| *** thorst_ has quit IRC | 00:14 | |
| *** thorst_ has joined #openstack-watcher | 00:15 | |
| *** thorst_ has quit IRC | 00:19 | |
| *** thorst_ has joined #openstack-watcher | 00:34 | |
| *** thorst_ has joined #openstack-watcher | 00:35 | |
| *** thorst_ has quit IRC | 00:43 | |
| *** eil397 has quit IRC | 01:00 | |
| *** eil397_ has quit IRC | 01:03 | |
| *** eil397_ has joined #openstack-watcher | 01:36 | |
| *** thorst_ has joined #openstack-watcher | 01:37 | |
| *** thorst_ has quit IRC | 02:25 | |
| *** thorst_ has joined #openstack-watcher | 02:25 | |
| *** eil397_ has quit IRC | 02:32 | |
| *** thorst_ has quit IRC | 02:34 | |
| *** hidekazu has joined #openstack-watcher | 02:52 | |
| *** thorst_ has joined #openstack-watcher | 03:04 | |
| *** thorst_ has quit IRC | 03:12 | |
| *** Yumeng has quit IRC | 03:15 | |
| *** thorst_ has joined #openstack-watcher | 03:29 | |
| *** thorst_ has quit IRC | 03:30 | |
| *** diga has joined #openstack-watcher | 03:52 | |
| *** thorst_ has joined #openstack-watcher | 04:31 | |
| *** thorst_ has quit IRC | 04:39 | |
| *** atuly has joined #openstack-watcher | 05:10 | |
| *** thorst_ has joined #openstack-watcher | 05:37 | |
| *** thorst_ has quit IRC | 05:45 | |
| *** thorst_ has joined #openstack-watcher | 06:42 | |
| *** acabot has quit IRC | 06:47 | |
| *** thorst_ has quit IRC | 06:50 | |
| *** alexchadin has joined #openstack-watcher | 07:15 | |
| openstackgerrit | Alexander Chadin proposed openstack/watcher: Add doc for workload-stabilization spec https://review.openstack.org/389101 | 07:23 |
|---|---|---|
| *** dtardivel has joined #openstack-watcher | 07:25 | |
| *** thorst_ has joined #openstack-watcher | 07:47 | |
| *** thorst_ has quit IRC | 07:55 | |
| *** brunograz has joined #openstack-watcher | 07:55 | |
| *** vincentfrancoise has joined #openstack-watcher | 08:15 | |
| *** vincentfrancois1 has joined #openstack-watcher | 08:15 | |
| *** vincentfrancois1 has left #openstack-watcher | 08:15 | |
| *** vincentfrancoise has quit IRC | 08:15 | |
| *** vincentfrancoise has joined #openstack-watcher | 08:16 | |
| *** atuly has quit IRC | 08:23 | |
| *** exploreshaifali has joined #openstack-watcher | 08:29 | |
| vincentfrancoise | alexchadin: morning | 08:33 |
| alexchadin | vincentfrancoise: good morning! | 08:33 |
| vincentfrancoise | alexchadin: I just replied to the comments you and hidekazu made about the cpu metrics via SNMP | 08:34 |
| vincentfrancoise | alexchadin: did you see https://review.openstack.org/#/c/393355/ ? | 08:35 |
| alexchadin | let me take a look | 08:36 |
| alexchadin | vincentfrancoise: do you speak about this? https://review.openstack.org/#/c/389101/ | 08:36 |
| alexchadin | vincentfrancoise: cause I don't see the thread about SNMP in tempest tests | 08:37 |
| dtardivel | pwd | 08:38 |
| vincentfrancoise | alexchadin: yes I was refering to my comments on the doc | 08:38 |
| alexchadin | I agree that we can replace CPU SNMP with compute node cpu percent | 08:40 |
| vincentfrancoise | alexchadin: cool :) | 08:41 |
| alexchadin | vincentfrancoise: I will mention it in new PS for doc | 08:41 |
| vincentfrancoise | alexchadin: the other way of doing it would be to have some sort of list of metrics that can be queried in order to get the CPU metrics | 08:42 |
| alexchadin | vincentfrancoise: oh | 08:42 |
| vincentfrancoise | alexchadin: and we would try each one of them one by one until we get something | 08:42 |
| vincentfrancoise | alexchadin: but that requires more work | 08:42 |
| alexchadin | vincentfrancoise: there is so many unstable things with ceilometer and its metrics | 08:42 |
| vincentfrancoise | alexchadin: I agree! | 08:43 |
| hidekazu | morning, alexchadin, vincentfrancoise, | 08:44 |
| vincentfrancoise | hidekazu: hi | 08:44 |
| alexchadin | vincentfrancoise: this mechanism can be used not only for one strategy, but for many strategies | 08:44 |
| alexchadin | hidekazu: hello! | 08:44 |
| vincentfrancoise | alexchadin: yes I agree, but this would also mean some extra work with a small blueprint maybe | 08:45 |
| alexchadin | +1 | 08:45 |
| vincentfrancoise | I'm off for a meeting, see you in a bit ;) | 08:46 |
| hidekazu | i am little sad because i was able to study a lot from workload standardization which needs SNMP plugin :) | 08:48 |
| alexchadin | vincentfrancoise: back to our topic - I would prefer to leave RAM metrics as required since I always want to set the focus on using CPU and RAM together. BTW, you can correct impact of each metric by using the weights and even use them separately. | 08:49 |
| alexchadin | hidekazu: do you have some problems with setting your SNMP env? | 08:50 |
| hidekazu | alexchadin: i used only cpu_util and compute.node.cpu.percent. | 08:52 |
| alexchadin | hidekazu: does it work for you?:) | 08:52 |
| alexchadin | hidekazu: I've uploaded new patch https://github.com/openstack/watcher/commit/ef0d133ba83005a69b31055ab4e6e11716a9f8fe | 08:53 |
| hidekazu | alexchadin: yes, today strategy worked! | 08:53 |
| *** thorst_ has joined #openstack-watcher | 08:54 | |
| alexchadin | hidekazu: great! | 08:54 |
| alexchadin | hidekazu: have you installed this patch https://github.com/openstack/watcher/commit/ef0d133ba83005a69b31055ab4e6e11716a9f8fe ? | 08:55 |
| hidekazu | alexchadin: i parameterized cpu_util_weight to 0.0005. | 08:55 |
| hidekazu | alexchadin: no, i have not installed. | 08:56 |
| alexchadin | hidekazu: It is strongly recommended to install this patch since it fixes problems with weights | 08:56 |
| hidekazu | alexchadin: thanks, i will try. | 08:58 |
| *** thorst_ has quit IRC | 08:59 | |
| *** exploreshaifali has quit IRC | 09:04 | |
| openstackgerrit | Alexander Chadin proposed openstack/python-watcherclient: [WIP] Add functional tests to watcherclient https://review.openstack.org/389181 | 09:05 |
| *** khamtamtun has joined #openstack-watcher | 09:07 | |
| *** khamtamtun has quit IRC | 09:10 | |
| openstackgerrit | Alexander Chadin proposed openstack/watcher: Add doc for workload-stabilization spec https://review.openstack.org/389101 | 09:22 |
| vincentfrancoise | alexchadin: Yes, I agree that RAM should stay by default | 09:28 |
| vincentfrancoise | alexchadin: it's just that if someone like hidekazu doesn't have SNMP metrics available, they should still be able to use the strategy. | 09:28 |
| vincentfrancoise | alexchadin: there is another thing I want to discuss with you regarding your strategy but also all strategies in Watcher in general | 09:32 |
| alexchadin | vincentfrancoise: yes? | 09:32 |
| vincentfrancoise | alexchadin: as I was trying to make your strategy and workload_balance work by using a tempest test | 09:32 |
| vincentfrancoise | alexchadin: I realised that running the strategy just after creating a new VM makes the strategy fail | 09:33 |
| vincentfrancoise | alexchadin: that's because you don't have any measurement yet for new VMs | 09:33 |
| alexchadin | yes-yes | 09:33 |
| vincentfrancoise | alexchadin: so IMHO, we should cover this case in all strategies to make sure that it doesn't fail if there is just one VM that doesn't have any measurements | 09:34 |
| vincentfrancoise | alexchadin: something like this was done in basic_consolidation | 09:35 |
| vincentfrancoise | alexchadin: https://github.com/openstack/watcher/blob/master/watcher/decision_engine/strategy/strategies/basic_consolidation.py#L238-L244 | 09:35 |
| alexchadin | instead of set 100 to host_avg_cpu_util, in case of workload stabilization strategy we can not to take into account failed vms since it was just created and don't make a lot of loads | 09:37 |
| alexchadin | it don't make* | 09:37 |
| vincentfrancoise | alexchadin: yes, I agree that setting it to 100 is not what we want in your case | 09:38 |
| alexchadin | vincentfrancoise: I will create the patch:) | 09:38 |
| vincentfrancoise | alexchadin: it's just an example, and I didn't think about what was the good way of doing it too | 09:38 |
| vincentfrancoise | alexchadin: it's just that I think that it's something we need somehow | 09:39 |
| *** hidekazu has left #openstack-watcher | 09:41 | |
| alexchadin | vincentfrancoise: https://bugs.launchpad.net/watcher/+bug/1640096 | 09:46 |
| openstack | Launchpad bug 1640096 in watcher "workload stabilization cannot be executed" [Medium,Confirmed] - Assigned to Alexander Chadin (joker946) | 09:46 |
| *** exploreshaifali has joined #openstack-watcher | 09:53 | |
| *** thorst_ has joined #openstack-watcher | 09:57 | |
| *** thorst_ has quit IRC | 10:05 | |
| *** alexchadin has quit IRC | 10:15 | |
| openstackgerrit | howard lee proposed openstack/watcher: Use oslo_log instead of logging https://review.openstack.org/394882 | 10:20 |
| *** exploreshaifali has quit IRC | 10:43 | |
| diga | Hi All | 10:56 |
| diga | vincentfrancoise: Hi | 10:56 |
| diga | vincentfrancoise: about today's meeting, it is in the 14:00UTC today ? | 10:58 |
| vincentfrancoise | diga: hi | 10:58 |
| vincentfrancoise | diga: meetings are on wednesdays ;) | 10:58 |
| diga | sorry, yes forgot | 10:58 |
| diga | vincentfrancoise: Today I joined office from long holiday, so missed the schedule :) | 10:59 |
| vincentfrancoise | diga: no problem :) | 11:00 |
| vincentfrancoise | diga: I'm off for lunch, so see you later | 11:00 |
| *** thorst_ has joined #openstack-watcher | 11:03 | |
| *** thorst_ has quit IRC | 11:09 | |
| *** Kevin_Zheng has quit IRC | 11:18 | |
| openstackgerrit | Alexander Chadin proposed openstack/watcher: Add doc for workload-stabilization spec https://review.openstack.org/389101 | 12:00 |
| *** khamtamtun has joined #openstack-watcher | 12:00 | |
| *** khamtamtun has quit IRC | 12:05 | |
| *** khamtamtun has joined #openstack-watcher | 12:05 | |
| *** khamtamtun has quit IRC | 12:07 | |
| *** exploreshaifali has joined #openstack-watcher | 12:12 | |
| *** khamtamtun has joined #openstack-watcher | 12:17 | |
| *** thorst_ has joined #openstack-watcher | 12:50 | |
| openstackgerrit | Merged openstack/python-watcherclient: Add Python 3.5 classifier and venv https://review.openstack.org/390703 | 12:53 |
| sballe_ | morning | 12:54 |
| openstackgerrit | Merged openstack/python-watcherclient: Updated from global requirements https://review.openstack.org/388332 | 12:55 |
| openstackgerrit | Vincent Françoise proposed openstack/watcher: Add audit.strategy events https://review.openstack.org/394482 | 12:58 |
| openstackgerrit | Vincent Françoise proposed openstack/watcher: Implemented audit.create notification https://review.openstack.org/389765 | 12:58 |
| openstackgerrit | Vincent Françoise proposed openstack/watcher: Implemented audit.delete notification https://review.openstack.org/393703 | 12:58 |
| openstackgerrit | Vincent Françoise proposed openstack/watcher: Add audit.planner events https://review.openstack.org/394948 | 12:58 |
| *** khamtamtun has quit IRC | 13:02 | |
| *** lamt has joined #openstack-watcher | 13:21 | |
| *** jwcroppe has quit IRC | 13:35 | |
| *** jwcroppe has joined #openstack-watcher | 13:36 | |
| *** jwcroppe has quit IRC | 13:40 | |
| *** khamtamtun has joined #openstack-watcher | 13:42 | |
| *** khamtamtun has quit IRC | 13:43 | |
| *** openstackgerrit has quit IRC | 13:48 | |
| *** openstackgerrit has joined #openstack-watcher | 13:48 | |
| openstackgerrit | Vincent Françoise proposed openstack/watcher: Added notifications documentation page https://review.openstack.org/387543 | 13:55 |
| openstackgerrit | Vincent Françoise proposed openstack/watcher: Added support for versioned notifications https://review.openstack.org/387542 | 13:55 |
| *** jwcroppe has joined #openstack-watcher | 13:57 | |
| openstackgerrit | Vincent Françoise proposed openstack/watcher: Add audit.strategy events https://review.openstack.org/394482 | 14:01 |
| openstackgerrit | Vincent Françoise proposed openstack/watcher: Implemented audit.create notification https://review.openstack.org/389765 | 14:01 |
| openstackgerrit | Vincent Françoise proposed openstack/watcher: Implemented audit.update notification https://review.openstack.org/360008 | 14:01 |
| openstackgerrit | Vincent Françoise proposed openstack/watcher: Add audit.planner events https://review.openstack.org/394948 | 14:01 |
| openstackgerrit | Vincent Françoise proposed openstack/watcher: Implemented audit.delete notification https://review.openstack.org/393703 | 14:01 |
| *** exploreshaifali has quit IRC | 14:16 | |
| *** absubram has joined #openstack-watcher | 14:39 | |
| *** absubram_ has joined #openstack-watcher | 14:41 | |
| *** absubram has quit IRC | 14:44 | |
| *** absubram_ is now known as absubram | 14:44 | |
| *** absubram has quit IRC | 15:22 | |
| *** diga has quit IRC | 15:54 | |
| *** diga has joined #openstack-watcher | 15:58 | |
| *** eil397_ has joined #openstack-watcher | 16:10 | |
| *** eil397_ has quit IRC | 16:19 | |
| *** eil397_ has joined #openstack-watcher | 17:05 | |
| openstackgerrit | Vincent Françoise proposed openstack/watcher: Remove stale notification code https://review.openstack.org/395087 | 17:09 |
| vincentfrancoise | eil397_: hi | 17:12 |
| vincentfrancoise | eil397_: I saw that you tried to get into contact with one of us yesterday evening | 17:14 |
| vincentfrancoise | eil397_: Most of the Watcher contributors are working in the UTC+1 timezone so people had left already | 17:15 |
| *** eil397 has joined #openstack-watcher | 17:58 | |
| *** dtardivel has quit IRC | 18:34 | |
| *** diga has quit IRC | 19:14 | |
| *** eil397 has left #openstack-watcher | 19:30 | |
| *** eil397 has joined #openstack-watcher | 19:31 | |
| *** thorst_ has quit IRC | 22:32 | |
| *** thorst_ has joined #openstack-watcher | 22:55 | |
| *** thorst_ has quit IRC | 23:07 | |
| openstackgerrit | Alex Schultz proposed openstack/puppet-watcher: Deprecate rabbitmq connection parameters https://review.openstack.org/395244 | 23:12 |
| *** lamt has quit IRC | 23:16 | |
| openstackgerrit | Alex Schultz proposed openstack/puppet-watcher: Deprecate rabbitmq connection parameters https://review.openstack.org/395244 | 23:17 |
| *** absubram has joined #openstack-watcher | 23:21 | |
| absubram | hi.. what time is the meeting tomorrow? 9am UTC or 2pm UTC? | 23:21 |
| *** jwcroppe has quit IRC | 23:27 | |
| *** jwcroppe has joined #openstack-watcher | 23:27 | |
| *** jwcroppe has quit IRC | 23:31 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!