| *** thorst has quit IRC | 00:01 | |
| *** thorst has joined #openstack-watcher | 00:20 | |
| *** xenogear has quit IRC | 00:30 | |
| *** thorst_ has joined #openstack-watcher | 00:33 | |
| *** xenogear has joined #openstack-watcher | 00:33 | |
| *** thorst has quit IRC | 00:36 | |
| *** xenogear has quit IRC | 00:37 | |
| *** thorst_ has quit IRC | 02:57 | |
| *** thorst has joined #openstack-watcher | 02:57 | |
| *** jwcroppe has joined #openstack-watcher | 03:02 | |
| *** thorst has quit IRC | 03:06 | |
| *** jwcroppe has quit IRC | 03:08 | |
| *** thorst has joined #openstack-watcher | 04:03 | |
| *** esberglu has joined #openstack-watcher | 04:09 | |
| *** thorst has quit IRC | 04:10 | |
| *** esberglu has quit IRC | 04:12 | |
| *** apoorv has joined #openstack-watcher | 04:13 | |
| *** xenogear has joined #openstack-watcher | 04:22 | |
| *** jwcroppe has joined #openstack-watcher | 05:06 | |
| *** thorst has joined #openstack-watcher | 05:08 | |
| *** jwcroppe has quit IRC | 05:11 | |
| *** thorst has quit IRC | 05:16 | |
| *** apoorv_ has joined #openstack-watcher | 05:25 | |
| *** apoorv_ has quit IRC | 05:51 | |
| *** apoorv has quit IRC | 05:51 | |
| *** apoorv has joined #openstack-watcher | 05:55 | |
| *** thorst has joined #openstack-watcher | 06:14 | |
| *** thorst has quit IRC | 06:21 | |
| *** vtech has joined #openstack-watcher | 06:37 | |
| *** vtech has quit IRC | 06:42 | |
| *** vtech has joined #openstack-watcher | 07:00 | |
| *** brunograz_ has quit IRC | 07:06 | |
| *** brunograz has joined #openstack-watcher | 07:07 | |
| *** jwcroppe has joined #openstack-watcher | 07:09 | |
| *** jwcroppe has quit IRC | 07:14 | |
| *** thorst has joined #openstack-watcher | 07:19 | |
| openstackgerrit | Merged openstack/watcher: Added missing support for resource states in unicode format in VM workload consolidation strategy https://review.openstack.org/301123 | 07:24 |
|---|---|---|
| *** thorst has quit IRC | 07:26 | |
| openstackgerrit | Merged openstack/watcher: Invalid states for Action Plan in the glossary https://review.openstack.org/297215 | 07:27 |
| *** vincentfrancoise has joined #openstack-watcher | 07:29 | |
| openstackgerrit | Merged openstack/watcher: Disabled PATCH, POST and DELETE for /actions https://review.openstack.org/281873 | 07:30 |
| openstackgerrit | Merged openstack/watcher: Added information on plugin mechanism to glossary https://review.openstack.org/296873 | 07:52 |
| *** alexchadin has joined #openstack-watcher | 08:03 | |
| *** alexchadin has quit IRC | 08:19 | |
| *** thorst has joined #openstack-watcher | 08:24 | |
| *** thorst has quit IRC | 08:30 | |
| openstackgerrit | Vincent Françoise proposed openstack/watcher-dashboard: Added Goals and Strategies to Dashboard https://review.openstack.org/301516 | 08:43 |
| *** jwcroppe has joined #openstack-watcher | 09:11 | |
| *** jwcroppe has quit IRC | 09:17 | |
| *** thorst has joined #openstack-watcher | 09:28 | |
| jed56 | agenda for tomorrow meeting https://wiki.openstack.org/wiki/Watcher_Meeting_Agenda#04.2F05.2F2016 | 09:34 |
| jed56 | please feel free to add items :) | 09:34 |
| *** thorst has quit IRC | 09:36 | |
| *** xenogear has quit IRC | 09:38 | |
| *** anem0ne has quit IRC | 09:58 | |
| *** vtech has quit IRC | 10:01 | |
| *** alexchadin has joined #openstack-watcher | 10:01 | |
| alexchadin | hi | 10:01 |
| vincentfrancoise | hi | 10:01 |
| alexchadin | I can handle this https://bugs.launchpad.net/watcher/+bug/1532843 | 10:02 |
| openstack | Launchpad bug 1532843 in watcher "Can create Audit with invalid parameters" [Low,In progress] | 10:02 |
| *** vtech has joined #openstack-watcher | 10:02 | |
| alexchadin | if it still actual | 10:03 |
| vincentfrancoise | yes it is, so you can go for it ;) | 10:03 |
| *** alexchadin has quit IRC | 10:04 | |
| *** vtech has quit IRC | 10:12 | |
| apoorv | my controller and compute node have different version of nova...how to solve this? | 10:12 |
| jed56 | hello apporv | 10:18 |
| jed56 | IMO, you could set the version in the watcher configuration file | 10:18 |
| jed56 | apoorv: | 10:18 |
| apoorv | but compute node has no watcher module | 10:19 |
| *** thorst has joined #openstack-watcher | 10:33 | |
| *** thorst has quit IRC | 10:41 | |
| *** vtech has joined #openstack-watcher | 10:46 | |
| *** jwcroppe has joined #openstack-watcher | 11:14 | |
| *** jwcroppe has quit IRC | 11:19 | |
| *** thorst has joined #openstack-watcher | 11:44 | |
| *** thorst_ has joined #openstack-watcher | 11:45 | |
| *** thorst has quit IRC | 11:48 | |
| jed56 | apoorv: what do you mean ? | 12:04 |
| jed56 | the watcher applier is in charge to interact with the nova | 12:08 |
| apoorv | so nova version depends on watcher configuration? | 12:10 |
| jed56 | exaclty | 12:12 |
| vincentfrancoise | apoorv: watcher should be compatible with nova API v2 | 12:12 |
| jed56 | https://www.irccloud.com/pastebin/Cppt3euA/ | 12:13 |
| jed56 | In the watcher configuration file | 12:13 |
| jed56 | you can specify the version of nova | 12:13 |
| vincentfrancoise | so the exact version of nova you are using does not matter much as long as it is not too old | 12:13 |
| jed56 | you can look at the https://factory.b-com.com/www/watcher/doc/watcher/deploy/conf-files.html | 12:13 |
| apoorv | will different versions create a problem? | 12:18 |
| vincentfrancoise | as long as they are running the same API version it should be fine | 12:20 |
| apoorv | i am getting an error that version 1.20 of service not supported in logs of n-cond | 12:22 |
| vincentfrancoise | I guess their internal API changed | 12:23 |
| vincentfrancoise | so you probably need upgrade then | 12:24 |
| apoorv | how to do that? | 12:24 |
| vincentfrancoise | I personlly never had such issue with nova | 12:26 |
| vincentfrancoise | when I upgrade my devstack, I just re-run a ./stack.sh with the latest versions | 12:26 |
| gzhai2 | vincentfrancoise:hi | 12:27 |
| vincentfrancoise | so I may not be of great help when it comes to this: have you tried asking them on their #openstack-nova IRC channel? | 12:27 |
| vincentfrancoise | gzhai2: hi | 12:28 |
| apoorv | actually no | 12:28 |
| gzhai2 | vincentfrancoise: I had holiday leave on Monday, sorry for missing your ping. | 12:28 |
| vincentfrancoise | gzhai2: no problem | 12:28 |
| vincentfrancoise | I was just willing to discuss the various comments you made on my changesets | 12:29 |
| gzhai2 | ok | 12:29 |
| vincentfrancoise | about the whole ID/UUID thing on goals and the comment you made this morning as well | 12:29 |
| gzhai2 | yes. | 12:30 |
| gzhai2 | ID is in fact the unique name of the goal, right? | 12:31 |
| vincentfrancoise | yes | 12:31 |
| gzhai2 | When define a new goal. You need a strategy return a new goal name, right? | 12:31 |
| gzhai2 | rather than define a new goal object. | 12:31 |
| vincentfrancoise | yes via the get_goal_id() class method | 12:32 |
| vincentfrancoise | and all these names are collected as read-only data inside the DB to be exposed in the API so we can show the available choices in the command line and in Watcher Dashboard | 12:33 |
| gzhai2 | ID is ok. But deep relationship between strategy and goal is a little bit odd. Just my 2 cents. | 12:33 |
| gzhai2 | So DB is always sync with the code after restart? | 12:34 |
| vincentfrancoise | yes | 12:34 |
| vincentfrancoise | see https://review.openstack.org/#/c/298808/ | 12:34 |
| gzhai2 | But I saw your code try to keep old stuff in DB, right? | 12:34 |
| vincentfrancoise | I do not delete, only update indeed | 12:34 |
| vincentfrancoise | because you have a foreign key on audit templates so I don't want to have a goal/strategy disappearing in case we remove something | 12:35 |
| vincentfrancoise | this is to avoid having stale audit templates | 12:36 |
| *** apoorv has quit IRC | 12:36 | |
| gzhai2 | It's common in openstack project. How to delete one object referenced by another? | 12:37 |
| gzhai2 | always Prevent deleting, or prevent when referenced, or just delete. | 12:37 |
| vincentfrancoise | so you think I should actually try to remove old goals if not referenced when I do the registering of goals/strategies? | 12:39 |
| gzhai2 | What if we just remove actioin-plan of one audit? | 12:40 |
| vincentfrancoise | in that case it's fine because an action plan has a foreign key onto the audit | 12:41 |
| gzhai2 | Goal can't find its strategy? | 12:43 |
| vincentfrancoise | not sure what you mean by this | 12:46 |
| vincentfrancoise | audits and action plan lose any sense of goal/strategy | 12:47 |
| gzhai2 | you said action plan has a foreighn key onto audit. so can just delete action plan. | 12:47 |
| vincentfrancoise | so they are isolated from the staleness issue | 12:47 |
| vincentfrancoise | yes but my problem is on the audit-template level | 12:48 |
| vincentfrancoise | it references a goal and a strategy (optional) | 12:48 |
| vincentfrancoise | so if I delete either one or the other it can complain about having a broken FK constraint | 12:49 |
| gzhai2 | ok. Nice to have orphan goal deleting if you have time. | 12:50 |
| gzhai2 | In our code, any requirement to look for strategy from a goal? | 12:51 |
| vincentfrancoise | there is one | 12:52 |
| vincentfrancoise | in https://review.openstack.org/#/c/296636/6/watcher/decision_engine/strategy/selection/default.py | 12:53 |
| vincentfrancoise | we want to provide a way for an admin to automatically select the best suited strategy given a goal | 12:53 |
| vincentfrancoise | so we actually need to know which strategy achieves the given goal | 12:54 |
| vincentfrancoise | hence the FK on goals | 12:54 |
| gzhai2 | So this 'goals' OPTS provides goal:strategy mapping, right? | 12:56 |
| gzhai2 | And it's duplcated info that we can find via registration. | 12:57 |
| vincentfrancoise | currently yes, but I am actually trying to replace this one-to-one mapping to get a one-to-many mapping within the DB | 12:57 |
| vincentfrancoise | so the 'goals' opt is getting removed | 12:58 |
| gzhai2 | ok. we only depends on DB for this mapping. | 12:58 |
| vincentfrancoise | yes exactly | 12:58 |
| gzhai2 | And DB is synced with the code upon restart, right? | 12:58 |
| vincentfrancoise | yes | 12:59 |
| gzhai2 | reasonable. | 12:59 |
| *** vtech has quit IRC | 13:00 | |
| gzhai2 | is goal object loaded by stevedore? I may misunderstand Jean's comments. | 13:00 |
| gzhai2 | in future? | 13:00 |
| *** vtech has joined #openstack-watcher | 13:00 | |
| vincentfrancoise | no it's not meant to, at least for now | 13:01 |
| vincentfrancoise | but since goals are defined in strategies and because strategies are loaded via stevedore, we can already say that goals are pluggable | 13:02 |
| vincentfrancoise | it's just that it is more abstract | 13:02 |
| *** brunograz has quit IRC | 13:02 | |
| *** brunograz has joined #openstack-watcher | 13:03 | |
| gzhai2 | In future, how to find all strategies for one goal? | 13:04 |
| vincentfrancoise | via a DB request to get all strategies filtered by goal ID or externally, through the API | 13:06 |
| vincentfrancoise | so my main question now | 13:07 |
| jed56 | hello gzhai2 | 13:07 |
| gzhai2 | jed56:hello | 13:08 |
| vincentfrancoise | is that you (and tpeoplesvacation) were skeptical about using goal IDs as strings and also you wanted UUIDs | 13:08 |
| jed56 | gzhai2 : IMHO, you can read the spec https://github.com/openstack/watcher-specs/blob/master/specs/mitaka/approved/get-goal-from-strategy.rst | 13:08 |
| gzhai2 | vincentfrancoise: Not now. string is ok. | 13:09 |
| gzhai2 | But string as ID is a little bit misleading here. | 13:09 |
| gzhai2 | I think tepoples has same feeling:) | 13:09 |
| vincentfrancoise | I agree but saying goal name does not show the uniqueness of the string | 13:10 |
| jed56 | gzhai2 What is an id of you ? :) | 13:10 |
| jed56 | for you | 13:10 |
| gzhai2 | anything unique rather than name. | 13:10 |
| gzhai2 | :) | 13:10 |
| gzhai2 | name could be same, but ID are different. | 13:11 |
| jed56 | in which case they are different in your opinion ? :) | 13:12 |
| *** jwcroppe has joined #openstack-watcher | 13:17 | |
| jed56 | IMHO, we need to add an UUID in the goal because we need to save the history | 13:20 |
| jed56 | if we add the soft delete something which is currently missing | 13:20 |
| *** jwcroppe has quit IRC | 13:21 | |
| *** sballe_ has quit IRC | 13:21 | |
| *** pav0 has quit IRC | 13:21 | |
| *** harlowja has quit IRC | 13:21 | |
| *** tkaczynski has quit IRC | 13:21 | |
| *** brunograz has quit IRC | 13:21 | |
| *** logan- has quit IRC | 13:21 | |
| *** edleafe has quit IRC | 13:21 | |
| *** openstackgerrit has quit IRC | 13:21 | |
| *** gzhai2 has quit IRC | 13:21 | |
| *** mordred has quit IRC | 13:21 | |
| *** thorst_ has quit IRC | 13:21 | |
| *** ChanServ has quit IRC | 13:21 | |
| *** vtech has quit IRC | 13:21 | |
| *** mestery has quit IRC | 13:22 | |
| *** pkoniszewski has quit IRC | 13:22 | |
| *** aspiers has quit IRC | 13:22 | |
| *** vincentfrancoise has quit IRC | 13:22 | |
| *** acabot has quit IRC | 13:22 | |
| jed56 | gzhai, we will update the specification with vincentfrancoise | 13:22 |
| *** sballe__ has joined #openstack-watcher | 13:24 | |
| *** brunograz has joined #openstack-watcher | 13:24 | |
| *** vtech has joined #openstack-watcher | 13:24 | |
| *** thorst_ has joined #openstack-watcher | 13:24 | |
| *** vincentfrancoise has joined #openstack-watcher | 13:24 | |
| *** openstackgerrit has joined #openstack-watcher | 13:24 | |
| *** pav0 has joined #openstack-watcher | 13:24 | |
| *** aspiers has joined #openstack-watcher | 13:24 | |
| *** logan- has joined #openstack-watcher | 13:24 | |
| *** harlowja has joined #openstack-watcher | 13:24 | |
| *** pkoniszewski has joined #openstack-watcher | 13:24 | |
| *** mestery has joined #openstack-watcher | 13:24 | |
| *** mordred has joined #openstack-watcher | 13:24 | |
| *** gzhai2 has joined #openstack-watcher | 13:24 | |
| *** acabot has joined #openstack-watcher | 13:24 | |
| *** edleafe has joined #openstack-watcher | 13:24 | |
| *** tkaczynski has joined #openstack-watcher | 13:24 | |
| *** ChanServ has joined #openstack-watcher | 13:24 | |
| *** wolfe.freenode.net sets mode: +o ChanServ | 13:24 | |
| *** sballe__ is now known as sballe_ | 13:24 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/watcher: Updated from global requirements https://review.openstack.org/300782 | 13:44 |
| *** brunograz has quit IRC | 13:50 | |
| *** brunograz has joined #openstack-watcher | 13:51 | |
| *** vtech has quit IRC | 13:54 | |
| *** vtech has joined #openstack-watcher | 13:55 | |
| openstackgerrit | zhangguoqing proposed openstack/watcher-dashboard: update test-requirements to get horizon faster https://review.openstack.org/301684 | 13:56 |
| openstackgerrit | zhangguoqing proposed openstack/watcher-dashboard: Fix api test about AuditTemplate https://review.openstack.org/301692 | 14:09 |
| *** esberglu has joined #openstack-watcher | 14:57 | |
| *** jwcroppe has joined #openstack-watcher | 15:03 | |
| *** vtech has quit IRC | 15:11 | |
| *** vtech has joined #openstack-watcher | 15:19 | |
| openstackgerrit | Jean-Emile DARTOIS proposed openstack/watcher-specs: Open specs for Newton https://review.openstack.org/301753 | 15:22 |
| *** vtech has quit IRC | 15:22 | |
| *** vtech has joined #openstack-watcher | 15:23 | |
| openstackgerrit | Jean-Emile DARTOIS proposed openstack/watcher-specs: Open specs for Newton https://review.openstack.org/301753 | 15:29 |
| openstackgerrit | Jean-Emile DARTOIS proposed openstack/watcher-specs: Open specs for Newton https://review.openstack.org/301753 | 15:35 |
| openstackgerrit | Merged openstack/watcher: Updated from global requirements https://review.openstack.org/300782 | 15:41 |
| openstackgerrit | Jean-Emile DARTOIS proposed openstack/watcher-specs: Achieved goal should be returned by each strategy Today, there is no way to know what goal is achieved by a given strategy. The mapping between goal and strategies is done in the Watcher configuration. https://review.openstack.org/301774 | 15:42 |
| openstackgerrit | Jean-Emile DARTOIS proposed openstack/watcher-specs: Achieved goal should be returned by each strategy https://review.openstack.org/301774 | 15:43 |
| *** jwcroppe has quit IRC | 15:48 | |
| *** wootehfoot has joined #openstack-watcher | 15:55 | |
| openstackgerrit | Jean-Emile DARTOIS proposed openstack/watcher-specs: Achieved goal should be returned by each strategy Today, there is no way to know what goal is achieved by a given strategy. The mapping between goal and strategies is done in the Watcher configuration. https://review.openstack.org/301774 | 16:12 |
| *** vtech has quit IRC | 16:16 | |
| openstackgerrit | Jean-Emile DARTOIS proposed openstack/watcher-specs: Achieved goal should be returned by each strategy https://review.openstack.org/301774 | 16:19 |
| openstackgerrit | Jean-Emile DARTOIS proposed openstack/watcher-specs: Open specs for Newton https://review.openstack.org/301753 | 16:23 |
| openstackgerrit | Jean-Emile DARTOIS proposed openstack/watcher-specs: Achieved goal should be returned by each strategy https://review.openstack.org/301774 | 16:24 |
| openstackgerrit | Jean-Emile DARTOIS proposed openstack/watcher-specs: Achieved goal should be returned by each strategy https://review.openstack.org/301774 | 16:28 |
| *** wootehfoot has quit IRC | 16:32 | |
| *** wootehfoot has joined #openstack-watcher | 16:44 | |
| *** vincentfrancoise has quit IRC | 17:00 | |
| *** wootehfoot has quit IRC | 17:06 | |
| *** wootehfoot has joined #openstack-watcher | 17:06 | |
| *** vtech has joined #openstack-watcher | 17:36 | |
| *** vtech has quit IRC | 17:38 | |
| *** vtech has joined #openstack-watcher | 18:23 | |
| *** vtech has quit IRC | 20:41 | |
| *** thorst_ has quit IRC | 21:11 | |
| *** thorst_ has joined #openstack-watcher | 21:12 | |
| *** thorst_ has quit IRC | 21:16 | |
| *** thorst_ has joined #openstack-watcher | 21:32 | |
| *** thorst_ has quit IRC | 21:37 | |
| *** jwcroppe has joined #openstack-watcher | 21:42 | |
| *** jwcroppe has quit IRC | 22:13 | |
| *** esberglu has quit IRC | 23:03 | |
| *** thorst has joined #openstack-watcher | 23:24 | |
| *** thorst_ has joined #openstack-watcher | 23:28 | |
| *** thorst has quit IRC | 23:30 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!