Wednesday, 2017-03-22

*** _d34dh0r53_ has quit IRC00:14
*** d34dh0r53 has joined #openstack-watcher00:20
*** karthikpr has quit IRC00:42
*** harlowja has joined #openstack-watcher00:42
*** karthikpr has joined #openstack-watcher00:42
*** karthikpr has quit IRC00:46
*** zhurong has joined #openstack-watcher00:53
*** karthikpr has joined #openstack-watcher00:53
*** karthikpr has quit IRC00:57
*** karthikpr has joined #openstack-watcher01:11
*** karthikpr has quit IRC01:15
*** sanfern has quit IRC01:19
*** karthikpr has joined #openstack-watcher01:40
*** sanfern has joined #openstack-watcher02:55
*** yuanying has quit IRC03:05
openstackgerritlicanwei proposed openstack/watcher master: exception when running 'watcher service list'
*** harlowja has quit IRC03:35
*** adisky_ has joined #openstack-watcher04:14
*** yuanying has joined #openstack-watcher04:18
*** zhurong has quit IRC04:37
*** zhurong has joined #openstack-watcher05:32
*** karthikpr has quit IRC05:43
*** karthikpr has joined #openstack-watcher05:43
*** karthikpr has quit IRC05:47
openstackgerritAnanth Narayan S proposed openstack/watcher-specs master: Support visualizing workload fingerprints
openstackgerritAnanth Narayan S proposed openstack/watcher-specs master: Support workload fingerprinting
openstackgerritAnanth Narayan S proposed openstack/watcher master: Support workload fingerprint generation
openstackgerritSanthosh Fernandes proposed openstack/watcher master: Add period input parameter to vm workload consolidation and outlet temp control strategy.
*** alexchadin has joined #openstack-watcher06:59
*** hidekazu has joined #openstack-watcher07:53
*** zhurong has quit IRC08:00
alexchadinchrisspencer_: hi08:01
openstackgerrityanxubin proposed openstack/watcher master: Remove log translations
*** zhurong has joined #openstack-watcher08:11
*** sanfern has quit IRC08:25
*** sanfern has joined #openstack-watcher08:26
*** vincentfrancoise has joined #openstack-watcher08:31
alexchadinhi sanfern08:39
vincentfrancoisealexchadin: hi08:40
alexchadinhi vincentfrancoise08:40
vincentfrancoisealexchadin: just as a quick note for today's meeting08:40
sanfernhi alexchadin08:40
vincentfrancoisealexchadin: can you add an annoucement for the time change which means people need to be careful for next week?08:40
alexchadinsanfern: I've seen that you have had a discussion with Prudhvi Rao Shedimbi and would like to take watcher-multi-datasource bp08:41
sanfernalexchadin, what is the time line?08:42
alexchadinvincentfrancoise: there is no time change in our country. you will have +1 hour?08:42
vincentfrancoisealexchadin: yeah08:43
vincentfrancoisealexchadin: good for you BTW, time change is always a mess08:43
alexchadinsanfern: you mean milestone target?08:43
sanfernalexchadin, yes08:43
alexchadinvincentfrancoise: of course, I'll do announcement08:43
alexchadinsanfern: pike-308:43
sanfernalexchadin, ok I can take it08:44
alexchadinvincentfrancoise: there was cancellation of winter time change about 4 or 5 years ago08:45
alexchadinvincentfrancoise: so we don't care about it anymore :)08:45
sanfernalexchadin, should we wait till then gnocchi support?08:46
alexchadinsanfern: yeah, I suggest to end work with gnocchi and then that bp08:47
alexchadinthen take*08:47
*** pksingh has joined #openstack-watcher08:47
pksinghvincentfrancoise: Hello,08:47
sanfernalexchadin, ok08:48
alexchadinfolks, I'll be back in 20 minutes08:48
*** karthikpr has joined #openstack-watcher08:53
*** karthikpr has quit IRC08:58
vincentfrancoisepksingh: will be back in 20mn too09:09
pksinghvincentfrancoise: sure09:09
pksinghvincentfrancoise: i had a question related to test_model unit test09:10
*** alexchadin has quit IRC09:20
*** alexchadin has joined #openstack-watcher09:27
*** yuanying has quit IRC09:31
vincentfrancoisepksingh: i'm back09:40
pksinghvincentfrancoise: i had a question related to test case you already pointed out yesterday09:40
pksinghvincentfrancoise: you were correct, dict is not matching09:41
vincentfrancoisepksingh: :p09:41
pksinghvincentfrancoise: how about using, instead of comapring XML strings09:41
vincentfrancoisepksingh: if it works I'm good with it09:42
pksinghvincentfrancoise: XML part is just for testing correct?09:42
vincentfrancoisepksingh: as long as there's something that works09:42
vincentfrancoisepksingh: it's also used for debugging09:43
pksinghvincentfrancoise: i mean to say, we convert to/from xml only for unit testing and debugging?09:43
vincentfrancoisepksingh: it's logged at the beginning and at the end of all strategies09:43
vincentfrancoisepksingh: yeah pretty much09:43
pksinghvincentfrancoise: why dont we use other serialization format like yaml or json?09:44
vincentfrancoisepksingh: because I happened to do it this way and it stuck :p09:44
pksinghvincentfrancoise: OK, its little bit hard for me when you involve complex structure in xml09:46
vincentfrancoisepksingh: that feeling depends on who you speak to09:47
pksinghvincentfrancoise: :)09:48
vincentfrancoisepksingh: the need was to be able to produce a human readable representation of the CDM09:48
vincentfrancoisepksingh: the CDMs are serialized as part of the guru meditation report as well BTW09:48
pksinghvincentfrancoise: OK, i will check if i can find other way to compare the graphs09:49
vincentfrancoisepksingh: then for the tests09:49
vincentfrancoisepksingh: the idea was to simplify their readability09:50
vincentfrancoisepksingh: because trust me, reading nested for loops with slight changes between scenarios was getting too hard to debug09:51
*** zhurong has quit IRC09:52
pksinghvincentfrancoise: i understood the problem, may be json was easy to program but less verbose than xml09:53
vincentfrancoisepksingh: the thing I like with XML in that case over JSON is that each node is typed.09:54
pksinghvincentfrancoise: true09:54
*** alexchadin has quit IRC10:07
*** yuanying has joined #openstack-watcher10:11
openstackgerritTuan Luong-Anh proposed openstack/python-watcherclient master: Indicating the location tests directory in oslo_debug_helper
pksinghvincentfrancoise: in file,, there are instances which do not belong to any node10:40
pksinghvincentfrancoise: any specific reason to test these?10:40
vincentfrancoisepksingh: yes10:40
vincentfrancoisepksingh: for a short while (~10s) a new instance is not yet physically created and Nova still has to schedule its creation10:41
vincentfrancoisepksingh: Wacher receives notification of its existence at this stage too10:41
vincentfrancoisepksingh: so we have to represent it somehow10:41
pksinghvincentfrancoise: ok, what if we dont put in CDM?10:42
vincentfrancoisepksingh: graph-wise, that means an edgeless node10:42
vincentfrancoisepksingh: that's arguable I agree, but what would be the incentive towards removing this piece of information?10:43
pksinghvincentfrancoise: i was thinking we should capture the notification when node is created and assigned to some node10:44
vincentfrancoisepksingh: yes but I wanted to show in the data model instances that were in a failed state because the scheduling failed10:46
vincentfrancoisepksingh: in which case they will never be  physically created10:46
pksinghvincentfrancoise: would they help in achieving any goal?10:46
vincentfrancoisepksingh: people might need it so when I implemented it I decided to show it in the first place10:47
vincentfrancoisepksingh: I don't know eeither way, so having the information is the one state I knew would statisfy both10:47
pksinghvincentfrancoise: ok10:47
vincentfrancoisepksingh: although I agree it adds some noise if you don't actually need it10:48
pksinghvincentfrancoise: BTW isomorphism thing is working :)10:48
vincentfrancoisepksingh: cool :)10:48
*** pksingh has quit IRC10:50
*** sanfern has quit IRC10:54
*** karthikpr has joined #openstack-watcher11:39
*** karthikpr has quit IRC11:43
*** sanfern has joined #openstack-watcher11:50
*** alexchadin has joined #openstack-watcher11:54
*** alexchadin has quit IRC12:10
*** alexchadin has joined #openstack-watcher12:12
*** alexchadin has quit IRC12:17
*** alexchadin has joined #openstack-watcher12:19
alexchadinvincentfrancoise: ping12:33
adisky_alexchadin: hi12:51
alexchadinhi adisky_12:52
adisky_alexchadin: I found that python-watcherclient is not included in rdo repo of  python-openstackclient12:53
adisky_and we don't have rmp package for python-watcher-client12:54
adisky_alexchadin: I think u must be busy in meeting preps, we will discuss later on12:54
alexchadinadisky_: let's add it to meeting agenda12:55
adisky_alexchadin: ok12:55
pshedimbalexchadin, Where is the meeting? Or am I early by an hour?13:02
alexchadinpshedimb: Meeting will be in one hour :)13:02
pshedimbAh okay. Daylight savings confused me ;)13:03
*** dtardivel has joined #openstack-watcher13:14
alexchadinfolks, here is agenda:
openstackgerritAnanth Narayan S proposed openstack/watcher-dashboard master: [WIP] Support visualization of workload fingerprint
alexchadinif you won't mind, I will start meeting from discussions13:29
sballe_sounds good13:37
*** alexchadin has quit IRC13:49
*** alexchadin has joined #openstack-watcher13:55
chrisspencer_alexchadin: hi13:55
alexchadinhi chrisspencer_13:56
alexchadinmeeting in 4 minutes!13:56
alexchadinhi sballe_13:56
vincentfrancoisewhich chan?13:56
*** ostroverkhov has joined #openstack-watcher13:57
vincentfrancoisealexchadin: thx13:57
*** karthikpr has joined #openstack-watcher14:04
*** karthikpr has quit IRC14:07
*** karthikpr has joined #openstack-watcher14:09
*** pksingh has joined #openstack-watcher14:48
*** ostroverkhov has quit IRC15:00
alexchadinwelcome here, zigo!15:00
alexchadinwhat is your time zone?15:01
zigoI live near Grenoble, in the mountains...15:01
alexchadinI live near Moscow, non-stop city :)15:03
alexchadinzigo: let's discuss packaging tomorrow15:03
chrisspencer_dtardivel: I hear you might have some experience with monasca, I've got some questions for you15:05
dtardivelHi chrisspencer_: Yes I deployed 1 year ago monasca in our lab ;)15:06
dtardivelchrisspencer_: hi, yes I deployed 1 year ago monasca in our lab ;)15:07
chrisspencer_alright, hoping you've seen the issues I'm running into15:08
sballe_dtardivel: is an expert :)15:08
chrisspencer_dtardivel: We've got a devstack setup and are using the monasca devstack plugin as well.  When stacking we run into a failure with monasca-thresh and the service is in the failed state15:09
openstackgerritPradeep Kumar Singh proposed openstack/watcher master: Prevent the migration of VM with 'optimize' False in VM metadata
pksinghchrisspencer_: similar to
pksinghvincentfrancoise: i updated patch set, please review when you have time :)15:12
chrisspencer_pksingh: yes, I came across your bug while trying to find answers.  It is also the same as:
pksinghchrisspencer_: I am not sure, but i read some where that it is due to memory issue, may be increasing RAM will solve it15:14
pksinghchrisspencer_: i could not allocate more ram since my laptop has only 16 GB15:14
dtardivelchrisspencer_: I did not deploy Monasca on devstack, but directly on our lab cluster15:15
chrisspencer_dtardivel: alright,  I'm not convinced it's a devstack issue, but it may be as pksingh said that it's a ram issue.  sballe_ do you know how much ram these systems have?15:16
pksinghchrisspencer_: according to Vagrant file, its 12GB
sballe_chrisspencer_: Not of hand15:18
chrisspencer_pksingh: I think the system we are running on has 256G, so that should be fine15:22
pksinghchrisspencer_: may be you can disable monasca-thresh, if you dont want it , thats a quick fix ;)15:24
vincentfrancoisepksingh: will do tomorrow I guess15:32
pksinghvincentfrancoise: sure, no problem15:32
pksinghvincentfrancoise: thanks for all your hel :)15:32
pksinghvincentfrancoise: help*15:33
*** pksingh has quit IRC15:34
*** harlowja has joined #openstack-watcher16:06
*** alexchadin has quit IRC16:22
openstackgerritMargarita Shakhova proposed openstack/watcher master: Add endpoint_type option for openstack clients. Interface type internalURL' is used as a default value.
*** vincentfrancoise has quit IRC16:49
openstackgerritMargarita Shakhova proposed openstack/watcher master: Add endpoint_type option for openstack clients. Interface type internalURL' is used as a default value.
*** wootehfoot has joined #openstack-watcher17:29
*** karthikpr has quit IRC17:33
*** karthikpr has joined #openstack-watcher17:54
*** sanfern has quit IRC18:49
*** sanfern has joined #openstack-watcher18:49
*** sanfern has quit IRC18:49
*** sanfern has joined #openstack-watcher18:50
*** sanfern has quit IRC18:50
*** sanfern has joined #openstack-watcher18:51
*** sanfern has quit IRC18:51
*** harlowja has quit IRC18:59
*** harlowja has joined #openstack-watcher19:00
*** harlowja has quit IRC19:01
*** harlowja has joined #openstack-watcher19:45
*** harlowja has quit IRC20:18
*** karthikpr has quit IRC20:20
*** karthikpr has joined #openstack-watcher20:20
*** karthik__ has joined #openstack-watcher20:23
*** karthikpr has quit IRC20:25
*** karthik__ has quit IRC20:27
*** karthikpr has joined #openstack-watcher20:45
*** wootehfoot has quit IRC21:09
*** wootehfoot has joined #openstack-watcher21:14
*** wootehfoot has quit IRC21:17
*** dtardivel has quit IRC21:20
*** karthikpr has quit IRC21:29
*** harlowja has joined #openstack-watcher21:33
*** pshedimb has quit IRC21:38
*** wootehfoot has joined #openstack-watcher21:42
*** karthikpr has joined #openstack-watcher22:54
*** karthikpr has quit IRC23:30
*** hidekazu has quit IRC23:31

Generated by 2.14.0 by Marius Gedminas - find it at!