| *** vanham has joined #openstack-zaqar | 01:08 | |
| vanham | Hey yall. Good night. | 01:08 |
|---|---|---|
| vanham | I'm trying to setup zaqar here as per https://github.com/openstack/zaqar/blob/master/README.rst | 01:08 |
| vanham | But when I try to run zaqar-server -v it gives me the following http://paste.openstack.org/show/133991/ | 01:09 |
| vanham | (something like object has no attribute 'PKI_ASN1_FORM) | 01:09 |
| *** X019 has joined #openstack-zaqar | 01:19 | |
| *** vanham has quit IRC | 01:30 | |
| *** achanda has quit IRC | 01:39 | |
| *** achanda has joined #openstack-zaqar | 01:43 | |
| *** davideagnello has joined #openstack-zaqar | 02:12 | |
| *** achanda has quit IRC | 02:13 | |
| *** jeffrey4l has joined #openstack-zaqar | 02:28 | |
| *** X019 has quit IRC | 02:32 | |
| *** X019 has joined #openstack-zaqar | 02:44 | |
| *** bradjones has quit IRC | 02:44 | |
| *** fifieldt has quit IRC | 02:45 | |
| openstackgerrit | Jeffrey Zhang proposed openstack/zaqar: Remove the outdated openstack common modules https://review.openstack.org/133886 | 02:45 |
| *** bradjones has joined #openstack-zaqar | 02:46 | |
| openstackgerrit | Jeffrey Zhang proposed openstack/zaqar: pooling conf is using a wrong deprecated conf https://review.openstack.org/132937 | 02:47 |
| *** boris-42 has quit IRC | 03:07 | |
| *** achanda has joined #openstack-zaqar | 03:14 | |
| *** achanda has quit IRC | 03:19 | |
| *** jeffrey4l has quit IRC | 03:51 | |
| *** jeffrey4l has joined #openstack-zaqar | 04:03 | |
| *** achanda has joined #openstack-zaqar | 04:53 | |
| *** davideagnello has quit IRC | 05:16 | |
| *** davideagnello has joined #openstack-zaqar | 05:16 | |
| *** duanhf has joined #openstack-zaqar | 05:26 | |
| *** sgotliv has joined #openstack-zaqar | 05:26 | |
| *** sgotliv has quit IRC | 06:20 | |
| *** davideagnello has quit IRC | 06:30 | |
| *** jeffrey4l has quit IRC | 06:46 | |
| *** exploreshaifali has joined #openstack-zaqar | 06:47 | |
| exploreshaifali | flaper87: around? | 06:57 |
| *** achanda has quit IRC | 07:02 | |
| *** jeffrey4l has joined #openstack-zaqar | 07:06 | |
| *** miqui_ has quit IRC | 07:22 | |
| *** jeffrey4l has quit IRC | 07:37 | |
| *** exploreshaifali has quit IRC | 07:39 | |
| *** exploreshaifali has joined #openstack-zaqar | 07:41 | |
| *** jeffrey4l has joined #openstack-zaqar | 07:43 | |
| *** exploreshaifali has quit IRC | 08:02 | |
| *** exploreshaifali has joined #openstack-zaqar | 08:05 | |
| *** dynarro has joined #openstack-zaqar | 08:08 | |
| *** fifieldt has joined #openstack-zaqar | 08:09 | |
| *** exploreshaifali has quit IRC | 08:16 | |
| *** davideagnello has joined #openstack-zaqar | 08:18 | |
| *** davideagnello has quit IRC | 08:23 | |
| *** duanhf has quit IRC | 08:23 | |
| *** reed has joined #openstack-zaqar | 09:13 | |
| *** sgotliv has joined #openstack-zaqar | 09:20 | |
| *** flaper87 has quit IRC | 09:43 | |
| *** flaper87 has joined #openstack-zaqar | 09:43 | |
| *** jeffrey4l has quit IRC | 09:50 | |
| *** doraly_ has joined #openstack-zaqar | 09:51 | |
| *** sgotliv has quit IRC | 09:57 | |
| *** jeffrey4l has joined #openstack-zaqar | 10:04 | |
| *** achanda has joined #openstack-zaqar | 10:06 | |
| *** achanda has quit IRC | 10:10 | |
| *** boris-42 has joined #openstack-zaqar | 10:33 | |
| *** dynarro has quit IRC | 10:38 | |
| *** doraly_ has quit IRC | 10:38 | |
| *** exploreshaifali has joined #openstack-zaqar | 10:52 | |
| exploreshaifali | flaper87, around? | 11:01 |
| flaper87 | exploreshaifali: hey hey | 11:01 |
| flaper87 | yeah, not for long, almost lunch time | 11:02 |
| flaper87 | but I'll be back later | 11:02 |
| exploreshaifali | flaper87, all right | 11:02 |
| flaper87 | exploreshaifali: what's up? how are you doing? Hope you had a great weekend | 11:02 |
| exploreshaifali | have a happy lunch | 11:02 |
| exploreshaifali | flaper87, yes! | 11:02 |
| exploreshaifali | checkout https://etherpad.openstack.org/p/exploreshaifali-opw-split-layers | 11:03 |
| exploreshaifali | from line 16 | 11:03 |
| * flaper87 clicks | 11:03 | |
| flaper87 | exploreshaifali: 'sup ? | 11:05 |
| flaper87 | exploreshaifali: btw, feel free to submit a patch for that code, we can discuss over reviews too | 11:05 |
| flaper87 | it's probably easier to write comments there | 11:05 |
| flaper87 | exploreshaifali: you can mark the patch as work-in-progress | 11:05 |
| flaper87 | ok brb | 11:05 |
| exploreshaifali | flaper87, okay... what is 'sup? | 11:06 |
| *** flwang1 has joined #openstack-zaqar | 11:14 | |
| flwang1 | flaper87: morning | 11:15 |
| openstackgerrit | Merged openstack/zaqar: Let the test client serialize request's body https://review.openstack.org/134210 | 11:31 |
| openstackgerrit | Shaifali Agrawal proposed openstack/zaqar: Split Control and Data planes of Storage layer https://review.openstack.org/134910 | 11:37 |
| *** vkmc has joined #openstack-zaqar | 11:38 | |
| vkmc | morniiiiiiiiing | 11:40 |
| flwang1 | vkmc: morning | 11:43 |
| vkmc | hey :) | 11:43 |
| *** openstackgerrit has quit IRC | 11:48 | |
| *** openstackgerrit has joined #openstack-zaqar | 11:49 | |
| exploreshaifali | morning vkmc :) | 11:50 |
| vkmc | hi exploreshaifali! | 11:53 |
| vkmc | saw you started working on your bp | 11:53 |
| vkmc | kudos! | 11:53 |
| exploreshaifali | \o/ | 11:53 |
| exploreshaifali | flaper87, vkmc wrote first post http://shaifaliagrawal.wordpress.com/2014/11/16/switch-on-to-openstack/ | 11:54 |
| vkmc | that's so great! | 11:55 |
| * vkmc follows exploreshaifali's blog | 11:56 | |
| exploreshaifali | :) | 11:56 |
| flaper87 | flwang1: morning | 11:56 |
| flaper87 | vkmc: morning | 11:56 |
| * flaper87 is back | 11:57 | |
| flaper87 | exploreshaifali: awesome! | 11:57 |
| exploreshaifali | flaper87, :D | 11:57 |
| vkmc | flaper87, 'sup dud :) | 11:57 |
| flaper87 | exploreshaifali: awesome post | 11:59 |
| flaper87 | vkmc: helllllllllllllllooooooooooooooooooooooooo | 11:59 |
| flaper87 | vkmc: all good, you? | 11:59 |
| flaper87 | hope you had a great w/e | 11:59 |
| vkmc | flaper87, good too! yeah it was great :) I went to the beach hehe | 12:00 |
| flwang1 | flaper87: just post a comment for the remove queue spec | 12:00 |
| vkmc | flaper87, yours? | 12:00 |
| flaper87 | exploreshaifali: is your blog in openstack's planet? | 12:00 |
| flwang1 | flaper87: now working on a poc of notification | 12:00 |
| flaper87 | flwang1: thanks for reviewing the specs | 12:00 |
| flaper87 | we'll discuss them in today's meeting | 12:01 |
| flaper87 | I'll probably send an email to the mailing list about that one | 12:01 |
| exploreshaifali | flaper87, Thanks!, I added my blog in OPW planet but not in openstack planet | 12:01 |
| flaper87 | exploreshaifali: if not, here's how you can add it https://wiki.openstack.org/wiki/AddingYourBlog | 12:01 |
| flwang1 | flaper87: it would be nice | 12:01 |
| exploreshaifali | ok :) | 12:01 |
| flwang1 | flaper87: as for the notification's subscription body, I'm still keen to know your thoughts | 12:02 |
| flaper87 | did you read my last comments? | 12:03 |
| flaper87 | oh, you also replied | 12:03 |
| *** dynarro has joined #openstack-zaqar | 12:03 | |
| *** doraly_ has joined #openstack-zaqar | 12:03 | |
| *** doraly_ has quit IRC | 12:05 | |
| *** dynarro has quit IRC | 12:05 | |
| *** dynarro has joined #openstack-zaqar | 12:05 | |
| *** dynarro_ has joined #openstack-zaqar | 12:06 | |
| flwang1 | flaper87: yep | 12:07 |
| flaper87 | flwang1: replied, I'll update the spec | 12:07 |
| flwang1 | flaper87: cool | 12:07 |
| flwang1 | flaper87: thank you | 12:07 |
| flaper87 | flwang1: fWIW, the s/queue/topics/ change is parallel and completely optional | 12:07 |
| flaper87 | we don't have to do it, I do see value in it | 12:07 |
| flaper87 | if you read my last comments, I proposed keeping the URL and just changing the way it works under the hood | 12:07 |
| flaper87 | that'd allow people to use it in the URL *and* the message body. | 12:08 |
| flwang1 | flaper87: yep, i noticed that | 12:08 |
| flaper87 | taking small steps, basically | 12:08 |
| *** sgotliv has joined #openstack-zaqar | 12:09 | |
| vkmc | flaper87, this should be targeted to k-2 https://review.openstack.org/#/c/134015/2/specs/kilo/migrate-to-topics.rst | 12:09 |
| flaper87 | vkmc: probably, that's something I'd like to discuss in today's meeting | 12:10 |
| flwang1 | vkmc: I think so | 12:10 |
| flaper87 | vkmc: pls, add a comment there | 12:11 |
| vkmc | flaper87, will do | 12:11 |
| flaper87 | the thing is, it's quite a change, leaving it for k-2 means we might introduce some aweful regresions late in the game | 12:11 |
| flaper87 | but generally, yeah, I agree with you both | 12:11 |
| vkmc | we should focus on end persistent transport, notifications and fifo for k-1 | 12:11 |
| flaper87 | k-2 sounds better for a change that is actually optional | 12:11 |
| flaper87 | vkmc: totally agree | 12:11 |
| vkmc | but yeah, lets discuss about this later today :) | 12:12 |
| vkmc | I'll leave some comments in the specs, which btw, are great... thanks for working on those flaper87 | 12:12 |
| flwang1 | flaper87: I think I will miss the meeting | 12:17 |
| flaper87 | flwang1: :( | 12:17 |
| flwang1 | it's UTC 3:00 right? | 12:17 |
| flaper87 | flwang1: anything you want to say now? | 12:18 |
| flaper87 | flwang1: yeah | 12:18 |
| flaper87 | flwang1: pls, drop your comments on those reviews before the meeting | 12:18 |
| flwang1 | http://www.worldtimebuddy.com/?pl=1&lid=100,2179537&h=100 | 12:20 |
| flaper87 | flwang1: also, it'd be great if you could review these patches before you go to sleep https://review.openstack.org/#/q/status:open+project:openstack/zaqar+branch:master+topic:bug/1383604,n,z | 12:20 |
| flaper87 | flwang1: yeah, I know it's late for you | 12:20 |
| flaper87 | flwang1: next week's is 21 UTC, FWIW | 12:20 |
| flaper87 | vkmc: is your spec ready for review? | 12:21 |
| flwang1 | flaper87: sure, I will review them before sleep :) | 12:21 |
| flaper87 | http://www.theregister.co.uk/2013/11/14/amazon_kinesis_launch/ | 12:22 |
| vkmc | flaper87, sure... I have to polish the work items, but maybe you can give me some pointers there | 12:22 |
| flaper87 | http://azure.microsoft.com/en-gb/services/event-hubs/ | 12:22 |
| flaper87 | vkmc: awesome, will do | 12:22 |
| flwang1 | flaper87: vkmc: i'm going to sleep, have a nice day, guys | 12:31 |
| flaper87 | flwang1: g'night! | 12:33 |
| flaper87 | flwang1: thanks | 12:33 |
| *** flwang1 has quit IRC | 12:33 | |
| vkmc | flwang1, nite :) | 12:33 |
| vkmc | flaper87, lemme know when you have a moment to discuss about the notifications spec | 12:34 |
| vkmc | or if you prefer doing that during the meeting | 12:34 |
| flaper87 | vkmc: we can do it now, I'm in review specs mode before the meeting | 12:35 |
| vkmc | flaper87, cool... well, I'm not sure about subscriptions handling | 12:37 |
| vkmc | first, a suscription has to be created | 12:37 |
| vkmc | and then you access them from /v2/subscriptions/{s}/listen | 12:37 |
| vkmc | wouldn't make more sense to have subscriptions as a topic/queue attribute? | 12:38 |
| vkmc | kickban me if I'm saying nonsense | 12:38 |
| flaper87 | that's just if you subscribe using a persistent transport (the listen step, I mean) | 12:38 |
| flaper87 | vkmc: it's not nonsense, you know there are no dumb questions but dumb answers ;) | 12:39 |
| vkmc | kewl | 12:39 |
| vkmc | well, for the case of non persistent transports | 12:39 |
| flaper87 | vkmc: I still have to update the spec but the topic will be part of the subscription body | 12:39 |
| vkmc | the accesing is slightly different, but the same | 12:39 |
| flaper87 | topic/queue | 12:39 |
| vkmc | /v2/subscribe | 12:39 |
| flaper87 | lets just talk about queues until the change happens - assuming it will. | 12:39 |
| vkmc | it will | 12:39 |
| flaper87 | Here's what I had in mind, lemme know what you think | 12:40 |
| vkmc | k | 12:40 |
| *** dynarro has quit IRC | 12:40 | |
| *** dynarro_ has quit IRC | 12:40 | |
| flaper87 | I want to subscribe my mobile APN to messages going into queue A | 12:40 |
| flaper87 | I send a POST request to /v2/subscriptions/ with a body that contains: | 12:40 |
| flaper87 | {"queue": ..., "listener": ..., "hooks?": ...} | 12:41 |
| flaper87 | Every time a messages goes into queue A the notifications trigger will be called and it'll notify all the listeners | 12:41 |
| flaper87 | Does that make sense? | 12:42 |
| vkmc | yeah | 12:42 |
| vkmc | I was expecting suscriptions to be inside the queues endpoint though | 12:43 |
| vkmc | a client subscribes to a queue | 12:43 |
| vkmc | you don't need the client to send a 'I want to subscribe to queueu A' message to the server | 12:43 |
| flaper87 | mmh, that's a good point, btw. | 12:44 |
| flaper87 | mmh, what if I want to subscribe listener L to multiple queues ? | 12:44 |
| vkmc | well yeah, in that case perhaps your approach is better | 12:45 |
| flaper87 | Mind adding a comment for that ? | 12:45 |
| vkmc | sure | 12:45 |
| flaper87 | I'll comment back and hopefully kgriffs|afk will take a look as well | 12:45 |
| vkmc | in the case a listener L want to subscripbe to queue A, B and C, then L will have to POST messages for all the queues | 12:45 |
| vkmc | making it less performant probably | 12:46 |
| vkmc | but, but,... we have to check if leaving subscriptions control to each of the queues makes things easier to handle | 12:47 |
| *** sgotliv has quit IRC | 12:47 | |
| vkmc | in your approach, we should keep a data structure with all the subscriptions and make sure it is updated | 12:48 |
| vkmc | ah well >.> | 12:48 |
| flaper87 | yeah, I kinda prefer your proposal because we could re-use the metadata model | 12:49 |
| *** cpallares has joined #openstack-zaqar | 12:49 | |
| flaper87 | "model | 12:50 |
| flaper87 | " | 12:50 |
| *** dynarro_ has joined #openstack-zaqar | 12:50 | |
| *** dynarro has joined #openstack-zaqar | 12:50 | |
| vkmc | flaper87, I'll leave a comment in the spec so we can discuss it with someone else | 12:50 |
| vkmc | now... lets hide from cpallares | 12:50 |
| flaper87 | but mmh, I wonder if that's clear enough API-wise | 12:50 |
| * vkmc hides | 12:50 | |
| * flaper87 hides | 12:50 | |
| vkmc | what do you mean... clear enough API-wise? | 12:50 |
| * cpallares looks around and doesn't see anyone | 12:51 | |
| flaper87 | /v2/queue_name/subscriptions/ | 12:51 |
| flaper87 | /v2/notifications/subscriptions/ | 12:51 |
| * vkmc thinks 'success!' | 12:51 | |
| vkmc | flaper87, that is subjective | 12:52 |
| flaper87 | sorry, I got distracted | 12:52 |
| vkmc | for me the first one is clearer because I was expecting it to be like that | 12:52 |
| flaper87 | what I wanted to say is: The think I liked about the later is that it keeps both things separate | 12:52 |
| vkmc | yeah | 12:52 |
| flaper87 | but again, I think I prefer your proposal | 12:52 |
| vkmc | but... should those be separated? | 12:53 |
| vkmc | lets check some other messaging solutions | 12:53 |
| flaper87 | and it fits perfectly with the "lets mix these 2 things together" kind of idea we had | 12:53 |
| vkmc | yeah | 12:53 |
| flaper87 | in AMQP, you just listen on a queue what changes is what messages are sent to that queue | 12:54 |
| flaper87 | based on the routing key, etc | 12:54 |
| vkmc | I'm afraid that if we remove the queue concept as a container that would stop making sense | 12:54 |
| vkmc | how would the endpoint look when we remove queues? | 12:54 |
| *** exploreshaifali has quit IRC | 12:55 | |
| flaper87 | vkmc: exactly | 12:56 |
| flaper87 | I think I was thinking in a queue-less scenario when I wrote that spec | 12:57 |
| vkmc | it makes sense | 12:57 |
| * flaper87 thinks: It's funny that queue-less kinda sounds like clueless | 12:57 | |
| vkmc | no moar queuein' up! | 12:58 |
| vkmc | AMQP 1.0 http://qpid.apache.org/releases/qpid-proton-0.8/protocol-engine/python/api/proton.Messenger-class.html#subscribe | 12:59 |
| flaper87 | LOL | 13:00 |
| vkmc | :x | 13:00 |
| vkmc | ./v2/subscribe | 13:00 |
| vkmc | we don't have to forget that AMQP 1.0 != Zaqar though | 13:01 |
| *** sgotliv has joined #openstack-zaqar | 13:01 | |
| vkmc | we should make things the best we can do for our use cases and stuff | 13:01 |
| openstackgerrit | Flavio Percoco proposed openstack/zaqar: Claim post requires a body https://review.openstack.org/134214 | 13:02 |
| flaper87 | vkmc: exactly | 13:03 |
| flaper87 | although, we should consider amqp 1.0 is basically topic based | 13:03 |
| flaper87 | vkmc: https://review.openstack.org/#/c/134214/ <- just updated commit message | 13:03 |
| vkmc | yeah of course | 13:04 |
| vkmc | so | 13:04 |
| vkmc | how would Zaqar look like when we get rid of queues? | 13:04 |
| vkmc | IMO notifications would be tightly coupled to that scenario | 13:04 |
| vkmc | s/queue/topic | 13:05 |
| flaper87 | vkmc: that's still an open question | 13:08 |
| flaper87 | vkmc: as in, 2s... | 13:08 |
| vkmc | http://docs.aws.amazon.com/sns/latest/api/API_Subscribe.html subscriptions in SNS | 13:09 |
| flaper87 | vkmc: https://review.openstack.org/#/c/129192/2/specs/kilo/notification-api.rst,cm <- flwang comments here | 13:09 |
| vkmc | yeap, saw those | 13:09 |
| flaper87 | vkmc: but AWS has 2 separate services | 13:09 |
| vkmc | yeah | 13:09 |
| flaper87 | not sure how good it'd be to follow it semantic wise | 13:09 |
| flaper87 | vkmc: back to notifications | 13:09 |
| vkmc | right now we are more comparable to SNS than SQS | 13:09 |
| flaper87 | actually no, I mixed thse | 13:09 |
| flaper87 | anyway | 13:10 |
| flaper87 | what I wanted to say is that we may want to keep supporting queues in the URL and just change the logic under-the-hood | 13:10 |
| vkmc | oic | 13:10 |
| flaper87 | We could allow both, queue's in the URL and in the message body | 13:10 |
| flaper87 | and encourage people to move away from URL queue's calls | 13:10 |
| flaper87 | but I think that'll create lots of confussion | 13:11 |
| vkmc | I'm afraid so | 13:11 |
| flaper87 | IMHO, we should either go all in or abstain from doing it | 13:11 |
| vkmc | exactly | 13:12 |
| vkmc | something to have in mind is that... we are no longer the 'queueing' system | 13:12 |
| vkmc | if we keep the concept of 'queue' in our API, that would bring some misconceptions | 13:12 |
| flaper87 | right | 13:14 |
| vkmc | 'ya know | 13:14 |
| vkmc | we should be careful about that | 13:15 |
| *** dynarro_ has quit IRC | 13:19 | |
| *** dynarro_ has joined #openstack-zaqar | 13:19 | |
| *** doraly__ has joined #openstack-zaqar | 13:20 | |
| *** dynarro has quit IRC | 13:20 | |
| cpallares | It's colder here than it was in London. It's -3 :'( | 13:33 |
| vkmc | that's low :o | 13:34 |
| vkmc | we are expecting 32 C for today, wanna come? :) | 13:34 |
| * cpallares is freezing | 13:34 | |
| cpallares | vkmc: I'm on the next plane to Bahia Blanca. | 13:35 |
| vkmc | hahaha | 13:35 |
| *** jeffrey4l has quit IRC | 13:41 | |
| *** jeffrey4l has joined #openstack-zaqar | 13:41 | |
| flaper87 | I've been thinking about that | 13:43 |
| flaper87 | I mean, I love winter | 13:44 |
| flaper87 | I probably loe winter more than summer | 13:44 |
| flaper87 | but summer is nicer for "outside" activities | 13:44 |
| flaper87 | which are a "must have" for remotees | 13:44 |
| cpallares | flaper87: You've been thinking about going to Bahia blanca? | 13:44 |
| cpallares | flaper87: Or about -3 weather? | 13:44 |
| flaper87 | cpallares: no, that was a half-written idea :P | 13:44 |
| flaper87 | I've been thinking about going to southern countries during northern winter | 13:45 |
| *** davideagnello has joined #openstack-zaqar | 13:45 | |
| vkmc | chasing the sun | 13:45 |
| flaper87 | or I should probably just move back to Venezuela and forget about seasons | 13:45 |
| flaper87 | erm, fuck no, I'd rather travel every 6 months | 13:45 |
| vkmc | nah | 13:45 |
| flaper87 | not that I don't travel already | 13:45 |
| vkmc | you don't want to do that | 13:45 |
| vkmc | lol | 13:45 |
| flaper87 | s/travel/move/ | 13:46 |
| flaper87 | I could also go to Israel | 13:46 |
| vkmc | I want to move to New Zealand as flwang | 13:46 |
| cpallares | flaper87: Southern California. Sunny all day everyday. | 13:46 |
| vkmc | and live in a hobbit house | 13:46 |
| * cpallares wants to live a hobbit house | 13:47 | |
| cpallares | vkmc: http://hobbithouses.tumblr.com/ | 13:47 |
| flaper87 | vkmc: you know you can have a hobbit house there and have penguins coming for tea every day, right? | 13:47 |
| vkmc | <3 | 13:47 |
| flaper87 | cpallares: but, but, but, it's Cali... :/ | 13:47 |
| flaper87 | Well, I could totally live in Santa Barbara | 13:48 |
| cpallares | LOL | 13:48 |
| *** kragniz has joined #openstack-zaqar | 13:48 | |
| cpallares | flaper87: Weren't you complaining about Santa Barbara? | 13:48 |
| flaper87 | cpallares: no no, SFO | 13:48 |
| cpallares | Or was it Santa Clara? | 13:48 |
| flaper87 | and Santa Clara | 13:48 |
| flaper87 | I would never live in SFO and I'm sick of conference organized in Santa Clara | 13:48 |
| flaper87 | there's NOTHING to do there | 13:48 |
| vkmc | flaper87, building a hobbit house in Argentina would be like going out in your front yard and yelling 'hey, I want to get mugged' | 13:49 |
| flaper87 | I mean, do people even live there? There's just 1 Hyatt, 1 Hilton and VTA | 13:49 |
| flaper87 | at least the VTA has free wifi | 13:49 |
| flaper87 | ..... | 13:49 |
| *** davideagnello has quit IRC | 13:49 | |
| cpallares | vkmc: haha | 13:50 |
| flaper87 | vkmc: LOOOOOOOOOL | 13:50 |
| kragniz | this seems to be where the cool kids live | 13:50 |
| vkmc | cpallares, similar to getting of the plane in Ezeiza with a Red Hat | 13:51 |
| flaper87 | kragniz: this is the coolest channel EVER | 13:51 |
| flaper87 | stay and see for yourself | 13:51 |
| cpallares | vkmc: how is your red hat? did it survive? | 13:51 |
| cpallares | vkmc: Needless to say your cookies did not survive. | 13:51 |
| kragniz | flaper87: may I come and live here if I have nothing to do with messaging? :D | 13:51 |
| cpallares | vkmc: But mostly because I ate them. | 13:52 |
| flaper87 | kragniz: absolutely | 13:52 |
| kragniz | flaper87: cool beans! | 13:52 |
| flaper87 | you'll definitely end up having something to do with messaging | 13:52 |
| vkmc | cpallares, it did! I lost the pin though :( | 13:52 |
| flaper87 | ;) | 13:52 |
| flaper87 | we're good at giving others things to do | 13:52 |
| kragniz | flaper87: D: | 13:52 |
| cpallares | vkmc: ah but that's the coolest part | 13:52 |
| vkmc | cpallares, haha hope you enjoyed those | 13:52 |
| flaper87 | kragniz: does swift re-write ring any bell ? | 13:52 |
| kragniz | vkmc: I want a hobbit hole | 13:52 |
| kragniz | flaper87: D: | 13:52 |
| vkmc | kragniz, feel free to lurk here... its something we frequently do here | 13:53 |
| vkmc | :D | 13:53 |
| vkmc | kragniz, +1 for the hobbit hole! haha | 13:53 |
| flaper87 | I like hobbit holes but as bad as it sounds, when I see them I kind of remember teletubbies' house http://www.talbotfarmlandscapes.co.uk/wp-content/files_mf/telebubbies_682_1002431a.jpg | 13:54 |
| kragniz | vkmc: they look so cosy | 13:54 |
| cpallares | flaper87: LOL | 13:54 |
| kragniz | flaper87: heh | 13:54 |
| vkmc | flaper87, you didn't compare LOTR with Teletubbies | 13:54 |
| vkmc | >.> | 13:54 |
| flaper87 | vkmc: no no, I compared Bilbo's house to Teletubbies' | 13:55 |
| flaper87 | Obviously, the later must have been fans of LOTR | 13:55 |
| kragniz | flaper87: and by extension the rest of the shire! | 13:55 |
| *** kgriffs|afk is now known as kgriffs | 13:55 | |
| flaper87 | I'm just making a point here. The idea is so cool that teletubbies adopted it | 13:56 |
| vkmc | oh no, kgriffs is back | 13:56 |
| flaper87 | Thing is, they added speaking vacuum cleaners to it | 13:56 |
| flaper87 | and other disgusting things | 13:56 |
| cpallares | flaper87: And their tummies are TVs, no? | 13:57 |
| vkmc | flaper87, seems you are a real Teletubbies fan | 13:57 |
| kragniz | we know flaper87's true passions now | 13:57 |
| cpallares | vkmc: He's secretly a fan | 13:57 |
| * kgriffs lurks | 13:57 | |
| vkmc | kragniz, +2 we do! | 13:57 |
| cpallares | vkmc: He's had time to think about their Tolkien archiitecture influence. | 13:57 |
| flaper87 | LOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOL | 13:57 |
| * flaper87 has a younger siblins as cpallares does | 13:57 | |
| vkmc | cpallares, yeah... he have done research on that | 13:58 |
| flaper87 | >.> | 13:58 |
| vkmc | yeah... sure... younger siblings | 13:58 |
| cpallares | vkmc: He's had time to formulate theories | 13:58 |
| vkmc | cpallares, he probably has a notebook with diagrams and stuff | 13:58 |
| kragniz | flaper87: how do you see the parallels between the sun and the eye of sauron? | 13:58 |
| flaper87 | I just wish sauron would've targetted them instead of the Hobbits | 13:58 |
| flaper87 | kragniz: dude, have you seen the face of that sun? It's scarier thatn Sauron's eye | 13:59 |
| flaper87 | iiuuugh | 13:59 |
| flaper87 | than* | 13:59 |
| flaper87 | vkmc: cpallares LOOOL, I do | 13:59 |
| * flaper87 looks around | 13:59 | |
| flaper87 | kgriffs: welcome back to the best channel ever | 14:00 |
| flaper87 | kgriffs: meeting in 1h | 14:00 |
| cpallares | flaper87: Have you had time to formulate why they're all different heights? Is there a symbolism between hobbits, dwarfs, elves, and wizards? :P | 14:00 |
| cpallares | kgriffs: What's your stance on the teletubbies and their appropriation of hobbit houses? | 14:00 |
| kgriffs | lol | 14:00 |
| kgriffs | teletubbies always gave me the creeps | 14:00 |
| vkmc | its a really creepy... tv show? | 14:01 |
| kgriffs | I prefer to not think of them corrupting the Tolkien world | 14:01 |
| kgriffs | :p | 14:01 |
| kgriffs | I lump teletubbies in with barney and other children's shows that should be banned from the planet for the good of all mankind. | 14:03 |
| flaper87 | cpallares: re #2, the just liked it and they didn't think about the fact that they are way taller. This is the reason their house is bigger than the Hobbit's version | 14:03 |
| *** sriram has joined #openstack-zaqar | 14:03 | |
| flaper87 | kgriffs: +2 | 14:03 |
| flaper87 | :P | 14:03 |
| cpallares | vkmc: Looking at a youtube video of it I'd say it's more like wtf than creepy https://www.youtube.com/watch?v=TmQ--bESNXA | 14:04 |
| kragniz | someone is doing research | 14:05 |
| kragniz | help us | 14:05 |
| vkmc | don't make me click on that | 14:05 |
| vkmc | lol | 14:05 |
| flaper87 | cpallares: you are not really expecting me to believe that you hadn't seen the show ever | 14:05 |
| flaper87 | I won't ever believe that | 14:05 |
| kragniz | to be fair, I've only seen clips of it | 14:05 |
| cpallares | flaper87: haha unlike you I actually had young siblings who occasionally watched the show. | 14:06 |
| flaper87 | cpallares: I dooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo have younger siblings | 14:06 |
| flaper87 | kragniz: LUCKY YOU! | 14:07 |
| flaper87 | kragniz: but I won't forget that and be prepared for the next summit. | 14:07 |
| kragniz | flaper87: the bits I saw scarred me for life, though | 14:07 |
| flaper87 | kragniz: tequila + teletubbies | 14:07 |
| cpallares | flaper87: OH GOD NO | 14:07 |
| vkmc | that is a bad idea | 14:07 |
| vkmc | horrible idea | 14:07 |
| cpallares | vkmc: ALWAYS | 14:07 |
| kragniz | flaper87: that doesn't sound good D: | 14:07 |
| flaper87 | cpallares: vkmc don't worry, we'll just watch him suffering | 14:08 |
| cpallares | lol | 14:08 |
| flaper87 | which makes it a good idea for us | 14:08 |
| vkmc | flaper87, we could do something like the Clockwork Orange | 14:08 |
| * kragniz has regrets for joining this channel | 14:08 | |
| vkmc | nah that is too mean | 14:09 |
| flaper87 | kragniz: come one, you'll forget those regrets when we'll start messing with kgriffs | 14:09 |
| flaper87 | vkmc: LOOOL | 14:09 |
| flaper87 | s/one/on/ | 14:09 |
| kragniz | flaper87: heh | 14:10 |
| cpallares | It's always kgriffs fault. | 14:10 |
| flaper87 | kragniz: ^ THIS | 14:10 |
| cpallares | The tequila idea that night was kgriffs fault. | 14:10 |
| * kragniz shakes his fist at kgriffs | 14:10 | |
| cpallares | kgriffs is just a bad influence on all of us. | 14:10 |
| vkmc | Clockwork Orange's eye scene http://media.tumblr.com/tumblr_m3cvklT7Yz1qhfkar.jpg (not suitable for sensitive people) | 14:11 |
| * kgriffs can't help it | 14:11 | |
| vkmc | ah well | 14:11 |
| * vkmc goes back reviewing | 14:11 | |
| * kragniz goes back to figuring out sqlalchemy-migrate wizardry | 14:13 | |
| cpallares | /me goes back to playing in the snow | 14:14 |
| kragniz | cpallares: jelly :( | 14:15 |
| *** mpanetta has joined #openstack-zaqar | 14:25 | |
| *** jchai has joined #openstack-zaqar | 14:25 | |
| *** ametts has joined #openstack-zaqar | 14:28 | |
| *** cpallares has quit IRC | 14:33 | |
| *** cpallares has joined #openstack-zaqar | 14:34 | |
| vkmc | flaper87, https://review.openstack.org/#/c/132348/1 +2 +A? | 14:40 |
| *** malini has joined #openstack-zaqar | 14:44 | |
| *** cpallares has quit IRC | 14:46 | |
| *** cpallares has joined #openstack-zaqar | 14:47 | |
| flaper87 | vkmc: yeah | 14:50 |
| vkmc | k | 14:53 |
| vkmc | meeting in 4'! | 14:56 |
| vkmc | flaper87, I just noticed that cannot +2 +A, I'm not core in zaqar-specs | 14:58 |
| flaper87 | vkmc: oh, mmh | 14:59 |
| flaper87 | why is there a separate group for that? | 14:59 |
| vkmc | dunno | 14:59 |
| flaper87 | it's kgriffs fault for sure | 14:59 |
| vkmc | you tell me | 14:59 |
| vkmc | haha | 14:59 |
| vkmc | probably yes | 14:59 |
| flaper87 | meeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeting | 14:59 |
| vkmc | MEETING! >.< | 15:00 |
| flaper87 | dynarro_: cpallares vkmc kgriffs kragniz flwang zhiyan EVERYONE! ^^^^^^^^^^^^^^ | 15:00 |
| flaper87 | join #openstack-meeting-3 | 15:00 |
| cpallares | hey you got the channel right :P | 15:00 |
| cpallares | flaper87: But it's actually #openstack-meeting-alt-3-alt-2 | 15:00 |
| flaper87 | cpallares: LOL, I think you spent way too much time in Paris | 15:00 |
| flaper87 | ;) | 15:00 |
| flaper87 | :D | 15:01 |
| flaper87 | cpallares: LOOOL | 15:01 |
| vkmc | jeffrey4l, #openstack-meeting-3 | 15:03 |
| vkmc | wpf, #openstack-meeting-3 | 15:03 |
| jeffrey4l | i am in | 15:03 |
| vkmc | :) cool | 15:06 |
| *** exploreshaifali has joined #openstack-zaqar | 15:16 | |
| kragniz | totes joining this | 15:18 |
| vkmc | exploreshaifali, join us on the team meeting :) #openstack-meeting-3 | 15:19 |
| *** amitgandhinz has joined #openstack-zaqar | 15:20 | |
| zhiyan | vkmc: https://review.openstack.org/#/c/103825/ - the spec for osprofiler idea | 15:23 |
| vkmc | hey zhiyan! :) | 15:23 |
| vkmc | thanks for that | 15:23 |
| zhiyan | vkmc: hey!, sorry for the delay sending for the link | 15:24 |
| vkmc | not a problem | 15:24 |
| zhiyan | vkmc: last week i'm a little busy to handle my filmily issue. | 15:25 |
| zhiyan | vkmc: ^^ any ideas pls let me know. and guide me in zaqar pls ! | 15:25 |
| zhiyan | thanks! | 15:25 |
| vkmc | zhiyan, I'll check it after the team meeting (if you want to join, its happening in #openstack-meeting-3) | 15:25 |
| vkmc | zhiyan, and of course, it would be great to give you a hand with Zaqar :) | 15:26 |
| zhiyan | vkmc: BIG thanks! | 15:26 |
| vkmc | :) my pleasure | 15:28 |
| zhiyan | :)) | 15:29 |
| *** achanda has joined #openstack-zaqar | 15:49 | |
| * vkmc picks the minutes | 15:59 | |
| kragniz | flaper87: what did I get volunteered for this time D: | 16:00 |
| kgriffs | vkmc: thanks! | 16:01 |
| vkmc | btw, I setted up the etherpad | 16:02 |
| vkmc | for the serialization protocols selection https://etherpad.openstack.org/p/zaqar-serialization-protocol-ws | 16:02 |
| vkmc | there is nothing relevant yet | 16:02 |
| *** davideagnello has joined #openstack-zaqar | 16:07 | |
| *** achanda has quit IRC | 16:09 | |
| *** reed has quit IRC | 16:13 | |
| *** reed has joined #openstack-zaqar | 16:13 | |
| *** davideagnello has quit IRC | 16:28 | |
| * flaper87 needs to write some code, really | 16:31 | |
| kragniz | flaper87: you're a manager now | 16:32 |
| kragniz | flaper87: welcome to the ruling class | 16:32 |
| flaper87 | kragniz: omg, of all the insults you could've used, you had to use that one | 16:33 |
| kragniz | flaper87: I went there | 16:33 |
| flaper87 | damn, that one was a very low kick | 16:33 |
| * kragniz makes flaper87's fingers type code | 16:34 | |
| vkmc | lunch time >:) | 16:50 |
| kragniz | vkmc: the evil face makes it sound like your lunch is nefarious in some way | 16:51 |
| openstackgerrit | Flavio Percoco proposed openstack/zaqar: Update oslo-config-generation code https://review.openstack.org/135012 | 17:00 |
| *** reed has quit IRC | 17:02 | |
| openstackgerrit | Shaifali Agrawal proposed openstack/zaqar: Split Control and Data planes of Storage layer https://review.openstack.org/134910 | 17:04 |
| flaper87 | yay, I submitted a patch | 17:10 |
| flaper87 | >.> | 17:10 |
| *** reed has joined #openstack-zaqar | 17:15 | |
| *** sgotliv has quit IRC | 17:19 | |
| *** dynarro_ has quit IRC | 17:28 | |
| *** doraly__ has quit IRC | 17:28 | |
| exploreshaifali | flaper87, I too submitted a patch ;) | 17:32 |
| kragniz | patches everywhere! \o/ | 17:33 |
| exploreshaifali | \o/ | 17:33 |
| *** davideagnello has joined #openstack-zaqar | 17:56 | |
| exploreshaifali | if possible pleaset review https://review.openstack.org/#/c/134910/ so that I can start for step 2 | 18:06 |
| exploreshaifali | flaper87, ^ | 18:06 |
| vkmc | exploreshaifali, you explained the different steps in an etherpad? | 18:14 |
| vkmc | make sure to make commit messages more general because those are going to be read by people that are not familiar with your internship | 18:14 |
| exploreshaifali | vkmc, I will keep that in mind regarding commit messages :) | 18:15 |
| vkmc | sure :) | 18:15 |
| *** amitgandhinz has quit IRC | 18:15 | |
| exploreshaifali | vkmc, in ether pad same steps were explained | 18:15 |
| vkmc | great | 18:15 |
| exploreshaifali | vkmc, https://etherpad.openstack.org/p/exploreshaifali-opw-split-layers | 18:16 |
| vkmc | exploreshaifali, cool, I'm taking a look | 18:19 |
| exploreshaifali | yeah :) | 18:19 |
| vkmc | something I'm not sure about is L84 | 18:19 |
| vkmc | I mean... you want to separate MANAGEMENT from MESSAGES | 18:20 |
| vkmc | and by summing the options you are duplicating conf params | 18:20 |
| exploreshaifali | vkmc, L84 or L94 here https://review.openstack.org/#/c/134910/2/zaqar/queues/storage/mongodb/options.py | 18:22 |
| vkmc | L84 in your change | 18:22 |
| vkmc | MESSAGE_MONGODB_OPTIONS = COMMON_OPTIONS + ( ... | 18:22 |
| exploreshaifali | ok got it | 18:23 |
| vkmc | I'm taking a look to the zaqar-scratch etherpad | 18:23 |
| vkmc | I might be missing something | 18:23 |
| *** achanda has joined #openstack-zaqar | 18:25 | |
| exploreshaifali | there are many options which are common in both, such options are in COMMON_OPTIONS | 18:25 |
| *** cpallares has quit IRC | 18:25 | |
| exploreshaifali | at one place | 18:26 |
| *** cpallares has joined #openstack-zaqar | 18:26 | |
| *** jchai is now known as jchai_afk | 18:27 | |
| exploreshaifali | and then created 2 different variables to hold all the options needed for corresponding plane | 18:27 |
| exploreshaifali | vkmc, ^* | 18:27 |
| exploreshaifali | vkmc, also if there is surety that MANAGEMENT_MONGODB_OPTIONS will not going to alter in future | 18:29 |
| exploreshaifali | we can get rid of L83 too | 18:29 |
| *** achanda has quit IRC | 18:30 | |
| vkmc | that sounds good but | 18:32 |
| vkmc | for one plane you are using the common options | 18:32 |
| vkmc | and for the other one, the common options plus the partitions conf | 18:32 |
| vkmc | one is the subset of the other | 18:33 |
| vkmc | options should be disjoint sets | 18:33 |
| exploreshaifali | it is because one need more options than other | 18:33 |
| exploreshaifali | oh they should be disjoint set! | 18:34 |
| vkmc | that is what I want to understand :) if one if for management and the other is for messages, then the options should be different | 18:34 |
| vkmc | but maybe I'm missing something | 18:35 |
| *** achanda has joined #openstack-zaqar | 18:35 | |
| vkmc | ah well, apart from that doubt I have the change looks good exploreshaifali | 18:35 |
| vkmc | let's wait for other reviewers comments | 18:35 |
| exploreshaifali | as per as I understood, all these options are specifying details of a database for zaqar.conf file | 18:35 |
| vkmc | exactly yes :) | 18:36 |
| exploreshaifali | so now if we have two sections of mongodb one for management database and one for message database | 18:36 |
| exploreshaifali | we will be having some options common | 18:37 |
| *** jchai_afk is now known as jchai | 18:37 | |
| exploreshaifali | but their values could be different | 18:37 |
| exploreshaifali | that is the problem | 18:37 |
| vkmc | yup | 18:37 |
| vkmc | options would overlap | 18:37 |
| exploreshaifali | okay, all the alteration one want to do in zaqar.conf, should be done manually | 18:38 |
| exploreshaifali | right? | 18:38 |
| vkmc | yeah | 18:38 |
| vkmc | you deploy Zaqar by setting the required confs in zaqar.conf | 18:38 |
| exploreshaifali | yes | 18:39 |
| exploreshaifali | hmm... so we can have few common options like database-name different | 18:39 |
| exploreshaifali | that is it should not be in common | 18:39 |
| exploreshaifali | other few options like max_entry and max_retry can be in common | 18:40 |
| exploreshaifali | variable | 18:40 |
| vkmc | so... for what I can understand from https://etherpad.openstack.org/p/zaqar-scratch | 18:43 |
| vkmc | (wait for it) lol | 18:45 |
| exploreshaifali | :D | 18:46 |
| vkmc | yeah still doesn't make much sense for me | 18:47 |
| vkmc | flaper87, kgriffs ^ | 18:47 |
| exploreshaifali | one option is there | 18:47 |
| exploreshaifali | let me try to put it on ethrpad | 18:47 |
| vkmc | sure | 18:47 |
| exploreshaifali | vkmc, stucked need to think more | 18:51 |
| exploreshaifali | vkmc, from where all these options get value? | 18:52 |
| vkmc | exploreshaifali, lets wait for flaper87's feedback on this :) | 18:53 |
| vkmc | exploreshaifali, all those options are crunched in zaqar.conf | 18:53 |
| exploreshaifali | like there is an option `'ssl_keyfile` | 18:53 |
| vkmc | some of them have default values, some of them not | 18:53 |
| exploreshaifali | it does not have default value | 18:53 |
| vkmc | but the user is the one who was to initialize those according to their settings | 18:53 |
| exploreshaifali | and in zaqar.conf it will be commented | 18:53 |
| exploreshaifali | fine | 18:54 |
| kgriffs | vkmc: what doesn't make sense? | 18:54 |
| exploreshaifali | then we can do somethind like | 18:54 |
| vkmc | kgriffs, the separation of MONGODB group in MANAGEMENT and MESSAGES | 18:54 |
| vkmc | I was checking this etherpad https://etherpad.openstack.org/p/zaqar-scratch | 18:54 |
| vkmc | it probably makes sense, but I don't understand it :| | 18:55 |
| kgriffs | oh | 18:57 |
| exploreshaifali | can we do something like just create a function that will set detault values of different options | 18:57 |
| kgriffs | iirc, the idea was that most options are pretty generic and can be shared regardless of whether the DB connection is being set up for the admin data or the message data | 18:57 |
| exploreshaifali | from line 5 | 18:57 |
| kgriffs | basically, connection specific options that are defined by the python driver for said data stores are the same | 18:58 |
| exploreshaifali | kgriffs, https://review.openstack.org/#/c/134910/2/zaqar/queues/storage/mongodb/options.py | 18:59 |
| vkmc | kgriffs, yeah, that makes sense | 18:59 |
| kgriffs | then there are options that are specific to the "application layer", i.e., our project's business logic | 18:59 |
| kgriffs | so that's the basic idea | 18:59 |
| *** amitgandhinz has joined #openstack-zaqar | 19:00 | |
| * kgriffs clicks link | 19:00 | |
| *** flwang1 has joined #openstack-zaqar | 19:01 | |
| *** malini1 has joined #openstack-zaqar | 19:01 | |
| *** flwang1 has quit IRC | 19:03 | |
| *** malini has quit IRC | 19:03 | |
| kgriffs | vkmc: ok, so what is the concern with this patch? | 19:03 |
| vkmc | kgriffs, IIRC that would generate two groups with common options duplicated in zaqar.conf | 19:04 |
| vkmc | is it like that? | 19:05 |
| kgriffs | yes, that is correct. But you really are configuring two different backends, one for management and one for messages | 19:05 |
| kgriffs | you should be able to configure them with different settings | 19:05 |
| vkmc | oic | 19:05 |
| kgriffs | I suppose we could have a "common" section and "inherit | 19:06 |
| vkmc | that is what I was missing then :) | 19:06 |
| exploreshaifali | kgriffs, but I think one option database name should also be not in common variable | 19:06 |
| kgriffs | but I think that would actually be more confusing | 19:06 |
| exploreshaifali | database name should also be different for both management and messages isn't it? | 19:07 |
| exploreshaifali | I didn't got your point regarding 'commom' and 'inherit' | 19:07 |
| kgriffs | exploreshaifali: right, if the operator wants to point both drivers to the same DB cluster, it would be prudent to use different DB names/namespaces | 19:08 |
| kgriffs | exploreshaifali: re inherit | 19:08 |
| kgriffs | I was just saying we could define a common section in the config where you could put options that are common across driver types | 19:09 |
| kgriffs | but then you would have to have a way to override them for each driver | 19:09 |
| exploreshaifali | okay | 19:09 |
| kgriffs | and in the end I think it is simpler to just say you need to configure all the options independently | 19:09 |
| kgriffs | especially since I think we are going to recommend using different backend clusters for each anyway | 19:10 |
| exploreshaifali | yes this would be more easy | 19:10 |
| kgriffs | so I think the patch as-is makes sense | 19:10 |
| kgriffs | also keep in mind that for every DB there may not be both management and message drivers. | 19:11 |
| kgriffs | aaaaanyway | 19:11 |
| vkmc | it does :) thanks exploreshaifali and sorry for the confusion I caused | 19:11 |
| kgriffs | one question I had though | 19:11 |
| vkmc | I learned a few things ;) | 19:11 |
| kgriffs | was if we need to clone those options | 19:11 |
| exploreshaifali | vkmc, not at all it made things clear to me too | 19:12 |
| kgriffs | or does oslo config make copies when you register the options? | 19:12 |
| exploreshaifali | kgriffs, why do we need to clone the options? | 19:13 |
| exploreshaifali | just to have a recovery? | 19:13 |
| kgriffs | let me just grep the 2,448 lines of cfg.py | 19:13 |
| kgriffs | (srsly. you couldn't break that module up to make it any smaller?) | 19:14 |
| exploreshaifali | sure | 19:14 |
| vkmc | IIRC oslo clone those kgriffs | 19:14 |
| vkmc | when those are registered | 19:14 |
| kgriffs | https://github.com/openstack/oslo.config/blob/master/oslo/config/cfg.py#L1805 | 19:15 |
| kgriffs | doesn't look like it | 19:16 |
| kgriffs | question is, does the state get stored directly by those opts... | 19:16 |
| vkmc | now I don't know for sure :) | 19:19 |
| kgriffs | hmmm, there is this namespace thingy... | 19:20 |
| kgriffs | https://github.com/openstack/oslo.config/blob/master/oslo/config/cfg.py#L2118 | 19:20 |
| kgriffs | https://github.com/openstack/oslo.config/blob/master/oslo/config/cfg.py#L2323 | 19:21 |
| kgriffs | OK, looks like the Opt class proxies to an instance of _Namespace class | 19:22 |
| kgriffs | https://github.com/openstack/oslo.config/blob/master/oslo/config/cfg.py#L1414 | 19:22 |
| * vkmc watches how kgriffs brain works | 19:22 | |
| exploreshaifali | lol vkmc | 19:23 |
| kgriffs | and yeah, that in turn just proxies to config parser instance | 19:23 |
| kgriffs | https://github.com/openstack/oslo.config/blob/master/oslo/config/cfg.py#L1529 | 19:23 |
| kgriffs | TLDR; We can share instances of Opt since state is actually maintained by this _Namespace class | 19:24 |
| vkmc | so... options are not cloned but the state is handled differently according to the namespace | 19:26 |
| vkmc | in this case, the group name | 19:26 |
| vkmc | hopefully that's what kgriffs mean | 19:30 |
| vkmc | lol | 19:30 |
| kgriffs | yeah, I'm still looking | 19:30 |
| vkmc | thanks kgriffs :) | 19:30 |
| kgriffs | I'm still fuzzy re the relationship between groups and namespace | 19:30 |
| kgriffs | it seems like namespace may just include all groups actually | 19:30 |
| kgriffs | so maybe namespace is "config file" vs. "cli options"? | 19:31 |
| kgriffs | i don't know | 19:31 |
| kgriffs | :p | 19:31 |
| kgriffs | anyway, I think we know enough for now | 19:31 |
| vkmc | let's try and see what happens (?) | 19:32 |
| vkmc | zhiyan, I checked out the OSProfiler spec | 19:39 |
| vkmc | zhiyan, it looks great! | 19:39 |
| vkmc | zhiyan, and I think Zaqar would be really beneficiated from it | 19:39 |
| vkmc | I'll write the spec so other devs can review it | 19:39 |
| vkmc | probably I'll have it done by tomorrow | 19:39 |
| exploreshaifali | kgriffs, whenver you get time, please give a thought to flaper87's last comment here https://bugs.launchpad.net/python-zaqarclient/+bug/1358889 | 19:43 |
| *** jchai is now known as jchai_afk | 19:56 | |
| kgriffs | exploreshaifali: commented | 20:00 |
| exploreshaifali | kragniz, thanks, will look at it :) | 20:01 |
| exploreshaifali | oh sorry kgriffs ^^ | 20:01 |
| *** jchai_afk is now known as jchai | 20:09 | |
| *** sgotliv has joined #openstack-zaqar | 20:14 | |
| *** sgotliv has quit IRC | 20:14 | |
| *** sgotliv has joined #openstack-zaqar | 20:14 | |
| openstackgerrit | Victoria MartÃnez de la Cruz proposed openstack/zaqar-specs: Add a persistent transport alternative https://review.openstack.org/134567 | 20:21 |
| *** flwang1 has joined #openstack-zaqar | 20:25 | |
| *** sgotliv has quit IRC | 20:43 | |
| *** exploreshaifali has quit IRC | 20:45 | |
| openstackgerrit | Victoria MartÃnez de la Cruz proposed openstack/zaqar-specs: Add a persistent transport alternative https://review.openstack.org/134567 | 20:48 |
| *** openstackgerrit has quit IRC | 20:49 | |
| *** openstackgerrit has joined #openstack-zaqar | 20:49 | |
| openstackgerrit | Flavio Percoco proposed openstack/zaqar: Update oslo-config-generation code https://review.openstack.org/135012 | 20:50 |
| flaper87 | vkmc: kgriffs ^ pls :) it's an easy review | 20:50 |
| flaper87 | I think claims are broken in the sqlalchemy driver | 20:51 |
| flaper87 | claims expiration | 20:51 |
| flaper87 | oh lord, I hate this driver | 20:51 |
| flaper87 | oh wait, mmh | 20:54 |
| flaper87 | damn | 20:54 |
| *** achanda has quit IRC | 20:54 | |
| * vkmc hides | 20:55 | |
| *** davideagnello has quit IRC | 20:59 | |
| *** sgotliv has joined #openstack-zaqar | 21:00 | |
| flaper87 | kgriffs: do you recall why we're doing this? https://git.openstack.org/cgit/openstack/zaqar/tree/zaqar/queues/storage/sqlalchemy/messages.py#n302 | 21:01 |
| flaper87 | I mean, if the claim expired and someone is deleting the message, why are we raising that exception? | 21:01 |
| *** davideagnello has joined #openstack-zaqar | 21:01 | |
| flaper87 | ah wait, there's an if statement missing there | 21:02 |
| flaper87 | I htink | 21:03 |
| kgriffs | looking | 21:03 |
| flaper87 | yeah, htink... >.> | 21:03 |
| kgriffs | i vaguely recall this being part of an effort to make all the drivers behave consistently | 21:06 |
| flaper87 | the error there is that we're not checking if the claim has expired | 21:07 |
| flaper87 | which we do for the mongo driver | 21:07 |
| kgriffs | oic | 21:07 |
| flaper87 | damn, I hate this driver | 21:07 |
| kgriffs | aren't we about to kill it? | 21:07 |
| kgriffs | :) | 21:07 |
| kgriffs | or do we need to keep it around after all? | 21:07 |
| flaper87 | no no, I already wrote a spec for that | 21:08 |
| flaper87 | :P | 21:08 |
| flaper87 | https://review.openstack.org/#/c/134248/ | 21:08 |
| kgriffs | the downside is people will say they don't want to deploy anything other than mysql or pg | 21:08 |
| flaper87 | one thing though, I think we mentioned something about keeping it around for the management thing | 21:08 |
| kgriffs | flaper87: yeah, that's true | 21:08 |
| kgriffs | but that is a small subset of the current driver | 21:08 |
| flaper87 | we can keep it outside the code base | 21:08 |
| flaper87 | we don't need to delete the code | 21:08 |
| flaper87 | seriously, we ain't have enough resource to maintain this driver | 21:09 |
| kgriffs | that was my next question | 21:09 |
| flaper87 | if someone wants to step up, I think we'll be more than happy to keep it | 21:09 |
| flaper87 | resources* | 21:09 |
| kgriffs | BTW, keystone middleware is still using memcached directly, not dogpile? | 21:11 |
| kgriffs | seems like that is what I saw last time I checked | 21:11 |
| kgriffs | I'd like to get All the Things able to use Redis | 21:11 |
| flaper87 | I think they are already using dogpile | 21:12 |
| kgriffs | keystone service is | 21:12 |
| flaper87 | wah ok | 21:12 |
| kgriffs | but I'm not so sure about the middleware | 21:12 |
| flaper87 | then no, it's not | 21:12 |
| kgriffs | if only I could clone myself I'd be happy to work on that. :p | 21:13 |
| kgriffs | it would be nice to sort of get Redis into lots of OS deployments | 21:13 |
| kgriffs | then projects can leverage it in interesting ways without running into the "we don't want to have to manage one more DB thing" pushback | 21:14 |
| kgriffs | IMHO, it is a good candidate for "that other thing" beyond MySQL | 21:14 |
| kgriffs | whereas memcached is fairly limited, Redis can do a lot more and would unlock some interesting use cases for projects, not just Zaqar. | 21:15 |
| *** malini1 has quit IRC | 21:15 | |
| kgriffs | </soapbox> | 21:15 |
| flaper87 | LOL | 21:15 |
| flaper87 | but yeah, I agree with | 21:15 |
| flaper87 | you | 21:15 |
| flaper87 | :P | 21:16 |
| flaper87 | btw, I dunno how the sqlalchemy thing passes tests | 21:16 |
| flaper87 | really | 21:16 |
| flaper87 | it's dark magic | 21:16 |
| kgriffs | lol | 21:16 |
| kgriffs | no, it's Alchemy | 21:16 |
| kgriffs | OS X dictionary defines Alchemy as " a seemingly magical process of transformation, creation, or combination" | 21:17 |
| kgriffs | lol | 21:17 |
| kgriffs | dark magic indeed. :p | 21:17 |
| *** fifieldt_ has joined #openstack-zaqar | 21:22 | |
| vkmc | a magical process of joining stuff together and wasting tons of resources | 21:22 |
| kgriffs | I think if we add a swift driver then that can be our "i don't like mongo" alternative | 21:23 |
| kgriffs | as well as our "i don't want to deploy anything else than what I already have" answer | 21:23 |
| *** fifieldt has quit IRC | 21:23 | |
| flaper87 | really, I dunno how this driver works | 21:23 |
| flaper87 | we don't even check on claims expire time | 21:23 |
| flaper87 | when getting messages | 21:23 |
| kgriffs | speaking of which, do we have a spec for the swift work? | 21:23 |
| flaper87 | we just check on the id | 21:23 |
| flaper87 | kgriffs: zhiyan will work on one | 21:24 |
| mpanetta | kgriffs: You don't like mongo? *sniff sniff* :P | 21:24 |
| kgriffs | oh, I like it | 21:24 |
| notmyname | http://2.bp.blogspot.com/-YhcOntV_57M/UslUK8g059I/AAAAAAAAPy4/f8IZZVotRxQ/s1600/put-that-evil-on-me.gif | 21:24 |
| flaper87 | I had a really quick chat with notmyname about it | 21:24 |
| flaper87 | notmyname: LOL | 21:24 |
| flaper87 | It might work if we figure out how to align all plates on December 31st at midnight (PST) before the firworks happen in China | 21:25 |
| flaper87 | kgriffs: ^ | 21:25 |
| flaper87 | planets* | 21:25 |
| flaper87 | fireworks* | 21:25 |
| flaper87 | YES, I'M HUNGRY | 21:25 |
| kgriffs | lol | 21:25 |
| flaper87 | and that's the easy part | 21:26 |
| flaper87 | :P | 21:26 |
| kgriffs | heh | 21:26 |
| kgriffs | so, the heat guys were keen on the idea of using swift | 21:26 |
| notmyname | or are they keen on not deploying mongo? | 21:26 |
| kgriffs | were there other people asking for it besides them? | 21:26 |
| kgriffs | notmyname: idk, good question | 21:26 |
| kgriffs | I think they are already using swift and it just seemed like a natural progression? | 21:27 |
| * kgriffs is admittedly fuzzy on the details | 21:27 | |
| *** achanda has joined #openstack-zaqar | 21:28 | |
| *** exploreshaifali has joined #openstack-zaqar | 21:30 | |
| flaper87 | kgriffs: btw, i think we've a problem. That patch that did some magic with the logging might have enabled more logging than it should | 21:30 |
| flaper87 | kgriffs: https://review.openstack.org/#/c/135012/ <- now >.> | 21:31 |
| flaper87 | it's an EASY one | 21:31 |
| flaper87 | :D | 21:31 |
| kgriffs | bash no mor? srlsy? | 21:31 |
| flaper87 | YEAH! | 21:31 |
| kgriffs | http://stream1.gifsoup.com/view1/1603965/bill-and-ted-air-guitar-o.gif | 21:32 |
| flaper87 | and no more sample file to keep update | 21:32 |
| flaper87 | (well that one already landed) | 21:32 |
| flaper87 | hahah | 21:32 |
| * flaper87 wants a new badge everytime he does something like: #0day +13, -218 | 21:32 | |
| kgriffs | hmm | 21:32 |
| * flaper87 wants a new badge everytime he does something like: +13, -218 | 21:32 | |
| kgriffs | https://review.openstack.org/#/c/135012/2/etc/oslo-config-generator/zaqar.conf | 21:32 |
| kgriffs | how do I know all the right namespaces are included? ;) | 21:33 |
| flaper87 | kgriffs: I copied them from our previous file | 21:33 |
| flaper87 | :P | 21:33 |
| kgriffs | wfm | 21:33 |
| flaper87 | *cough* you should know them all, btw *cough* | 21:33 |
| * flaper87 doesn't even know what a namespace is | 21:33 | |
| * vkmc stopped reading for a while now | 21:34 | |
| kgriffs | flaper87: looks like tempest is borked | 21:36 |
| kgriffs | https://review.openstack.org/#/c/135012 | 21:36 |
| flaper87 | I hate it when tests force me to do the right thing | 21:38 |
| flaper87 | jsut rechecked, it can't be related to the patch | 21:39 |
| * flaper87 should've read the logs | 21:39 | |
| flaper87 | :s | 21:39 |
| *** reed has quit IRC | 21:41 | |
| *** reed has joined #openstack-zaqar | 21:42 | |
| *** reed has quit IRC | 21:42 | |
| *** jchai is now known as jchai_afk | 21:44 | |
| *** reed has joined #openstack-zaqar | 21:52 | |
| *** amitgandhinz has quit IRC | 21:58 | |
| *** amitgandhinz has joined #openstack-zaqar | 21:58 | |
| *** sriram has quit IRC | 22:07 | |
| *** jchai_afk is now known as jchai | 22:08 | |
| *** kgriffs is now known as kgriffs|afk | 22:19 | |
| *** mpanetta has quit IRC | 22:20 | |
| *** exploreshaifali has quit IRC | 22:20 | |
| *** davideagnello has quit IRC | 22:36 | |
| *** davideagnello has joined #openstack-zaqar | 22:36 | |
| *** mpanetta has joined #openstack-zaqar | 22:39 | |
| *** cpallares has quit IRC | 22:46 | |
| *** jchai has quit IRC | 22:50 | |
| *** mpanetta has quit IRC | 23:02 | |
| *** vkmc has quit IRC | 23:11 | |
| *** kgriffs|afk is now known as kgriffs | 23:19 | |
| *** ametts has quit IRC | 23:23 | |
| *** kgriffs is now known as kgriffs|afk | 23:29 | |
| *** amitgandhinz has quit IRC | 23:37 | |
| *** reed has quit IRC | 23:41 | |
| *** davideagnello has quit IRC | 23:48 | |
| *** davideagnello has joined #openstack-zaqar | 23:53 | |
| *** achanda has quit IRC | 23:57 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
