*** hidekazu has joined #openstack-watcher | 00:02 | |
*** thorst_ has joined #openstack-watcher | 00:42 | |
*** zhurong has joined #openstack-watcher | 00:44 | |
*** thorst_ has quit IRC | 00:58 | |
*** thorst_ has joined #openstack-watcher | 02:57 | |
*** thorst_ has quit IRC | 02:57 | |
openstackgerrit | rajiv proposed openstack/watcher-dashboard: Remove link to modindex https://review.openstack.org/424967 | 03:55 |
---|---|---|
openstackgerrit | rajiv proposed openstack/watcher-dashboard: Remove link to modindex https://review.openstack.org/424967 | 04:05 |
openstackgerrit | Prashanth Hari proposed openstack/watcher-specs: Audit tag in VM Metadata https://review.openstack.org/391374 | 04:28 |
*** diga has joined #openstack-watcher | 04:41 | |
*** thorst_ has joined #openstack-watcher | 04:59 | |
*** thorst_ has quit IRC | 05:03 | |
*** adisky_ has joined #openstack-watcher | 05:04 | |
openstackgerrit | Zeyu Zhu proposed openstack/python-watcherclient: Remove useless utf-8 coding https://review.openstack.org/425006 | 06:53 |
*** thorst_ has joined #openstack-watcher | 06:59 | |
*** thorst_ has quit IRC | 07:04 | |
*** pshedimb has quit IRC | 07:37 | |
*** pshedimb has joined #openstack-watcher | 07:39 | |
*** alexchadin has joined #openstack-watcher | 07:45 | |
*** vincentfrancoise has joined #openstack-watcher | 08:25 | |
openstackgerrit | Merged openstack/python-watcherclient: Remove useless utf-8 coding https://review.openstack.org/425006 | 08:34 |
openstackgerrit | Merged openstack/watcher-dashboard: Remove link to modindex https://review.openstack.org/424967 | 08:35 |
openstackgerrit | David TARDIVEL proposed openstack/watcher: Reduce the call of nova.servers.list https://review.openstack.org/406018 | 08:37 |
openstackgerrit | David TARDIVEL proposed openstack/watcher: Documentation update https://review.openstack.org/424187 | 08:43 |
openstackgerrit | Merged openstack/watcher: Updated from global requirements https://review.openstack.org/424630 | 08:45 |
*** thorst_ has joined #openstack-watcher | 09:00 | |
*** thorst_ has quit IRC | 09:05 | |
openstackgerrit | David TARDIVEL proposed openstack/watcher: Documentation update https://review.openstack.org/424187 | 09:07 |
*** zhurong has quit IRC | 09:14 | |
*** efoley has joined #openstack-watcher | 09:21 | |
*** efoley_ has joined #openstack-watcher | 09:23 | |
*** efoley has quit IRC | 09:27 | |
openstackgerrit | Vincent Françoise proposed openstack/watcher: Fix multinode tempest test failure https://review.openstack.org/419544 | 09:30 |
*** dtardivel has joined #openstack-watcher | 09:35 | |
openstackgerrit | OpenStack Proposal Bot proposed openstack/python-watcherclient: Updated from global requirements https://review.openstack.org/423691 | 09:37 |
openstackgerrit | Vincent Françoise proposed openstack/watcher: Fix multinode tempest test failure https://review.openstack.org/419544 | 09:40 |
*** adisky_ has quit IRC | 09:49 | |
alexchadin | acabot: hi | 10:09 |
acabot | alexchadin : hi | 10:09 |
alexchadin | acabot: what is the status of nova-policies bp? I've found out that this BP has been removed | 10:11 |
*** pksingh has joined #openstack-watcher | 10:12 | |
pksingh | vincentfrancoise: Hello, | 10:12 |
acabot | yes I moved it as obsolete as it needs to be rewritten according to the new placement API provided by Nova | 10:12 |
acabot | alexchadin : this is something we have to discuss at PTG | 10:12 |
vincentfrancoise | pksingh: morning | 10:13 |
alexchadin | acabot: thx | 10:13 |
pksingh | vincentfrancoise: Good morning, | 10:13 |
pksingh | vincentfrancoise: how's the day? | 10:13 |
vincentfrancoise | pksingh: well, since our gate is now working, I'm great :) | 10:14 |
pksingh | vincentfrancoise: haha | 10:14 |
vincentfrancoise | pksingh: how are you getting on with the watcher codebase BTW? | 10:16 |
*** openstackgerrit has quit IRC | 10:17 | |
pksingh | vincentfrancoise: yesterday i looked into the CDM, and understood it | 10:17 |
pksingh | vincentfrancoise: lot of datastructures | 10:18 |
*** alexchad_ has joined #openstack-watcher | 10:18 | |
pksingh | vincentfrancoise: Next part is to understand the strategies | 10:19 |
vincentfrancoise | pksingh: you don't need to go through all strategies | 10:19 |
vincentfrancoise | pksingh: I only know about half of them really | 10:20 |
pksingh | vincentfrancoise: which one should i start? | 10:20 |
vincentfrancoise | pksingh: if you understand dummy and thenn basic_consolidation then you're good | 10:20 |
pksingh | vincentfrancoise: ok great | 10:20 |
vincentfrancoise | pksingh: the rest is bonus or depends on what you want to achieve | 10:20 |
pksingh | vincentfrancoise: can i run any strategy on single node devstack? | 10:21 |
vincentfrancoise | pksingh: dummy actually | 10:21 |
*** alexchadin has quit IRC | 10:21 | |
pksingh | vincentfrancoise: what is the result of strategy, what i understand is it generate a action plan, which contains action(s) | 10:23 |
pksingh | vincentfrancoise: right? | 10:23 |
vincentfrancoise | pksingh: yup | 10:23 |
*** openstackgerrit has joined #openstack-watcher | 10:24 | |
openstackgerrit | Vincent Françoise proposed openstack/watcher: Added Actuator Strategy https://review.openstack.org/425110 | 10:24 |
vincentfrancoise | pksingh: the exact workflow is described in the doc ;) | 10:24 |
vincentfrancoise | pksingh: let me give you the link | 10:24 |
vincentfrancoise | pksingh: http://docs.openstack.org/developer/watcher/deploy/user-guide.html | 10:24 |
pksingh | vincentfrancoise: so, when these action(s) are executed, just after strategy is finished or we can get postpone them or later? | 10:24 |
pksingh | vincentfrancoise: thanks a lot | 10:24 |
pksingh | vincentfrancoise: i will check this | 10:25 |
vincentfrancoise | pksingh: By default it's a 2-step process | 10:25 |
vincentfrancoise | pksingh: the execution of an audit is composed of 2 subtasks | 10:25 |
vincentfrancoise | pksingh: first you execute the strategy which generates a "solution" | 10:26 |
vincentfrancoise | pksingh: a solution contains an *unordered* set of actions (although that will be made configurable at some point) | 10:27 |
vincentfrancoise | pksingh: then this solution is passed on to a planner | 10:27 |
vincentfrancoise | pksingh: the planner is responsible to produce the action plan by ordering (as a workflow/graph) the provided actions | 10:27 |
vincentfrancoise | pksingh: the Watcher Applier is then the one to consume/execute the action plans | 10:28 |
pksingh | vincentfrancoise: so we can decide when to apply the plans right? | 10:29 |
vincentfrancoise | pksingh: yes | 10:29 |
vincentfrancoise | pksingh: Watcher can be used as an advisor without ever applying an action plan | 10:30 |
pksingh | vincentfrancoise: so after some time when we decide to apply the plan, the system condition may not be the same as when strategy was executed, makes sense? | 10:30 |
pksingh | vincentfrancoise: ok | 10:31 |
vincentfrancoise | pksingh: yes, staleness is an issue that we are aware and we are making things quite strict on this matter | 10:31 |
vincentfrancoise | pksingh: but for now you hav to be careful about this factor | 10:32 |
pksingh | vincentfrancoise: one way to avoid this is to run audit in auto mode, i think | 10:33 |
*** sanfern has joined #openstack-watcher | 10:33 | |
vincentfrancoise | pksingh: auto trigger mode is also a matter of trust | 10:33 |
vincentfrancoise | pksingh: so we need both to work | 10:33 |
vincentfrancoise | pksingh: https://blueprints.launchpad.net/watcher/+spec/stale-action-plan | 10:34 |
pksingh | vincentfrancoise: ok, will see it | 10:35 |
pksingh | vincentfrancoise: today is team meeting correct | 10:36 |
vincentfrancoise | pksingh: I'm not sure it'll make it for tomorrow though | 10:36 |
vincentfrancoise | pksingh: yes | 10:36 |
vincentfrancoise | pksingh: in 3.5 hours | 10:36 |
pksingh | vincentfrancoise: 3.5 hours later, correct? | 10:37 |
pksingh | vincentfrancoise: ohh you replied | 10:37 |
pksingh | vincentfrancoise: thanks | 10:37 |
pksingh | vincentfrancoise: it seems there is lot of work required to be done, if watcher team needs any help in development i am ready | 10:39 |
pksingh | vincentfrancoise: otherwise i have started to understand the watcher at my own pace | 10:40 |
vincentfrancoise | pksingh: there is one issue I saw this morning that requires quite a bit of refactoring that you can go for if you want | 10:41 |
vincentfrancoise | pksingh: I didn't create a ticket and that's more like a nice-to-have than a bug really | 10:42 |
pksingh | vincentfrancoise: ok sure, please tell me, i am more than happy | 10:42 |
*** vmahe has joined #openstack-watcher | 10:42 | |
vincentfrancoise | pksingh: ok so let me explain it to you | 10:43 |
vincentfrancoise | pksingh: take a look at https://review.openstack.org/#/c/425110/ | 10:43 |
vincentfrancoise | pksingh: if you have a look at https://review.openstack.org/#/c/425110/1/watcher/tests/decision_engine/strategy/strategies/test_actuator.py@33 | 10:43 |
pksingh | vincentfrancoise: yes | 10:44 |
vincentfrancoise | pksingh: you see that input_parameters (which are additional options that you specify to the audit and are passed down to the strategy) | 10:44 |
pksingh | vincentfrancoise: ok | 10:44 |
*** vmahe_ has joined #openstack-watcher | 10:45 | |
vincentfrancoise | pksingh: and these input_parameters are not injected at the construction of the Strategy objects | 10:45 |
vincentfrancoise | pksingh: instead they are set after instantiation | 10:45 |
pksingh | vincentfrancoise: correct | 10:45 |
vincentfrancoise | pksingh: that forces us to do https://review.openstack.org/#/c/425110/1/watcher/decision_engine/strategy/strategies/actuation.py@88 | 10:46 |
vincentfrancoise | pksingh: to do some lazyloading | 10:46 |
pksingh | vincentfrancoise: ok | 10:46 |
vincentfrancoise | pksingh: and we actually "hope" it's going to be there when we get to it | 10:46 |
pksingh | vincentfrancoise: through the getter? | 10:47 |
vincentfrancoise | pksingh: so the idea would be to make it part of the Loadable interface | 10:47 |
vincentfrancoise | pksingh: https://github.com/openstack/watcher/blob/master/watcher/common/loader/loadable.py#L32 | 10:47 |
pksingh | vincentfrancoise: ok | 10:48 |
vincentfrancoise | pksingh: all plugins in Watcher inherit from this class that acts like an interface | 10:48 |
pksingh | vincentfrancoise: k | 10:48 |
vincentfrancoise | pksingh: this is a simple problem but it requires a lot of refactoring | 10:49 |
* vmahe_ | 10:49 | |
pksingh | vincentfrancoise: agree | 10:49 |
vincentfrancoise | pksingh: but you don't really add a new feature so that should make things easier | 10:49 |
pksingh | vincentfrancoise: i think we can create an issue in launcpad and make it wish-list and i will start working on it | 10:50 |
vincentfrancoise | pksingh: yes that's what I was about to ask you | 10:50 |
pksingh | vincentfrancoise: does that make sense? | 10:50 |
vincentfrancoise | pksingh: this way you'll learn the whole process, and will also make you go through all Watcher plugins | 10:51 |
pksingh | vincentfrancoise: may be it will be easy for you to phrase the details for the issue, so plaese could you open the issue for me? | 10:51 |
vincentfrancoise | pksingh: and if you're not sure about something, just ask ;) | 10:52 |
vincentfrancoise | pksingh: just copy/paste the discussion in the ticket that'll be enough | 10:52 |
pksingh | vincentfrancoise: thanks for help | 10:52 |
pksingh | sure | 10:52 |
vincentfrancoise | pksingh: and also provide the link from eavesdrop (you know about it?) | 10:53 |
pksingh | vincentfrancoise: link for the chat | 10:53 |
vincentfrancoise | pksingh: this link http://eavesdrop.openstack.org/irclogs/%23openstack-watcher/%23openstack-watcher.2017-01-25.log.html#t2017-01-25T10:41:51 | 10:54 |
pksingh | vincentfrancoise: yes, i know it | 10:55 |
pksingh | vincentfrancoise: i raised https://bugs.launchpad.net/watcher/+bug/1659259 | 10:59 |
openstack | Launchpad bug 1659259 in watcher "make input_parameters part of the Loadable interface" [Undecided,New] | 10:59 |
pksingh | vincentfrancoise: can you please make it wish-list | 10:59 |
vincentfrancoise | pksingh: done | 10:59 |
pksingh | vincentfrancoise: thanks | 10:59 |
vincentfrancoise | pksingh: I'm off for lunch, see you later ;) | 11:00 |
pksingh | vincentfrancoise: sure, will meet in meeting | 11:00 |
*** thorst_ has joined #openstack-watcher | 11:01 | |
*** vincentfrancoise has quit IRC | 11:04 | |
pksingh | acabot: hello, can i point singlenode devstack to existing multinode openstack cloud? | 11:04 |
pksingh | acabot: single node watcher devstack^^ | 11:05 |
*** thorst_ has quit IRC | 11:05 | |
*** pksingh has quit IRC | 11:28 | |
*** alexchad_ has quit IRC | 11:42 | |
*** sanfern has quit IRC | 11:45 | |
*** alexchadin has joined #openstack-watcher | 11:47 | |
openstackgerrit | Merged openstack/python-watcherclient: Updated from global requirements https://review.openstack.org/423691 | 11:59 |
openstackgerrit | Merged openstack/watcher: Fix building of model with a scoped exclusion rule https://review.openstack.org/424668 | 11:59 |
openstackgerrit | David TARDIVEL proposed openstack/watcher: Documentation update https://review.openstack.org/424187 | 12:11 |
*** thorst_ has joined #openstack-watcher | 12:44 | |
*** vincentfrancoise has joined #openstack-watcher | 13:12 | |
*** sanfern has joined #openstack-watcher | 13:13 | |
openstackgerrit | Vincent Françoise proposed openstack/watcher: Fix multinode tempest test failure https://review.openstack.org/419544 | 13:18 |
acabot | pksingh : hi, I'm not sure to understand what you want to do | 13:37 |
openstackgerrit | Merged openstack/watcher: Graph cluster model instead of mapping one https://review.openstack.org/418434 | 13:54 |
openstackgerrit | Merged openstack/watcher: Remove obsolete Resource element https://review.openstack.org/418974 | 13:54 |
sballe_ | morning | 13:57 |
vincentfrancoise | sballe_: morning | 13:58 |
acabot | weekly meeting starts on #openstack-meeting-4 | 14:01 |
*** diga has quit IRC | 14:20 | |
openstackgerrit | YumengBao proposed openstack/watcher-specs: Remove voluptuous and Use JSONSchema as our only JSON validation tool https://review.openstack.org/425204 | 14:32 |
openstackgerrit | Merged openstack/watcher-specs: Add improvements to the planner and workflow mechanisms https://review.openstack.org/385871 | 14:33 |
openstackgerrit | YumengBao proposed openstack/watcher-specs: Remove voluptuous and Use JSONSchema as our only JSON validation tool https://review.openstack.org/425204 | 14:40 |
openstackgerrit | YumengBao proposed openstack/watcher-specs: Remove voluptuous and Use JSONSchema as our only JSON validation tool https://review.openstack.org/425204 | 14:43 |
vincentfrancoise | alexchadin: for the metrics memo -> http://paste.openstack.org/show/596460/ | 14:58 |
alexchadin | vincentfrancoise: thx! | 14:58 |
vincentfrancoise | hvprash: I'm off for a coffee break and I'll be back for our 10mn chat? | 14:59 |
hvprash | vincentfrancoise, just ping me when you are back | 15:14 |
sballe_ | acabot | 15:15 |
sballe_ | what what category would a Watcher strategy go into? | 15:15 |
*** vmahe_ has quit IRC | 15:24 | |
acabot | sballe_ : you mean summit talks category ? | 15:44 |
sballe_ | yes | 15:45 |
sballe_ | I am thinking architecture and Operation | 15:45 |
acabot | Architecture & Operations > Ops Tools | 15:46 |
acabot | would be probably the best | 15:46 |
sballe_ | ok thx | 15:47 |
*** vmahe has left #openstack-watcher | 15:49 | |
openstackgerrit | Alexander Chadin proposed openstack/watcher-specs: Add fixes to planner-storage-action-plan spec https://review.openstack.org/425261 | 16:13 |
*** alexchadin has quit IRC | 16:14 | |
openstackgerrit | David TARDIVEL proposed openstack/watcher: check the state of action plan https://review.openstack.org/391383 | 16:16 |
*** vincentfrancoise has quit IRC | 17:21 | |
*** efoley_ has quit IRC | 18:26 | |
*** pshedimb has quit IRC | 19:01 | |
*** pshedimb has joined #openstack-watcher | 19:05 | |
*** dtardivel has quit IRC | 20:14 | |
openstackgerrit | OpenStack Proposal Bot proposed openstack/watcher-dashboard: Updated from global requirements https://review.openstack.org/425399 | 20:32 |
*** wootehfoot has joined #openstack-watcher | 22:27 | |
*** thorst_ has quit IRC | 22:28 | |
*** wootehfoot has quit IRC | 23:15 | |
*** hidekazu has quit IRC | 23:52 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!