| Eva-i | vkmc: okay | 00:00 |
|---|---|---|
| *** malini has joined #openstack-zaqar | 00:08 | |
| *** malini has quit IRC | 00:10 | |
| Eva-i | vkmc: also I think python-zaqarclient is just a user of Zaqar server API. Client's API and server's API can't be versioned together, they have even different nature. For example, many possible changes in Zaqar server web API can require only client's inner workings change and will not affect the client python API. | 00:11 |
| *** malini has joined #openstack-zaqar | 00:12 | |
| Eva-i | *client's python API | 00:12 |
| *** kgriffs is now known as kgriffs|afk | 00:15 | |
| Eva-i | vkmc: okay, we can close the topic. Maybe the reason that I find compatibility important is because that in my childhood there were incidents when I applied a patch to computer game and sometimes found my saved games not working. It was very sad. =) | 00:16 |
| vkmc | Eva-i, so... we always followed that schema across all OpenStack clients... the client is only a way to interact with the server and in server version changes we add the required support in the client side | 00:18 |
| vkmc | Eva-i, perhaps we should handle versioning for the client differently... I haven't bumped into a case in which I'd need to add something to the client that would require changing the version and it's not something that changed in the server | 00:19 |
| vkmc | but it can happen maybe | 00:19 |
| vkmc | some food for thought | 00:19 |
| Eva-i | vkmc: this patch is impacting the client API even if when there was no change from the server side to adapt to it. https://review.openstack.org/#/c/245526/ | 00:19 |
| vkmc | Eva-i, sure, it's ok, it's always good to stop and think why we are doing things and if they make sense | 00:19 |
| Eva-i | *even if there was | 00:20 |
| vkmc | let me see, I missed that one | 00:20 |
| *** malini has quit IRC | 00:21 | |
| flwang | Eva-i: i like questions from you, you know, some times they trigger me to think some corners i never touch | 00:21 |
| flwang | Eva-i: pls keep asking | 00:21 |
| Eva-i | vkmc: so the client API can actually change when Zaqar server API hasn't changed at all | 00:22 |
| Eva-i | flwang: oh, okay | 00:22 |
| flwang | Eva-i: let's forget client API | 00:23 |
| flwang | client just provide a interface so that the consumer can use zaqar rest api easily | 00:23 |
| flwang | if you're talking about lib layer | 00:23 |
| Eva-i | API is an interface | 00:24 |
| Eva-i | flwang: what python-zaqar client does: it converts Zaqar server's web API into python/command-line API. | 00:25 |
| flwang | Eva-i: i agree :) | 00:26 |
| Eva-i | oh, I just realized that python-zaqarclient has even two it's own APIs :o | 00:26 |
| flwang | 'Client API' is not a good term in openstack world | 00:26 |
| flwang | we don't use it, actually | 00:26 |
| Eva-i | flwang: what do you use? | 00:27 |
| Eva-i | middleware API? | 00:28 |
| flwang | we just say client, haha | 00:30 |
| flwang | because cli is just a very thin layer | 00:31 |
| flwang | TBH, i'm not really sure what you're looking for | 00:31 |
| flwang | generally, client should be compatible with server(REST API) | 00:32 |
| flwang | and the compatible means the version matching | 00:32 |
| Eva-i | yes | 00:32 |
| Eva-i | it should be adapted to be able to use all supported REST API versions | 00:33 |
| Eva-i | but should the client try to be compatible with the code that relies on it? | 00:35 |
| *** tej_ has quit IRC | 00:37 | |
| *** tej has quit IRC | 00:37 | |
| Eva-i | or it can change sometimes, but how the people who use python-zaqarclient will know will be the new version of python-zaqarclient compatible with their code or not? Usually people look at major versions. | 00:38 |
| flwang | as for the first question, i would say yes, but you know, we're still growing | 00:39 |
| flwang | or i would say we're stilling changing | 00:39 |
| flwang | changing is not bad | 00:39 |
| flwang | incompatible is not always bad | 00:40 |
| flwang | just depends if the changing's direction is better or worse | 00:40 |
| Eva-i | flwang: if not many people are currently using Zaqar and python-zaqarclient, then we can change them drastically | 00:40 |
| flwang | if we're fixing some old/bad things | 00:40 |
| Eva-i | without telling people hints about compatibility | 00:41 |
| flwang | Eva-i: we have release note | 00:41 |
| flwang | did you mean the changing from noauth to keystone auth for library? | 00:43 |
| flwang | if there are many users jump and beat me, i will be very happy :) | 00:43 |
| flwang | because that means we have a lot of users | 00:43 |
| flwang | anyway, i think i have mentioned several times about the reason i prefer keystone auth instead of noauth | 00:44 |
| flwang | i believe all the user using zaqar is an openstack user | 00:44 |
| flwang | and keystone is the only ideantiy service of openstack | 00:45 |
| flwang | at least i believe no user is using noauth for zaqar | 00:45 |
| Eva-i | flwang: I personally think that every minor change that affects the interface should be made somehow visible for users, even if it's just a change from "noauth" to "keystone". | 00:46 |
| Eva-i | flwang: just wanted to say that | 00:46 |
| flwang | Eva-i: i agree | 00:46 |
| Eva-i | flwang: of course serious users will not notice anything in this particular change, they are already using keystone for auth. | 00:46 |
| *** sonuk has quit IRC | 00:47 | |
| ryansb | ^^ agree with eva | 00:47 |
| ryansb | so the client has to keep up with the REST API (or what good would it be!) | 00:48 |
| flwang | Eva-i: ryansb: i totally agree, just like we all agree with the good things. | 00:48 |
| flwang | but | 00:48 |
| ryansb | but the Python API (for people who `import zaqarclient`) is also important and should be kept stable as much as we can | 00:48 |
| openstackgerrit | Merged openstack/zaqar: Fix duplicate auth_section issue https://review.openstack.org/253220 | 00:48 |
| ryansb | it will sometimes have to change due to server changes, and that's ok | 00:49 |
| Eva-i | flwang: I just wanted to say that maybe it would be nice in the future to address compatibility issues with more thought | 00:49 |
| ryansb | \o/ yay that's merged | 00:49 |
| flwang | ryansb: it merged YEAH! | 00:49 |
| ryansb | Eva-i: yes, you're right | 00:49 |
| Eva-i | flwang: so the people will not be scared of updating | 00:49 |
| flwang | Eva-i: yep | 00:49 |
| flwang | let me continuing the 'but' | 00:49 |
| ryansb | just rechecked the two dependent patches | 00:49 |
| Eva-i | yes, continue =) | 00:49 |
| flwang | but, we're a small team, sometimes you have to balance the growth and something else | 00:50 |
| flwang | we can't make anything perfect at the same time | 00:50 |
| ryansb | I think the way to go for the Python API is to be liberal about increasing the version when we break something. As you say, we're still evolving so there *will* be changes | 00:50 |
| ryansb | and folks can always cap their requirements | 00:50 |
| flwang | ryansb: yes | 00:51 |
| flwang | Eva-i: we haven't release a new release, so we have time to improve | 00:51 |
| flwang | but again, i agree with you | 00:51 |
| flwang | ryansb: thank you very much | 00:52 |
| Eva-i | flwang: also what do you think, should I use in zaqar-bench just "noauth" option or implement the thing mdnadeem has proposed https://review.openstack.org/#/c/251023/4/zaqar/bench/producer.py | 00:54 |
| ryansb | Eva-i: I'd say with ks | 00:55 |
| ryansb | since every "real" deploy is going to have it | 00:56 |
| flwang | Eva-i: i'm a little bit confused | 00:56 |
| ryansb | so we should include it in benchmarks | 00:56 |
| flwang | Eva-i: does it work with noauth? | 00:56 |
| Eva-i | flwang: with current version of python-zaqarclient it will work if "noauth" will be explicitly passed as option | 00:57 |
| Eva-i | ryansb: with ks? | 00:57 |
| ryansb | yeah, keystone | 00:57 |
| flwang | Eva-i: then, i'm ok with it. | 00:59 |
| flwang | but just like ryansb said, if we can support keystone auth, it would be great | 00:59 |
| Eva-i | flwang: then it's better for me to implement keystone support for zaqar-bench =) | 00:59 |
| flwang | but it's not the blocker of your patch | 01:00 |
| Eva-i | ryansb: ^ | 01:00 |
| Eva-i | flwang: should I separate these two patches? | 01:00 |
| Eva-i | *changes | 01:00 |
| flwang | Eva-i: it would be nice, keystone auth support is a 'new' change. as for your current patch, you're fix an existing issue | 01:00 |
| flwang | so they're different and should be tracked by separated patches | 01:01 |
| *** Qiming has joined #openstack-zaqar | 01:01 | |
| Eva-i | flwang: oki | 01:01 |
| Eva-i | flwang: can be _setup_auth_params from python-zaqarclient be reused in zaqar-bench? :o | 01:01 |
| ryansb | Eva-i: wait, which two? | 01:02 |
| ryansb | the "get zaqar-bench working" one and the "add keystone" change? | 01:02 |
| Eva-i | flwang: maybe credentials should be just options for zaqar-bench? | 01:02 |
| ryansb | yeah, I'd say that'd be two patches | 01:02 |
| Eva-i | ryansb: yes | 01:02 |
| Eva-i | ryansb: alright | 01:02 |
| flwang | Eva-i: thanks | 01:04 |
| Eva-i | I'll make zaqar-bench use os_client_config.OpenStackConfig(), so it will be able to get credentials from clouds.yaml/environment variables | 01:14 |
| flwang | Eva-i: sounds a good idea | 01:15 |
| Eva-i | so if user will want noauth, it will set environment variable OS_AUTH_STRATEGY=noauth | 01:16 |
| Eva-i | flwang: okay, than it's all decided | 01:16 |
| *** kgriffs|afk is now known as kgriffs | 01:16 | |
| Eva-i | *then | 01:16 |
| ryansb | ok, https://review.openstack.org/#/c/252154/ and https://review.openstack.org/#/c/252638/ are almost past the first set of checks, passing so far | 01:17 |
| ryansb | I'm logging off for the night, so flwang (or whoever else is core in a tz where it's daytime) please shepherd those along for M-1 | 01:18 |
| ryansb | Night zaqarians | 01:18 |
| Eva-i | ryansb: good night | 01:18 |
| flwang | ryansb: sure, thank you so much | 01:20 |
| *** lei-zh1 has joined #openstack-zaqar | 01:27 | |
| *** lei-zh1 has left #openstack-zaqar | 01:27 | |
| openstackgerrit | Merged openstack/python-zaqarclient: Pool delete cli support for v2 https://review.openstack.org/250236 | 01:35 |
| openstackgerrit | Merged openstack/python-zaqarclient: Pool list cli support for v2 https://review.openstack.org/250232 | 01:35 |
| openstackgerrit | Merged openstack/python-zaqarclient: Pool & flavor cli implementation for v2 https://review.openstack.org/243995 | 01:35 |
| openstackgerrit | Merged openstack/python-zaqarclient: Remove py26 support https://review.openstack.org/252270 | 01:36 |
| openstackgerrit | Merged openstack/zaqar: Add unreleased notes for releasenotes tree https://review.openstack.org/252638 | 01:42 |
| openstackgerrit | Fei Long Wang proposed openstack/python-zaqarclient: Refactor the test for v2 queues and messages https://review.openstack.org/245415 | 01:55 |
| openstackgerrit | Merged openstack/python-zaqarclient: delete python bytecode before every test run https://review.openstack.org/252181 | 01:58 |
| openstackgerrit | Fei Long Wang proposed openstack/python-zaqarclient: Add support for flavor list v1 https://review.openstack.org/248502 | 01:59 |
| *** sonuk has joined #openstack-zaqar | 03:04 | |
| *** dmowrer has joined #openstack-zaqar | 03:07 | |
| *** dmowrer has quit IRC | 03:11 | |
| *** kgriffs is now known as kgriffs|afk | 03:21 | |
| *** flwang1 has quit IRC | 03:46 | |
| *** VeggieMeat_ is now known as VeggieMeat | 04:13 | |
| *** dmowrer has joined #openstack-zaqar | 05:34 | |
| *** dmowrer has quit IRC | 05:38 | |
| *** rcernin has joined #openstack-zaqar | 06:12 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/zaqar: Imported Translations from Zanata https://review.openstack.org/250621 | 06:25 |
| *** rcernin has quit IRC | 06:59 | |
| *** rcernin has joined #openstack-zaqar | 07:07 | |
| *** sonuk has quit IRC | 07:58 | |
| -openstackstatus- NOTICE: Tox tests are broken at the moment. From openstack-infra we are working to fix them. Please don't approve changes until we notify that tox tests work again. | 09:32 | |
| *** Qiming has quit IRC | 10:00 | |
| *** openstackgerrit has quit IRC | 10:32 | |
| *** openstackgerrit has joined #openstack-zaqar | 10:32 | |
| *** Qiming has joined #openstack-zaqar | 10:56 | |
| *** Qiming has quit IRC | 11:55 | |
| *** dmowrer has joined #openstack-zaqar | 12:13 | |
| *** dmowrer has quit IRC | 12:47 | |
| *** dmowrer has joined #openstack-zaqar | 12:56 | |
| *** njohnsto_ has quit IRC | 13:19 | |
| *** david-lyle has quit IRC | 13:43 | |
| *** dmowrer has quit IRC | 13:43 | |
| *** dmowrer has joined #openstack-zaqar | 13:44 | |
| *** csoukup has joined #openstack-zaqar | 13:53 | |
| *** sriram has joined #openstack-zaqar | 14:04 | |
| *** sriram has quit IRC | 14:06 | |
| *** sriram has joined #openstack-zaqar | 14:08 | |
| *** mpanetta has joined #openstack-zaqar | 14:17 | |
| *** dmowrer has quit IRC | 14:22 | |
| *** amitgandhinz has joined #openstack-zaqar | 14:32 | |
| *** njohnston has joined #openstack-zaqar | 14:37 | |
| *** malini has joined #openstack-zaqar | 14:39 | |
| *** dmowrer has joined #openstack-zaqar | 14:39 | |
| *** malini1 has joined #openstack-zaqar | 14:41 | |
| *** malini has quit IRC | 14:44 | |
| *** dynarro has joined #openstack-zaqar | 14:49 | |
| *** sriram has quit IRC | 14:52 | |
| *** sriram has joined #openstack-zaqar | 14:53 | |
| *** ametts has joined #openstack-zaqar | 15:00 | |
| *** tej has joined #openstack-zaqar | 15:01 | |
| *** tej_ has joined #openstack-zaqar | 15:01 | |
| *** kgriffs|afk is now known as kgriffs | 15:02 | |
| *** csoukup has quit IRC | 15:10 | |
| *** dynarro has quit IRC | 15:12 | |
| *** sonuk has joined #openstack-zaqar | 15:18 | |
| *** csoukup has joined #openstack-zaqar | 15:32 | |
| *** dmowrer has quit IRC | 15:43 | |
| *** dmowrer has joined #openstack-zaqar | 15:44 | |
| *** dynarro has joined #openstack-zaqar | 15:52 | |
| *** csoukup has quit IRC | 15:57 | |
| *** tej_ has quit IRC | 16:11 | |
| *** tej has quit IRC | 16:11 | |
| *** dmowrer has quit IRC | 16:23 | |
| *** dynarro has quit IRC | 16:23 | |
| *** lei-zh1 has joined #openstack-zaqar | 16:24 | |
| *** tej has joined #openstack-zaqar | 16:32 | |
| *** tej_ has joined #openstack-zaqar | 16:32 | |
| *** dmowrer has joined #openstack-zaqar | 16:34 | |
| *** tej_ has quit IRC | 16:40 | |
| *** tej has quit IRC | 16:40 | |
| *** david-lyle has joined #openstack-zaqar | 16:49 | |
| *** rcernin has quit IRC | 16:52 | |
| -openstackstatus- NOTICE: The earlier JJB bug which disrupted tox-based job configurations has been reverted and applied; jobs seem to be running successfully for the past two hours. | 16:54 | |
| *** david-lyle has quit IRC | 17:13 | |
| *** dmowrer has quit IRC | 17:17 | |
| *** dmowrer has joined #openstack-zaqar | 17:17 | |
| *** shakamunyi has quit IRC | 17:21 | |
| *** lei-zh1 has quit IRC | 17:21 | |
| *** dmowrer has quit IRC | 17:22 | |
| *** dmowrer has joined #openstack-zaqar | 17:22 | |
| *** sonuk has quit IRC | 17:23 | |
| *** dmowrer has quit IRC | 17:27 | |
| *** dmowrer has joined #openstack-zaqar | 17:27 | |
| *** dmowrer has quit IRC | 17:31 | |
| *** dmowrer has joined #openstack-zaqar | 17:32 | |
| *** tej has joined #openstack-zaqar | 17:40 | |
| *** tej_ has joined #openstack-zaqar | 17:40 | |
| *** tej has quit IRC | 17:51 | |
| *** amitgandhinz has quit IRC | 17:51 | |
| *** tej has joined #openstack-zaqar | 17:51 | |
| *** tej_ has quit IRC | 17:52 | |
| *** malini1 has quit IRC | 17:52 | |
| *** tej_ has joined #openstack-zaqar | 17:52 | |
| *** malini has joined #openstack-zaqar | 17:57 | |
| *** malini1 has joined #openstack-zaqar | 17:59 | |
| *** malini has quit IRC | 18:01 | |
| *** dmowrer has quit IRC | 18:18 | |
| *** dmowrer has joined #openstack-zaqar | 18:33 | |
| *** tej has quit IRC | 18:35 | |
| *** tej_ has quit IRC | 18:35 | |
| *** flwang1 has joined #openstack-zaqar | 18:52 | |
| *** malini1 has quit IRC | 18:52 | |
| *** amitgandhinz has joined #openstack-zaqar | 19:00 | |
| *** david-ly_ has joined #openstack-zaqar | 19:06 | |
| *** malini has joined #openstack-zaqar | 19:07 | |
| *** tej_ has joined #openstack-zaqar | 19:11 | |
| *** tej has joined #openstack-zaqar | 19:11 | |
| *** tej has quit IRC | 19:11 | |
| *** tej_ has quit IRC | 19:11 | |
| *** malini1 has joined #openstack-zaqar | 19:14 | |
| *** malini has quit IRC | 19:17 | |
| *** dmowrer has quit IRC | 19:27 | |
| *** flwang1 has quit IRC | 19:36 | |
| *** david-ly_ is now known as david-lyle | 19:39 | |
| *** dmowrer has joined #openstack-zaqar | 19:47 | |
| *** dmowrer has quit IRC | 19:51 | |
| *** shakamunyi has joined #openstack-zaqar | 20:12 | |
| *** mpanetta has quit IRC | 20:21 | |
| *** malini1 has quit IRC | 20:29 | |
| ryansb | vkmc: I didn't see that https://review.openstack.org/#/c/248354/1 had been merged | 20:37 |
| ryansb | in the future can you wait for tests, I'd meant to set that as WIP | 20:38 |
| vkmc | ryansb, sorry for that, will do the next time | 20:50 |
| ryansb | no worries - M1 was coming up fast but I really like to "force" tests | 20:50 |
| vkmc | you are totally right yeah | 20:52 |
| ryansb | I went to write the tests and was like "wait, that merged?" | 20:55 |
| vkmc | hehe | 20:58 |
| openstackgerrit | Ryan Brown proposed openstack/python-zaqarclient: Expose CLI commands for v2 https://review.openstack.org/253704 | 21:15 |
| *** boris-42_ has joined #openstack-zaqar | 21:28 | |
| *** sonuk has joined #openstack-zaqar | 21:30 | |
| *** sriram has quit IRC | 21:35 | |
| openstackgerrit | Ryan Brown proposed openstack/python-zaqarclient: Delete CLI for CheckQueueExistence which is <=v1.1 only https://review.openstack.org/253711 | 21:38 |
| openstackgerrit | Ryan Brown proposed openstack/python-zaqarclient: Delete CLI for Get/SetQueueMetadata which is <=v1.1 only https://review.openstack.org/253712 | 21:38 |
| openstackgerrit | Ryan Brown proposed openstack/python-zaqarclient: Expose CLI commands for v2 https://review.openstack.org/253704 | 21:43 |
| *** njohnston has quit IRC | 22:07 | |
| *** amitgandhinz has quit IRC | 22:10 | |
| *** david-ly_ has joined #openstack-zaqar | 22:29 | |
| *** david-ly_ is now known as david-lyle_ | 22:31 | |
| *** david-lyle has quit IRC | 22:32 | |
| *** ametts has quit IRC | 22:35 | |
| *** sonuk has quit IRC | 22:35 | |
| openstackgerrit | Ryan Brown proposed openstack/python-zaqarclient: First pass at CLI for signed URLs https://review.openstack.org/253731 | 22:59 |
| *** malini has joined #openstack-zaqar | 23:13 | |
| *** tej_ has joined #openstack-zaqar | 23:34 | |
| *** tej has joined #openstack-zaqar | 23:34 | |
| *** openstack has joined #openstack-zaqar | 23:37 | |
| *** tej has quit IRC | 23:38 | |
| *** tej_ has quit IRC | 23:38 | |
| *** malini has quit IRC | 23:41 | |
| *** tej has joined #openstack-zaqar | 23:42 | |
| *** tej_ has joined #openstack-zaqar | 23:45 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!