Tuesday, 2014-07-22

*** dmsimard_away is now known as dmsimard00:09
*** dmsimard is now known as dmsimard_away00:20
*** tongli has quit IRC00:20
*** tongli has joined #openstack-ceilometer00:21
*** dmsimard_away is now known as dmsimard00:25
*** nati_ueno has quit IRC00:37
*** ddieterly has joined #openstack-ceilometer00:42
*** dmsimard is now known as dmsimard_away00:47
*** ccrouch has joined #openstack-ceilometer00:49
*** ddieterly has quit IRC00:55
*** llu has joined #openstack-ceilometer00:58
*** ddieterly has joined #openstack-ceilometer00:59
*** _cjones_ has quit IRC01:05
*** _cjones_ has joined #openstack-ceilometer01:06
*** _cjones_ has quit IRC01:09
*** _cjones_ has joined #openstack-ceilometer01:10
*** _cjones_ has quit IRC01:14
*** changbl has quit IRC01:20
*** changbl has joined #openstack-ceilometer01:34
*** dmsimard_away is now known as dmsimard01:35
*** nosnos has joined #openstack-ceilometer01:40
*** dmsimard is now known as dmsimard_away01:42
*** thomasem has joined #openstack-ceilometer01:59
*** thomasem has quit IRC02:03
openstackgerritA change was merged to openstack/ceilometer: improve expirer performance for sql backend  https://review.openstack.org/9648902:10
*** _cjones_ has joined #openstack-ceilometer02:20
openstackgerritA change was merged to openstack/ceilometer: Move log alarms driver code to alarm tree  https://review.openstack.org/10794102:25
*** _cjones_ has quit IRC02:25
*** nati_ueno has joined #openstack-ceilometer02:29
*** sbfox has joined #openstack-ceilometer02:34
*** sbfox has quit IRC02:36
*** liusheng has quit IRC02:41
*** sbfox has joined #openstack-ceilometer02:45
*** tongli has quit IRC03:03
*** flwang has joined #openstack-ceilometer03:03
*** raymondr has joined #openstack-ceilometer03:06
*** raymondr has quit IRC03:10
*** ccrouch has quit IRC03:28
*** sbfox has quit IRC03:32
*** raymondr has joined #openstack-ceilometer03:33
*** sbfox has joined #openstack-ceilometer03:38
*** nosnos has quit IRC03:40
*** nati_ueno has quit IRC03:44
*** raymondr has quit IRC03:57
*** nosnos has joined #openstack-ceilometer04:06
*** shakamunyi has joined #openstack-ceilometer04:06
*** shakamunyi has quit IRC04:08
*** sbfox has quit IRC04:12
*** Longgeek has joined #openstack-ceilometer04:13
*** r3pl4y has quit IRC04:36
*** nati_ueno has joined #openstack-ceilometer04:50
*** sam___ has quit IRC05:14
*** _nadya_ has joined #openstack-ceilometer05:16
*** _nadya_ has quit IRC05:19
*** ildikov has joined #openstack-ceilometer05:19
*** sbfox has joined #openstack-ceilometer05:25
*** _nadya_ has joined #openstack-ceilometer05:43
*** psharma has joined #openstack-ceilometer05:43
openstackgerritRikimaru Honjo proposed a change to openstack/ceilometer: Add retry function for alarm REST notifier  https://review.openstack.org/9990505:54
*** _nadya_ has quit IRC05:55
*** raymondr has joined #openstack-ceilometer05:58
*** k4n0 has joined #openstack-ceilometer05:59
*** raymondr has quit IRC06:03
*** raymondr has joined #openstack-ceilometer06:06
openstackgerritOpenStack Proposal Bot proposed a change to openstack/ceilometer: Imported Translations from Transifex  https://review.openstack.org/10497406:10
*** raymondr has quit IRC06:10
*** _nadya_ has joined #openstack-ceilometer06:33
*** Qu310 has quit IRC06:37
*** _nadya_ has quit IRC06:42
*** _nadya_ has joined #openstack-ceilometer06:53
*** _nadya_ has quit IRC06:54
*** _nadya_ has joined #openstack-ceilometer06:58
*** harlowja is now known as harlowja_away07:04
*** r3pl4y has joined #openstack-ceilometer07:06
*** ityaptin has joined #openstack-ceilometer07:08
openstackgerritA change was merged to openstack/ceilometer: clean up event model  https://review.openstack.org/10198807:12
*** _nadya__ has joined #openstack-ceilometer07:16
*** _nadya_ has quit IRC07:16
openstackgerritRikimaru Honjo proposed a change to openstack/ceilometer: Add validate alarm_actions schema in alarm API  https://review.openstack.org/10733007:21
*** yatin has joined #openstack-ceilometer07:25
*** lvdongbing has joined #openstack-ceilometer07:28
*** sbfox has quit IRC07:30
*** regebro has joined #openstack-ceilometer07:33
regebroHey all! I have a question.07:35
regebroIn the tuskar-ui we have ceilometer graphs per node. That all works nicely.07:35
regebroBut we would also want to have on the overview page some graphs with like a general overview.07:36
regebroIs that possible? Would it be possible to get say a cpu percentage for all deployed nodes in a group?07:36
ildikovreg07:45
ildikovregebro: hi07:45
nsajeregebro: why not?07:45
regebrohi!07:45
ildikovregebro: if you can identify somehow that those nodes are in one group, then it should be possible07:46
regebronsaje: The second question is then *how*. :-07:46
ildikovregebro: the REST API of Ceilometer supports complex query functionalities: http://docs.openstack.org/developer/ceilometer/webapi/v2.html#complex-query07:47
ildikovregebro: the question is that do you have the group information in metadata or any other field of a sample or not07:47
nsajeregebro: if you know the resource_ids of all the nodes, then you don't even need complex queries07:48
regebronsaje: I can get the resource_id's sure.07:49
regebroCan I just pass in a list or something?07:49
ildikovregebro: with complex query you can specify a list with using the 'in' relation07:49
ildikovregebro: with simple query you need as many requests as many resource ids you have07:50
ildikovregebro: it doesn't support any 'or' or 'in' relation07:50
nsajeregebro: ildikov: yep, sorry, I thought simple queries supported or relation07:50
regebroildikov: yeah, one request per resource is surely going to be way to slow.07:51
ildikovregebro: or you retrieve all the information in one request with the simple query and then do the filtering on your side07:51
*** k4n0 has quit IRC07:51
ildikovnsaje: no, it does not, that is why complex query was added, TBH I wanted a simple or relation ;)07:51
regebroildikov: I've tried that and failed. I don't get anything back at all.07:51
ildikovregebro: with SQL backends the releases before Juno suffer from performance problems07:52
ildikovregebro: I haven't tested it with complex query, maybe the situation is better with that, it depends on your db backend as most of the work is done there07:53
ildikovregebro: I mean in the DB07:53
regebroildikov: Well, even so, we'll need this to scale to pretty big situations, so a complex query must be better. I guess it won07:53
regebroI guess it won't actually aggregrate the data in any case, right?07:53
ildikovregebro: unfortunately the supprot for statistics is still on the todo list07:54
regebrook, thanks.07:54
ildikovregebro: so it will work now only for samples and alarms without any aggregation07:54
regebroI think we'll look into properly this next week. :-)07:55
ildikovregebro: you can start with the docco I sent, there are some examples at the bottom of the page, if you have any questions, I will try to answer :)07:55
regebroWhen people with more ceilometer experience are back from Europython. :-)07:55
ildikovregebro: LOL, cool :)07:55
*** nacim has joined #openstack-ceilometer07:55
*** raymondr has joined #openstack-ceilometer07:58
nsajeildikov: If you have some time today, could you take a look at arithmetic transformer implementation? https://review.openstack.org/#/c/106302/08:00
nsajeildikov: I'd like to get it through the gate by juno-208:00
*** _nadya__ has quit IRC08:02
*** safchain has joined #openstack-ceilometer08:03
*** raymondr has quit IRC08:03
*** k4n0 has joined #openstack-ceilometer08:05
openstackgerritIlya Tyaptin proposed a change to openstack/ceilometer: Use None instead of mutables in test method params defaults  https://review.openstack.org/10838908:07
*** eoutin has joined #openstack-ceilometer08:10
*** IvanBerezovskiy has joined #openstack-ceilometer08:17
ildikovnsaje: sure, I will try, however I cannot promise anything regarding to the gate :(08:18
nsajeildikov: I think it's a bit better today, I've had two patches patch checks on first try08:19
nsaje*pass08:19
ildikovnsaje: earlier is better, when a milestone is sooo close and in general also, but well, we should stay within reality :)08:24
nsajeildikov: yes, but it would be good if the cores kept an eye out at least for the blueprint implementations marked as high priority for juno-2 :)08:26
ildikovnsaje: yes, in a perfect world, where they have a plenty of allocated time for that :)08:31
*** k4n0 has quit IRC08:32
*** k4n0 has joined #openstack-ceilometer08:49
*** cdent has joined #openstack-ceilometer09:07
*** k4n0 has quit IRC09:19
*** openstackgerrit has quit IRC09:31
*** openstackgerrit has joined #openstack-ceilometer09:32
DinaBelovafolks, does somebody know if Fabio Giannetti is somewhere around? :)09:35
DinaBelovaI mean his vacation and work on the HA :)09:36
*** k4n0 has joined #openstack-ceilometer09:38
*** alexpilotti has joined #openstack-ceilometer09:38
cdentDinaBelova: my understanding was that he was back sometime this week.09:41
cdentBut I'm not sure exactly when.09:41
DinaBelovacdent, hehe, ok, thanks :)09:43
openstackgerritA change was merged to openstack/ceilometer: Separate the console scripts  https://review.openstack.org/10829409:50
*** _nadya_ has joined #openstack-ceilometer09:52
nsajesince eglynn appears to be missing in action today, forgive me for re-pasting his plea from yesterday:09:52
nsaje5:37 PM <eglynn> ... we have a few juno-2 BPs just needing a +1 on gerrit to get them over the line09:52
nsaje5:37 PM <eglynn> (cdent) ironic notifications https://review.openstack.org/10548609:52
nsaje5:37 PM <eglynn> (nsaje) arithmetic transformer https://review.openstack.org/10630209:52
nsaje5:39 PM <eglynn> ^^^ so if any cores have some review cycle going spare today, it would great to get those two reviews over the line by EoD09:52
nsaje^^^ sileht jd__ ildikov _nadya_ llu09:52
ildikovDinaBelova: I guess later afternoon is better for his TZ09:53
silehtnsaje, :) I take a look09:53
nsajesileht: great, thanks!09:54
DinaBelovaildikov, yeah, for sure - I just was wondering due to the fact I was not available yesterday and dunno if something happened during this time :)09:54
*** dmsimard_away has quit IRC09:54
*** jogo has quit IRC09:54
*** shadower has quit IRC09:54
*** jogo has joined #openstack-ceilometer09:54
*** shadower has joined #openstack-ceilometer09:55
*** dmsimardz has joined #openstack-ceilometer09:56
*** dmsimardz is now known as dmsimard09:56
silehtnsaje, ildikov was faster than me :)09:58
*** nosnos has quit IRC09:58
ildikovnsaje: bad news, you will need a rebase, but I will take a look, when you're ready with it09:58
*** nosnos has joined #openstack-ceilometer09:58
nsajeildikov sileht : thanks guys, I'll rebase asap and'll ping you then09:59
ildikovsileht: yeap, he asked me earlier, it just took a while to check that patch :)09:59
*** nosnos has quit IRC10:03
*** regebro has quit IRC10:03
*** regebro has joined #openstack-ceilometer10:04
*** lvdongbing has quit IRC10:11
*** _nadya_ has quit IRC10:16
openstackgerritNejc Saje proposed a change to openstack/ceilometer: Multi meter arithmetic transformer  https://review.openstack.org/10630210:20
nsajeildikov: I only just noticed the comment you made on the variable name, so wait for a new patchset :)10:23
nsajeildikov: I can't use 'sample' though, since it shadows the sample module. What do you suggest, smpl? :)10:23
ildikovnsaje: ok :)10:23
ildikovnsaje: well, I would use _sample or smth like, but smpl can be good too10:24
ildikovnsaje: there others like meter, etc, in that file ;)10:24
ildikovnsaje: but as I wrote, it's a non-blocking issue10:25
ildikovnsaje: maybe I'm too maniac... :)10:26
nsajeildikov: no, it's a good point, let me fix it lest it remains unfixed for a long time10:26
ildikovnsaje: sure, thanks for fixing it now10:30
openstackgerritNejc Saje proposed a change to openstack/ceilometer: Multi meter arithmetic transformer  https://review.openstack.org/10630210:31
ildikovnsaje: would you hate me if I would ask you to do the same fix in the __call__ method that starts at line 133? just for the sake of consistency10:38
ildikovnsaje: and then I promise it's the last one!10:39
nsajeildikov: I'm just thinking about consistency, take a look at https://github.com/openstack/ceilometer/blob/master/ceilometer/transformer/conversions.py10:39
nsajesample is denoted by 's' mostly in the code, except in the signature of the interface function (handle_sample)10:40
nsajeildikov: I guess it'd be best do use the same approach here, for consistency?10:40
nsaje(by here I mean in the new code)10:41
ildikovyou mean the keep the 's'?10:41
nsajeyes, except in the function that implements the Transformer intreface10:42
nsajeIt is pretty self explanatory that if you have a function handle_sample(s), then the s denotes a sample :)10:42
ildikovnsaje: well, just because this is used in many parts of this code, I do not think that it is a good pattern that should be followed10:43
ildikovnsaje: I'm too old to be that naive that it can be corrected in the whole codebase, it's still an opensource community10:43
ildikovnsaje: the only thing I can do is to ask the authors to use normal variable names, as I like the clean code patterns and this one is useful, when you check a longer patch or you're totally new and would like to understand the code10:45
ildikovnsaje: so I will not -1 your patch even if you change back the '_sample' to 's', it was just a non-blocking comment for you to consider in the future, if you agree with me :)10:46
nsajeildikov: I do :)10:47
nsajeildikov: I'm actually thinking really hard what that 'm' in the __call__ really denotes10:47
nsajeildikov: so I can see the benefit of good naming first-hand :)10:48
ildikovnsaje: cool :)10:48
openstackgerritNejc Saje proposed a change to openstack/ceilometer: Multi meter arithmetic transformer  https://review.openstack.org/10630210:49
ildikovnsaje: there's still one 'm' in line 13810:50
nsajeildikov: darn, how the heck did I miss this one10:51
ildikovnsaje: it usually happens, when someone is in a hurry, I'm a master of these mistakes :)10:51
nsajeildikov: true :) did I botch anything else or is that the last 'm'? :D10:52
openstackgerritNejc Saje proposed a change to openstack/ceilometer: Multi meter arithmetic transformer  https://review.openstack.org/10630210:53
nsajeildikov: these are the 'I just search&replaced, no need to run tox' mistaked10:53
nsaje*mistakes10:53
ildikovnsaje: this last 'm' is on your bill ;)10:53
ildikovnsaje: yeap, now I'm fine with the fact that Murphy likes other people too, not just me :)10:54
ildikovnsaje: now it looks good to me, so sileht can have a look at, if he still feels like :)10:56
ildikovnsaje: I have to run to a quick meeting, thanks again for the quick trunaround on this really minor nit10:56
nsajeildikov: thanks for taking the time to review! laters!10:56
ildikovnsaje: np! laters! :)10:57
openstackgerritNejc Saje proposed a change to openstack/ceilometer: Add cumulative and gauge to aggregator transformer  https://review.openstack.org/10831510:59
*** eglynn has joined #openstack-ceilometer11:02
openstackgerritNejc Saje proposed a change to openstack/ceilometer: Adds Content-Type to alarm REST notifier  https://review.openstack.org/10834711:12
*** nosnos has joined #openstack-ceilometer11:13
openstackgerritIgor Degtiarov proposed a change to openstack/ceilometer: Implement redesigned separator in names of columns in HBase  https://review.openstack.org/10637611:25
*** ryanpetrello has quit IRC11:27
*** k4n0 has quit IRC11:28
*** ryanpetrello has joined #openstack-ceilometer11:32
*** ccrouch has joined #openstack-ceilometer11:34
*** nosnos has quit IRC11:40
*** tongli has joined #openstack-ceilometer11:48
*** psharma has quit IRC12:24
*** jdob has joined #openstack-ceilometer12:27
*** ccrouch has quit IRC12:31
openstackgerritA change was merged to openstack/ceilometer: Multi meter arithmetic transformer  https://review.openstack.org/10630212:36
*** ccrouch has joined #openstack-ceilometer12:45
*** gordc has joined #openstack-ceilometer12:48
eglynnnsaje: ^^^ \o/ :)12:50
nsajeeglynn: I'm in the middle of doing a ritual dance to the gating gods :)12:51
nsajeeglynn: they have shown us mercy today12:51
ildikovnsaje: LOL :)12:51
ildikovnsaje: you should be really thankful, I hope noone died for showing some blood to these gods ;)12:52
eglynnnsaje: quick let's propose a bunch of patches while they continue smiling upon us! ;)12:52
cdentdo not taunt happy fun gate gods!12:52
*** julim has joined #openstack-ceilometer12:56
*** ddieterly has quit IRC13:05
*** rwsu has joined #openstack-ceilometer13:08
*** heyongli has joined #openstack-ceilometer13:10
*** changbl has quit IRC13:23
*** russellb has joined #openstack-ceilometer13:24
russellbeglynn: o/  just wanted to check in on juno-2 whenever you have a minute13:30
*** heyongli has quit IRC13:30
eglynnrussellb: hey13:30
russellbhttps://launchpad.net/ceilometer/+milestone/juno-213:30
russellbthink all of those can merge by sometime tomorrow?13:30
eglynnrussellb: so gating has been slow due to the oslo-messaging 1.3.1 release still pending13:30
eglynnrussellb: but yeah, all BPs except possibly grenade show land13:31
* russellb nods13:31
eglynnrussellb: I'll start bumping in-progress bugs by EoD my time13:31
eglynnrussellb: i.e. midday your's13:31
russellbOK, ideally would like to tag sometime tomorrow, so we can talk again tomorrow to see how things look13:31
russellblet me know if i can do anything13:31
eglynnrussellb: BTW the grenade BP will not be included in the tag anyway, as the code isn't in our tree13:32
eglynnrussellb: thanks, all good!13:32
russellbah, one of those, yes13:32
russellbyou could potentially use the "ongoing" milestone for that13:32
eglynnrussellb: BTW I presume we're still skipping the project/release status meeting this evening?13:32
russellbdownside is it's easier to not see it13:33
russellbyep, still skipping meetings today13:33
eglynnrussellb: yeap, good point ... I'll touch base with the QA folks in any case13:33
russellbcool13:33
eglynnrussellb: (they were in conclave at mid-cycle all last week)13:33
russellbah yes :)13:33
russellbsounded fun, actually13:34
eglynnyeah, surprisingly for Darmstadt ;)13:34
eglynnDarmstadt != Vegas13:34
russellbheh, i just heard "germany"13:34
*** _nadya_ has joined #openstack-ceilometer13:35
eglynn:)13:35
russellband i also saw mention of biergarten.13:35
russellbso it couldn't have been that bad13:35
eglynnah yeah, I was just jesting ... no doubt they enjoyed the post-worldcup buzz also13:37
*** sbfox has joined #openstack-ceilometer13:37
* eglynn would prefer to have been in Darmstadt last week than Buenos Aires ;)13:38
openstackgerritA change was merged to openstack/ceilometer: Adds Content-Type to alarm REST notifier  https://review.openstack.org/10834713:42
openstackgerritOpenStack Proposal Bot proposed a change to openstack/ceilometer: Updated from global requirements  https://review.openstack.org/10302913:42
*** ddieterly has joined #openstack-ceilometer13:44
*** raymondr has joined #openstack-ceilometer13:45
*** prad has joined #openstack-ceilometer13:47
*** raymondr has quit IRC13:50
*** amalagon has joined #openstack-ceilometer13:50
*** raymondr has joined #openstack-ceilometer13:51
tongliDoes anyone know if there are any keystone message listeners to process keystone messages on the queue?13:51
tongliI mean to convert keystone messages to meters.13:51
*** rbowen has joined #openstack-ceilometer13:51
gordctongli: it was added recently.13:53
gordchttps://github.com/openstack/ceilometer/tree/master/ceilometer/identity13:53
gordctongli: you'll probably need to configure it for specific topics.13:53
tongligordon, I was talking about the notification handler.13:55
*** ddieterly has quit IRC13:55
tonglithe code actually convert the message to meter.13:55
tongliI think that code you point to only tells ceilometer to listen on keystone exchange.13:56
*** thomasem has joined #openstack-ceilometer13:56
tonglibut for messages in keystone exchange, there needs to be some code to handle the keystone messages13:56
*** thomasem has quit IRC13:57
gordctongli: right, you'll need to setup which messages you want to process...13:57
tonglisuch as event_type = "identity.authenticate"13:57
tongligordc, we already have a lot handlers for compute events. but not for keystone messages. even though Ceilometer listens on keystone exchange.13:58
gordctongli: so is the question whether anyone else is going to implement audit message handling portion?14:00
gordctongli: i don't think anyone is working on it currently.14:00
*** _nadya_ has quit IRC14:00
tongli@gordc, it is not even audit message, it is the keystone messages as a whole , there is no message handler to handle keystone messages on the queue.14:01
*** thomasem has joined #openstack-ceilometer14:03
gordctongli: ...right. which is why i mention you'll need setup/configure more than what currently exists.14:03
gordcyou can check with boris-42 to see if he plans to implement complete solution... he put in original commit14:04
tongligordc, thanks for the confirmation.14:04
gordctongli: np, if you look at commit message it mentions audit notification... not sure status on that.14:05
*** ddieterly has joined #openstack-ceilometer14:09
cdentildikov: I've responded to your ironic review comments  with some questions could you and perhaps eglynn hop back in quickly? https://review.openstack.org/#/c/105486  thanks!14:13
*** raymondr has quit IRC14:13
*** ccrouch has quit IRC14:14
*** r3pl4y has quit IRC14:15
*** _nadya_ has joined #openstack-ceilometer14:16
*** raymondr_ has joined #openstack-ceilometer14:18
ildikovcdent: I'm checking14:18
ildikovcdent: as for logging, gordc had a BP spec for a logging guide for Ceilometer14:19
dmsimardHey guys. I have a use case where I'd like to run multiple collectors on one server (several standalone rabbitmq instances). What do you believe would be the cleanest way to do that ?14:20
*** ccrouch has joined #openstack-ceilometer14:20
ildikovcdent: my concern there is that if we log something it should be useful for debugging, but that one line does not have too many information in it14:20
dmsimardI would probably have a ceilometer.conf file for each rabbitmq instance and launch a ceilometer-collector process/daemon per file14:21
ildikovgordc: if you have a few sec, could you take a look at the warning log in this file: https://review.openstack.org/#/c/105486/13/ceilometer/hardware/notifications/ipmi.py ?14:22
cdentyeah, I get that ildikov, I'm just not sure how best to resolve it14:26
*** mrmoje has joined #openstack-ceilometer14:27
ildikovcdent: yeap sure, I'm thinking on the best first step solution here14:28
jd__eglynn: if you missed it I fixed all the so-far-found-bugs in carbonara new aggregation code14:31
jd__just sayin'14:31
*** raymondr_ has quit IRC14:32
eglynnjd__: cool enough ... I was hoping to get to it yesterday, slipped into today14:32
jd__ack14:32
eglynnjd__: ... will defo give it a whirl before EoD14:32
*** raymondr_ has joined #openstack-ceilometer14:32
*** sbfox has quit IRC14:35
*** raymond__ has joined #openstack-ceilometer14:35
*** raymondr_ has quit IRC14:36
*** Longgeek_ has joined #openstack-ceilometer14:37
*** Longgeek_ has quit IRC14:37
openstackgerritOpenStack Proposal Bot proposed a change to openstack/ceilometer: Updated from global requirements  https://review.openstack.org/10302914:37
*** r3pl4y has joined #openstack-ceilometer14:39
*** Longgeek has quit IRC14:39
*** raymond__ has quit IRC14:40
*** raymondr_ has joined #openstack-ceilometer14:40
ildikovcdent: I tried to answer your questions, please let me know if I could or could not14:40
cdentthanks will take a look14:41
*** _nadya_ has quit IRC14:42
openstackgerritPradeep Kilambi proposed a change to openstack/ceilometer: Support for metering FWaaS  https://review.openstack.org/10687714:43
*** raymondr_ has quit IRC14:44
*** raymondr_ has joined #openstack-ceilometer14:45
cdentildikov, eglynn the advice on the logging is clear and I think I know what to do, but I'm still unsure about the test data. My preference at this point would be to keep the data as is and move it to another file14:48
cdentUsing a loop to generate it, may as well just generate the keys we care about and not any of the other stuff14:48
cdentin which case, what are we really testing?14:48
eglynncdent: isn't the python object created exactly the same, even if generated in a loop?14:49
eglynncdent: so wouldn't be testing exactly what we're testing currently?14:49
*** raymondr_ has quit IRC14:50
* cdent shrugs14:50
eglynn(... or do the sensor data differ in more subtle ways that I didn't catch on a quick scan by eye)14:50
*** raymondr_ has joined #openstack-ceilometer14:50
cdentit all seems a bit artificial, creating data14:50
cdenteven if the data is the same14:50
cdentit doesn't really matter, but it just strikes me as kind of cackhanded14:50
eglynnyeah14:51
ildikovcdent: I think it's better if it's readable14:51
cdentif I really wanted to do it the way I think it ought to be done I would take a _real_ sample from the ipmitool driver in ironic as the basis14:51
eglynnI guess a little artificialness is par for the course in a unit test?14:51
eglynn(esp. with the additional readability)14:51
ildikovcdent: and that is an example, so if we pretend that we had less sensor in our non-existing env used for example for the testing it should be a good deal14:52
cdenteglynn: not the way I do unit tests, but I sense there is a big difference between what I do and what is tradtionally done in openstack (what with mocks-- and everything)14:52
cdentildikov: I'm sorry I'm not really following that point?14:52
eglynncdent: OK taking actual ooutput from ipmitool would work also, you could just dump in a separate file as ildikov suggested?14:52
* eglynn though your test data was already indirectly taken from ipmitool?14:53
ildikovcdent: I meant if we use for testing only that data that we really check and not all the data from *every* sensor14:53
cdentthe point of the test is to make sure that it handles real data, right?14:53
cdentthat is: if we feed it data X, it doesn't blow up14:53
ildikoveglynn: cdent: but if we would like to keep all, I would definitely vote for a separate file14:53
cdentI'll go for a separate file14:54
eglynncdent: yeah agreed, and you have "real" data already, or?14:54
cdentreal-ish14:54
*** raymondr_ has quit IRC14:54
cdentthere's apparently a difference between the ipmitool that Haomeng used to create his sample and the one I've been using to run my manual tests14:54
eglynncdent: ... on the mocks issue, that pattern is used pretty extensively in openstack tests, but I don't see a huge need in this case14:54
cdent(mine currently breaks his code)14:55
*** raymondr_ has joined #openstack-ceilometer14:55
cdentyeah, was just moaning about mocks in general as a bad smell14:55
eglynncdent: ... i.e. sense the notification handler interaction is pretty much notification-in, sample-out, no other interactions?14:55
eglynn*since14:55
* cdent nods14:55
eglynncdent: +1 on using more "realistic" data if you can get it using Haomengs latest patch14:56
eglynncdent: WRT mocks, a-ha, I thought you'd be a fan of such things :)14:57
cdenthowever did I give you that impression?14:57
eglynncdent: bad smell ==> indirect interactions?14:57
ildikovcdent: sure, realistic data is good to have14:57
eglynnBTW we've a few other juno-2 BP-related reviews that need a +1 to get them over the line by EoD14:59
eglynn(prad) networking services *aaS: https://review.openstack.org/108476 https://review.openstack.org/103956 & https://review.openstack.org/10687714:59
*** raymondr_ has quit IRC14:59
eglynnso any cores with some available cycles, would be great to get those in by EoD14:59
*** raymondr_ has joined #openstack-ceilometer14:59
eglynnsileht: I wonder is the gate stuck on "Move hbase alarms driver code to alarm tree" https://review.openstack.org/107942 ?15:00
gordceglynn: EoD your time or mine?15:01
eglynngordc: your's :)15:01
gordceglynn: cool cool. will have a look in a bit.15:01
eglynngordc: thank you sir!15:01
pradthx gordc15:01
*** raymond__ has joined #openstack-ceilometer15:03
*** nati_ueno has quit IRC15:04
*** raymondr_ has quit IRC15:04
*** nati_ueno has joined #openstack-ceilometer15:05
*** rbowen has quit IRC15:05
*** raymondr_ has joined #openstack-ceilometer15:08
*** raymond__ has quit IRC15:08
* eglynn checks with the QA folks why https://review.openstack.org/107942 hasn't gated15:09
*** nacim has quit IRC15:09
*** nati_ueno has quit IRC15:10
*** raymondr_ has quit IRC15:12
*** raymondr_ has joined #openstack-ceilometer15:13
*** raymond__ has joined #openstack-ceilometer15:15
*** raymondr_ has quit IRC15:17
*** rbowen has joined #openstack-ceilometer15:21
*** promulo has quit IRC15:21
*** cdent_ has joined #openstack-ceilometer15:23
*** eoutin has quit IRC15:25
*** cdent has quit IRC15:25
*** cdent_ is now known as cdent15:25
*** yatin has quit IRC15:25
*** rbowen has quit IRC15:26
pradgordc, would you ok, if i addressed your comments as a follow up patch?15:27
gordcprad: sure. could you reply to comments first?15:30
gordcprad: i know the pain of rebasing.15:31
pradgordc, yea doing as we speak15:31
gordccool cool15:31
gordcprad: stupid question for you... what is ipsec in regards to vpnaas? is there a link somewhere?15:33
pradgordc, the ipsec conns metric gives us the vpn tunnels15:34
gordcprad: ah, i see. thanks for clarifying15:34
pradgordc, replied to your comments15:35
pradgordc, you'll probably have similar comments for vpn and fw patches.. I can submit a follow up patch addressing all three in one.15:36
gordcprad: cool cool. saves me hassle of commenting :)15:36
pradif thats ok with you.. yea since all three are dependent on each other.. rebase is gonna be hell :)15:36
openstackgerritNejc Saje proposed a change to openstack/ceilometer: Add cumulative and gauge to aggregator transformer  https://review.openstack.org/10831515:38
*** changbl has joined #openstack-ceilometer15:40
*** _nadya_ has joined #openstack-ceilometer15:48
openstackgerritChris Dent proposed a change to openstack/ceilometer: Implement consuming ipmi notifications from Ironic  https://review.openstack.org/10548615:58
*** eglynn is now known as eglynn-afk15:58
cdent^ that, eglynn and ildikov, is a bigger chanage than I would have preferred, but gets the logs more meaningful and the test data into it's own module15:58
*** _nadya_ has quit IRC15:58
*** IvanBerezovskiy has left #openstack-ceilometer16:04
*** promulo has joined #openstack-ceilometer16:06
*** sbfox has joined #openstack-ceilometer16:08
*** shengjiemin has quit IRC16:09
*** shengjiemin has joined #openstack-ceilometer16:12
*** ildikov has quit IRC16:14
*** eglynn-afk is now known as eglynn16:18
*** mrmoje has quit IRC16:21
*** _cjones_ has joined #openstack-ceilometer16:22
*** fnaval has joined #openstack-ceilometer16:24
*** drjones has joined #openstack-ceilometer16:24
*** _cjones_ has quit IRC16:27
*** sbfox1 has joined #openstack-ceilometer16:30
*** sbfox has quit IRC16:31
*** ildikov has joined #openstack-ceilometer16:34
*** ityaptin has quit IRC16:34
openstackgerritA change was merged to openstack/ceilometer: Use resource discovery for Network Services  https://review.openstack.org/10395616:37
openstackgerritOpenStack Proposal Bot proposed a change to openstack/ceilometer: Updated from global requirements  https://review.openstack.org/10302916:39
openstackgerritA change was merged to openstack/ceilometer: Add support for metering VPNaaS  https://review.openstack.org/10847616:40
eglynnprad: ^^^ \o/ :)16:41
pradyay, thx eglynn and others for reviews16:44
pradthx gordc for approving, i have the follow up commit addressing your comments coming up16:45
eglynnprad: fingers crossed https://review.openstack.org/106877 will gate at the first time of asking16:45
pradyea hopefully16:47
*** safchain has quit IRC16:53
openstackgerritPradeep Kilambi proposed a change to openstack/ceilometer: Follow up fixes to network services pollsters  https://review.openstack.org/10877016:57
pradgordc, DinaBelova, ^^ this commit should address your comments16:59
*** thomasem_ has joined #openstack-ceilometer17:00
*** sbfox1 has quit IRC17:01
*** sbfox has joined #openstack-ceilometer17:01
openstackgerritA change was merged to openstack/ceilometer: Support for metering FWaaS  https://review.openstack.org/10687717:02
*** nati_ueno has joined #openstack-ceilometer17:03
*** thomasem has quit IRC17:04
*** r3pl4y has quit IRC17:12
openstackgerritgordon chung proposed a change to openstack/ceilometer: fix indices in event tables  https://review.openstack.org/9150117:13
*** renatoarmani has joined #openstack-ceilometer17:18
*** harlowja_away is now known as harlowja17:19
*** renatoarmani has quit IRC17:27
*** raymond__ has quit IRC17:28
*** renatoarmani has joined #openstack-ceilometer17:30
*** eglynn is now known as eglynn-afk17:31
*** raymondr_ has joined #openstack-ceilometer17:38
*** raymondr_ has quit IRC17:42
*** raymondr_ has joined #openstack-ceilometer17:42
openstackgerritgordon chung proposed a change to openstack/ceilometer: skip lbaas polls when neutron not configured  https://review.openstack.org/10489217:43
*** raymondr_ has quit IRC17:47
*** sbfox has quit IRC17:53
*** ccrouch has quit IRC17:54
*** ccrouch has joined #openstack-ceilometer17:57
*** cdent has quit IRC18:02
*** raymondr_ has joined #openstack-ceilometer18:03
openstackgerritA change was merged to openstack/ceilometer: Implement consuming ipmi notifications from Ironic  https://review.openstack.org/10548618:14
*** terriyu has joined #openstack-ceilometer18:18
*** shengjiemin has quit IRC18:19
*** terriyu has quit IRC18:21
*** terriyu has joined #openstack-ceilometer18:21
*** Fabio_ has quit IRC18:24
gordcok. is ayoung owning it or do i have to move it in myself?18:31
*** mrmoje has joined #openstack-ceilometer18:33
*** renatoarmani has quit IRC18:34
openstackgerritPradeep Kilambi proposed a change to openstack/ceilometer: Follow up fixes to network services pollsters  https://review.openstack.org/10877018:34
*** sbfox has joined #openstack-ceilometer18:34
*** bhdeeps has joined #openstack-ceilometer18:34
*** renatoarmani has joined #openstack-ceilometer18:39
gordcerr... everyone can ignore that previous msg. figuring out how to use irc. :)18:43
*** jogo has left #openstack-ceilometer18:50
*** renatoar_ has joined #openstack-ceilometer18:55
*** renatoarmani has quit IRC18:56
*** shakamunyi has joined #openstack-ceilometer18:56
*** nati_ueno has quit IRC19:07
*** raymondr_ has quit IRC19:08
*** eglynn-afk has quit IRC19:11
openstackgerritPradeep Kilambi proposed a change to openstack/ceilometer: Update measurement docs for VPNaaS  https://review.openstack.org/10881419:12
*** ccrouch has quit IRC19:12
openstackgerritPradeep Kilambi proposed a change to openstack/ceilometer: Update measurement docs for FWaaS  https://review.openstack.org/10881519:15
*** raymondr has joined #openstack-ceilometer19:18
*** ccrouch has joined #openstack-ceilometer19:20
*** raymondr_ has joined #openstack-ceilometer19:20
*** raymondr has quit IRC19:22
*** renatoar_ has quit IRC19:26
openstackgerritA change was merged to openstack/ceilometer: Fix SQL exception getting statitics with metaquery  https://review.openstack.org/10215019:28
*** renatoar_ has joined #openstack-ceilometer19:29
tonglihello, boris and I were talking about turning keystone messages to meters.19:37
tonglikeystone can send a lot of messages onto the queue.19:37
tonglihere is the list of messages it can send.19:37
tonglihttp://docs.openstack.org/developer/keystone/event_notifications.html19:37
tongliI would like to get an idea on if you guys have any opinion on these messages being converted to one meter or different meters?19:38
tonglihello. anyone has any say on this?19:40
tongli@boris-42, it does not seem to be any takers, I think we probably can put this on the ceilometer next meeting agenda?19:41
gordctongli: you'd get a better response during European office hours.19:42
*** thomasem_ has quit IRC19:42
*** shakamunyi has quit IRC19:42
boris-42tongli +119:42
*** thomasem has joined #openstack-ceilometer19:42
boris-42or +2=)19:42
gordctongli: if it's just crud-type events i'd assume you could do something similar to what heat did19:43
tongli@gordc, yeah, let me put it on the ceilometer meeting agenda.19:43
gordchttps://github.com/openstack/ceilometer/blob/master/ceilometer/orchestration/notifications.py19:43
tongli@gordc, seems it create one meter per action.19:45
tongliit becomes stack.create, stack.update, etc.19:45
tongliwe can certainly do something like that.19:46
tongliBoris-42, are you ok with that?19:46
boris-42tongli yep seems ok19:49
tongli@boris-42, cool, let me create a patch for it. thanks.19:49
boris-42tongli just ping me for review19:49
boris-42tongli ok?)19:49
*** sdake_ has joined #openstack-ceilometer19:50
*** sdake has quit IRC19:51
tongli@boris-42, absolutely, will do that, thanks.19:54
*** renatoar_ has quit IRC19:58
*** thomasem has quit IRC19:59
*** thomasem has joined #openstack-ceilometer20:01
*** fnaval has quit IRC20:01
*** sdake_ has quit IRC20:03
*** sbfox has quit IRC20:06
*** fnaval has joined #openstack-ceilometer20:13
*** sdake has joined #openstack-ceilometer20:15
*** renatoarmani has joined #openstack-ceilometer20:16
*** eglynn-afk has joined #openstack-ceilometer20:19
*** sbfox has joined #openstack-ceilometer20:20
pradeglynn-afk, gordc, could you please reap prove this https://review.openstack.org/#/c/108770/ had a pep8 issue so pushed another version20:21
*** amalagon has quit IRC20:23
eglynn-afkprad: looking20:24
pradthx20:24
eglynn-afkprad: done20:24
pradty sir20:24
eglynn-afknp! :)20:25
*** eglynn-afk is now known as eglynn20:25
gordcprad: approved20:26
pradthx gordc !20:26
openstackgerritPradeep Kilambi proposed a change to openstack/ceilometer: Update measurement docs for VPNaaS  https://review.openstack.org/10881420:28
openstackgerritPradeep Kilambi proposed a change to openstack/ceilometer: Update measurement docs for FWaaS  https://review.openstack.org/10881520:29
openstackgerritPradeep Kilambi proposed a change to openstack/ceilometer: Update measurement docs for FWaaS  https://review.openstack.org/10881520:31
eglynnlast bug we need to land for juno-2: https://review.openstack.org/99905 just needs a +1 to get it over the line20:48
eglynn^^^ any cores still online?20:52
gordceglynn: looking now20:53
eglynngordc: thank you sir!20:53
openstackgerritA change was merged to openstack/pycadf: Remove docutils pin  https://review.openstack.org/10562620:54
openstackgerritA change was merged to openstack/pycadf: Bump hacking to version 0.9.2  https://review.openstack.org/10734620:57
*** jdob has quit IRC20:57
*** promulo has quit IRC21:12
*** sbfox1 has joined #openstack-ceilometer21:12
openstackgerritPradeep Kilambi proposed a change to openstack/ceilometer: Update measurement docs for VPNaaS  https://review.openstack.org/10881421:12
*** thomasem has quit IRC21:13
openstackgerritPradeep Kilambi proposed a change to openstack/ceilometer: Update measurement docs for FWaaS  https://review.openstack.org/10881521:14
*** sbfox1 has quit IRC21:14
*** sbfox1 has joined #openstack-ceilometer21:15
*** sbfox has quit IRC21:15
*** promulo has joined #openstack-ceilometer21:26
*** dmsimard is now known as dmsimard_away21:26
*** shakamunyi has joined #openstack-ceilometer21:33
openstackgerritA change was merged to openstack/ceilometer: Follow up fixes to network services pollsters  https://review.openstack.org/10877021:34
*** thomasem has joined #openstack-ceilometer21:44
*** shakamunyi has quit IRC21:56
*** sbfox1 has quit IRC21:58
*** tongli has quit IRC21:59
*** sbfox has joined #openstack-ceilometer21:59
*** rwsu has quit IRC22:02
*** changbl has quit IRC22:05
*** julim has quit IRC22:09
*** nati_ueno has joined #openstack-ceilometer22:18
*** terriyu has quit IRC22:18
*** renatoarmani has quit IRC22:33
*** ddieterly has quit IRC22:34
*** eglynn has quit IRC22:38
*** gordc has quit IRC22:38
*** raymondr_ has quit IRC22:40
*** ccrouch has quit IRC22:41
*** thomasem has quit IRC22:54
*** thomasem_ has joined #openstack-ceilometer22:54
*** thomasem_ has quit IRC22:59
openstackgerritA change was merged to openstack/python-ceilometerclient: Calculate a suitable column width for positional arguments  https://review.openstack.org/9788123:10
*** shengjiemin has joined #openstack-ceilometer23:10
*** prad has quit IRC23:12
*** thomasem has joined #openstack-ceilometer23:13
*** fnaval has quit IRC23:19
*** sbfox has quit IRC23:21
*** nati_ueno has quit IRC23:22
openstackgerritA change was merged to openstack/python-ceilometerclient: Improve --debug logging output  https://review.openstack.org/10207623:23
*** thomasem has quit IRC23:41
*** nati_ueno has joined #openstack-ceilometer23:49
*** nati_ueno has quit IRC23:54
*** fnaval has joined #openstack-ceilometer23:58
*** shengjiemin has quit IRC23:59

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