** yatin has quit IRC | 10:06 | |
| eglynn | sudipto: say for example, the alarm threshold was X and the statistic used is max over N periods | 10:08 |
| eglynn | sudipto: if the alarm fires, you don't know if just one resource crossed the thredhold over those periods | 10:09 |
| sudipto | eglynn: I was wondering if that would atleast give you a set of resources in that case, which have that alarm fired... | 10:09 |
| eglynn | sudipto: or indeed whether *different* resources crossed threshold over each of the N periods | 10:10 |
| eglynn | sudipto: so it sounds like an alarm over max/min in a group is not really what you need here (if identifying the individual resource is key to you) | 10:13 |
| sudipto | eglynn: Yeah. For me, the way i am looking at this problem is...let's say there are 30 people in a classroom and i ask, then after the exams - i trigger an alarm IF there are/is any student who scored more than a certain percentage. Then if that event is true, i also notify the students' name along with the event. I might be totally wrong with my interpretation, so please bear with me :) | 10:15 |
| sudipto | eglynn: I think the reason is we are dealing aggregation functions which are average/max/min - which only give out 1 value? | 10:16 |
| eglynn | sudipto: yes, one value, the actual max ... so if multiple students score higher than the percentage, that info would be lost | 10:17 |
| eglynn | sudipto: i.e. it's an exclusive maximum value | 10:17 |
| eglynn | sudipto: not all the values above some threshold X | 10:17 |
| eglynn | sudipto: so there are at least 2 aspects to this | 10:18 |
| eglynn | 1. the ceilo statistics API just reports a *single* max/min value for each period | 10:19 |
| eglynn | that's fair enough, as this is the meaning of "maximum" | 10:19 |
| eglynn | 2. the ceilo statistics API does not identify the resource associated with that maximum | 10:19 |
| eglynn | it's perhaps arguable that it could do so, though that wouln't be symmetric with true aggregate functions like sum,count,average | 10:20 |
| eglynn | 3. alarming over a the maximum in a group, when you really need to know *all* the resources above a threshold, is not a "natural" mapping of your problem IMO | 10:21 |
| eglynn | sudipto: ^^^ does that make sense? | 10:22 |
| sudipto | Yeah - that explains...Thanks. Hmm ok, then typically - if i want to detect a H/W failure kind of a situation via Ceilometer readings - is there a way to do that? | 10:23 |
| *** cdent has joined #openstack-ceilometer | 10:24 | |
| eglynn | sudipto: what kind of hardware failure? ... alarming in ceilometer is more about detecting when a statistic crosses a threshold | 10:40 |
| sudipto | eglynn: If there are number of memory check errors on the host beyond a threshold, i may want to fire an event and want to take an action based on that. And i am talking about a scenario that can have multiple hosts. | 10:46 |
| eglynn | sudipto: how would that memory-check-error-count metric be exposed? (via SNMP for example?) | 10:47 |
| *** SergeyLukjanov has quit IRC | 10:48 | |
| *** DinaBelova has quit IRC | 10:48 | |
| *** aignatov has quit IRC | 10:48 | |
| *** DinaBelova has joined #openstack-ceilometer | 10:48 | |
| *** aignatov has joined #openstack-ceilometer | 10:49 | |
| *** eglynn has quit IRC | 10:49 | |
| *** eglynn has joined #openstack-ceilometer | 10:50 | |
| *** SergeyLukjanov has joined #openstack-ceilometer | 10:50 | |
| *** Longgeek has quit IRC | 10:54 | |
| *** Longgeek has joined #openstack-ceilometer | 10:56 | |
| cmyster | eglynn: hey what's up ? | 11:07 |
| eglynn | cmyster: hey | 11:08 |
| cmyster | I have an issue that I am exploring, mind having a look ? | 11:08 |
| cmyster | seems like a configuration thing, and its a warning, but Its bothering me :) | 11:09 |
| eglynn | cmyster: what's the issue? | 11:09 |
| cmyster | http://paste.openstack.org/show/147817/ | 11:09 |
| cmyster | thats from api.log | 11:10 |
| cmyster | system seems to be up and running otherwise | 11:10 |
| sudipto | eglynn: Yeah. | 11:11 |
| eglynn | cmyster: that's odd, 404 on the token from keystone | 11:11 |
| cmyster | exacly | 11:11 |
| cmyster | I might find some more things in that log... | 11:11 |
| eglynn | cmyster: try restarting the API service with debug/verbose logging enabled in the ceilometer.conf | 11:12 |
| cmyster | on it | 11:12 |
| eglynn | cmyster: what's the keystone_authtoken.auth_uri configured as? | 11:13 |
| cmyster | sec, phone | 11:16 |
| eglynn | np | 11:17 |
| *** alexpilotti has joined #openstack-ceilometer | 11:17 | |
| cmyster | eglynn: that server's IP:5000 | 11:17 |
| *** yatin has joined #openstack-ceilometer | 11:18 | |
| *** _nadya_ has quit IRC | 11:18 | |
| eglynn | cmyster: k, that looks fine | 11:18 |
| cmyster | and its working (tested with telnet) | 11:18 |
| cmyster | working as in the target responds | 11:18 |
| *** deepthi has quit IRC | 11:18 | |
| sudipto | eglynn: Sorry for being away for a bit...and apologies for interrupting your conversation. Can you suggest a way, we could handle the the h/w failure use case based on the a metrics exposed via SNMP? | 11:19 |
| eglynn | sudipto: well firstly ceilometer would have to gather that particular metric | 11:20 |
| sudipto | eglynn: I have been able to populate it in the Ceilometer DB... | 11:20 |
| eglynn | sudipto: in kilo we plan to add a mechanism to allow any new SNMP metric to be gather, driven purely by config | 11:20 |
| eglynn | sudipto: what is the actual SNMP metric name? | 11:21 |
| eglynn | sudipto: second, if ceilometer gathers that metric, then ceilometer can alarm on it crossing a threshold | 11:22 |
| eglynn | sudipto: (subject to the restrictions discussed earlier, around maximum being a single value and the associated resource not being identified ... which suggests a per-host alarming strategy) | 11:23 |
| sudipto | eglynn: hmm ok, so we would have to set the alarming on every host...that's what my initial thought was...was just wondering if there's a better way. Also is there a way to make HEAT aware of this h/w based alarm? Since HEAT seems to work with alarms which are set of it's stack...(I might not be aware of any other way)... | 11:24 |
| sudipto | *set on it's. | 11:25 |
| eglynn | sudipto: heat uses alarms to drive autoscaling of *instances* | 11:25 |
| eglynn | sudipto: whereas you're more interested in host-level alarming, amiright? | 11:25 |
| sudipto | eglynn: Yeah | 11:26 |
| eglynn | sudipto: what would you envisage heat doing with the alarm notification in this case? | 11:26 |
| sudipto | eglynn: And i want heat to react to that...for it's auto-scale... | 11:26 |
| eglynn | sudipto: for example for Heat to migrate autoscaled instance off the problematic host? | 11:27 |
| sudipto | eglynn: like for example - if the host PFA is raised, i want HEAT to migrate the instances off. | 11:27 |
| eglynn | sudipto: I don't think heat has that logic currently | 11:27 |
| sudipto | eglynn: Exactly. | 11:27 |
| cmyster | something apart from ScalingPolicy ? | 11:27 |
| eglynn | sudipto: heat autoscaling logic IIUC is limited to spinning up new instances when busy, or shutting down existing instances when idle | 11:27 |
| cmyster | yes | 11:27 |
| sudipto | eglynn: Can you tell me if it's logical to have a different policy in this case? | 11:28 |
| cmyster | +1 -1 to resources in a group depending on alerts | 11:28 |
| eglynn | sudipto: (I'm open to correct on that, but I'm not aware of instance-migration smarts in the existing Heat autoscaling loigc) | 11:28 |
| eglynn | sudipto: it could be a useful thing for Heat to support in the future | 11:29 |
| sudipto | eglynn: This leans towards the scheduler doing a host evac in this case - isn't it? | 11:29 |
| eglynn | sudipto: check with the folks on #heat though to be 100% sure if instance migration isn't support in autoscaling | 11:29 |
| sudipto | I am thinking - how the scheduler would know of this alarm as to when this can be triggered so that it can empty a host...and migrate all the workloads out of it... | 11:30 |
| eglynn | sudipto: the nova scheduler? | 11:30 |
| eglynn | sudipto: ceilometer can trigger a POST on webhook when an alarm fires, but the receiver of that notification has to have the logic in place to do the expected thing | 11:31 |
| eglynn | sudipto: so logic to do the host evacuation would have to be built in to the receiver side | 11:32 |
| openstackgerrit | Merged openstack/ceilometer: Port to graduated library oslo.i18n https://review.openstack.org/137941 | 11:32 |
| cmyster | eglynn: might be something nasty... could be a selinux thing on my rhel machine I'll let you know ;) | 11:34 |
| eglynn | cmyster: a-ha. k | 11:35 |
| cmyster | I got suspicious when after re-connecting to the machine I was welcomed with ABRT message regarding ceilo-api | 11:36 |
| cdent | DinaBelova: yesterday's store_events changes ended up being documentation only: https://review.openstack.org/#/c/140018/ | 11:38 |
| DinaBelova | cdent, ah, well, anyway it's better | 11:39 |
| DinaBelova | cdent, thanks! | 11:39 |
| * cdent nods | 11:39 | |
| swamireddy | hi | 11:41 |
| swamireddy | eglynn: Can you plz review https://review.openstack.org/#/c/138349/? | 11:41 |
| eglynn | swamireddy: k, I'll try to get to it today | 11:42 |
| swamireddy | eglynn: Thanks... | 11:42 |
| sudipto | eglynn: thanks for that explanation :) | 11:46 |
| eglynn | sudipto: np! | 11:47 |
| openstackgerrit | Dina Belova proposed openstack/ceilometer: Move central agent code to the polling agent module https://review.openstack.org/127186 | 11:48 |
| openstackgerrit | Dina Belova proposed openstack/ceilometer: Make compute discovery pollster-based, not agent-level https://review.openstack.org/127185 | 11:48 |
| openstackgerrit | Dina Belova proposed openstack/ceilometer: Merge Central and Compute agents to *polling agent* https://review.openstack.org/124719 | 11:48 |
| *** fwdit has quit IRC | 11:50 | |
| openstackgerrit | Mehdi Abaakouk proposed stackforge/gnocchi: Allows to get aggregated measurements of metrics https://review.openstack.org/125572 | 11:54 |
| openstackgerrit | Mehdi Abaakouk proposed stackforge/gnocchi: Use oslotest as base class for unittests https://review.openstack.org/140307 | 11:54 |
| *** sudipto has quit IRC | 11:56 | |
| *** cdent has quit IRC | 11:59 | |
| openstackgerrit | Mehdi Abaakouk proposed stackforge/gnocchi: Allows to get aggregated measurements of metrics https://review.openstack.org/125572 | 12:00 |
| openstackgerrit | Mehdi Abaakouk proposed stackforge/gnocchi: Allows to get aggregated measurements of metrics https://review.openstack.org/125572 | 12:06 |
| *** eglynn has quit IRC | 12:12 | |
| *** alexpilotti has quit IRC | 12:46 | |
| cmyster | eglynn-regus: when you get back: https://bugzilla.redhat.com/show_bug.cgi?id=1172122 | 12:48 |
| *** ddieterly has joined #openstack-ceilometer | 12:50 | |
| *** eglynn has joined #openstack-ceilometer | 12:57 | |
| *** Longgeek has quit IRC | 12:58 | |
| *** Longgeek has joined #openstack-ceilometer | 13:04 | |
| *** yatin has quit IRC | 13:12 | |
| *** eglynn is now known as eglynn-lunch | 13:14 | |
| *** alexpilotti has joined #openstack-ceilometer | 13:25 | |
| *** _nadya_ has joined #openstack-ceilometer | 13:39 | |
| *** thomasem has joined #openstack-ceilometer | 13:43 | |
| *** gordc has joined #openstack-ceilometer | 13:43 | |
| *** thomasem has quit IRC | 13:46 | |
| *** ddieterly has quit IRC | 13:46 | |
| *** Longgeek_ has joined #openstack-ceilometer | 13:52 | |
| *** Longgeek has quit IRC | 13:52 | |
| openstackgerrit | Igor Degtiarov proposed openstack/ceilometer: Encompassing one source pollsters with common context https://review.openstack.org/139037 | 13:55 |
| *** julim has joined #openstack-ceilometer | 13:58 | |
| openstackgerrit | Ilya Tyaptin proposed openstack/ceilometer: Add test data generator via oslo messaging https://review.openstack.org/138804 | 14:02 |
| openstackgerrit | Dina Belova proposed stackforge/gnocchi: Add OpenTSDB storage driver https://review.openstack.org/107986 | 14:04 |
| *** zqfan has joined #openstack-ceilometer | 14:10 | |
| *** ddieterly has joined #openstack-ceilometer | 14:17 | |
| *** changbl has quit IRC | 14:20 | |
| *** k4n0 has quit IRC | 14:21 | |
| *** Longgeek_ has quit IRC | 14:22 | |
| *** ryanpetrello has joined #openstack-ceilometer | 14:26 | |
| *** blinky_ghost_ has joined #openstack-ceilometer | 14:28 | |
| *** Longgeek has joined #openstack-ceilometer | 14:29 | |
| blinky_ghost_ | HI all is it possible to use ceilometer to restart an instance when a compute node dies in another healthy node? | 14:30 |
| *** thomasem has joined #openstack-ceilometer | 14:30 | |
| eglynn-lunch | blinky_ghost_: if the node has completely died, you can't go a live migration, just a fresh restart | 14:32 |
| *** eglynn-lunch is now known as eglynn | 14:32 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/ceilometer: Updated from global requirements https://review.openstack.org/140049 | 14:32 |
| eglynn | blinky_ghost_: ... but in any case, see related discussion on this review https://review.openstack.org/#/c/130539/ | 14:32 |
| blinky_ghost_ | eglynn: Fresh restart it's fine by me. Does this review means it will only be available in Kilo? | 14:34 |
| eglynn | blinky_ghost_: yes, it's a work-in-progress | 14:34 |
| *** thomasem has quit IRC | 14:35 | |
| blinky_ghost_ | eglynn: do you know any other working solution to spawn an instance on the healthy node when the other dies? | 14:35 |
| *** thomasem has joined #openstack-ceilometer | 14:36 | |
| eglynn | blinky_ghost_: that's the only explicit effort to achieve that via ceilometer that I'm aware of | 14:37 |
| blinky_ghost_ | eglynn: ok thanks | 14:37 |
| *** Longgeek has quit IRC | 14:39 | |
| *** thomasem has quit IRC | 14:39 | |
| *** thomasem has joined #openstack-ceilometer | 14:40 | |
| *** Longgeek has joined #openstack-ceilometer | 14:44 | |
| claudiub | eglynn: hello | 14:44 |
| eglynn | claudiub: hi | 14:44 |
| claudiub | eglynn: do you have a few minutes to talk about some blueprints? | 14:45 |
| claudiub | inspector metrics | 14:45 |
| eglynn | claudiub: got an LP or ceilometer-specs link? | 14:45 |
| claudiub | eglynn: yeap: https://blueprints.launchpad.net/ceilometer/+spec/hyper-v-memory-metrics | 14:46 |
| claudiub | and also for disk iops: https://blueprints.launchpad.net/ceilometer/+spec/hyper-v-disk-iops-metrics | 14:46 |
| *** zqfan has quit IRC | 14:47 | |
| *** adelinatuvenie has joined #openstack-ceilometer | 14:47 | |
| eglynn | claudiub: my knowledge of hyper-v is limited, but both seem like reasonable ideas | 14:47 |
| eglynn | claudiub: are you planning to implement these for kilo? | 14:47 |
| claudiub | eglynn: indeed, for the memory metrics we already had it done for juno, but we did not have a blueprint for it | 14:48 |
| claudiub | eglynn: and it's just sitting in our repo, waiting to see the openstack light. :D | 14:48 |
| claudiub | eglynn: also, it's not a new feature, vmware already have it implemented, and it's quite simple to implement for hyper-v as well | 14:49 |
| eglynn | claudiub: a-ha, k ... so can you propose a more detailed discription of the feature in ceilometer-specs, then you should be good to go with proposing the code patch | 14:49 |
| eglynn | claudiub: on the subject of hyper-v, just a heads-up that we need to start thinking about 3rd party CI for features like this | 14:50 |
| eglynn | claudiub: i.e. features that can't be verified in the upstream gate | 14:50 |
| claudiub | eglynn: actually, that's a good point | 14:50 |
| claudiub | eglynn: i have written a tempest test for it some time ago | 14:50 |
| *** fnaval has joined #openstack-ceilometer | 14:51 | |
| claudiub | eglynn: and the tests were checking what metrics the compute nodes were publishing | 14:51 |
| eglynn | claudiub: presumably that tempest test can't be run in the upstream CI gate, or? | 14:52 |
| eglynn | claudiub: (... due to the lack of a hypver-v deployment upstream) | 14:52 |
| eglynn | claudiub: there's already this for nova, neutron testing against hyper-v https://wiki.openstack.org/wiki/ThirdPartySystems/Hyper-V_CI | 14:52 |
| claudiub | eglynn: it shouldn't be a problem, it's not hyper-v specific tempest test. I was just running the tests for hyper-v since we had a bug with the cpu metrics | 14:53 |
| *** alexpilotti_ has joined #openstack-ceilometer | 14:54 | |
| eglynn | claudiub: it would be great if we could glom onto the hyper-v_ci@microsoft.com tests for nova and get all these ceilometer features CI'd against real hyper-v | 14:55 |
| *** adelinatuvenie has quit IRC | 14:57 | |
| *** alexpilotti has quit IRC | 14:57 | |
| *** alexpilotti_ is now known as alexpilotti | 14:57 | |
| jd__ | sileht: are you working on a fix for the new i18n change in Ceilometer/Gnocchi dispatcher? | 14:58 |
| sileht | jd__, no | 14:58 |
| jd__ | ok will do | 14:58 |
| claudiub | eglynn: sure thing :) we'll talk about it and see what we can do. Also, I'll fetch the ceilometer-compute test I've written and retest it (hyper-v + kvm) | 14:59 |
| *** _nadya_ has quit IRC | 14:59 | |
| claudiub | eglynn: I'll keep you posted with the updates :) | 14:59 |
| eglynn | claudiub: cool, thanks! | 14:59 |
| *** k4n0 has joined #openstack-ceilometer | 15:01 | |
| *** shakamunyi has joined #openstack-ceilometer | 15:04 | |
| *** shakamunyi has quit IRC | 15:04 | |
| *** shakamunyi has joined #openstack-ceilometer | 15:05 | |
| *** rbak has joined #openstack-ceilometer | 15:11 | |
| *** _nadya_ has joined #openstack-ceilometer | 15:11 | |
| *** pradk has joined #openstack-ceilometer | 15:17 | |
| *** zqfan has joined #openstack-ceilometer | 15:20 | |
| openstackgerrit | Julien Danjou proposed stackforge/gnocchi: Fix Ceilometer dispatcher after i18n transition https://review.openstack.org/140364 | 15:20 |
| *** adelinatuvenie has joined #openstack-ceilometer | 15:21 | |
| *** shakamunyi has quit IRC | 15:25 | |
| *** shakamunyi has joined #openstack-ceilometer | 15:25 | |
| *** ViswaV has quit IRC | 15:28 | |
| *** ViswaV has joined #openstack-ceilometer | 15:29 | |
| *** fnaval has quit IRC | 15:48 | |
| *** fnaval has joined #openstack-ceilometer | 15:48 | |
| *** ryanpetrello has quit IRC | 15:49 | |
| *** fnaval has quit IRC | 15:53 | |
| *** amalagon has joined #openstack-ceilometer | 15:57 | |
| openstackgerrit | Julien Danjou proposed stackforge/gnocchi: Fix Ceilometer dispatcher after i18n transition https://review.openstack.org/140364 | 15:58 |
| openstackgerrit | Julien Danjou proposed stackforge/gnocchi: Remove mox from import exceptions https://review.openstack.org/140385 | 15:58 |
| openstackgerrit | Dina Belova proposed openstack/ceilometer: Move central agent code to the polling agent module https://review.openstack.org/127186 | 16:01 |
| openstackgerrit | Dina Belova proposed openstack/ceilometer: Merge Central and Compute agents to *polling agent* https://review.openstack.org/124719 | 16:01 |
| eglynn | influx guys in the news | 16:10 |
| eglynn | http://techcrunch.com/2014/12/08/errplane-snags-8-1m-to-continue-building-open-source-influxdb-time-database/ | 16:10 |
| gordc | eglynn: paul dix is definitely paying for the tab next time. ;) | 16:11 |
| eglynn | gordc: LOL :) | 16:11 |
| gordc | eglynn: i guess as a segue. any progress on the influxdb driver? i got asked that question a few mins/hours ago. | 16:16 |
| *** fnaval has joined #openstack-ceilometer | 16:18 | |
| openstackgerrit | Dina Belova proposed stackforge/gnocchi: Add OpenTSDB storage driver https://review.openstack.org/107986 | 16:19 |
| DinaBelova | hehe, influx in news will make them good karma | 16:20 |
| DinaBelova | :) | 16:20 |
| eglynn | gordc: yeah, I'm working on it in the background, in tandem with a few other things ... originally targetting kilo-1 | 16:20 |
| gordc | eglynn: cool cool. i think we were looking to try it out internally... have a few other items to take care of beforehand though so no rush. | 16:21 |
| *** changbl has joined #openstack-ceilometer | 16:28 | |
| openstackgerrit | gordon chung proposed openstack/ceilometer: implement notification coordination https://review.openstack.org/135414 | 16:29 |
| *** ryanpetrello has joined #openstack-ceilometer | 16:30 | |
| openstackgerrit | gordon chung proposed openstack/ceilometer: implement notification coordination https://review.openstack.org/135414 | 16:32 |
| *** shakamunyi has quit IRC | 16:33 | |
| openstackgerrit | gordon chung proposed openstack/ceilometer: ensure unique pipeline names https://review.openstack.org/139203 | 16:33 |
| *** k4n0 has quit IRC | 16:34 | |
| openstackgerrit | gordon chung proposed openstack/ceilometer: untie pipeline manager from samples https://review.openstack.org/140131 | 16:36 |
| alexpilotti | eglynn: claudiub we’re going to add Ceilometer to the Hyper-V CI, hopefully already in the next few weeks | 16:37 |
| eglynn | alexpilotti: that's really good news, thanks! | 16:38 |
| eglynn | alexpilotti: (... please keep me updated) | 16:38 |
| alexpilotti | eglynn: I will, sure! We already deploy all the ceilometer pieces on the devstack side | 16:39 |
| eglynn | alexpilotti: coolness, thanks :) | 16:39 |
| alexpilotti | eglynn: so it’s just a matter of enabling them on the compute side (Hyper-V) as well | 16:39 |
| alexpilotti | eglynn: and this should really not take too long :-) | 16:40 |
| eglynn | alexpilotti: excellent, we'll probably want to make assertions also that the hyper-v metrics are being gathered as expected | 16:41 |
| alexpilotti | eglynn: claudiub already wrote a ceilometer Temptest test for compute | 16:41 |
| openstackgerrit | Merged stackforge/gnocchi: Fix Ceilometer dispatcher after i18n transition https://review.openstack.org/140364 | 16:42 |
| alexpilotti | eglynn: do you already have a way in Tempest to assert on the existence of specific metrics? | 16:42 |
| *** ryanpetrello has quit IRC | 16:42 | |
| eglynn | alexpilotti: yeah, I probably just need to look over that tempest for completeness | 16:43 |
| alexpilotti | eglynn: cool tx! | 16:43 |
| openstackgerrit | Merged openstack/ceilometer-specs: Ceilometer RBAC Support for API https://review.openstack.org/112137 | 16:46 |
| *** shakamunyi has joined #openstack-ceilometer | 16:47 | |
| amalagon | ping DinaBelova | 17:02 |
| *** isviridov is now known as isviridov_away | 17:03 | |
| *** safchain has quit IRC | 17:05 | |
| *** _cjones_ has joined #openstack-ceilometer | 17:09 | |
| *** ala_ has quit IRC | 17:10 | |
| *** cdent has joined #openstack-ceilometer | 17:11 | |
| *** ryanpetrello has joined #openstack-ceilometer | 17:13 | |
| *** _nadya_ has quit IRC | 17:21 | |
| openstackgerrit | ZhiQiang Fan proposed openstack/ceilometer: Remove inspect_instances method from virt https://review.openstack.org/137874 | 17:31 |
| openstackgerrit | Julien Danjou proposed openstack/ceilometer: Switch to oslo.context library https://review.openstack.org/140412 | 17:35 |
| openstackgerrit | Julien Danjou proposed openstack/ceilometer: Switch to oslo.context library https://review.openstack.org/140412 | 17:39 |
| *** adelinatuvenie has quit IRC | 17:46 | |
| openstackgerrit | Julien Danjou proposed openstack/ceilometer: Switch to oslo.context library https://review.openstack.org/140412 | 17:46 |
| openstackgerrit | Julien Danjou proposed openstack/ceilometer: Switch to oslo.context library https://review.openstack.org/140412 | 17:47 |
| openstackgerrit | Julien Danjou proposed openstack/ceilometer: Switch to oslo.context library https://review.openstack.org/140412 | 17:48 |
| jd__ | bad day :D | 17:48 |
| *** ryanpetrello has quit IRC | 17:52 | |
| *** sbfox has joined #openstack-ceilometer | 17:52 | |
| *** ryanpetrello has joined #openstack-ceilometer | 17:54 | |
| *** ViswaV has quit IRC | 17:54 | |
| *** david-lyle_afk is now known as david-lyle | 17:58 | |
| *** packet has joined #openstack-ceilometer | 17:59 | |
| openstackgerrit | ZhiQiang Fan proposed openstack/ceilometer: Return a meaningful value or raise an excpetion for libvirt https://review.openstack.org/134966 | 18:02 |
| *** fabiog has joined #openstack-ceilometer | 18:04 | |
| *** ddieterly has quit IRC | 18:15 | |
| openstackgerrit | gordon chung proposed openstack/ceilometer: ensure unique pipeline names https://review.openstack.org/139203 | 18:16 |
| *** ddieterly has joined #openstack-ceilometer | 18:18 | |
| openstackgerrit | gordon chung proposed openstack/ceilometer: untie pipeline manager from samples https://review.openstack.org/140131 | 18:20 |
| openstackgerrit | Merged stackforge/gnocchi: Use oslotest as base class for unittests https://review.openstack.org/140307 | 18:24 |
| openstackgerrit | Merged stackforge/gnocchi: Fix doc8 errors https://review.openstack.org/138393 | 18:24 |
| *** eglynn has quit IRC | 18:29 | |
| *** harlowja_away is now known as harlowja | 18:30 | |
| *** blinky_ghost_ has quit IRC | 18:31 | |
| *** r-daneel has joined #openstack-ceilometer | 18:35 | |
| *** fwdit has joined #openstack-ceilometer | 18:36 | |
| *** Flcn has quit IRC | 18:39 | |
| *** shakamunyi has quit IRC | 18:39 | |
| *** sbfox has quit IRC | 18:45 | |
| *** eglynn has joined #openstack-ceilometer | 18:48 | |
| *** ViswaV has joined #openstack-ceilometer | 18:53 | |
| *** sbfox has joined #openstack-ceilometer | 18:55 | |
| *** jdandrea has quit IRC | 18:59 | |
| *** jdandrea has joined #openstack-ceilometer | 18:59 | |
| *** eglynn has quit IRC | 19:01 | |
| *** Flcn has joined #openstack-ceilometer | 19:02 | |
| *** ajayaa has joined #openstack-ceilometer | 19:07 | |
| *** zqfan has quit IRC | 19:07 | |
| ajayaa | gordc, Hi | 19:12 |
| ajayaa | Was about to go to bed and saw your comment. :) | 19:12 |
| gordc | ajayaa: :) | 19:13 |
| ajayaa | What is the difference between a sample and an event? | 19:13 |
| gordc | caught your attention | 19:13 |
| gordc | samples are essentially a derived value from an event. | 19:13 |
| ajayaa | In fact, So notifications are first tranformed into an an event first? | 19:14 |
| gordc | right... or events are just slightly filtered version of notification | 19:14 |
| gordc | i've been adding the definition here: https://github.com/openstack/ceilometer/blob/master/etc/ceilometer/event_definitions.yaml | 19:14 |
| gordc | based on the notifications you listed, i think they could be listed there as well. | 19:15 |
| gordc | ajayaa: i'm ok with approving your spec as is... mainly concerned whether you were still looking at it. | 19:15 |
| ajayaa | I have written code to catch the notifications and convert them into samples. Just the unit tests are yet to be written. | 19:16 |
| ajayaa | There were couple of issues with payload from magnetodb side. So I was fixing those. | 19:17 |
| gordc | ajayaa: cool cool. i'll approve spec... i'll probably add your notifications as events as well since i think they're better suited. | 19:17 |
| ajayaa | gordc, okay. One more thing, I see there are two tables in ceilometer schema in mysql. | 19:18 |
| ajayaa | event and event_type. | 19:18 |
| gordc | yep | 19:18 |
| ajayaa | Are the notifications going to be stored at both sample table and event table? | 19:19 |
| gordc | or well there's two more trait and trait_type i think. | 19:19 |
| gordc | notifications we receive are derived into multiple samples/events... if you have events enabled, then yes, you'll have entries in both tables. | 19:20 |
| *** Longgeek has quit IRC | 19:20 | |
| gordc | multiple samples and one event* | 19:20 |
| ajayaa | How do I try events out? I see identity operations are already listed in event_definitions.yaml. But today when I tries keystone operations they were only going to sample table. | 19:22 |
| ajayaa | gordc ^^ | 19:22 |
| gordc | you should add store_events=True to notification section in your conf file | 19:23 |
| gordc | ajayaa: https://github.com/openstack/ceilometer/blob/master/ceilometer/notification.py#L37-L40 | 19:24 |
| *** eglynn has joined #openstack-ceilometer | 19:26 | |
| ajayaa | okay. I will try it out. Thanks. | 19:26 |
| gordc | ajayaa: cool cool. | 19:28 |
| *** ajayaa has quit IRC | 19:30 | |
| openstackgerrit | Merged openstack/ceilometer-specs: Add spec for MagnetoDB metering support https://review.openstack.org/126335 | 19:31 |
| *** zul has quit IRC | 19:31 | |
| *** zul has joined #openstack-ceilometer | 19:34 | |
| *** zul has quit IRC | 19:35 | |
| *** zul has joined #openstack-ceilometer | 19:37 | |
| *** cdent has quit IRC | 19:43 | |
| *** exploreshaifali has joined #openstack-ceilometer | 19:47 | |
| *** _nadya_ has joined #openstack-ceilometer | 19:50 | |
| *** changbl has quit IRC | 19:53 | |
| *** changbl has joined #openstack-ceilometer | 20:00 | |
| *** changbl has quit IRC | 20:07 | |
| *** changbl has joined #openstack-ceilometer | 20:16 | |
| *** vishy has joined #openstack-ceilometer | 20:19 | |
| *** Longgeek has joined #openstack-ceilometer | 20:20 | |
| *** Longgeek has quit IRC | 20:26 | |
| *** sbfox has quit IRC | 20:59 | |
| *** _nadya_ has quit IRC | 21:00 | |
| *** changbl has quit IRC | 21:01 | |
| *** ViswaV has quit IRC | 21:02 | |
| *** ViswaV has joined #openstack-ceilometer | 21:03 | |
| *** changbl has joined #openstack-ceilometer | 21:03 | |
| *** harlowja is now known as harlowja_away | 21:04 | |
| *** ViswaV has quit IRC | 21:04 | |
| *** ViswaV has joined #openstack-ceilometer | 21:04 | |
| *** harlowja_away is now known as harlowja | 21:12 | |
| *** julim has quit IRC | 21:18 | |
| *** Longgeek has joined #openstack-ceilometer | 21:22 | |
| *** Longgeek has quit IRC | 21:26 | |
| *** nellysmitt has quit IRC | 21:28 | |
| *** ViswaV has quit IRC | 21:30 | |
| *** alexpilotti has quit IRC | 21:35 | |
| *** ViswaV has joined #openstack-ceilometer | 21:35 | |
| openstackgerrit | gordon chung proposed openstack/ceilometer: untie pipeline manager from samples https://review.openstack.org/140131 | 21:54 |
| *** harlowja has quit IRC | 22:02 | |
| *** ryanpetrello has quit IRC | 22:05 | |
| *** exploreshaifali has quit IRC | 22:06 | |
| *** eglynn has quit IRC | 22:10 | |
| *** alexpilotti has joined #openstack-ceilometer | 22:11 | |
| *** thomasem has quit IRC | 22:13 | |
| *** promulo has joined #openstack-ceilometer | 22:16 | |
| *** changbl has quit IRC | 22:23 | |
| *** ViswaV has quit IRC | 22:24 | |
| *** ViswaV has joined #openstack-ceilometer | 22:30 | |
| *** changbl has joined #openstack-ceilometer | 22:34 | |
| *** gordc has left #openstack-ceilometer | 22:36 | |
| *** gordc has quit IRC | 22:36 | |
| *** ViswaV has quit IRC | 22:37 | |
| *** ryanpetrello has joined #openstack-ceilometer | 22:37 | |
| *** harlowja has joined #openstack-ceilometer | 22:37 | |
| *** pradk has quit IRC | 22:41 | |
| *** ryanpetrello has quit IRC | 22:44 | |
| *** changbl has quit IRC | 22:45 | |
| *** ViswaV has joined #openstack-ceilometer | 22:46 | |
| *** ryanpetrello has joined #openstack-ceilometer | 22:46 | |
| *** Flcn has quit IRC | 22:48 | |
| *** Kennan has quit IRC | 22:53 | |
| *** ViswaV has quit IRC | 22:54 | |
| *** ViswaV has joined #openstack-ceilometer | 22:55 | |
| *** ViswaV has quit IRC | 22:55 | |
| *** ViswaV has joined #openstack-ceilometer | 22:56 | |
| *** ViswaV has quit IRC | 22:58 | |
| *** jaypipes has quit IRC | 23:01 | |
| *** alexpilotti has quit IRC | 23:02 | |
| *** shakamunyi has joined #openstack-ceilometer | 23:04 | |
| *** Longgeek has joined #openstack-ceilometer | 23:10 | |
| *** Longgeek has quit IRC | 23:15 | |
| *** mikedillion has joined #openstack-ceilometer | 23:16 | |
| *** mikedillion has quit IRC | 23:23 | |
| *** mikedillion has joined #openstack-ceilometer | 23:23 | |
| *** ddieterly has quit IRC | 23:25 | |
| *** sbfox has joined #openstack-ceilometer | 23:36 | |
| *** shakamunyi has quit IRC | 23:39 | |
| *** shakamunyi has joined #openstack-ceilometer | 23:45 | |
| *** ViswaV has joined #openstack-ceilometer | 23:50 | |
| *** shakamunyi has quit IRC | 23:54 | |
| *** harlowja has quit IRC | 23:56 | |
| *** rbak has quit IRC | 23:57 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!