15:00:33 #startmeeting Ceilometer 15:00:33 #meetingtopic Ceilometer 15:00:33 #chair nijaba 15:00:33 #link http://wiki.openstack.org/Meetings/MeteringAgenda 15:00:34 Meeting started Thu Dec 13 15:00:33 2012 UTC. The chair is nijaba. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:35 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:37 The meeting name has been set to 'ceilometer' 15:00:39 Current chairs: nijaba 15:00:44 ATTENTION: please keep discussion focussed on topic until we reach open discussion topic 15:00:44 Hello everyone! Show of hands, who is around for the ceilometer meeting? 15:00:44 o/ 15:00:53 o/ 15:00:55 o/ 15:01:04 hi 15:01:04 o/ 15:01:06 o/ 15:01:17 #topic actions from previous meeting 15:01:18 |o| 15:01:29 #topic asalkeld prepare a blueprint for v2 api "tightening" changes 15:01:45 o/ 15:01:51 anyone seen progress on this? 15:02:11 AFAIK asalkeld is working away on the flattening aspect 15:02:11 o/ 15:02:26 ok, Iĺl rea-action for next week 15:02:32 * eglynn is not sure whether flattening == tightening 15:02:37 #action asalkeld prepare a blueprint for v2 api "tightening" changes 15:02:49 #topic dhellmann replace ceilometer.readthedocs.org with a page directing readers to the new site 15:02:57 #info done 15:02:57 #link http://ceilometer.readthedocs.org/en/latest/ 15:03:07 #topic dhellmann experiment with passing arrays of parameters to wsme for queries 15:03:18 I have not seen much on this either. anyone? 15:03:32 nop 15:03:45 #action dhellmann experiment with passing arrays of parameters to wsme for queries 15:03:53 #topic eglynn prepare a blueprint for synaps integration 15:04:06 eglynn wrote: I'm trying to get them to commit to some level of participation and to contribute to an assessment of how the healthnmon APIs/concept map onto ceilo (or conversely the level of impedance) 15:04:27 eglynn: synaps? 15:04:27 yep, wot he said ;) 15:04:44 woops, I mixed up topics 15:04:53 wrote: I didn't get to filing the synaps-integration blueprints yet, still figuring some stuff out, but will have at least placeholders in place for our grizzly BP deadline 15:05:17 yep that about sums it up for now 15:05:34 I'm looking at Cassandra intregation as a first step 15:05:38 so we should mark this in progress, I guess 15:05:46 trying to unpick all the dependency issues ... 15:05:52 so progress, but still ongoing 15:06:01 BTW on the BPs 15:06:01 #info in progress 15:06:12 eglynn: possibly a wiki page for discussion? 15:06:14 can I clarify the deadline for grizzly? 15:06:25 yjiang5_away: yep, I'll do that ... 15:06:36 eglynn: thanks. 15:06:57 so is the deadline for grizzly approval Dec 15th? 15:06:57 eglynn: so the idea is that all blueprint that are targetted for grizzly, have an owner before the eo week 15:07:09 (or did I mis-remember that date ...) 15:07:17 OK end of this week 15:07:26 yes, but approval is automatic as long as someone has raised his hand sayong ï will do it 15:07:32 so Monday AM? ;) 15:07:44 and identify himself as such as assignee 15:07:56 moday am will be fine too 15:08:00 * eglynn tries to squeeze a little more time ;) 15:08:02 cool 15:08:22 we'll do a review and prioritization later in the meeting 15:08:44 #topic eglynn reach out to healthnmon project 15:08:51 eglynn wrote: I didn't get to filing the synaps-integration blueprints yet, still figuring some stuff out, but will have at least placeholders in place for our grizzly BP deadline 15:09:00 now that's on topic 15:09:17 nijaba: hehe 15:09:24 as above: I'm trying to get them to commit to some level of participation and to contribute to an assessment of how the healthnmon APIs/concept map onto ceilo (or conversely the level of impedance) 15:09:35 * eglynn is wary of a code dump 15:10:04 #info still trying to get commitment 15:10:07 (so I would like some commitment on participation) 15:10:09 yep 15:10:24 so re-action for next week I think 15:10:34 thanks for your efforts eglynn 15:10:35 thanks for your efforts eglynn 15:10:39 np! 15:10:48 nijaba: possibly some anysis of the feature duplication and feature we want on ceilometer will be helpful? 15:11:05 #action eglynn reach out to healthnmon project 15:11:24 nijaba: I mean for healthnmon. 15:11:25 yjiang5_away: yes, thatś should be a starting point 15:11:40 again, a wiki page would be welcome 15:11:52 yjiang5_away: yes, duplication is certainly there ... and yes, I'll set up a wiki for that also 15:11:53 nijaba: we can try to create one 15:12:03 great 15:12:13 Yes, we can help on that healthnmon wiki 15:12:32 #action yjiang5 create wiki page for healthnmon 15:12:57 #action eglynn, yjiang5_away, llu-laptop to work on a wiki page for analysis of healthmon duplication, etc.. 15:13:04 yjiang5_away: sorry :) 15:13:14 nijaba: :) 15:13:22 This was all from last week's action 15:13:25 Hi This is Divakar from Healthnmon 15:13:34 Divakar: hi 15:13:34 hello Divakar 15:14:05 hey Divikar, wlecome! 15:14:15 so whomever starts the analysis on the wiki should post the link to the ml 15:14:22 thank you.. I had shared the information on Healthnmon with eglynn 15:14:57 Divakar: did you see my latest mail on that thread? 15:15:12 Yes...I just did.. 15:15:18 Divakar: (trying to get a sense of participation on your side...) 15:15:35 Divakar: cool 15:15:53 shall we move on to the next topic? 15:16:15 Divakar: and great to see you on IRC, hope to see more of you guys! 15:16:25 nijaba: k 15:16:29 #topic Final blueprint review 15:16:34 Yeah.. this time works for me 15:16:39 looking at https://blueprints.launchpad.net/ceilometer 15:16:42 alternate weeks works.. 15:16:54 Divakar: great! 15:16:56 there are quite a few bp without an assignee 15:17:16 if you intend to commit working on one, now is the time toi mark yourself as the assignee 15:17:43 any non assigned bp next week will be retardgetted for H :) 15:17:44 I'll take https://blueprints.launchpad.net/ceilometer/+spec/remove-nova-imports 15:17:51 As shared earlier with you eglynn we are looking at single common Monitoring and alerting framework that will be part of base openstack with minimal replication 15:18:08 (done for libvirt, may still need some work for nova-notifier) 15:18:34 To start with implementation with stardardization of the framework, model and API which works for KVM hypervisor 15:18:36 I plan to take https://blueprints.launchpad.net/ceilometer/+spec/publisher-counters-frequency after I finished the multiple publisher work 15:18:36 of course we can retarget back later to grizzly, but I want to have a better vioew on what we are commited to deliver for g 15:18:59 please mark yourself as the assignee :) 15:19:14 I rememeber we can't mark ourselves 15:19:24 Brooklyn expressed an interest in https://blueprints.launchpad.net/ceilometer/+spec/horizon-plugin earlier today 15:19:32 only the core team can do that. 15:19:33 #info if you intend to commit working on one, now is the time toi mark yourself as the assignee, any non assigned bp next week will be retardgetted for H :) 15:19:50 so I assigned him to it, he'll take his initial ideas to the ML for discussion with UX experts 15:19:57 in this case, feel free to ping me to assign you. 15:20:03 nijaba: thanks! 15:20:39 https://blueprints.launchpad.net/ceilometer/+spec/merge-efforts-synaps is obviously mine too 15:20:45 (just assigned it to me) 15:20:59 thanks 15:20:59 I'll use it as an umbrella BP for the sunaps work 15:21:09 s/sunaps/synaps/ 15:21:22 we'd like to take https://blueprints.launchpad.net/ceilometer/+spec/monitoring-physical-devices 15:21:24 #info please ping nijaba to get assigned to a blueprint 15:21:43 #info and specify your launchpad id 15:22:09 doesn't https://blueprints.launchpad.net/ceilometer/+spec/user-api fall out of dhellman's v2 API work? 15:22:29 Our intent of healthnmon also caters towards monitoring physical devices as well.. 15:22:30 eglynn: no, but I have an assignee for it 15:22:33 eglynn: I think you need create several other BP for synaps integration, it's a big effort IMO :) 15:22:38 sorry gotta drop off now ... 15:22:44 my launchpad id is lianhao-lu 15:22:48 eglynn: cheers 15:22:53 I'll catch up via the IRC logs later! 15:22:56 thanks all! 15:23:06 data model defined should support physical devices as well 15:23:45 Divakar: yes, we should refine the data model. 15:23:59 Divakar: I think we can have more discussion on this through ml? 15:24:06 sure 15:24:26 Divakar: great. 15:25:27 now looking at https://launchpad.net/ceilometer/+milestone/grizzly-2 15:25:39 This looks good to me 15:26:03 any bp that we should add for this target? 15:26:13 which is Jan 10, btw 15:26:24 I just changed https://blueprints.launchpad.net/ceilometer/+spec/python-ceilometerclient to implemented 15:26:42 jd__: great, Ill change the target to g2 then 15:26:55 I think https://blueprints.launchpad.net/ceilometer/+spec/nova-independent-virt is implemented too, eglynn? 15:27:01 ah eglynn is AFK 15:27:21 duh, it was already g2 15:27:38 #action jd__ ask eglynn if https://blueprints.launchpad.net/ceilometer/+spec/nova-independent-virt is implemented 15:27:45 perfect 15:28:13 jd__: are you still confident about https://blueprints.launchpad.net/ceilometer/+spec/pollster-swift? 15:28:16 eglynn: ^^^ feel free to answer to me or change the status if/when you'll read the backlog :-) 15:28:47 nijaba: I've already a working basic thing, I'll send a patch early next week 15:29:03 I want to do some more tests on it first :) 15:29:06 jd__: great 15:30:01 any news on the implementation of the new rpc model for https://blueprints.launchpad.net/ceilometer/+spec/use-new-rpc-messsage 15:30:03 ? 15:30:29 never heard of anything about this oslo side 15:30:30 nijaba: anyone working on it? 15:30:51 nijaba: is the oslo blueprint implemented already? 15:31:05 russellb is on the oslo bp 15:31:23 it is marked as needs review 15:31:53 ah missed the dep graph 15:32:25 https://review.openstack.org/#/c/17965/ seems really recent 15:33:21 hum, the patch is against nova and not oslo, and has no tests 15:33:31 what about https://blueprints.launchpad.net/ceilometer/+spec/nova-cell-support ? any news on cell implementation? 15:33:55 jd__: ok, can you get in touch with russellb to get some more info? 15:34:27 #action jd__ get in touch with russellb to more info about https://blueprints.launchpad.net/ceilometer/+spec/use-new-rpc-messsage 15:34:35 hi 15:34:43 yeah, that was a fast action! 15:34:47 hi russellb :) 15:34:57 hello russellb 15:35:14 so ... that patch is very close, up for review, i think it's "done" 15:35:20 bit of a complication with ceilometer actually 15:35:28 russellb: I was looking at your patch https://review.openstack.org/#/c/17965/, why is it on nova and not oslo? 15:35:45 it's in oslo too, the nova patch is just up so that nova tests will run against it 15:36:04 https://review.openstack.org/#/c/17554/ 15:36:06 that's the oslo patch 15:36:21 russellb: ah great! 15:36:23 it's not going to be complete from the ceilometer perspective though, i'm afraid 15:36:35 it doesn't apply the message envelope to notifications right now 15:36:45 ok 15:37:02 because i was thinking about it, and notifications are to be consumed by applications outside of openstack, and it would break all of them 15:37:02 anything that's blocking doing that? 15:37:24 :-/ 15:37:39 has (other) oslo maintainers an opinion about this? 15:37:43 russellb: but isn't the change made so that we can prevent future breakage? 15:37:44 but i excluded notifications before i later updated the patch to not use the message envelope by default, so we can phase it in 15:37:44 s/s/ve/ 15:37:49 that *may* be sufficient ... 15:38:09 definitely interested in your thoughts on it 15:38:41 I'd suggest raising the question on the ml so more eyes get on it 15:38:57 good idea, i can do that 15:39:09 russellb: thanks! 15:39:11 thanks russellb 15:39:56 #action russellb raises the question about Oslo notifications versionning on the ml 15:39:57 :) 15:40:05 aside from that slightly significant complication, it's done :-p 15:40:07 hehe 15:40:22 that's a good first step, indeed 15:40:57 ok, thatś all for me on this topic of blueprints. anything else? 15:42:03 ok, moving on 15:42:09 #topic Open Discussion 15:42:16 hi i have something 15:42:24 zul: pelase shoot 15:42:37 so recently pecan was added to tools/pip-requires 15:42:46 which is fine (not packaged in ubuntu) 15:42:56 however it depends on a newer version of webob 15:43:20 1.2dev i believe which is incompatible with other openstack projects (ie: nova, glance, etc) 15:43:35 so makes me not happy 15:43:44 yes. That's bug https://bugs.launchpad.net/ceilometer/+bug/1089734. 15:43:45 Launchpad bug 1089734 in ceilometer "ceilometer-collector fail to start cause no ceilometer.collector extensions found" [Undecided,New] 15:44:02 zul: that's only for version 2 of the API which is not finished, so you can also comment out that deps and be fine 15:44:04 zul: dhellmann made most of these changes and is unfortunately on vaction this week. I would hope this would be fixed before g2 15:44:33 jd_: ok i can live with that 15:44:56 zul: you'll probably have ceilometer-api not working but ceilometer-api-v1 will be fine 15:45:08 nijaba: im on vacation starting tomorrow so ill put it on my todo list to talk to dhellman when i get back 15:45:16 zul: great! 15:45:52 So, we decided to have a Bug squashing day on Jan 4th. Any volunteer to organize it, as I will be on vacation that week 15:46:59 #action need to find a volunteer to organize bug squashing day on Jan 4th 15:47:07 nijaba: I'm available that day, so if any help needed, I'm available, although no experience on orgnize it 15:47:52 nijaba: I also could help on it if needed 15:47:57 yjiang5_away: thanks. Well hope to get more volunteers before next week 15:47:57 s/orginize/orgnizing/ 15:48:02 jiajun: ditto 15:48:05 nijaba: I can probably help too 15:48:10 \o/ 15:48:31 and I think dhellmann will be around too 15:48:43 don't want to volunteer him, but… :) 15:48:51 so I'll send an email on the ml and we can get more on this on the ml 15:49:25 any other topic, or should we end the meeting early? 15:49:56 I'm good 15:50:11 nijaba: k 15:50:14 counting down to from 3 15:50:19 2 15:50:23 1 15:50:35 #endmeeting