08:00:17 <alexchadin> #startmeeting watcher
08:00:18 <openstack> Meeting started Wed Sep 26 08:00:17 2018 UTC and is due to finish in 60 minutes.  The chair is alexchadin. Information about MeetBot at http://wiki.debian.org/MeetBot.
08:00:19 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
08:00:22 <openstack> The meeting name has been set to 'watcher'
08:01:19 <alexchadin> o/
08:01:42 <TuanVu> Hi Alex
08:01:45 <TuanVu> have a nice day :)
08:02:15 <alexchadin> TuanVu: you too!
08:02:20 <alexchadin> TuanVu: how are you?
08:02:39 <TuanVu> I'm doing fine, thank you very much :D
08:03:36 <TuanVu> I'm doing some tests for watcher live migration feature in my local environment
08:03:57 <alexchadin> TuanVu: what strategy do you use?
08:05:36 <TuanVu> I intend to test the ability to migrate VM from 1 compute node to another when host's workload (eg: CPU) is too high
08:06:10 <TuanVu> it should be "workload balance migration" strategy, right?
08:06:51 <alexchadin> TuanVu: I'd recommend you to use workload stabilization strategy
08:07:32 <alexchadin> TuanVu: we test this strategy on production clusters
08:08:17 <TuanVu> thank you, Alex
08:08:43 <TuanVu> however, I don't see your mentioned strategy on this list
08:08:45 <TuanVu> https://docs.openstack.org/watcher/latest/strategies/index.html
08:09:03 <TuanVu> could you please point out exactly it is which one?
08:09:25 <alexchadin> TuanVu: Workload Overload standard deviation
08:10:30 <TuanVu> awesome! thanks
08:11:07 <TuanVu> as I can see, it requires metrics from ceilometer
08:11:29 <TuanVu> are there anyway for it to work with metric from "Monasca"?
08:11:38 <alexchadin> TuanVu: that's right, it's recommended to use ceilometer+gnocchi
08:12:56 <alexchadin> TuanVu: I haven't tested this strategy with Monasca yet, I'm not sure if Monasca provides SNMP metrics
08:13:43 <TuanVu> thanks for your recommendation, I'll test this strategy in my environment
08:16:08 <alexchadin> TuanVu: do you have any other question regarding Watcher?
08:16:30 <TuanVu> that's all I have for now :)
08:17:52 <alexchadin> TuanVu: you don't use ceilometer+gnocchi in your env?
08:19:13 <TuanVu> yes, I do use ceilometer+gnocchi, but just want to test out monasca as well
08:21:09 <alexchadin> TuanVu: great :) this strategy works with CPU and RAM metrics. You can use them both or separately
08:21:41 <TuanVu> thank you, that's awesome!
08:25:12 <alexchadin> TuanVu: unfortunately, other core developers haven't come today on meeting
08:27:05 <TuanVu> so ... we can finish the meeting here, I guess?
08:29:00 <alexchadin> TuanVu: I suppose so
08:29:08 <alexchadin> TuanVu: thank you for attending!
08:29:37 <TuanVu> it's my pleasure :) Thanks for your sharing, Alex
08:29:57 <TuanVu> it means a lot for me, I really appreciate your kindness
08:30:15 <TuanVu> see you next time :)
08:30:23 <alexchadin> TuanVu: see you! :)
08:30:27 <alexchadin> #endmeeting