| *** nati_ueno has quit IRC | 00:01 | |
| *** matsuhashi has joined #openstack-metering | 00:33 | |
| *** mgagne has quit IRC | 00:37 | |
| *** nosnos has joined #openstack-metering | 01:20 | |
| *** mengxd has joined #openstack-metering | 01:37 | |
| *** rongze has joined #openstack-metering | 01:59 | |
| *** rongze has quit IRC | 02:18 | |
| *** rongze has joined #openstack-metering | 02:20 | |
| *** ArcTanSusan has joined #openstack-metering | 02:25 | |
| *** bingbu has joined #openstack-metering | 02:31 | |
| *** suo has joined #openstack-metering | 02:41 | |
| openstackgerrit | Lianhao Lu proposed a change to openstack/ceilometer: Add resource loader support https://review.openstack.org/58747 | 02:42 |
|---|---|---|
| openstackgerrit | Lianhao Lu proposed a change to openstack/ceilometer: Added resources support in pollster's interface https://review.openstack.org/58489 | 02:42 |
| *** rongze has quit IRC | 03:06 | |
| *** rongze has joined #openstack-metering | 03:06 | |
| *** rbrady has quit IRC | 03:55 | |
| *** rongze has quit IRC | 04:11 | |
| *** rongze has joined #openstack-metering | 04:17 | |
| *** urulama has joined #openstack-metering | 04:21 | |
| *** boris-42 has joined #openstack-metering | 04:49 | |
| *** rongze has quit IRC | 04:50 | |
| *** rongze has joined #openstack-metering | 05:08 | |
| *** ArcTanSusan has quit IRC | 05:27 | |
| *** nati_ueno has joined #openstack-metering | 05:44 | |
| *** urulama has quit IRC | 05:45 | |
| *** ArcTanSusan has joined #openstack-metering | 05:47 | |
| *** urulama has joined #openstack-metering | 05:47 | |
| *** ildikov has joined #openstack-metering | 05:59 | |
| openstackgerrit | Jenkins proposed a change to openstack/ceilometer: Imported Translations from Transifex https://review.openstack.org/57603 | 06:05 |
| *** sdake_ has joined #openstack-metering | 06:07 | |
| *** nadya has joined #openstack-metering | 06:13 | |
| *** nadya is now known as Guest20655 | 06:13 | |
| *** ChanServ changes topic to "This chan is deprecated, please join #openstack-ceilometer" | 06:16 | |
| *** ArcTanSusan has quit IRC | 06:16 | |
| *** ildikov has quit IRC | 06:20 | |
| *** sdake_ has quit IRC | 06:20 | |
| *** ArcTanSusan has joined #openstack-metering | 06:26 | |
| *** urulama has quit IRC | 06:50 | |
| *** Guest20655 has quit IRC | 06:52 | |
| *** rongze has quit IRC | 07:03 | |
| *** rongze has joined #openstack-metering | 07:06 | |
| *** rongze has quit IRC | 07:27 | |
| *** rongze has joined #openstack-metering | 07:28 | |
| *** nosnos_ has joined #openstack-metering | 07:33 | |
| *** nati_ueno has quit IRC | 07:34 | |
| *** nati_ueno has joined #openstack-metering | 07:35 | |
| *** nosnos has quit IRC | 07:37 | |
| *** rongze_ has joined #openstack-metering | 07:50 | |
| *** rongze has quit IRC | 07:53 | |
| *** rongze has joined #openstack-metering | 08:11 | |
| *** boris-42 has quit IRC | 08:13 | |
| *** rongze_ has quit IRC | 08:15 | |
| *** matsuhashi has quit IRC | 08:28 | |
| *** nprivalova has joined #openstack-metering | 08:37 | |
| *** shardy_afk is now known as shardy | 08:39 | |
| *** herndon has quit IRC | 08:42 | |
| *** tian has quit IRC | 08:42 | |
| *** llu has quit IRC | 08:43 | |
| *** tian has joined #openstack-metering | 08:44 | |
| *** llu has joined #openstack-metering | 08:44 | |
| *** rongze has quit IRC | 08:44 | |
| *** rongze has joined #openstack-metering | 08:45 | |
| *** SergeyLukjanov has joined #openstack-metering | 09:17 | |
| *** dperaza has joined #openstack-metering | 09:20 | |
| *** matsuhashi has joined #openstack-metering | 09:20 | |
| *** dperaza1 has quit IRC | 09:26 | |
| *** matsuhashi has quit IRC | 09:29 | |
| *** matsuhas_ has joined #openstack-metering | 09:31 | |
| *** boris-42 has joined #openstack-metering | 09:31 | |
| *** nosnos_ has quit IRC | 09:36 | |
| *** eglynn has joined #openstack-metering | 09:38 | |
| *** mengxd has quit IRC | 09:44 | |
| *** ruhe has joined #openstack-metering | 10:02 | |
| *** matsuhas_ has quit IRC | 10:12 | |
| *** ildikov has joined #openstack-metering | 10:14 | |
| *** matsuhashi has joined #openstack-metering | 10:17 | |
| *** ArcTanSusan has quit IRC | 10:19 | |
| *** bingbu has quit IRC | 10:19 | |
| *** nprivalova has quit IRC | 10:21 | |
| openstackgerrit | A change was merged to openstack/ceilometer: Imported Translations from Transifex https://review.openstack.org/57603 | 10:37 |
| *** nati_ueno has quit IRC | 10:42 | |
| *** nprivalova has joined #openstack-metering | 10:44 | |
| *** ruhe has quit IRC | 10:45 | |
| *** ruhe has joined #openstack-metering | 11:17 | |
| *** mengxd has joined #openstack-metering | 11:19 | |
| openstackgerrit | ChenZheng proposed a change to openstack/ceilometer: Add i18n warpping for all LOG messages https://review.openstack.org/54683 | 11:20 |
| *** SergeyLukjanov is now known as _SergeyLukjanov | 11:30 | |
| *** _SergeyLukjanov has quit IRC | 11:30 | |
| *** ruhe has quit IRC | 11:31 | |
| *** mengxd has quit IRC | 11:56 | |
| *** matsuhashi has quit IRC | 12:19 | |
| *** matsuhashi has joined #openstack-metering | 12:19 | |
| *** matsuhashi has quit IRC | 12:23 | |
| *** rongze has quit IRC | 12:41 | |
| *** SergeyLukjanov has joined #openstack-metering | 12:49 | |
| *** ruhe has joined #openstack-metering | 12:49 | |
| *** boris-42 has quit IRC | 13:07 | |
| *** mengxd has joined #openstack-metering | 13:11 | |
| *** mengxd has quit IRC | 13:11 | |
| *** rongze has joined #openstack-metering | 13:12 | |
| *** Shaan7 has joined #openstack-metering | 13:19 | |
| *** jd__ has left #openstack-metering | 13:22 | |
| *** rbrady has joined #openstack-metering | 13:24 | |
| *** suo has quit IRC | 13:27 | |
| *** rongze has quit IRC | 13:28 | |
| *** scroiset has left #openstack-metering | 13:36 | |
| *** ruhe has quit IRC | 13:45 | |
| *** rbrady has quit IRC | 14:11 | |
| *** herndon has joined #openstack-metering | 14:11 | |
| openstackgerrit | ChenZheng proposed a change to openstack/ceilometer: Add i18n warpping for all LOG messages https://review.openstack.org/54683 | 14:13 |
| *** nati_ueno has joined #openstack-metering | 14:13 | |
| *** jergerber has joined #openstack-metering | 14:17 | |
| *** thomasem has joined #openstack-metering | 14:21 | |
| lsmola | eglynn, hello, are yo uaround | 14:22 |
| eglynn | lsmola: I am | 14:22 |
| eglynn | lsmola: 'sup? | 14:22 |
| lsmola | eglynn, I am trying to build something like this http://people.redhat.com/~lsurette/OpenStack/Horizon%20Overview%20Pages_v1.0.pdf | 14:22 |
| lsmola | eglynn, do you think it's possible e.g. for current instance? | 14:23 |
| eglynn | lsmola: are those little trend lines what are known as sparklines? | 14:23 |
| lsmola | eglynn, basically if I do SUM of some appropriate period like 3 minutes, it gives me almost the time series i want | 14:24 |
| lsmola | eglynn, yes, this will be used as sparklines in tables too | 14:24 |
| eglynn | lsmola: so the "90 instances used, 10 remaining" ... that refers to quotas, right? | 14:24 |
| lsmola | eglynn, yes | 14:24 |
| lsmola | eglynn, (if you will make quotas lower, it will show values more then 100%, but let it be) | 14:25 |
| lsmola | eglynn, the thing is if i can rely on aggregation | 14:25 |
| eglynn | lsmola: k, so basically there's a mix here between two different accounting systems IIUC | 14:25 |
| eglynn | lsmola: ceilometer to determine the current usage | 14:26 |
| *** thomasem has quit IRC | 14:26 | |
| eglynn | lsmola: ... and nova os-quotas to determine the high-water-mark, correct? | 14:26 |
| lsmola | eglynn, let's say that it is polling for presence of all of them every 10 minutes, I have to basically rely on it will be collected like in 3 minutes, so I set the aggregation for 5minutes | 14:27 |
| lsmola | eglynn, yes | 14:27 |
| lsmola | eglynn, but basically every sample that is running wild, is miss-counted :-( | 14:28 |
| eglynn | lsmola: OK, so just thinking aloud for a second ... | 14:28 |
| lsmola | eglynn, I am thinking whether it would be better to actually get new meters | 14:28 |
| eglynn | lsmola: there's a possibility that the "remaining" portion reported may be slightly misleading | 14:28 |
| lsmola | eglynn, making samples for Instances_used (for project) and instances_total | 14:29 |
| eglynn | lsmola: as in reality, it the value in the quota_usages table in the nova DB that's used for the quota headroom determination | 14:29 |
| eglynn | lsmola: not the instances meters in ceilo | 14:29 |
| lsmola | eglynn, well I have decided to take those from nova, so it is always accurate, so I am taking just the timeseries from ceilometer | 14:29 |
| lsmola | eglynn, have to go on meeting | 14:30 |
| eglynn | lsmola: which one is not always accurate, the nova-maintained instance count? | 14:30 |
| eglynn | lsmola: a-ha, k, laters ... | 14:30 |
| lsmola | eglynn, nope the aggregation of ceilometer | 14:30 |
| lsmola | eglynn, i'' tkae laptop | 14:31 |
| *** ruhe has joined #openstack-metering | 14:31 | |
| eglynn | lsmola: so it would concern me reporting a "remaining" value that doesn't always match the concept of remaining headroom within the nova quotas logic | 14:31 |
| eglynn | lsmola: ok, I get it, the remaining value *does* always match the concept of remaining headroom within the nova quotas logic | 14:32 |
| eglynn | lsmola: but the trendline (populated from ceilo) may diverge from that? | 14:33 |
| *** lsmola_ has joined #openstack-metering | 14:35 | |
| *** lsmola has quit IRC | 14:35 | |
| lsmola_ | eglynn, sorry, got disconnected | 14:35 |
| lsmola_ | eglynn, so the aggregation of ceilometer is not correct | 14:35 |
| lsmola_ | eglynn, it is stored as many samples, showing just existence of the instances in time by 1 or 0 | 14:35 |
| lsmola_ | eglynn, and i want to extract the actual number of running instances in time for some project | 14:36 |
| lsmola_ | eglynn, which is not really reliable, as I have to time it precisely, so i correctly SUM only instances that belongs together | 14:37 |
| lsmola_ | eglynn, would it make sense to create new metrics, something like instances_count and instances_total, for project? | 14:38 |
| lsmola_ | eglynn, the same for other meters probably? | 14:39 |
| eglynn | lsmola_: hmmm, by "1 or 0" you mean either presence of a sample with volume=1 in the timeslice, or the absence of a sample in that timeslice? | 14:39 |
| eglynn | lsmola_: i.e. there's no actual samples with volume=0, right? | 14:39 |
| lsmola_ | eglynn, yes | 14:41 |
| lsmola_ | eglynn, hmm, not sure really, how it acts when the instance is suspended | 14:41 |
| *** ruhe is now known as ruhe_ | 14:41 | |
| lsmola_ | eglynn, i am now in the process of documenting that behaviour | 14:41 |
| *** ruhe_ has quit IRC | 14:42 | |
| eglynn | lsmola_: k, so there are at least 3 drivers for samples being ingested for that instances meter ... | 14:43 |
| eglynn | 1. compute agent polling (once every 600s out-of-the-box) | 14:43 |
| lsmola_ | eglynn, but the absence and the 0 will act the same when SUM is used | 14:43 |
| eglynn | 2. nova instance auditing, giving instance.exists notifications hourly or daily | 14:43 |
| lsmola_ | eglynn, though count of the samples would be different | 14:44 |
| eglynn | 3. instance deletion triggers a last sample for the instance IIUC | 14:44 |
| eglynn | so can you account for #1 & #2 with a SUM and group-by resource_id with period approximating the compute agent poilling? | 14:45 |
| eglynn | s/poilling/polling period/ | 14:45 |
| lsmola_ | eglynn, are the compute agent somehow synced? like do they send it all in the same time? | 14:46 |
| eglynn | lsmola_: no they are not synced, and in fact the configured periods could actually differ between agents | 14:47 |
| eglynn | lsmola_: so it would have to be the longer configure period, plus a fudge factor to handle drift | 14:47 |
| eglynn | (between agents) | 14:48 |
| lsmola_ | eglynn, if it's every 10minutes, and I can rely on all samples are send in 5 minutes, then I would set it around 6 minutes | 14:48 |
| eglynn | lsmola_: why not go longer, say 15mins, then have the group-by resource_id exclude the dups? | 14:49 |
| eglynn | dupes? ... duplicates I meant, i.e. multiple samples in the period for the same instance | 14:49 |
| *** lsmola has joined #openstack-metering | 14:51 | |
| lsmola | eglynn, sorry, got disconnected again | 14:51 |
| eglynn | replay ... | 14:52 |
| eglynn | [14:42] <eglynn> lsmola_: why not go longer, say 15mins, then have the group-by resource_id exclude the dups? | 14:52 |
| eglynn | [14:42] <eglynn> dupes? ... duplicates I meant, i.e. multiple samples in the period for the same instance | 14:52 |
| lsmola | eglynn, hmm | 14:52 |
| *** lsmola_ has quit IRC | 14:53 | |
| lsmola | eglynn, when I use SUM | 14:53 |
| lsmola | eglynn, and group by resource_id, it will SUM them, not exclude them, right? | 14:53 |
| lsmola | eglynn, though if i am able to exclude duplicates, it's basically solved | 14:54 |
| lsmola | eglynn, I just have to pick long enough period, to be sure every sample got the chance | 14:54 |
| lsmola | eglynn, right now, I am using group by project_id | 14:55 |
| eglynn | lsmola: yeah, that's what I was thinking ... i.e you get the sum of the grouped-by samples, not the grouped-by sum of the samples | 14:55 |
| lsmola | eglynn, is there a way to say unique resource_id ? | 14:55 |
| *** sandywalsh has quit IRC | 14:55 | |
| lsmola | eglynn, hmm, not sure I get how that could be done, i need to do it in one query btw.. | 14:56 |
| eglynn | lsmola: you can have multiple groupby fields | 14:56 |
| eglynn | lsmola: so I'm thinking you groupby *both* project_id and resource_id | 14:57 |
| lsmola | eglynn, ok, I thought that will just do the same | 14:57 |
| * lsmola intensively thinks | 14:57 | |
| lsmola | eglynn, well that would work if i could make for resource id different operation | 14:58 |
| eglynn | lsmola: there's still a problem though, how to account for recent instance deletion that occurred within the last period? | 14:58 |
| eglynn | lsmola: "make for resource id different operation"? | 14:58 |
| lsmola | eglynn, well it should be alright, always show as the latest sample actual info from nova | 14:59 |
| lsmola | eglynn, wel if it will be SUM of group_by resource_id and project_id, it will also sum the duplicate instances, right? | 15:00 |
| lsmola | eglynn, so something like Unique constraint on resource_id has to be used | 15:00 |
| lsmola | eglynn, hm, now I don't know how i meant the latter | 15:01 |
| eglynn | lsmola: yeah, you may be right ... I think I understand what you're getting at | 15:02 |
| eglynn | lsmola: let me do a little experimentation to figure out how this can be made work | 15:02 |
| lsmola | eglynn, is there something liek Unique? | 15:02 |
| lsmola | eglynn, excellent | 15:03 |
| eglynn | lsmola: nothing like unqiue currently, but let me see if I can figure out a way to get what you need | 15:03 |
| eglynn | lsmola: I'll give you a shout later (or email) | 15:03 |
| lsmola | eglynn, ok, great | 15:03 |
| lsmola | eglynn, great, thank you very much | 15:04 |
| *** boris-42 has joined #openstack-metering | 15:04 | |
| *** gordc has joined #openstack-metering | 15:06 | |
| *** sandywalsh has joined #openstack-metering | 15:09 | |
| ildikov | eglynn, lsmola: can you update me too with the results of the experimentation? this topic looks interesting for my API improvement blueprint | 15:11 |
| eglynn | ildikov: cool, will do! | 15:11 |
| ildikov | eglynn: thanks :) | 15:12 |
| *** ruhe has joined #openstack-metering | 15:13 | |
| lsmola | eglynn, hmm from mongo doc, i think we could use SUM of distinct('resource_id', 'value') Group BY period, project_id | 15:13 |
| eglynn | lsmola: yep, I was thinking along exactly those lines ... would obviously require API extentions to expose | 15:14 |
| lsmola | eglynn, or, more group_by in a pipeline, if that is possible, though that makes my brain hurt :-) | 15:14 |
| lsmola | eglynn, ok | 15:15 |
| lsmola | ildikov, do you think it would be possible to add support of "distinct" ? | 15:15 |
| lsmola | ildikov, so we can make beautiful queries like this? :-) | 15:16 |
| eglynn | lsmola: gotta divert off to sort out some internal stuff for a few mins, I'll back to this issue in a while ... | 15:16 |
| lsmola | eglynn, np | 15:17 |
| ildikov | lsmola: we had a short brainstorming session this week about what we should support in reach queries | 15:17 |
| lsmola | ildikov, the use case is for new Horizon overview page http://people.redhat.com/~lsurette/OpenStack/Horizon%20Overview%20Pages_v1.0.pdf | 15:18 |
| ildikov | lsmola: it is the beginning of the process, but we would like to implement a modular and extendable solution, to be able to support functionalities like this | 15:18 |
| lsmola | ildikov, http://ask-openstackux.rhcloud.com/question/59/improvements-to-horizon-overview/ | 15:18 |
| lsmola | ildikov, hmm, cool | 15:18 |
| ildikov | lsmola: that is why I asked you for update, to collect some real life UCs for the further design of this part of the API | 15:18 |
| lsmola | ildikov, though that means that it will take longer time right? | 15:19 |
| lsmola | ildikov, cause we would like those new overview pages and sparklines to land in I | 15:19 |
| *** ruhe has quit IRC | 15:19 | |
| *** sphoorti has joined #openstack-metering | 15:20 | |
| ildikov | the first draft of the improvement is targeted to I-2 | 15:20 |
| ildikov | lsmola: but if we have valid UCs, then it is easier to include it into our implementation even into Icehouse | 15:21 |
| lsmola | ildikov, ok, cool | 15:24 |
| *** ruhe has joined #openstack-metering | 15:25 | |
| *** rongze has joined #openstack-metering | 15:29 | |
| ildikov | lsmola: sorry, I was interrupted | 15:40 |
| *** herndon has quit IRC | 15:41 | |
| ildikov | lsmola: thanks for the docs, I will look into them and I will think it over, how this can fit into our solution | 15:42 |
| lsmola | ildikov, ok, excellent | 15:43 |
| lsmola | ildikov, do you have some BP for this that i can follow? | 15:43 |
| ildikov | lsmola: please inform me, if there are any new use cases/requirements related to this area | 15:43 |
| ildikov | lsmola: https://blueprints.launchpad.net/ceilometer/+spec/complex-filter-expressions-in-api-queries | 15:44 |
| ildikov | lsmola: you can find a link in the blueprint with some details of the idea | 15:45 |
| ildikov | lsmola: feel free to write some comments | 15:45 |
| lsmola | ildikov, ok, cool, I will definitely read that | 15:46 |
| ildikov | lsmola: I would only ask to use Authorship colors on the etherpad | 15:46 |
| lsmola | ildikov, just out of my Head there is still sample-api bp, + some bp I've created that should allow ordering and limiting | 15:47 |
| ildikov | lsmola: ok, cool :) | 15:47 |
| lsmola | ildikov, those are basically things that would make my life easier regarding Horizon integration:-) | 15:47 |
| lsmola | ildikov, I have it tracked here as dependencies https://blueprints.launchpad.net/horizon/+spec/ceilometer-api-enhancements | 15:48 |
| ildikov | lsmola: you will see on the etherpad, that the main idea is to use POST requests and use the body to send query filters to the API in JSON format | 15:49 |
| ildikov | lsmola: as we plan to use the request body, it gives us some freedome, which we do not have in case of the query in GET | 15:50 |
| lsmola | ildikov, oh very cool, looking forward to use that for Alarms | 15:51 |
| ildikov | lsmola: we also planned to use a separate query resource, in order to avoid messing up the current solution and having a fresh start | 15:52 |
| ildikov | lsmola: the proof of concept implementation is started, hopefully we will have something to show soon | 15:53 |
| lsmola | ildikov, yeah that might make sense | 15:53 |
| lsmola | ildikov, great!! | 15:53 |
| ildikov | lsmola: hopefully it will ;) | 15:56 |
| openstackgerrit | Ilya Tyaptin proposed a change to openstack/ceilometer: Fix for get_statistics with postgresql https://review.openstack.org/59204 | 15:56 |
| *** SergeyLukjanov is now known as _SergeyLukjanov | 15:56 | |
| *** ruhe has quit IRC | 15:58 | |
| *** giroro_ has quit IRC | 16:01 | |
| *** Ruetobas has joined #openstack-metering | 16:03 | |
| *** shengyao is now known as shengyao_afk | 16:03 | |
| *** Ruetobas has quit IRC | 16:07 | |
| *** Ruetobas has joined #openstack-metering | 16:07 | |
| *** nprivalova has quit IRC | 16:12 | |
| *** fnaval has quit IRC | 16:20 | |
| *** mgagne has joined #openstack-metering | 16:27 | |
| *** mgagne has joined #openstack-metering | 16:27 | |
| *** nati_ueno has quit IRC | 16:28 | |
| eglynn | darn and drats ... looks like we have a regression in python-ceiloclient 1.0.7 :( | 16:29 |
| eglynn | this changed line specifically http://goo.gl/QfneZz | 16:29 |
| eglynn | causes sample-list with prettytable-0.6.1-1 on RHEL to fail with "Invalid field name: None!" | 16:30 |
| eglynn | looking at the review https://review.openstack.org/47331 | 16:30 |
| eglynn | I don't I fully understand the motivation for sortby=None | 16:30 |
| eglynn | because the samples are already sorted in ts order by the API, None == "don't re-sort", or? | 16:31 |
| eglynn | sileht, gordc: ^^^ do you remember? | 16:31 |
| eglynn | I'm guessing that newer versions of prettytable allow sortby=None | 16:32 |
| eglynn | but the requirement is still PrettyTable>=0.6,<0.8 so the so should still work against 0.6.1 on RHEL | 16:33 |
| eglynn | meh! methinks I'll need to cut a 1.0.8 with a fix, and quick ... | 16:34 |
| gordc | eglynn: i don't remember what the purpose of sortby was | 16:35 |
| eglynn | I think the original code set sortby=0 which sorted on resource_id (first field in the list) | 16:35 |
| eglynn | the patch seems to want to change this to display samples in timestamp order, relying on the API layer pre-sorting in that order instead or re-sorting on the client side | 16:36 |
| gordc | i see... i thought prettytable just formatted the output... it sorts as well? | 16:36 |
| eglynn | gordc: yeah, it does | 16:37 |
| eglynn | gordc: unless you tell it to sortby None, in which case it barfs on RHEL :( | 16:38 |
| gordc | oh, wasn't aware of that. | 16:39 |
| *** sdake_ has joined #openstack-metering | 16:40 | |
| *** sdake_ has quit IRC | 16:40 | |
| *** sdake_ has joined #openstack-metering | 16:40 | |
| sileht | eglynn, the bad new is that is a oslo.cliutils issue | 16:46 |
| sileht | :q | 16:47 |
| sileht | oups | 16:47 |
| openstackgerrit | Ilya Tyaptin proposed a change to openstack/ceilometer: Fix for get_statistics with postgresql https://review.openstack.org/59214 | 16:47 |
| *** nati_ueno has joined #openstack-metering | 16:47 | |
| eglynn | sileht: so not specific to that version of prettytable? | 16:47 |
| sileht | eglynn, I just think of where we can fix that in openstack | 16:48 |
| eglynn | sileht: a-ha, I see ... | 16:48 |
| sileht | eglynn, I haven't yet find when prettytable introduce sortby=None | 16:49 |
| sileht | eglynn, https://code.google.com/p/prettytable/source/detail?r=87 | 16:53 |
| sileht | eglynn, None is allowed since 0.7 | 16:54 |
| eglynn | sileht: yep, good spot ... but since we still only require >=0.6, it's a regression to rely on that | 16:54 |
| gordc | eglynn: isn't jenkins suppose to auto patch our requirements? | 16:55 |
| eglynn | gordc: it grabs the newest in pypi within the declared range | 16:56 |
| eglynn | gordc: but the distros are free to use any version within that range if declared | 16:57 |
| gordc | seems like it was bumped to >=0.7 in requirements: https://github.com/openstack/requirements/blob/master/global-requirements.txt#L52 | 16:57 |
| eglynn | gordc: a-ha, sorry didn't spot that | 16:57 |
| sileht | eglynn, gordc but oslo-incubator is not yet up2date https://github.com/openstack/oslo-incubator/blob/master/requirements.txt#L8 | 16:58 |
| eglynn | sileht, gordc: yeah, I dunno if that auto-patching was turned off recently? | 16:58 |
| eglynn | (certainly it was working a month or two ago ...) | 16:58 |
| gordc | sileht: ah i see, are we dependent on that? or can we just bump up our requirements regardless. | 16:58 |
| gordc | i guess it could break cliutils if we do... | 16:59 |
| sileht | I think only the requirements repo is important for us | 16:59 |
| * sileht is looking for the latest auto requriements update | 16:59 | |
| gordc | eglynn, sileht: i haven't seen an auto-patch in a while... i doubt we're that up to date that it doesn't need to autopatch for us. | 17:00 |
| eglynn | well either way, I'll need to cut python-ceilocli 1.0.8 ... either with a new prettytable requirement or a work-around | 17:00 |
| * eglynn is tending towards a work-around at this stage | 17:01 | |
| *** ildikov has quit IRC | 17:01 | |
| sileht | altest autoupdate https://github.com/openstack/python-ceilometerclient/commit/8e341cc620352ba07fb80d81a32ba01199e7d4cb | 17:01 |
| eglynn | (as the global requirements update is less than a week old https://github.com/openstack/requirements/commit/45b937a2 and prolly hasn't been packaged on multiple distros) | 17:02 |
| *** SergeyLukjanov has joined #openstack-metering | 17:02 | |
| sileht | eglynn, hum thanks, I guess the debian backport repo have the same issue, the wheezy version is 0.6.1 too | 17:04 |
| eglynn | sileht: yeah, I'm thinking work-around is the safest course of action for now ... I'll propose something at the w/e | 17:05 |
| sileht | eglynn, ok cool | 17:05 |
| sileht | eglynn, ping me on RIC for quick review | 17:06 |
| sileht | s/RIC/IRC | 17:06 |
| eglynn | sileht: cool, thanks! | 17:06 |
| sileht | ubuntu backport have 0.6.1 too, everybody is impacted :) | 17:07 |
| openstackgerrit | gordon chung proposed a change to openstack/ceilometer: update docs to adjust for naming change https://review.openstack.org/59220 | 17:21 |
| gordc | sileht: eglynn: just an fyi, i asked in infra, the requirements auto-patch job was disabled due to a bug | 17:22 |
| sileht | gordc, ack, thx | 17:26 |
| *** nadya_ has joined #openstack-metering | 17:35 | |
| *** sdake_ has quit IRC | 17:37 | |
| eglynn | sileht: FYI here's the kind of work-around I had in mind https://gist.github.com/anonymous/7709208 | 17:41 |
| eglynn | ... unfortunately requires changes to cliutils to expose the reversesort, that's not ideal :( | 17:41 |
| openstackgerrit | gordon chung proposed a change to openstack/ceilometer: synch with oslo-incubator https://review.openstack.org/59222 | 17:42 |
| *** ruhe has joined #openstack-metering | 17:43 | |
| gordc | eglynn: yeah... will probably take longer to get that merged in oslo and then transfer over. | 17:43 |
| eglynn | gordc: yep, but otherwise we lose the most recent sample first ordering | 17:44 |
| *** ruhe has quit IRC | 17:47 | |
| openstackgerrit | gordon chung proposed a change to openstack/ceilometer: synch with oslo-incubator https://review.openstack.org/59222 | 17:51 |
| openstackgerrit | gordon chung proposed a change to openstack/python-ceilometerclient: sync with oslo-incubator https://review.openstack.org/59225 | 17:54 |
| openstackgerrit | gordon chung proposed a change to openstack/ceilometer: sync with oslo-incubator https://review.openstack.org/59222 | 17:55 |
| *** nadya_ has quit IRC | 17:55 | |
| *** ildikov has joined #openstack-metering | 17:56 | |
| *** eglynn has quit IRC | 18:04 | |
| *** jergerber has quit IRC | 18:15 | |
| *** rongze has quit IRC | 18:29 | |
| *** ruhe has joined #openstack-metering | 18:35 | |
| *** sandywalsh has quit IRC | 18:36 | |
| *** rongze has joined #openstack-metering | 18:38 | |
| *** eglynn has joined #openstack-metering | 18:41 | |
| *** sphoorti has left #openstack-metering | 18:44 | |
| *** eglynn has quit IRC | 18:57 | |
| *** ruhe has quit IRC | 19:01 | |
| *** rongze has quit IRC | 19:16 | |
| *** rongze has joined #openstack-metering | 19:34 | |
| *** jergerber has joined #openstack-metering | 19:40 | |
| *** rongze has quit IRC | 19:41 | |
| *** nati_ueno has quit IRC | 19:41 | |
| *** eglynn has joined #openstack-metering | 19:44 | |
| *** eglynn has quit IRC | 19:54 | |
| *** sdake_ has joined #openstack-metering | 20:01 | |
| *** sdake_ has quit IRC | 20:01 | |
| *** sdake_ has joined #openstack-metering | 20:01 | |
| *** ArcTanSusan has joined #openstack-metering | 20:02 | |
| *** rongze has joined #openstack-metering | 20:11 | |
| *** boris-42 has quit IRC | 20:13 | |
| *** rongze has quit IRC | 20:20 | |
| *** ArcTanSusan has quit IRC | 20:31 | |
| *** nadya_ has joined #openstack-metering | 20:35 | |
| *** nadya_ has quit IRC | 20:39 | |
| *** ruhe has joined #openstack-metering | 20:45 | |
| *** ArcTanSusan has joined #openstack-metering | 20:46 | |
| *** rongze has joined #openstack-metering | 20:46 | |
| *** sdake_ has quit IRC | 20:47 | |
| *** mgagne has quit IRC | 20:48 | |
| openstackgerrit | gordon chung proposed a change to openstack/ceilometer: sync with oslo-incubator https://review.openstack.org/59222 | 20:52 |
| *** rongze has quit IRC | 20:52 | |
| *** eglynn has joined #openstack-metering | 20:58 | |
| *** gordc has quit IRC | 20:58 | |
| *** eglynn has quit IRC | 21:03 | |
| *** eglynn has joined #openstack-metering | 21:14 | |
| *** ruhe is now known as ruhe_ | 21:19 | |
| *** ruhe_ is now known as ruhe | 21:19 | |
| *** ruhe has quit IRC | 21:21 | |
| *** eglynn has quit IRC | 21:32 | |
| *** ildikov has quit IRC | 21:48 | |
| *** rongze has joined #openstack-metering | 21:48 | |
| *** rongze has quit IRC | 21:54 | |
| *** SergeyLukjanov has quit IRC | 22:02 | |
| *** sdake_ has joined #openstack-metering | 22:12 | |
| *** ArcTanSusan has quit IRC | 22:29 | |
| *** ArcTanSusan has joined #openstack-metering | 22:31 | |
| *** rongze has joined #openstack-metering | 22:50 | |
| *** rongze has quit IRC | 22:56 | |
| *** eglynn has joined #openstack-metering | 23:08 | |
| *** ArcTanSusan has quit IRC | 23:16 | |
| *** eglynn has quit IRC | 23:48 | |
| *** rongze has joined #openstack-metering | 23:52 | |
| *** rongze has quit IRC | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!