| *** achanda has joined #openstack-zaqar | 00:00 | |
| *** achanda has quit IRC | 00:04 | |
| *** Qiming has quit IRC | 00:07 | |
| *** achanda has joined #openstack-zaqar | 00:12 | |
| *** diga has joined #openstack-zaqar | 00:13 | |
| openstackgerrit | Fei Long Wang proposed openstack/zaqar: Fix duplicated notification https://review.openstack.org/218072 | 00:39 |
|---|---|---|
| *** malini has joined #openstack-zaqar | 01:01 | |
| *** Qiming has joined #openstack-zaqar | 01:06 | |
| *** achanda has quit IRC | 01:09 | |
| *** diga has quit IRC | 01:20 | |
| *** malini has quit IRC | 02:03 | |
| *** achanda has joined #openstack-zaqar | 02:04 | |
| *** achanda has quit IRC | 02:14 | |
| *** achanda has joined #openstack-zaqar | 02:23 | |
| *** achanda has quit IRC | 02:49 | |
| *** achanda has joined #openstack-zaqar | 03:02 | |
| *** achanda has quit IRC | 03:08 | |
| *** stanchan has joined #openstack-zaqar | 03:41 | |
| *** wanghao has joined #openstack-zaqar | 03:47 | |
| *** csoukup has joined #openstack-zaqar | 04:04 | |
| *** csoukup has quit IRC | 04:05 | |
| *** akanksha has quit IRC | 04:18 | |
| *** achanda has joined #openstack-zaqar | 04:42 | |
| *** flwang has quit IRC | 05:08 | |
| *** achanda has quit IRC | 06:47 | |
| *** exploreshaifali has joined #openstack-zaqar | 06:50 | |
| *** dynarro has joined #openstack-zaqar | 07:21 | |
| *** achanda has joined #openstack-zaqar | 07:24 | |
| therve | Hola! | 07:32 |
| therve | flwang1, Heya | 07:42 |
| *** achanda has quit IRC | 07:51 | |
| dynarro | flaper87, I don't understand something about the comment you made in this patch https://review.openstack.org/#/c/209238/5/zaqarclient/queues/v2/api.py | 08:04 |
| dynarro | flaper87, is not that I deleted the method `get_message` I just made it raise an exception for v2 | 08:05 |
| openstackgerrit | Thomas Herve proposed openstack/zaqar: Notifications over websocket https://review.openstack.org/206080 | 08:12 |
| * flaper87 here now | 08:16 | |
| flaper87 | dynarro: mmh, but probably the schema shouldn't have the `get_messages` key since it's not going to be used anyway | 08:17 |
| dynarro | flaper87, so you think it should be actually deleted? | 08:19 |
| openstackgerrit | Thomas Herve proposed openstack/zaqar: Support various paths for signed URLs https://review.openstack.org/215638 | 08:21 |
| flaper87 | dynarro: from the *schema*, yes. | 08:23 |
| dynarro | flaper87, got it thank you :) | 08:24 |
| flaper87 | I'd expect to see a `V2.schema.pop('...')` right after the class definition | 08:24 |
| *** diga has joined #openstack-zaqar | 08:24 | |
| dynarro | flaper87, yeah makes sense | 08:25 |
| therve | flaper87, I have some questions on what's the eventlet integration in zaqar | 08:37 |
| flaper87 | therve: shoot | 08:45 |
| therve | flaper87, Well, do we have any? | 08:45 |
| *** diga has quit IRC | 08:45 | |
| therve | I can't find when the patching happens | 08:45 |
| flaper87 | therve: we don't and we've fought hard to not have any | 08:45 |
| flaper87 | I'd personally prefer to keep it that way | 08:46 |
| flaper87 | That's one of the reasons we also adopted taskflow | 08:46 |
| flaper87 | to have other options besides eventlet | 08:46 |
| therve | So that's the reason why the greenlet executor doesn't work | 08:46 |
| flaper87 | therve: yeah, I figured | 08:46 |
| flaper87 | I guess the notifications code should patch stuff | 08:46 |
| therve | flaper87, Though I don't know why it worked before | 08:46 |
| therve | Or just use the thread one? | 08:47 |
| flaper87 | therve: do you think we can go back to taskflow? :D | 08:47 |
| flaper87 | therve: it used the thread one, I believe | 08:47 |
| therve | flaper87, No it defaults to the greenlet one | 08:47 |
| flaper87 | therve: taskflow? | 08:47 |
| flaper87 | in that case, taskflow does all the patching itself | 08:47 |
| therve | Ah, ok | 08:47 |
| *** exploreshaifali has quit IRC | 08:48 | |
| therve | Well we can run taskflow in a thread | 08:48 |
| therve | But I'd rather just use the threadpool executor everywhere, unless we see something useful in greenlet executor | 08:48 |
| flaper87 | therve: that sounds good to me | 08:49 |
| flaper87 | also | 08:50 |
| flaper87 | I'm wondering if we should probably just patch things in the notifier for now and do this in Mitaka. I'd like to see a spec that covers the addition of workers | 08:50 |
| therve | Cool will do that then | 08:50 |
| therve | Arg | 08:50 |
| flaper87 | I'm cool with both, honestly | 08:51 |
| flaper87 | I'm trying to think what's the sanest way forward | 08:51 |
| flaper87 | therve: thoughts ? | 08:51 |
| therve | flaper87, I think for the asyncio runner using threads is saner | 08:51 |
| flaper87 | ok, lets go with threadpool executor and all that | 08:51 |
| flaper87 | we'll revisit this in Mitaka | 08:52 |
| therve | +1 | 08:52 |
| therve | By workers you mean external processes? | 08:52 |
| flaper87 | therve: yeah... it's just a random thought | 08:52 |
| flaper87 | having external workers is something I've also fought to keep away | 08:53 |
| therve | I'd like to have a better way to run those for sure | 08:53 |
| flaper87 | because that means you need a way to communicate between them | 08:53 |
| therve | The business of talking to arbitrary URLs is a tough one | 08:53 |
| flaper87 | yeah | 08:53 |
| openstackgerrit | Thomas Herve proposed openstack/zaqar: Use thread executor all the time for notifier https://review.openstack.org/219609 | 08:55 |
| therve | I'd like to ask Joshua why there is no error when we use the greenlet executor without patching though | 08:56 |
| therve | I would expect the error to be raised in that case | 08:56 |
| openstackgerrit | Thomas Herve proposed openstack/zaqar: Support various paths for signed URLs https://review.openstack.org/215638 | 09:28 |
| *** dynarro has quit IRC | 09:37 | |
| *** flwang has joined #openstack-zaqar | 10:11 | |
| *** flwang has quit IRC | 10:22 | |
| *** malini has joined #openstack-zaqar | 10:22 | |
| *** diga has joined #openstack-zaqar | 10:27 | |
| openstackgerrit | Thomas Herve proposed openstack/zaqar: Use thread executor all the time for notifier https://review.openstack.org/219609 | 10:30 |
| *** Qiming has quit IRC | 10:54 | |
| *** malini has quit IRC | 11:28 | |
| * ryansb is late to the party | 11:34 | |
| ryansb | flaper87: for the swift backend eventlet is pretty much a requirement | 11:35 |
| ryansb | because the latency in swift + the number of disparate requests mean that it'll be horrible if we do everything serially | 11:35 |
| flaper87 | ryansb: right, we had talked about it and I forgot | 11:46 |
| flaper87 | ryansb: but again, that's specific to the backend | 11:46 |
| flaper87 | What I mean is, the swift backend will take care of patching and doing its stuff | 11:47 |
| flaper87 | although all these distributed patching calls might end up being a problem | 11:47 |
| * flaper87 shrugs | 11:47 | |
| flaper87 | definitely something we need to dive in during Mitaka | 11:47 |
| ryansb | I feel like each backend sometimes-maybe doing patching calls is a recipe for disaster | 11:48 |
| ryansb | not like, near future disaster | 11:48 |
| ryansb | but in a couple cycles, everything just implodes | 11:48 |
| *** Qiming_ has joined #openstack-zaqar | 11:49 | |
| therve | ryansb, Could we make async requests to swift? | 12:07 |
| therve | If we're gonna pick something I'd rather pick asyncio rather than eventlet, if we can | 12:08 |
| ryansb | aio would also work | 12:08 |
| ryansb | I don't have a preference, but I don't think tulip is in the global reqs | 12:09 |
| ryansb | so we would have to get it in there if we want to use it in python 2 | 12:10 |
| therve | tulip is not a thing | 12:10 |
| therve | trollius is though, and we already use it | 12:10 |
| ryansb | ah, I thought the backport was called tulip | 12:11 |
| ryansb | well then, excellent | 12:11 |
| ryansb | but yeah, I recall that tulip was the name for the early py3 version of asyncIO | 12:12 |
| ryansb | anyways, if that'd be preferable, then certainly it'd work | 12:12 |
| therve | Right, it was just a code name, never something released | 12:12 |
| therve | ryansb, We already use trollius for websocket | 12:12 |
| *** david-lyle has quit IRC | 12:27 | |
| *** david-lyle has joined #openstack-zaqar | 12:28 | |
| flaper87 | therve:++ | 12:34 |
| flaper87 | (sorry for stepping out of the convo, kinda doing PTL things now T_T) | 12:34 |
| flaper87 | YAY, FF week | 12:34 |
| ryansb | \o/ | 12:47 |
| openstackgerrit | Merged openstack/zaqar: Add support for flavors in sqlalchemy https://review.openstack.org/218291 | 12:48 |
| *** diga has quit IRC | 13:01 | |
| *** sriram1 has joined #openstack-zaqar | 13:23 | |
| *** AAzza_ has joined #openstack-zaqar | 13:25 | |
| *** AAzza has quit IRC | 13:27 | |
| openstackgerrit | Merged openstack/zaqar: s/update/_update/ sqlalchemy's pool controller https://review.openstack.org/218292 | 13:27 |
| *** AAzza_ has quit IRC | 13:30 | |
| *** AAzza has joined #openstack-zaqar | 13:30 | |
| *** amitgandhinz has joined #openstack-zaqar | 13:38 | |
| *** dynarro has joined #openstack-zaqar | 13:44 | |
| *** ametts has joined #openstack-zaqar | 13:51 | |
| *** malini has joined #openstack-zaqar | 14:09 | |
| *** sriram1 has quit IRC | 14:29 | |
| *** sriram1 has joined #openstack-zaqar | 14:29 | |
| *** sriram1 is now known as sriram | 14:29 | |
| openstackgerrit | Victoria Martinez de la Cruz proposed openstack/zaqar: Refactoring to make Websocket driver available in v2 https://review.openstack.org/218430 | 14:35 |
| *** kgriffs|afk is now known as kgriffs | 14:47 | |
| *** malini has quit IRC | 15:08 | |
| *** Qiming_ has quit IRC | 15:13 | |
| *** malini has joined #openstack-zaqar | 15:14 | |
| *** malini has quit IRC | 15:20 | |
| *** malini has joined #openstack-zaqar | 15:21 | |
| openstackgerrit | Claudiu Belu proposed openstack/zaqar: Adds oslo_cache dependency https://review.openstack.org/219374 | 15:50 |
| *** amitgandhinz has quit IRC | 15:51 | |
| *** ametts has quit IRC | 15:54 | |
| openstackgerrit | Claudiu Belu proposed openstack/zaqar: Adds oslo_cache dependency https://review.openstack.org/219374 | 16:00 |
| flaper87 | vkmc: thanks for updating WS | 16:03 |
| flaper87 | it's green | 16:03 |
| flaper87 | I'll review asap | 16:03 |
| *** mpanetta has joined #openstack-zaqar | 16:03 | |
| flaper87 | We'll land that and emails in RC1 | 16:03 |
| flaper87 | hopefully policy as well | 16:04 |
| *** dynarro has quit IRC | 16:17 | |
| *** dynarro has joined #openstack-zaqar | 16:22 | |
| *** dynarro has quit IRC | 16:33 | |
| *** malini has quit IRC | 16:41 | |
| vkmc | coolio | 16:56 |
| vkmc | will check the reviews for emails and policy | 16:56 |
| ryansb | can we have like, a dashboard of "we want to land in RC1" | 16:59 |
| ryansb | or a list/etherpad/something | 16:59 |
| *** malini has joined #openstack-zaqar | 17:17 | |
| *** malini has quit IRC | 17:21 | |
| *** stanchan has quit IRC | 17:22 | |
| *** achanda has joined #openstack-zaqar | 17:24 | |
| *** sriram has quit IRC | 17:53 | |
| *** sriram has joined #openstack-zaqar | 17:55 | |
| *** malini has joined #openstack-zaqar | 17:56 | |
| *** malini has quit IRC | 18:03 | |
| *** malini has joined #openstack-zaqar | 18:04 | |
| *** ametts has joined #openstack-zaqar | 18:08 | |
| vkmc | ryansb++ | 18:12 |
| *** csoukup has joined #openstack-zaqar | 18:16 | |
| *** amitgandhinz has joined #openstack-zaqar | 18:18 | |
| *** kgriffs is now known as kgriffs|afk | 18:21 | |
| *** kgriffs|afk is now known as kgriffs | 18:30 | |
| *** kgriffs is now known as kgriffs|afk | 18:41 | |
| *** kgriffs|afk is now known as kgriffs | 19:28 | |
| * ryansb engages review mode | 19:30 | |
| *** amitgandhinz has quit IRC | 19:38 | |
| *** malini has quit IRC | 19:39 | |
| openstackgerrit | Claudiu Belu proposed openstack/zaqar: Adds oslo_concurrency dependency https://review.openstack.org/219374 | 19:51 |
| *** amitgandhinz has joined #openstack-zaqar | 20:04 | |
| *** stanchan has joined #openstack-zaqar | 20:22 | |
| *** stanchan has quit IRC | 20:27 | |
| *** achanda has quit IRC | 21:08 | |
| flwang1 | ryansb: could you pls review this one https://review.openstack.org/#/c/219609/2 ? | 21:14 |
| ryansb | yeah | 21:14 |
| flwang1 | i think that's why the gate is not really stable for notification | 21:14 |
| flwang1 | ryansb: cool, cheers man | 21:14 |
| *** sriram has quit IRC | 21:26 | |
| openstackgerrit | Merged openstack/zaqar: Use thread executor all the time for notifier https://review.openstack.org/219609 | 21:35 |
| *** achanda has joined #openstack-zaqar | 21:37 | |
| openstackgerrit | Fei Long Wang proposed openstack/zaqar: Add email driver for notification https://review.openstack.org/202374 | 21:38 |
| openstackgerrit | Fei Long Wang proposed openstack/zaqar: Fix duplicated notification https://review.openstack.org/218072 | 21:39 |
| *** achanda has quit IRC | 21:47 | |
| *** achanda has joined #openstack-zaqar | 21:54 | |
| *** amitgandhinz has quit IRC | 21:56 | |
| *** achanda has quit IRC | 22:24 | |
| *** achanda_ has joined #openstack-zaqar | 22:24 | |
| *** ametts has quit IRC | 22:40 | |
| *** mpanetta has quit IRC | 22:46 | |
| *** csoukup has quit IRC | 22:47 | |
| *** kgriffs is now known as kgriffs|afk | 22:52 | |
| *** malini has joined #openstack-zaqar | 23:05 | |
| *** achanda_ has quit IRC | 23:12 | |
| *** malini has quit IRC | 23:13 | |
| *** malini has joined #openstack-zaqar | 23:15 | |
| *** malini has quit IRC | 23:16 | |
| *** malini has joined #openstack-zaqar | 23:19 | |
| *** malini has quit IRC | 23:20 | |
| *** achanda has joined #openstack-zaqar | 23:21 | |
| *** Qiming_ has joined #openstack-zaqar | 23:26 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/zaqar: Updated from global requirements https://review.openstack.org/219506 | 23:50 |
| *** achanda has quit IRC | 23:52 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!