| *** wanghao has quit IRC | 00:00 | |
| *** wanghao has joined #openstack-zaqar | 00:02 | |
| *** JAHoagie has quit IRC | 00:09 | |
| Eva-i | flwang: sure. Can you explain why such naming "show_default_attributes_for_queue-3d87333752484c87.yaml" of files, particularly the number in the end? | 00:46 |
|---|---|---|
| Eva-i | flwang: some kind of hash | 00:46 |
| Eva-i | flwang: nevermind, I have found the answer ;) | 01:02 |
| flwang | Eva-i: any updates for the install-guide work? | 01:16 |
| Eva-i | flwang: no | 01:16 |
| flwang | Eva-i: when can we get a patch like this https://review.openstack.org/#/c/325389/ ? | 01:17 |
| Eva-i | flwang: in the end of the week, I think | 01:18 |
| flwang | this week? | 01:18 |
| Eva-i | flwang: this week | 01:18 |
| flwang | awesome, thanks | 01:18 |
| openstackgerrit | Eva Balycheva proposed openstack/zaqar: Make queues lazy in subscriptions https://review.openstack.org/320749 | 01:30 |
| flwang | Eva-i: you raised a good topic in the release note of your feature | 01:41 |
| flwang | when delete a queue, if the subscriptions should be deleted as well | 01:41 |
| flwang | and the answer is yes i think just like what we did for messages | 01:42 |
| flwang | SNS is doing the same thing http://docs.aws.amazon.com/sns/latest/dg/CleanUp.html | 01:42 |
| Eva-i | flwang: oki, thank you. So you would like to delete all subscriptions on queue delete? | 01:45 |
| flwang | Eva-i: yes, https://blueprints.launchpad.net/zaqar/+spec/clear-subscription-after-delete-queue | 01:47 |
| Eva-i | flwang: this thing is backward incompatible. We will probably need a config option? | 01:48 |
| flwang | Eva-i: it's ok for me | 01:49 |
| *** shu-mutou has joined #openstack-zaqar | 01:49 | |
| Eva-i | flwang: oki | 01:50 |
| Eva-i | wxy: wanghao: flaper87: vkmc: ryansb: hello, dear zaqarians, see this bp: https://blueprints.launchpad.net/zaqar/+spec/clear-subscription-after-delete-queue. What do you think about this idea? | 01:51 |
| Eva-i | flwang: should I remove a sentence from my patch release notes about still working subscriptions? Or it will be removed on patch that will implement your blueprint? | 01:52 |
| flwang | Eva-i: it's ok. to keep it for now until we fix it. | 01:53 |
| Eva-i | flwang: alright, thank you | 01:53 |
| flwang | +2ed | 01:53 |
| Eva-i | flwang: for some reason I can't render releasenotes locally. Releasenotes for the current cycle are empty.. We should wait for jenkins in my patch and see how it will render. | 01:55 |
| Eva-i | flwang: thanks | 01:55 |
| flwang | ok | 01:55 |
| Eva-i | flwang: ok, it renders fine, but I found an error in my description. | 02:07 |
| Eva-i | flwang: will fix it now | 02:07 |
| flwang | Eva-i: ok | 02:08 |
| openstackgerrit | Eva Balycheva proposed openstack/zaqar: Make queues lazy in subscriptions https://review.openstack.org/320749 | 02:08 |
| *** Kevin_Zheng has quit IRC | 02:21 | |
| Eva-i | flwang: is will be bad if we make zaqar send notifications after messages getting to storage driver? It's in pipeline stage's definition that they must execute before storage driver? | 03:08 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/zaqar: Updated from global requirements https://review.openstack.org/322589 | 03:46 |
| flwang | Eva-i: are you talking about adding msg id into notification? | 04:17 |
| Eva-i | flwang: yes | 04:20 |
| flwang | why do you think it's bad to send notification after the msg saved into db? | 04:20 |
| Eva-i | flwang: nothing wrong with it. There's a problem that the notification driver is implemented as pipeline stage. If we move it's execution after storage driver, there will be no point of notification driver being the pipeline stage - being turned on by normal option would be much better. | 04:30 |
| flwang | i can see your point | 04:32 |
| flwang | but if we totally refactor current architecture, the change is too big | 04:34 |
| flwang | so what i'm looking for is keep the notifier in the pipeline but can get the message id | 04:34 |
| Eva-i | flwang: I started implementing pipeline stage that just sets special flag. Storage driver notices this flag and sends notification. But it feels wrong. | 04:36 |
| Eva-i | flwang: it's possible, but it feels wrong | 04:36 |
| flwang | Eva-i: IMHO | 04:37 |
| flwang | pipeline should do something like this stage 1 -> stage 2 -> stage 3 -> .... | 04:37 |
| flwang | and the stage N+1 should be able to get the output of stage N | 04:37 |
| Eva-i | flwang: hm, I'm very unsure it's true for Zaqar | 04:39 |
| *** JAHoagie has joined #openstack-zaqar | 04:39 | |
| flwang | Eva-i: so what's your current way? | 04:40 |
| Eva-i | flwang: I explained my current way, but I don't like it. As you know there's also other ways: 1. Big refactoring 1.1 Making output pipeline (after storage layer). 1.2 Making notification driver normal. 2. Somehow make generation of IDs before storage layer. MongoDB accepts custom IDs, but I'm not sure if it will affect performance (MongoDB is probably optimized to work incremental IDs, not with random) and how we avoid collisions. | 04:47 |
| flwang | #2 is also a big change, IMHO | 04:51 |
| flwang | can't we improve the current pipeline like this ? | 04:52 |
| flwang | (16:37:32) flwang: pipeline should do something like this stage 1 -> stage 2 -> stage 3 -> .... | 04:52 |
| flwang | (16:37:49) flwang: and the stage N+1 should be able to get the output of stage N | 04:52 |
| Eva-i | flwang: I think it's possible | 04:53 |
| flwang | Eva-i: that's what i want to see, IMHO, though i haven't dig into the code, so i have no idea if it's 'huge' change | 04:54 |
| Eva-i | flwang: but if users have created their own pipelines, they might be incompatible. But maybe it's possible to make this change compatible. | 04:54 |
| flwang | but that's the ideal solution for me | 04:54 |
| flwang | user have created their own pipelines? no | 04:55 |
| flwang | i don't think there is any user using zaqar and created a pipeline | 04:55 |
| flwang | and | 04:55 |
| flwang | our pipeline is not a plugin for end user, IMHO | 04:55 |
| Eva-i | flwang: I think I should read more about pipelines to be sure your way is good. I think I will check out how it's made in keystone.. | 04:55 |
| Eva-i | flwang: from the code it feels like pipeline is set to accept plugins. | 04:57 |
| *** GB21 has joined #openstack-zaqar | 04:57 | |
| Eva-i | flwang: okay, I'm going to sleep now | 04:57 |
| Eva-i | flwang: thanks for today | 04:57 |
| flwang | Eva-i: yes, for sure, otherwise, we won't have subscription plugin | 04:57 |
| flwang | i mean the notifier plugin | 04:58 |
| flwang | but again, i don't think it's designed for end user | 04:58 |
| Eva-i | flwang: yes, it's designed for programmers and cloud admins. | 04:58 |
| Eva-i | i think | 04:59 |
| flwang | even not cloud admins, at least based on my knowledge | 04:59 |
| flwang | good night, i have to go | 04:59 |
| *** rcernin|off has joined #openstack-zaqar | 05:15 | |
| *** ameade has quit IRC | 05:24 | |
| *** wxy has quit IRC | 05:25 | |
| *** aimeeu has quit IRC | 05:25 | |
| *** boris-42 has quit IRC | 05:25 | |
| *** zhiyan has quit IRC | 05:25 | |
| *** rcernin|off has quit IRC | 05:26 | |
| *** JAHoagie has quit IRC | 05:26 | |
| *** wxy has joined #openstack-zaqar | 05:28 | |
| *** aimeeu has joined #openstack-zaqar | 05:32 | |
| *** ameade has joined #openstack-zaqar | 05:32 | |
| *** wxy has quit IRC | 05:33 | |
| *** zhiyan has joined #openstack-zaqar | 05:33 | |
| *** boris-42 has joined #openstack-zaqar | 05:34 | |
| *** wxy has joined #openstack-zaqar | 05:35 | |
| *** JAHoagie has joined #openstack-zaqar | 05:54 | |
| *** pt_15 has quit IRC | 05:56 | |
| *** itisha has quit IRC | 06:00 | |
| *** GB21 has quit IRC | 06:03 | |
| *** wxy has quit IRC | 06:08 | |
| *** wxy has joined #openstack-zaqar | 06:11 | |
| *** rcernin|off has joined #openstack-zaqar | 06:14 | |
| *** GB21 has joined #openstack-zaqar | 06:19 | |
| *** rcernin|off is now known as rcernin | 06:21 | |
| *** GB21 has quit IRC | 06:42 | |
| *** GB21 has joined #openstack-zaqar | 07:00 | |
| *** pcaruana has joined #openstack-zaqar | 07:09 | |
| *** JAHoagie has quit IRC | 07:35 | |
| *** tesseract has joined #openstack-zaqar | 07:56 | |
| *** openstackgerrit has quit IRC | 08:03 | |
| *** openstackgerrit has joined #openstack-zaqar | 08:03 | |
| *** stanchan has joined #openstack-zaqar | 08:56 | |
| *** Ravikiran_K has joined #openstack-zaqar | 09:16 | |
| *** Ravikiran_K has quit IRC | 09:29 | |
| *** shu-mutou has quit IRC | 09:30 | |
| *** stanchan has quit IRC | 10:17 | |
| openstackgerrit | Yatin Kumbhare proposed openstack/zaqar: Use is_valid_ipv4 from oslo.utils https://review.openstack.org/327601 | 11:11 |
| *** GB21 has quit IRC | 11:37 | |
| *** mpanetta has joined #openstack-zaqar | 11:49 | |
| *** mpanetta has quit IRC | 11:53 | |
| *** GB21 has joined #openstack-zaqar | 11:57 | |
| *** GB21 has quit IRC | 12:16 | |
| *** wanghao has quit IRC | 12:20 | |
| *** wanghao has joined #openstack-zaqar | 12:20 | |
| *** itisha has joined #openstack-zaqar | 13:44 | |
| *** ametts has joined #openstack-zaqar | 13:49 | |
| *** JAHoagie has joined #openstack-zaqar | 13:58 | |
| *** JAHoagie has quit IRC | 14:04 | |
| *** mpanetta has joined #openstack-zaqar | 14:04 | |
| *** mpanetta has quit IRC | 14:09 | |
| *** GB21 has joined #openstack-zaqar | 14:11 | |
| *** GB21 has quit IRC | 14:21 | |
| *** wanghao has quit IRC | 14:41 | |
| *** pcaruana has quit IRC | 14:41 | |
| *** wanghao has joined #openstack-zaqar | 14:42 | |
| *** JAHoagie has joined #openstack-zaqar | 14:55 | |
| *** JAHoagie has quit IRC | 15:00 | |
| openstackgerrit | Alan Pevec proposed openstack/zaqar: Relocates cli options registration from bootstrap to server https://review.openstack.org/327268 | 15:30 |
| *** tesseract has quit IRC | 15:30 | |
| *** rcernin has quit IRC | 15:32 | |
| *** GB21 has joined #openstack-zaqar | 15:33 | |
| *** srwilkers has joined #openstack-zaqar | 16:07 | |
| *** GB21 has quit IRC | 16:12 | |
| *** srwilkers has quit IRC | 16:23 | |
| *** JAHoagie has joined #openstack-zaqar | 16:47 | |
| *** srwilkers has joined #openstack-zaqar | 17:43 | |
| *** srwilkers has quit IRC | 17:51 | |
| *** srwilkers has joined #openstack-zaqar | 17:52 | |
| *** itisha has quit IRC | 18:02 | |
| *** itisha has joined #openstack-zaqar | 18:03 | |
| *** kgriffs has quit IRC | 18:07 | |
| *** mpanetta has joined #openstack-zaqar | 18:07 | |
| *** kgriffs has joined #openstack-zaqar | 18:07 | |
| *** mpanetta has quit IRC | 18:11 | |
| *** stanchan has joined #openstack-zaqar | 18:12 | |
| *** mpanetta has joined #openstack-zaqar | 18:16 | |
| *** mpanetta has quit IRC | 18:21 | |
| *** mpanetta has joined #openstack-zaqar | 18:22 | |
| *** tqtran has joined #openstack-zaqar | 18:24 | |
| *** srwilkers has quit IRC | 18:24 | |
| *** stanchan has quit IRC | 18:39 | |
| *** stevebaker has quit IRC | 18:55 | |
| *** stevebaker has joined #openstack-zaqar | 18:55 | |
| *** srwilkers has joined #openstack-zaqar | 18:58 | |
| *** srwilkers has quit IRC | 19:19 | |
| *** srwilkers has joined #openstack-zaqar | 19:20 | |
| *** srwilkers has quit IRC | 19:23 | |
| openstackgerrit | Zane Bitter proposed openstack/zaqar-specs: Spec for Mistral Notifications https://review.openstack.org/318202 | 19:36 |
| *** srwilkers has joined #openstack-zaqar | 19:50 | |
| *** srwilkers has quit IRC | 20:11 | |
| *** rcernin has joined #openstack-zaqar | 20:35 | |
| *** ametts has quit IRC | 20:52 | |
| *** mpanetta has quit IRC | 21:00 | |
| *** rcernin has quit IRC | 21:15 | |
| *** njohnston is now known as njohnston|afk | 21:49 | |
| openstackgerrit | Steve Linabery proposed openstack/zaqar: WIP: Relocates cli options registration from bootstrap to server https://review.openstack.org/327268 | 22:49 |
| openstackgerrit | Steve Linabery proposed openstack/zaqar: Relocates cli options registration from bootstrap to server https://review.openstack.org/327268 | 22:53 |
| *** davideagnello has joined #openstack-zaqar | 23:03 | |
| *** davideagnello has quit IRC | 23:04 | |
| openstackgerrit | Fei Long Wang proposed openstack/zaqar: Add release notes https://review.openstack.org/326704 | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!