| *** mpanetta has joined #openstack-zaqar | 00:13 | |
| *** mpanetta has quit IRC | 00:17 | |
| *** cpallares has quit IRC | 02:35 | |
| *** achanda has joined #openstack-zaqar | 02:56 | |
| *** mpanetta has joined #openstack-zaqar | 03:14 | |
| *** mpanetta has quit IRC | 03:19 | |
| *** achanda has quit IRC | 03:39 | |
| *** achanda has joined #openstack-zaqar | 03:39 | |
| *** achanda has quit IRC | 03:40 | |
| *** achanda has joined #openstack-zaqar | 03:40 | |
| *** achanda has quit IRC | 03:44 | |
| *** vkmc has quit IRC | 03:55 | |
| *** achanda has joined #openstack-zaqar | 04:23 | |
| *** exploreshaifali has joined #openstack-zaqar | 04:50 | |
| *** amalagon has joined #openstack-zaqar | 04:54 | |
| *** amalagon has quit IRC | 04:54 | |
| *** achanda has quit IRC | 05:00 | |
| *** amalagon has joined #openstack-zaqar | 05:35 | |
| *** sgotliv has joined #openstack-zaqar | 05:36 | |
| *** amalagon has quit IRC | 05:39 | |
| *** achanda has joined #openstack-zaqar | 06:35 | |
| *** sgotliv has quit IRC | 06:43 | |
| *** achanda has quit IRC | 06:59 | |
| *** achanda has joined #openstack-zaqar | 07:00 | |
| *** achanda has quit IRC | 07:05 | |
| exploreshaifali | flaper87, gooood morning | 07:10 |
|---|---|---|
| exploreshaifali | flaper87, please can you review https://review.openstack.org/#/c/134910/ | 07:10 |
| exploreshaifali | issue with dsvm-redis | 07:10 |
| flaper87 | exploreshaifali: gooooooooooood morning | 07:21 |
| flaper87 | :) | 07:21 |
| flaper87 | sure thing, I'll review it in a bit | 07:21 |
| exploreshaifali | flaper87, all right ! | 07:22 |
| exploreshaifali | :) | 07:22 |
| flaper87 | exploreshaifali: you need to configure the deprecated options for redis too | 07:27 |
| flaper87 | :) | 07:27 |
| flaper87 | exploreshaifali: just like you did for mongodb | 07:27 |
| exploreshaifali | flaper87, oh yes !! | 07:28 |
| exploreshaifali | flaper87, I have few question too | 07:28 |
| flaper87 | exploreshaifali: go ahead | 07:28 |
| exploreshaifali | test file - `tests/etc/wsgi_mongodb` | 07:28 |
| flaper87 | exploreshaifali: btw, do you know how to debug gate issues? | 07:28 |
| exploreshaifali | is for message databse only | 07:29 |
| exploreshaifali | right ? | 07:29 |
| exploreshaifali | gate issues ? | 07:29 |
| flaper87 | exploreshaifali: yes, that's for the data plane | 07:29 |
| flaper87 | exploreshaifali: gate issue -> the dsvm-redis gate is not working. What's going on? | 07:29 |
| flaper87 | :P | 07:29 |
| exploreshaifali | and the test firl `tests/etc/wsgi_mongodb_pooled.conf` is for control plane | 07:29 |
| flaper87 | exploreshaifali: in that one you need to configure both, control and data plane | 07:30 |
| exploreshaifali | flaper87, yes I don't know much about it why that dsvm issues occuring | 07:30 |
| flaper87 | exploreshaifali: you may want to watch this: https://www.youtube.com/watch?v=fowBDdLGBlU | 07:30 |
| exploreshaifali | flaper87, okay :) | 07:31 |
| exploreshaifali | and here https://review.openstack.org/#/c/134910/8/tests/etc/wsgi_mongodb_pooled.conf | 07:31 |
| exploreshaifali | I want to configure both | 07:31 |
| exploreshaifali | data and plane storage ? | 07:31 |
| flaper87 | exploreshaifali: correct | 07:31 |
| flaper87 | exploreshaifali: do you know why? | 07:31 |
| exploreshaifali | I think it should not because it is related to pool stuff | 07:32 |
| exploreshaifali | we need only management part for pools and queues | 07:32 |
| flaper87 | exploreshaifali: here's a general rule. The data plane should *always* be configured otherwise messages won't go anyway | 07:32 |
| flaper87 | The management plane is required to enable pools and flavors | 07:33 |
| exploreshaifali | flaper87, what exactly a flavor is ? | 07:33 |
| exploreshaifali | I don't understand it well | 07:34 |
| exploreshaifali | any page from where I can read about them | 07:34 |
| flaper87 | exploreshaifali: well, actually, the message plane might not be needed if one uses pools and flavors but we need to discuss this a bit further | 07:35 |
| flaper87 | exploreshaifali: sure, 2 secs | 07:35 |
| flaper87 | exploreshaifali: I wrote about pools here: http://blog.flaper87.com/post/zaqar-pools-explained/ | 07:35 |
| flaper87 | exploreshaifali: flavors are a user-facing resource that exposes pools in a consumable way | 07:36 |
| exploreshaifali | flaper87, just one more - here https://review.openstack.org/#/c/134910/8/zaqar/queues/storage/utils.py we don't need management store | 07:36 |
| flaper87 | ok, that description is actually very confusing | 07:36 |
| flaper87 | in other words, flavors allow you to choose a group of pools based on their capabilities | 07:36 |
| flaper87 | exploreshaifali: no, IIRC, the management store is not needed there | 07:37 |
| exploreshaifali | okay - flavors are the combinations of capabilities | 07:37 |
| flaper87 | exploreshaifali: correct | 07:37 |
| flaper87 | damn, I hate that dynamic_conf function | 07:37 |
| exploreshaifali | hahaha :D | 07:38 |
| exploreshaifali | flaper87, okay.. Thanks !! | 07:39 |
| exploreshaifali | flaper87, btw I am having my college exams from Dec 4 to 11 | 07:40 |
| exploreshaifali | so may be I won't be much active in those days | 07:40 |
| flaper87 | exploreshaifali: ok ok, thanks for letting me know | 07:41 |
| flaper87 | :D | 07:41 |
| flaper87 | exploreshaifali: and very good luck with that | 07:42 |
| flaper87 | you've got a whole bunch of zaqarians supporting you | 07:42 |
| exploreshaifali | flaper87, Thanks a loooooot !!! | 07:42 |
| exploreshaifali | yes you all are there | 07:42 |
| exploreshaifali | I know ;) | 07:42 |
| *** exploreshaifali has quit IRC | 07:56 | |
| *** yfujioka has joined #openstack-zaqar | 08:00 | |
| *** dynarro has joined #openstack-zaqar | 08:00 | |
| *** sgotliv has joined #openstack-zaqar | 08:36 | |
| *** ekarlso- has quit IRC | 08:52 | |
| *** ekarlso- has joined #openstack-zaqar | 09:10 | |
| *** sgotliv has quit IRC | 09:30 | |
| openstackgerrit | Doraly Navarro proposed openstack/python-zaqarclient: Gets 'flavor' data if the resource exists https://review.openstack.org/137635 | 09:49 |
| *** openstackgerrit has quit IRC | 09:50 | |
| *** openstackgerrit has joined #openstack-zaqar | 09:50 | |
| *** sgotliv has joined #openstack-zaqar | 10:07 | |
| openstackgerrit | Flavio Percoco proposed openstack/zaqar-specs: Move approved specs under an approved dir https://review.openstack.org/138030 | 10:24 |
| openstackgerrit | Merged openstack/zaqar-specs: Move approved specs under an approved dir https://review.openstack.org/138030 | 10:34 |
| flaper87 | flwang: ping | 10:39 |
| flaper87 | flwang: how are you doing? | 10:39 |
| flaper87 | Can we sync on the notification work? | 10:39 |
| flaper87 | How's it going? | 10:39 |
| *** yfujioka has quit IRC | 10:51 | |
| *** dynarro has quit IRC | 11:01 | |
| *** exploreshaifali has joined #openstack-zaqar | 11:07 | |
| *** vkmc has joined #openstack-zaqar | 11:23 | |
| vkmc | good morning! | 11:25 |
| *** sgotliv has quit IRC | 11:28 | |
| vkmc | flaper87, hey o/ | 11:38 |
| *** sgotliv has joined #openstack-zaqar | 11:40 | |
| flaper87 | vkmc: hey hey | 12:08 |
| flaper87 | vkmc: goooooooood morning | 12:08 |
| vkmc | :) | 12:09 |
| *** exploreshaifali has quit IRC | 12:12 | |
| vkmc | addings the missing functional tests for v1 means a lot of duplication | 12:18 |
| vkmc | should we add them, considering that v1 is freezed? | 12:18 |
| vkmc | https://bugs.launchpad.net/zaqar/+bug/1246757 | 12:18 |
| vkmc | flaper87, your thoughts are welcome ^ | 12:18 |
| vkmc | wrt coverage, is the right thing to do | 12:19 |
| flaper87 | what's missing ? | 12:19 |
| flaper87 | I thought we had everything | 12:19 |
| * flaper87 reads the fucking bug | 12:19 | |
| vkmc | *v1* | 12:20 |
| flaper87 | aahh, fuck v1 | 12:20 |
| vkmc | pools tests | 12:20 |
| flaper87 | :D | 12:20 |
| vkmc | I don't mind adding them | 12:20 |
| vkmc | but I don't want to add more code if its not needed | 12:20 |
| flaper87 | please, don't copy paste what I said in the bug | 12:20 |
| flaper87 | :P | 12:20 |
| vkmc | I won't :) | 12:20 |
| flaper87 | nah, no need, lets advise people to move to v1.1 | 12:20 |
| vkmc | all right | 12:20 |
| flaper87 | https://review.openstack.org/#/c/137635/3/zaqarclient/tests/queues/flavor.py,cm | 12:21 |
| flaper87 | vkmc: ^^^ you commented on the wrong line | 12:21 |
| flaper87 | >.> | 12:21 |
| vkmc | flaper87, I know, sorry | 12:21 |
| flaper87 | lets wait for dynarro to come back and update the patch | 12:21 |
| vkmc | sure | 12:21 |
| flaper87 | vkmc: I know you know, I just wanted to point it out | 12:22 |
| flaper87 | because it's monday | 12:22 |
| flaper87 | and monday is not suppose to be nice | 12:22 |
| flaper87 | :P | 12:22 |
| vkmc | let's make Mondays unfriendly | 12:22 |
| vkmc | by the way | 12:22 |
| vkmc | would you be working on persistent transports? | 12:23 |
| flaper87 | if there's anything I can help with, yeah :) | 12:23 |
| vkmc | I'm asking because you assigned yourself to the blueprint | 12:23 |
| flaper87 | no, wait, I changed that | 12:23 |
| flaper87 | I added myself as approver | 12:23 |
| * flaper87 checks again | 12:24 | |
| flaper87 | I hate launchpad | 12:24 |
| vkmc | not cool dud | 12:24 |
| flaper87 | really, I hate it | 12:24 |
| vkmc | hahahhaa | 12:24 |
| flaper87 | I swear I put my name in the approvers field | 12:24 |
| flaper87 | I think it just got it backwards | 12:24 |
| flaper87 | vkmc: done :D | 12:24 |
| vkmc | stop blaming the software :P | 12:24 |
| vkmc | ahahaha | 12:24 |
| vkmc | ok | 12:25 |
| flaper87 | vkmc: you stop blaming me >.> | 12:25 |
| flaper87 | :P | 12:25 |
| * flaper87 is a cool dud | 12:25 | |
| vkmc | its Monday | 12:25 |
| vkmc | you said that | 12:25 |
| vkmc | :P | 12:25 |
| flaper87 | right | 12:26 |
| *** bradjones has quit IRC | 12:26 | |
| vkmc | ok | 12:26 |
| *** bradjones has joined #openstack-zaqar | 12:26 | |
| vkmc | I drop some notes on websockets yesterday | 12:26 |
| vkmc | and now I was reading the code for the POC | 12:27 |
| flaper87 | vkmc: cool beans, I really gotta finish this nova patch to enter the review hell asap | 12:32 |
| vkmc | flaper87, hang on there | 12:33 |
| vkmc | :) | 12:33 |
| vkmc | flaper87, btw, do you know in which version of ubuntu they changed mongodb for mongod? | 13:12 |
| vkmc | and they are no longer supporting mongodb | 13:13 |
| flaper87 | erm, no idea | 13:13 |
| flaper87 | :/ | 13:13 |
| flaper87 | vkmc: what version are you using? are you having issues w/ devstack? | 13:14 |
| vkmc | flaper87, well yeah, devstack is broken right now with... ubuntu 12.04 and mongo 2.4.9 | 13:15 |
| vkmc | I'm about to submit a fix | 13:15 |
| flaper87 | I think we're not supporting 12.04 anymore | 13:16 |
| vkmc | also... we are defaulting to mongo and we removed pymongo from requirements | 13:17 |
| vkmc | sorry, 14.04 | 13:17 |
| flaper87 | ah ok | 13:17 |
| flaper87 | are we using mongodb as the default? | 13:17 |
| flaper87 | I don't recall | 13:17 |
| flaper87 | probably in devstack, yes | 13:17 |
| flaper87 | but not in the codebase | 13:17 |
| vkmc | yeah | 13:17 |
| vkmc | devstack | 13:17 |
| flaper87 | the devstack has to install pymongo itself | 13:17 |
| flaper87 | :D | 13:17 |
| vkmc | huh, it didn't do that yesterday | 13:18 |
| vkmc | I'll add it if its missing | 13:18 |
| flaper87 | it'd be weird | 13:18 |
| flaper87 | also, how is the gate passing if 14.04 is broken? | 13:18 |
| vkmc | good point | 13:19 |
| vkmc | it must be my deployment... my mongo version or something | 13:19 |
| vkmc | pymongo is there | 13:20 |
| vkmc | its definitely my deployment :) | 13:20 |
| flaper87 | vkmc: Happy Monday | 13:20 |
| *** dynarro has joined #openstack-zaqar | 13:20 | |
| vkmc | flaper87, same for you | 13:20 |
| flaper87 | :P | 13:21 |
| vkmc | you should choose mentees better :P | 13:21 |
| flaper87 | vkmc: no no, I like you, you make it easy to mess with you | 13:22 |
| openstackgerrit | Doraly Navarro proposed openstack/python-zaqarclient: Gets 'flavor' data if the resource exists https://review.openstack.org/137635 | 13:26 |
| *** dynarro has quit IRC | 13:27 | |
| flaper87 | vkmc: ^ | 13:27 |
| flaper87 | I don't mean to stress you with reviews but I'd like to make that gate voting | 13:28 |
| flaper87 | and propose it for zaqar too | 13:28 |
| vkmc | sure thing | 13:28 |
| vkmc | there | 13:30 |
| *** dynarro has joined #openstack-zaqar | 13:38 | |
| dynarro | vkmc: I guess you were right when you said that bug was boring hahaha | 13:40 |
| flaper87 | dynarro: LOOOOOOOOOOL, I guess you weren't when you volunteered to work on it | 13:41 |
| flaper87 | :P | 13:41 |
| flaper87 | vkmc: btw, you asked for more music recommendations: http://open.spotify.com/artist/1yDbenrXWhaQij3SCpsXe7 | 13:41 |
| flaper87 | "How to destroy Angles" | 13:41 |
| flaper87 | kragniz: ^ | 13:41 |
| flaper87 | dude you haven't even said hi today | 13:41 |
| flaper87 | you should be ashamed | 13:41 |
| kragniz | ;_: | 13:41 |
| kragniz | I'm sorrrrrry! | 13:41 |
| flaper87 | Anyway, that's Trent Reznor (NiN dude) and his wife | 13:42 |
| kragniz | I'll start again | 13:42 |
| kragniz | good morning, everyone! \o/ | 13:42 |
| flaper87 | Also, Alessandro Cortini | 13:42 |
| flaper87 | vkmc: Alessandro Cortini is the guy from SONOIO (his solo project) | 13:42 |
| flaper87 | he's was also part of NiN | 13:42 |
| flaper87 | kragniz: GOOOOOOOOOOOOOOOOOOOOOOD MORNING! | 13:43 |
| flaper87 | hope you're doing great today | 13:43 |
| flaper87 | despite it being Monday | 13:43 |
| kragniz | flaper87: I'm doing well! | 13:43 |
| flaper87 | kragniz: despite this? https://twitter.com/flaper87/status/539414505797988352 | 13:43 |
| kragniz | except for cycling though a cloud on the way to the office | 13:44 |
| flaper87 | kragniz: damn :( | 13:44 |
| kragniz | it felt like a cloud, at least | 13:44 |
| flaper87 | LOOOOL | 13:44 |
| flaper87 | Was it yellow ? | 13:44 |
| flaper87 | or yellowish | 13:44 |
| kragniz | I should keep a towel at my desk | 13:44 |
| dynarro | flaper87: I regret LOL | 13:44 |
| kragniz | no, very grey | 13:44 |
| flaper87 | kragniz: ok, then it was either a cloud or a pigeon | 13:45 |
| vkmc | dynarro, I warned you haha! | 13:45 |
| vkmc | flaper87, thanks :D | 13:45 |
| kragniz | flaper87: it was a very big pigeon if it was one of those | 13:45 |
| vkmc | kragniz, o/ | 13:45 |
| kragniz | vkmc: \o | 13:45 |
| flaper87 | kragniz: Don't you guys have a weird thing in a lake called Ness ? | 13:48 |
| kragniz | the loch ness monster? | 13:48 |
| flaper87 | I'm pretty sure you also have big 90% transparent pigeons | 13:48 |
| kragniz | flaper87: yeah, but normally you see the feet | 13:49 |
| flaper87 | ok, that's a question. If they fly, why do they need feet ? | 13:49 |
| flaper87 | They should just land when they need to rest | 13:49 |
| flaper87 | and then fly | 13:49 |
| kragniz | good point | 13:49 |
| kragniz | or have four wings | 13:49 |
| flaper87 | That way, we wouldn't ahve gazillions of pigeons in all the fucking squares in the world | 13:50 |
| kragniz | when one pair needs to rest, just use the others | 13:50 |
| flaper87 | right | 13:50 |
| openstackgerrit | Merged openstack/python-zaqarclient: Gets 'flavor' data if the resource exists https://review.openstack.org/137635 | 13:51 |
| flaper87 | https://twitter.com/flaper87/status/539416947361054721 | 13:54 |
| kragniz | haha | 13:54 |
| flaper87 | vkmc: if you're interested in a specific song, I'd recommend this one: http://open.spotify.com/track/0FdDewrvAF7LCd5Vf2FjgR | 13:55 |
| * vkmc clicks | 13:55 | |
| flaper87 | https://www.youtube.com/watch?v=Oq3pDuJeMqQ | 13:56 |
| flaper87 | that's the video | 13:56 |
| flaper87 | vkmc: kragniz ^ | 13:56 |
| kragniz | flaper87: ty! | 13:56 |
| flaper87 | it's really good (both video and song) | 13:56 |
| * vkmc listening A Drowing | 13:56 | |
| flaper87 | And you can see the three of them | 13:56 |
| vkmc | its really good... feels like NIN after some sleep pills (for now) | 13:57 |
| flaper87 | hahaha, well, it's basically NIN... The best 2 guys from NIN | 13:57 |
| vkmc | yeah | 13:58 |
| vkmc | Mariqueen has a great voice, she is up to the game! must be hard to be Raznor's wife haha | 14:00 |
| *** sriram has joined #openstack-zaqar | 14:07 | |
| *** jchai has joined #openstack-zaqar | 14:14 | |
| *** mpanetta has joined #openstack-zaqar | 14:23 | |
| vkmc | (╯°□°)╯︵ ┻━┻ | 14:31 |
| flaper87 | (╯°□°)╯︵ ┻━┻ | 14:36 |
| sriram | \_o_/ | 14:37 |
| * sriram tries to catch the thrown tables. | 14:38 | |
| *** ametts has joined #openstack-zaqar | 14:38 | |
| flaper87 | https://twitter.com/flaper87/status/539429080853385216 | 14:42 |
| * kragniz is envious of flaper87 | 14:43 | |
| kragniz | pls share | 14:43 |
| kragniz | pls | 14:43 |
| * flaper87 sends some gummy bears to kragniz vkmc and sriram | 14:44 | |
| vkmc | :D | 14:45 |
| vkmc | omg dud | 14:45 |
| vkmc | that's an excesive amount of gummy stuff | 14:46 |
| vkmc | not only bears | 14:46 |
| kragniz | it's the not-bears that look extra yummy | 14:46 |
| * sriram does a pole vault to catch gummy bears | 14:48 | |
| * sriram creates olympic record! | 14:48 | |
| * sriram crowd goes wild | 14:48 | |
| sriram | :P | 14:48 |
| flaper87 | gummy stuff is never enough | 14:51 |
| openstackgerrit | Flavio Percoco proposed openstack/zaqar: Add capabilities property to the DataDriver https://review.openstack.org/135637 | 14:52 |
| flaper87 | MEETING IN 10 | 14:52 |
| vkmc | meeting in 5? | 14:52 |
| vkmc | oh | 14:52 |
| kragniz | COFFEE IN 8 | 14:53 |
| vkmc | lol | 14:53 |
| flaper87 | 7 | 14:53 |
| kragniz | 6 | 14:54 |
| flaper87 | 5 | 14:55 |
| kragniz | 4 | 14:56 |
| flaper87 | fuck | 14:56 |
| flaper87 | Error(NaN) | 14:56 |
| flaper87 | 3 | 14:57 |
| flaper87 | dynarro: vkmc kragniz sriram kgriffs|afk start joining #openstack-meeting-3 | 14:57 |
| flaper87 | 2 | 14:58 |
| kragniz | > implying I'm not already there | 14:58 |
| *** cpallares has joined #openstack-zaqar | 15:00 | |
| cpallares | o/ | 15:05 |
| vkmc | cpallares, \o | 15:07 |
| vkmc | meeting in #openstack-meeting-3 | 15:07 |
| *** exploreshaifali has joined #openstack-zaqar | 15:14 | |
| *** amitgandhinz has joined #openstack-zaqar | 15:34 | |
| *** kgriffs|afk is now known as kgriffs | 15:36 | |
| *** exploreshaifali has quit IRC | 15:39 | |
| vkmc | flaper87, we have in 30' the meeting of websockets | 15:58 |
| flaper87 | vkmc: yup | 15:59 |
| vkmc | I forgot lunchtime so I hope its short | 15:59 |
| flaper87 | kgriffs: there's a hangout with Keith in 30' | 15:59 |
| flaper87 | want to join? | 15:59 |
| kgriffs | not sure... what's the agenda? | 16:00 |
| flaper87 | kgriffs: status updates | 16:01 |
| kgriffs | sure, I'll listen in | 16:02 |
| vkmc | I'm a bit bothered that I don't have a clear architecture idea yet | 16:03 |
| kgriffs | vkmc: would you like to discuss? | 16:10 |
| vkmc | kgriffs, if you have some spare time, of course I'd | 16:10 |
| kgriffs | so where did things end up as far as basic decisions? | 16:11 |
| kgriffs | we are settled on websockets | 16:11 |
| kgriffs | but other than that? | 16:12 |
| * kgriffs wishes he had more time to keep up with the specs | 16:12 | |
| vkmc | we are going to use ws4py to implement this | 16:12 |
| vkmc | and msgpack as a serialization protocol | 16:12 |
| vkmc | we also have two pocs to consider | 16:12 |
| vkmc | my main concern is... considering current Zaqar arch | 16:14 |
| vkmc | we will have to design this transport in a different way that the base | 16:15 |
| vkmc | which is highly coupled to wsgi | 16:16 |
| * kgriffs looks | 16:17 | |
| *** dynarro has quit IRC | 16:18 | |
| kgriffs | by "base" do you mean this? https://github.com/openstack/zaqar/blob/master/zaqar/queues/transport/base.py | 16:19 |
| vkmc | yeap | 16:20 |
| vkmc | well in fact, not that | 16:20 |
| vkmc | the mapping between the transport operations and the storage | 16:20 |
| * flaper87 back | 16:21 | |
| kgriffs | I feel the urge to draw | 16:21 |
| * kgriffs realizes he left his ipad at home | 16:21 | |
| kgriffs | vkmc: so, the storage layer exposes an interface | 16:22 |
| vkmc | oh a draw would be cool | 16:22 |
| vkmc | yeah | 16:22 |
| kgriffs | are you saying the storage driver interface does not map well to a non-HTTP transport? | 16:22 |
| * kgriffs wishes for a shared vizio for the Nth time | 16:23 | |
| vkmc | I'm not sure yet | 16:24 |
| flaper87 | http://whiteboardfox.com | 16:24 |
| kgriffs | oooh | 16:24 |
| kgriffs | http://whiteboardfox.com/26405-2092-7552 | 16:25 |
| flaper87 | ;) | 16:25 |
| kgriffs | man, I really need my ipad for this | 16:25 |
| flaper87 | it works well on the iad | 16:25 |
| flaper87 | iPad | 16:25 |
| * flaper87 has tested it | 16:25 | |
| kgriffs | OK, I give up. Can't do this with a trackpad | 16:27 |
| vkmc | jajaja | 16:27 |
| vkmc | hahaha* | 16:27 |
| flaper87 | vkmc: can you join the Hangout and add me? | 16:27 |
| flaper87 | The invited my redhat account | 16:28 |
| vkmc | sure | 16:28 |
| vkmc | they invited my personal domain account so I'll try | 16:28 |
| flaper87 | oh, mmh | 16:28 |
| flaper87 | great | 16:28 |
| vkmc | let me invite you | 16:31 |
| vkmc | https://plus.google.com/hangouts/_/symantec40.com/zaqar-websocket | 16:31 |
| vkmc | click there | 16:32 |
| *** exploreshaifali has joined #openstack-zaqar | 16:33 | |
| kgriffs | mind if I listen in? | 16:34 |
| vkmc | join us :) | 16:34 |
| *** achanda has joined #openstack-zaqar | 16:43 | |
| exploreshaifali | hey vkmc \o | 16:53 |
| *** jimhoagland has joined #openstack-zaqar | 17:09 | |
| *** amalagon has joined #openstack-zaqar | 17:14 | |
| *** amalagon has quit IRC | 17:15 | |
| *** amalagon has joined #openstack-zaqar | 17:16 | |
| kgriffs | o/ | 17:20 |
| kgriffs | vkmc: so, msgpack vs. protobuf. | 17:21 |
| vkmc | hey exploreshaifali o/ | 17:21 |
| vkmc | jimhoagland, o/ | 17:21 |
| vkmc | kgriffs, yeah | 17:21 |
| kgriffs | did you say that you did look at performance already between the two? | 17:21 |
| vkmc | I did | 17:21 |
| vkmc | the situation with that is quite confusing though | 17:22 |
| vkmc | msgpack claims to be 4 times more faster than protobuf | 17:22 |
| vkmc | but some benchmarking shows that its not true | 17:22 |
| vkmc | let me get some links | 17:22 |
| kgriffs | oic | 17:22 |
| vkmc | http://theburningmonk.com/2011/12/performance-test-binary-serializers-part-ii/ | 17:23 |
| vkmc | its quite old though, it might have changed | 17:23 |
| kgriffs | i didn't see any info on the pad/wiki about it so wanted to ask | 17:23 |
| vkmc | yeah, I didn't update it | 17:23 |
| kgriffs | I think it is sufficient to know that "speed-wise there’s not much separating the two" | 17:24 |
| vkmc | I digged more into it when I saw the POC was made using protobuf | 17:24 |
| vkmc | yeah | 17:24 |
| vkmc | that's why I asked if there were another reason | 17:24 |
| vkmc | to use protobuf | 17:24 |
| kgriffs | they liked the compact format | 17:24 |
| ekarlso- | vkmc: what was the websocket stuff ? | 17:24 |
| kgriffs | and they were familiar with it already | 17:24 |
| kgriffs | i imagine msgpack can't be that different in terms of compactness | 17:25 |
| vkmc | kgriffs, yeah :) | 17:25 |
| vkmc | nope, its the same | 17:25 |
| kgriffs | did you look at that as well? | 17:25 |
| vkmc | and its more natural maybe... because its more similar to JSON | 17:25 |
| *** exploreshaifali has quit IRC | 17:25 | |
| kgriffs | which is more similar to JSON? | 17:26 |
| vkmc | msgpack | 17:26 |
| kgriffs | got it | 17:26 |
| kgriffs | ok, so it sounds like we should be OK with msgpack then | 17:26 |
| vkmc | yeah | 17:26 |
| vkmc | we also had a first approach on adopting msgpack | 17:26 |
| kgriffs | next up... STOMP | 17:26 |
| kgriffs | flaper87: ping | 17:26 |
| vkmc | bbl | 17:26 |
| flaper87 | pong | 17:27 |
| flaper87 | kgriffs: ^ | 17:27 |
| vkmc | ekarlso-, we will implement a websocket driver | 17:27 |
| flaper87 | :P | 17:27 |
| kgriffs | so, I saw your comment | 17:27 |
| kgriffs | on the spec re STOMP | 17:27 |
| flaper87 | kgriffs: what comment ? | 17:27 |
| flaper87 | ah ok | 17:27 |
| flaper87 | :P | 17:27 |
| kgriffs | so, it doesn't go over HTTP | 17:27 |
| kgriffs | is is just designed to be HTTP-like | 17:27 |
| kgriffs | so you would run it over TCP if I understand correctly | 17:27 |
| kgriffs | right? | 17:27 |
| kgriffs | http://stomp.github.io/stomp-specification-1.2.html#Abstract | 17:28 |
| flaper87 | oh, mmh | 17:28 |
| kgriffs | the reason I suggested taking another look was STOMP defines some of the stuff we need and we don't have to figure out a new protocol | 17:29 |
| kgriffs | like, it has a notion of "headers" | 17:29 |
| kgriffs | heart beat | 17:29 |
| flaper87 | right | 17:29 |
| kgriffs | subscribing and getting push messages | 17:29 |
| flaper87 | that's probably a good fit too | 17:29 |
| kgriffs | content type | 17:29 |
| flaper87 | not sure where I got that http thing from | 17:29 |
| kgriffs | by comparison, MQTT is more opinionated | 17:30 |
| kgriffs | but it seemed like STOMP was fairly generic and we might be able to use it | 17:30 |
| flaper87 | kgriffs: I'll take a look at STOMP again, I'm sorry for the misleading comment | 17:31 |
| flaper87 | IIRC, there are libraries supporting STOMP for the browser | 17:31 |
| kgriffs | no problem. I just think we should take a fair look. If we can use it, great, because it saves us time - we don't have to figure out how to do error handling, push messages, etc. | 17:32 |
| flaper87 | kgriffs: agreed... plus it's already used elsewhere | 17:33 |
| *** sgotliv has quit IRC | 17:34 | |
| kgriffs | kk, I gotta run for a bit | 17:35 |
| jimhoagland | kgriffs, vkmc, as we mentioned on the call, we are fine with whatever encoding makes the most sense. If you are looking at encode/decode performance in Python, you should be sure the benchmark used a 2.6.x version of the protobuf library as is reportedly much faster. | 17:40 |
| *** reed has joined #openstack-zaqar | 17:51 | |
| flaper87 | jimhoagland: I think we're after performance but also adoption | 17:51 |
| flaper87 | The broader the support is for a specific encoding, the better | 17:52 |
| vkmc | jimhoagland, ok, thanks, we will keep you updated with any changes | 18:00 |
| flaper87 | vkmc: did you push your food down your throat ? | 18:02 |
| flaper87 | (says the guy that can have lunch in 5 mins ) | 18:03 |
| flaper87 | s/have/finish/ | 18:03 |
| vkmc | flaper87, I'm trained to eat very fast over the keyboard | 18:03 |
| vkmc | I put it on my cv | 18:03 |
| vkmc | now I'm craving for candy | 18:05 |
| *** jimhoagland has quit IRC | 18:05 | |
| * flaper87 gives vkmc some gummy berries | 18:05 | |
| kragniz | vkmc: there is a vending machine rather close to me | 18:05 |
| vkmc | omg OMG! nom nom nom | 18:05 |
| kragniz | I don't think it has a good effect on my health | 18:05 |
| vkmc | oh I would be eating so much sugar if I were you >.> | 18:06 |
| vkmc | I think I'm going to do as flaper87... keep a stash of candy | 18:07 |
| kragniz | I have that, too | 18:07 |
| flaper87 | vkmc: good girl, goooooood girl | 18:07 |
| kragniz | or had, until I ate it all | 18:08 |
| *** achanda has quit IRC | 18:18 | |
| flaper87 | Guys, btw, remember to follow the hacksummit.org | 18:21 |
| flaper87 | Obviously, if you want | 18:21 |
| flaper87 | :D | 18:21 |
| kragniz | flaper87: is it any good so far? | 18:21 |
| flaper87 | I missed the first one but it was an intro to the summit | 18:22 |
| flaper87 | the ongoing one is about Azure | 18:22 |
| flaper87 | I'm not interested in that one | 18:23 |
| flaper87 | There's no schedule so, figuring it out as we go | 18:23 |
| vkmc | there is an agenda | 18:25 |
| vkmc | #fail | 18:28 |
| flaper87 | what? | 18:29 |
| flaper87 | I mean, there's a speakers agenda | 18:29 |
| vkmc | they mailed it | 18:29 |
| flaper87 | not topics | 18:29 |
| flaper87 | With titles and stop? | 18:29 |
| flaper87 | s/stop/stuff/ | 18:29 |
| flaper87 | OMG | 18:29 |
| flaper87 | I need help | 18:29 |
| flaper87 | ahhh | 18:29 |
| flaper87 | they sent it today | 18:29 |
| vkmc | TODAY 11AM-12PM PST - Culture and Happiness in Virtual Teams - FLOYD MARINESCU | 18:29 |
| vkmc | seems like a title for a presentation | 18:30 |
| vkmc | coming up next | 18:30 |
| vkmc | haha | 18:30 |
| flaper87 | yeah | 18:30 |
| flaper87 | brb | 18:30 |
| vkmc | I wanna hear Alex | 18:30 |
| *** jimhoagland has joined #openstack-zaqar | 18:34 | |
| *** davideagnello has joined #openstack-zaqar | 18:37 | |
| *** sgotliv has joined #openstack-zaqar | 18:40 | |
| *** sgotliv has quit IRC | 18:45 | |
| vkmc | poor Microsoft dud... 'I have to pay for my Azure account, they wouldn't give a free account for me' | 18:47 |
| kragniz | ;_; | 18:48 |
| vkmc | :'| | 18:48 |
| * flaper87 back | 18:53 | |
| vkmc | bet on Javascript | 18:54 |
| *** davideagnello has quit IRC | 19:00 | |
| *** davideagnello has joined #openstack-zaqar | 19:01 | |
| vkmc | Culture and Happiness in Virtual Teams next | 19:06 |
| vkmc | we should totally give some suggestions there | 19:06 |
| vkmc | 'it's everything about poptarts' | 19:07 |
| *** davideagnello has quit IRC | 19:07 | |
| *** davideagnello has joined #openstack-zaqar | 19:08 | |
| cpallares | kgriffs, mpanetta: You should check out the sparkfun cyber monday deals :D https://www.sparkfun.com/CyberMonday2014 | 19:14 |
| kragniz | cpallares: weee! | 19:14 |
| * kgriffs clicks | 19:14 | |
| mpanetta | cpallares: Did so last night :) | 19:14 |
| cpallares | mpanetta: Bought anything? | 19:15 |
| mpanetta | cpallares: Nah, didn't see anything I coud not live without | 19:15 |
| mpanetta | I think I have too many projects not finished to start something new ;) | 19:15 |
| cpallares | mpanetta: Me too, but I'm debating on buying my littler brother a makey-makey. | 19:16 |
| mpanetta | Oh cool | 19:16 |
| cpallares | mpanetta: I mean, I've been wanting to play with one since forever, but $50 seems quite unreasonable :P | 19:17 |
| mpanetta | What did they discount it to? | 19:17 |
| cpallares | mpanetta: It's 40% off. | 19:17 |
| mpanetta | Oh damn! | 19:17 |
| mpanetta | wow | 19:18 |
| cpallares | lol | 19:18 |
| cpallares | mpanetta: https://www.sparkfun.com/products/11511 | 19:18 |
| cpallares | Just for this hour. | 19:18 |
| flaper87 | vkmc: I'm not likuing that talk much | 19:20 |
| vkmc | yeah me neither | 19:20 |
| vkmc | we are a level 5 community, that is what I could learn | 19:20 |
| flaper87 | if this dude invites us to infoq again, I'll consume the bandwidth by yelling at him through it | 19:25 |
| *** sgotliv has joined #openstack-zaqar | 19:25 | |
| * cpallares wonders what flaper87 and vkmc are talking about | 19:27 | |
| flaper87 | cpallares: hacksummit.org | 19:28 |
| flaper87 | it's live now | 19:28 |
| vkmc | ahaha | 19:28 |
| vkmc | yeah | 19:28 |
| flaper87 | InfoQ CEO is talking about virtual teams | 19:28 |
| vkmc | I stopped listening | 19:28 |
| flaper87 | and he's invited the audience to go and read InfoQ a gazillion times | 19:28 |
| kragniz | flaper87: that doesn't sound great | 19:33 |
| *** davideagnello has quit IRC | 19:33 | |
| *** davideagnello has joined #openstack-zaqar | 19:35 | |
| *** davideagnello has quit IRC | 19:35 | |
| *** davideagnello has joined #openstack-zaqar | 19:35 | |
| flwang | flaper87: ping | 19:51 |
| *** jchai is now known as jchai_afk | 19:53 | |
| *** exploreshaifali has joined #openstack-zaqar | 19:58 | |
| *** jimhoagland has quit IRC | 20:03 | |
| exploreshaifali | vkmc, are you able to access https://hacksummit.org/ ? | 20:05 |
| kragniz | exploreshaifali: they're getting DDOSed | 20:06 |
| cpallares | exploreshaifali: If you refresh long enough you'll get through | 20:07 |
| exploreshaifali | kragniz, oh! ok | 20:07 |
| exploreshaifali | cpallares, I did that many times but same response :( | 20:08 |
| cpallares | exploreshaifali: Ah, well it was happening to me as well, maybe the talk that's on right now is more popular? | 20:08 |
| kragniz | exploreshaifali: I think you can direcly connect | 20:09 |
| kragniz | or at least I've heard that | 20:09 |
| exploreshaifali | cpallares, ya that could be a case :) | 20:09 |
| exploreshaifali | kragniz, directly connect ? | 20:09 |
| kragniz | to the hangout | 20:10 |
| exploreshaifali | kragniz, okay, let me try :) | 20:11 |
| *** jchai_afk is now known as jchai | 20:11 | |
| *** davideagnello has quit IRC | 20:11 | |
| cpallares | exploreshaifali: https://www.youtube.com/watch?v=b1i865DVYGU#t=428 | 20:14 |
| exploreshaifali | cpallares, ooohh Thaks a looot :) | 20:14 |
| *** wpf has quit IRC | 20:14 | |
| exploreshaifali | *thanks | 20:14 |
| *** wpf has joined #openstack-zaqar | 20:16 | |
| *** davideagnello has joined #openstack-zaqar | 20:21 | |
| vkmc | http://www.crowdcast.io/e/hack_summit3 mirror | 20:29 |
| flaper87 | omg, this dude is going to make me cry | 20:37 |
| flaper87 | .... of pain | 20:37 |
| flaper87 | flwang: pong | 20:38 |
| flwang | flaper87: i have a standup meeting in 5 minutes, will you be online after 20 min? | 20:38 |
| flwang | flaper87: I'd like to sync the notification | 20:39 |
| flaper87 | that's entirely, possibly, maybe | 20:39 |
| flwang | ok, a quick question is kgriffs is not a fan of splitting the notification service and queues | 20:40 |
| flaper87 | flwang: you mean, he prefers to keep everything under queues ? | 20:40 |
| flwang | flaper87: that means he seems prefer to keep the notification related stuff in current queues folder | 20:40 |
| flwang | yes | 20:40 |
| flwang | kgriffs: are u around? | 20:41 |
| flwang | and I'm almost convinced :) | 20:41 |
| flaper87 | mmh but then what's the queue's package for? | 20:42 |
| flaper87 | should we get rid of it? | 20:42 |
| *** jimhoagland has joined #openstack-zaqar | 20:42 | |
| flaper87 | we created it because we wanted to split notifications and queues | 20:42 |
| flaper87 | but we also said we wanted them to be completely separate services | 20:42 |
| flwang | flaper87: it's a good question, for both of the cases, we may need some refactoring work for the structure | 20:42 |
| flwang | yep, kgriffs thinks the notification case is most like one of the supported cases of queue | 20:43 |
| *** flwang1 has joined #openstack-zaqar | 20:43 | |
| kgriffs | here | 20:45 |
| flwang | kgriffs: cool | 20:45 |
| flwang | kgriffs: would you mind talking about the notification service structure? | 20:45 |
| kgriffs | sorry, I recently got assigned like three more projects at Rackspace so I'm spread pretty thin lately. :p | 20:45 |
| flwang | brb | 20:45 |
| kgriffs | flaper87: so, question about taskflow | 20:46 |
| flaper87 | kgriffs: shoot | 20:46 |
| kgriffs | so, with taskflow what are my options for farming out work? | 20:48 |
| kgriffs | iirc, you can do it in-process (gevent/eventlet/tulip) | 20:48 |
| sriram | you can look at conductors. they basically pull stuff off a jobboard and run it in an engine. | 20:49 |
| kgriffs | sriram: is that part of taskflow or is this a different project? | 20:49 |
| sriram | its part of taskflow | 20:49 |
| sriram | Thats the abc: https://github.com/openstack/taskflow/blob/master/taskflow/conductors/base.py | 20:50 |
| sriram | It would be like a producer-consumer scenario. | 20:51 |
| sriram | where your conductor acts as the consumer. | 20:51 |
| kgriffs | ok, so I am assuming I can run some engines in a separate process from my web server? | 20:52 |
| sriram | yes. | 20:52 |
| sriram | I was able to run 8, through multiprocessing. | 20:52 |
| kgriffs | flaper87: so your proposal was to implement the message delivery agents as taskflow engines? | 20:52 |
| sriram | Ohh, yeah that could work. | 20:53 |
| *** jimhoagland has left #openstack-zaqar | 20:53 | |
| flaper87 | kgriffs: yeah | 20:53 |
| kgriffs | not sure if I have the terminology right | 20:53 |
| flaper87 | you can run engines in a separate process if you want but that may require an external thing to send the message | 20:53 |
| sriram | since the conductors are always running, anytime something new is posted on the jobboard, they act on it immediately. -> notification. | 20:53 |
| *** jimhoagland has joined #openstack-zaqar | 20:53 | |
| sriram | sounds really neat. | 20:54 |
| kgriffs | ok | 20:54 |
| flaper87 | You can run tasks concurrently in the same process and distributedly in more processes | 20:54 |
| kgriffs | the business logic for the task - how is that defined? Stevedore plugin for engines or what? | 20:54 |
| kgriffs | (by logic, I mean the actual code that would handle pushing to webhook, SMS, APN, etc.) | 20:55 |
| flaper87 | kgriffs: you need to implement a Task class | 20:55 |
| flaper87 | IIRC, they're loadable through stevedore | 20:55 |
| kgriffs | ok, so what Zaqar would need to deliver is some task classes | 20:56 |
| sriram | its usually inside an execute function, inside a class that inherits Task | 20:56 |
| flaper87 | kgriffs: right | 20:56 |
| sriram | if i'm not mistaken. | 20:56 |
| kgriffs | well, 1-2 "out of the box" and then operators can add more | 20:56 |
| flaper87 | sriram: correct | 20:56 |
| flaper87 | kgriffs: right | 20:56 |
| kgriffs | should those be their own package (installable standalone) | 20:56 |
| kgriffs | or would you just | 20:56 |
| kgriffs | pip install zaqar | 20:56 |
| kgriffs | then that installs the entry points | 20:57 |
| flaper87 | I think you should get the built-in ones with zaqar | 20:57 |
| * kgriffs is thinking out loud | 20:57 | |
| flaper87 | which will install the entry_points | 20:57 |
| flaper87 | as we do for the storage/transport drivers | 20:57 |
| kgriffs | ok, so I may have a taskflow engine farm that just runs these tasks. For those I would still pip install zaqar even though I don't intend to use the API server | 20:58 |
| kgriffs | flaper87: I think we should have separate packages, but use a zaqar.* namespace to unify them | 20:59 |
| * kgriffs ducks | 20:59 | |
| *** amitgandhinz has quit IRC | 21:00 | |
| kgriffs | flwang: ok, so it seems that we can group the notifications implementation in two | 21:00 |
| kgriffs | first group is the API - adding the ability for a user to CRUD subscriptions to a queue or topic | 21:00 |
| flaper87 | wait, you still need a process that is running the whole thing | 21:00 |
| flaper87 | which is part of zaqar | 21:01 |
| flaper87 | it could be zaqar-worker | 21:01 |
| flaper87 | I dunno | 21:01 |
| *** amitgandhinz has joined #openstack-zaqar | 21:01 | |
| kgriffs | oh | 21:01 |
| kgriffs | I see | 21:01 |
| flaper87 | but I think we need to dig more into that | 21:01 |
| kgriffs | you need to bootstrap taskflow for that worker farm | 21:01 |
| kgriffs | because taskflow is purely a library? | 21:01 |
| flaper87 | because I'd hate to tell people, "Hey, this is a messaging service but if you want to scale notifications you need to install rabbitmq to have distributed workers" | 21:01 |
| kgriffs | it doesn't provide a taskflow-agent | 21:02 |
| flaper87 | kgriffs: correct, it's just a library | 21:02 |
| kgriffs | ok | 21:02 |
| kgriffs | so if you don't want to run these tasks on your web heads directly | 21:02 |
| kgriffs | you need some zaqar-notifications-agent process or something | 21:02 |
| * flwang back | 21:02 | |
| flwang | sorry, what's the notification-agent for? | 21:03 |
| flaper87 | kgriffs: right | 21:03 |
| flwang | am i missing something? | 21:03 |
| kgriffs | flwang: the proposal is to use taskflow to distribute tasks | 21:03 |
| kgriffs | so | 21:03 |
| flwang | kgriffs: yep, I know | 21:04 |
| kgriffs | ok | 21:04 |
| kgriffs | so taskflow is a library | 21:04 |
| flwang | kgriffs: you mean we many need some remote processor? | 21:04 |
| flwang | installed in remote node? | 21:04 |
| kgriffs | yeah | 21:04 |
| flwang | kgriffs: ah, ok, yep, it's possible | 21:05 |
| flaper87 | kgriffs: do you think running it in the webheads won't scale enough? | 21:05 |
| kgriffs | because a large deployment won't want to push stuff to, say, SMTP or webhook from the webheads, but will want to farm that out to a worker pool | 21:05 |
| kgriffs | flaper87: exactly | 21:05 |
| kgriffs | well, kind of | 21:05 |
| flaper87 | kgriffs: mmh | 21:05 |
| kgriffs | I know for a fact some operators like to keep their processes fairly pure | 21:06 |
| flwang | kgriffs: flaper87: could you target a distributed agent in stage 2? | 21:06 |
| kgriffs | it is easier from a capacity management and a diagnostics standpoint | 21:06 |
| flaper87 | flwang: sure | 21:06 |
| kgriffs | so they don't want webheads also doing notifciations | 21:06 |
| flwang | given it's a built in support by taskflow | 21:06 |
| *** dynarro has joined #openstack-zaqar | 21:06 | |
| flaper87 | kgriffs: mmh | 21:06 |
| flaper87 | I'm worried about what an Zaqar deployment will look like | 21:07 |
| kgriffs | I don't think it is something we have to force | 21:07 |
| kgriffs | we can definitely have the "easy button" installation where the webheads also push the notifications | 21:07 |
| kgriffs | but as a second iteration, we will need to have some kind of story around distributed agents | 21:08 |
| kgriffs | (as an alternative option) | 21:08 |
| kgriffs | flwang: so yeah, we could develop this iteratively as you suggested | 21:09 |
| *** dynarro has quit IRC | 21:10 | |
| flwang | kgriffs: it would be good | 21:10 |
| kgriffs | really all that is needed for the second iteration is creating that agent to bootstrap taskflow, tell it where to start getting jobs from | 21:11 |
| kgriffs | and operators could run farms of these agents | 21:11 |
| flaper87 | kgriffs: ok, but I think we should get straight what we'll recommend as a communication channel for that distribution | 21:11 |
| flaper87 | I understand that but what are we going to recommend for that | 21:12 |
| flaper87 | ? | 21:12 |
| kgriffs | first, quick question | 21:13 |
| kgriffs | does taskflow choose an engine based on job type or something? | 21:13 |
| kgriffs | basically, I want to have a group of notification agents dedicated to email vs. webhooks, for example | 21:14 |
| kgriffs | flaper87: what are the options for channels? | 21:14 |
| flaper87 | http://docs.openstack.org/developer/taskflow/workers.html | 21:15 |
| * kgriffs looks | 21:15 | |
| flaper87 | oslo.messaging, basically | 21:15 |
| kgriffs | fine, I'll RTFM | 21:15 |
| kgriffs | :p | 21:15 |
| flaper87 | haha, no no | 21:16 |
| flaper87 | I didn't mean that | 21:16 |
| flaper87 | :P | 21:16 |
| kgriffs | :) | 21:16 |
| flaper87 | kgriffs: ^ ^ ^ ^ | 21:16 |
| flaper87 | :P | 21:16 |
| kgriffs | OK, so what do we need to decide? | 21:19 |
| kgriffs | wouldn't we just embed the worker config in the zaqar-notification-agent INI ? | 21:21 |
| kgriffs | http://docs.openstack.org/developer/taskflow/workers.html#id1 | 21:21 |
| kgriffs | then the operator can set it up however they like | 21:21 |
| kgriffs | and we can refer them to the taskflow/oslo.messaging docs | 21:22 |
| *** wpf1 has joined #openstack-zaqar | 21:28 | |
| *** fesp has joined #openstack-zaqar | 21:28 | |
| flwang | kgriffs: it works for me | 21:28 |
| *** amitgandhinz has quit IRC | 21:33 | |
| *** wpf has quit IRC | 21:35 | |
| *** flaper87 has quit IRC | 21:35 | |
| *** fesp is now known as flaper87 | 21:35 | |
| *** amitgandhinz has joined #openstack-zaqar | 21:35 | |
| kgriffs | kk. but that would be a "round two" project (zaqar-notification-agent) | 21:35 |
| kgriffs | first thing is to just assume everything is handled in the web server process | 21:36 |
| kgriffs | flwang: is there an etherpad for notes on this topic, btw? | 21:36 |
| flwang | kgriffs: nope, but i'm happy to create one to track our discussion | 21:37 |
| kgriffs | ok, could you make one real quick? | 21:37 |
| kgriffs | I'd like to discuss directory structure | 21:37 |
| kgriffs | would be easier on a pad | 21:37 |
| kgriffs | :p | 21:37 |
| flwang | https://etherpad.openstack.org/p/zaqar-notification | 21:38 |
| flwang | not sure if we should leverage an existed notification etherpad | 21:38 |
| flwang | I believe we have 1 or 2 etherpads about notification | 21:38 |
| flwang | flaper87: if you still around, please join https://etherpad.openstack.org/p/zaqar-notification | 21:49 |
| kgriffs | flwang: ok, so I think we have a good proposal there. Just need to see what flaper87 thinks | 22:00 |
| flwang | kgriffs: right | 22:00 |
| flwang | kgriffs: I will push the guy to get his opinion | 22:00 |
| kgriffs | :) | 22:00 |
| flwang | kgriffs: thank you for the awesome proposal | 22:00 |
| kgriffs | no problem, thanks for helping figure it out | 22:01 |
| flwang | kgriffs: it makes more sense for me | 22:01 |
| *** sriram has quit IRC | 22:09 | |
| *** amitgandhinz has quit IRC | 22:14 | |
| *** amitgandhinz has joined #openstack-zaqar | 22:15 | |
| *** exploreshaifali has quit IRC | 22:26 | |
| *** jchai has quit IRC | 22:37 | |
| *** amitgandhinz has quit IRC | 23:07 | |
| *** jimhoagland has quit IRC | 23:27 | |
| *** mpanetta has quit IRC | 23:30 | |
| *** echevemaster has joined #openstack-zaqar | 23:30 | |
| *** jimhoagland has joined #openstack-zaqar | 23:41 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!