| *** bknudson has joined #openstack-oslo | 00:00 | |
| *** dims has joined #openstack-oslo | 00:00 | |
| lifeless | harlowja: rich comparison operators | 00:00 |
|---|---|---|
| harlowja | ah, hmmm | 00:01 |
| harlowja | possibly | 00:01 |
| *** rrepaka has quit IRC | 00:01 | |
| harlowja | did that change in 2.7 | 00:01 |
| *** sdake has quit IRC | 00:01 | |
| lifeless | harlowja: well, you're not really running 2.6 | 00:02 |
| lifeless | you're running 2.6++++ | 00:02 |
| harlowja | rhel special 2.6 | 00:02 |
| harlowja | lol | 00:02 |
| harlowja | :) | 00:02 |
| harlowja | ultra-special | 00:02 |
| harlowja | :-/ | 00:02 |
| harlowja | aka http://paste.openstack.org/show/220286/ | 00:03 |
| harlowja | or newer | 00:03 |
| harlowja | hasn't exploded that source rpm in a while, lol | 00:04 |
| *** achanda has quit IRC | 00:04 | |
| * harlowja doesn't want to look at libvirt++ | 00:12 | |
| harlowja | that one scares me | 00:12 |
| lifeless | do you have software collections in your RHEL? | 00:12 |
| harlowja | ya, | 00:13 |
| harlowja | 2.7 is there, some virtualenv work has happened to make that possible | 00:13 |
| lifeless | maybe you could use that to get 2.7 | 00:13 |
| harlowja | yup | 00:13 |
| lifeless | cause | 00:13 |
| lifeless | 2.6 ain't supported by openstack no more :) | 00:13 |
| harlowja | yup | 00:13 |
| harlowja | its happening, slowly but surely, lol | 00:14 |
| * harlowja hasn't been to involved in all that | 00:14 | |
| harlowja | can't be everywhere, ha | 00:14 |
| *** arnaud____ has joined #openstack-oslo | 00:16 | |
| *** sputnik13 has quit IRC | 00:19 | |
| *** arnaud____ has quit IRC | 00:20 | |
| lifeless | ok so | 00:23 |
| lifeless | I'm going to context switch this out | 00:23 |
| lifeless | do you think rrpaka will be back online today ? | 00:24 |
| harlowja | lifeless possibly | 00:26 |
| harlowja | def in PST office ours though | 00:26 |
| harlowja | usually always around | 00:26 |
| *** tsekiyama has quit IRC | 00:27 | |
| lifeless | k | 00:28 |
| lifeless | just this sounded panic/urgent | 00:28 |
| lifeless | I have another 4 hours work in me today, more or less | 00:28 |
| harlowja | i have way to fix it, just not ideal, force 'pbr==0.10' in setup.py files | 00:28 |
| lifeless | then tomorrow is a total mess | 00:28 |
| harlowja | and it only seems like oslo.config is doing this | 00:28 |
| harlowja | so get some sleep | 00:29 |
| harlowja | it will be ok :) | 00:29 |
| harlowja | we'll get this figured out | 00:29 |
| lifeless | if that works; one thing to be aware of is https://github.com/pypa/pip/issues/2666 | 00:29 |
| lifeless | harlowja: its midday :) - I'm waaaay ahead of you timewise ;) | 00:30 |
| harlowja | lol | 00:30 |
| lifeless | tomorrow my meetings start at 8am | 00:30 |
| harlowja | damn austrialia | 00:30 |
| lifeless | and go for hours | 00:30 |
| lifeless | harlowja: NZ :) LOL | 00:30 |
| harlowja | ya, that | 00:30 |
| lifeless | harlowja: and I have a doctors appt too | 00:30 |
| harlowja | where gandalf is | 00:30 |
| * lifeless shakes head | 00:30 | |
| harlowja | ha | 00:30 |
| harlowja | lifeless these version numbers get sucked into our rpm building via anvil (until the 2.7 stuff happens) | 00:31 |
| harlowja | so ravi ran to me when something weird going on | 00:31 |
| lifeless | yah | 00:31 |
| lifeless | so if you have pbr 0.11 installed | 00:31 |
| harlowja | and figured out how to force earlier version of pbr, but then wanted him to make sure he contacted ye pbr master | 00:31 |
| lifeless | and ==0.10 in setup.py | 00:31 |
| lifeless | setuptools will barf | 00:31 |
| harlowja | kk | 00:31 |
| lifeless | because 2666 | 00:31 |
| harlowja | i think its pulling it down automatically, didn't see pbr in the install | 00:31 |
| harlowja | *when did pip freeze | 00:31 |
| lifeless | k | 00:32 |
| harlowja | but if that happens | 00:32 |
| harlowja | i'll know :) | 00:32 |
| lifeless | so yeah using pbr 0.10 isn't the end of the world | 00:32 |
| harlowja | or get asked, ha | 00:32 |
| harlowja | ya, just to figure out --version | 00:32 |
| harlowja | then most of the rest happens in rpm land | 00:33 |
| harlowja | anyways, thx lifeless :) | 00:36 |
| openstackgerrit | James Polley proposed openstack-dev/pbr: WIP: Teach pbr to read extras from setup.cfg https://review.openstack.org/181795 | 00:36 |
| *** rrepaka has joined #openstack-oslo | 00:42 | |
| *** ozamiatin has joined #openstack-oslo | 00:42 | |
| *** tchaypo has joined #openstack-oslo | 00:42 | |
| tchaypo | lifeless: oh hai | 00:42 |
| lifeless | hai | 00:43 |
| lifeless | so | 00:43 |
| lifeless | tchaypo: some thoughts put there | 00:47 |
| lifeless | tchaypo: I realise that this is sketching territory | 00:47 |
| lifeless | tchaypo: one thing I think you should aim for in tests when doing sketch/exploratory work is clarity over DRY and pithiness | 00:49 |
| tchaypo | I agree :) It all seemed very clear to me at the time | 00:50 |
| tchaypo | re: six.moves - I could, but it’d be the first use of it here that I can see | 00:52 |
| tchaypo | existing code uses the try: else: pattern. I’d prefer to stick with that for now and do the move to six.moves seperately | 00:53 |
| *** ChanServ changes topic to "#openstack-oslo" | 00:56 | |
| -openstackstatus- NOTICE: Gerrit has been downgraded to version 2.8 due to the issues observed today. Please report further problems in #openstack-infra. | 00:56 | |
| *** rrepaka has quit IRC | 00:57 | |
| tchaypo | I’m not sure what you mean by “this setup is axiomatic”…. although I have realised I’m not actually testing the bit of code I wanted to test. good job james. | 00:58 |
| *** takedakn has joined #openstack-oslo | 00:58 | |
| lifeless | tchaypo: re clariyr | 01:02 |
| lifeless | its very narrow in your tests | 01:02 |
| lifeless | thats great for testing corner cases | 01:02 |
| lifeless | its not great for telling a story | 01:02 |
| lifeless | axiomatic - tautological might be better | 01:02 |
| lifeless | six.moves - this is new code, it has nothing to do with the existing code | 01:02 |
| lifeless | the pbr code can't use six. pbr.tests can. | 01:02 |
| *** achanda has joined #openstack-oslo | 01:03 | |
| openstackgerrit | Joshua Harlow proposed openstack/taskflow: Use hash path lookup vs path finding https://review.openstack.org/182132 | 01:05 |
| *** prad has quit IRC | 01:05 | |
| *** achanda has quit IRC | 01:08 | |
| *** mtanino has quit IRC | 01:10 | |
| *** achanda has joined #openstack-oslo | 01:11 | |
| openstackgerrit | James Polley proposed openstack-dev/pbr: WIP: Teach pbr to read extras from setup.cfg https://review.openstack.org/181795 | 01:14 |
| openstackgerrit | Joshua Harlow proposed openstack/taskflow: Use hash path lookup vs path finding https://review.openstack.org/182132 | 01:20 |
| *** takedakn1 has joined #openstack-oslo | 01:20 | |
| *** takedakn has quit IRC | 01:21 | |
| tchaypo | lifeless: I’ve updated the tests so that they’re now calling the code I wanted to test | 01:24 |
| tchaypo | I think that might address your concern - the function that’s being tested is now called a the top of each test | 01:25 |
| *** redrobot has quit IRC | 01:25 | |
| *** stevemar has joined #openstack-oslo | 01:25 | |
| tchaypo | imo that makes it a clear delinieation between the uninteresting setup and the one call that we actually want to exercise | 01:25 |
| *** redrobot has joined #openstack-oslo | 01:26 | |
| *** redrobot is now known as Guest20306 | 01:26 | |
| *** dstanek has quit IRC | 01:26 | |
| *** dstanek has joined #openstack-oslo | 01:26 | |
| lifeless | tchaypo: I haven't looked yet, need to run a chore | 01:27 |
| lifeless | tchaypo: but please put the inputs to the test within the test | 01:27 |
| lifeless | tchaypo: so its clear whats being acted on | 01:28 |
| tchaypo | kk, talk later. I think the tests are looking for the right data; they might be reformatted for clarity but I don’t think what they’re testing is going to change | 01:37 |
| tchaypo | so I’ll go ahead and start on the code to actually put the data in place | 01:38 |
| openstackgerrit | James Polley proposed openstack-dev/pbr: WIP: Teach pbr to read extras from setup.cfg https://review.openstack.org/181795 | 01:38 |
| *** browne has quit IRC | 01:39 | |
| *** zzzeek has quit IRC | 01:40 | |
| *** ashleighfarnham has joined #openstack-oslo | 01:44 | |
| openstackgerrit | Joshua Harlow proposed openstack/taskflow: Use hash path lookup vs path finding https://review.openstack.org/182132 | 01:44 |
| openstackgerrit | Joshua Harlow proposed openstack/taskflow: Use hash path lookup vs path finding https://review.openstack.org/182132 | 01:47 |
| *** takedakn1 has quit IRC | 01:47 | |
| openstackgerrit | Joshua Harlow proposed openstack/taskflow: Use hash path lookup vs path finding https://review.openstack.org/182132 | 01:49 |
| *** sigmavirus24 is now known as sigmavirus24_awa | 01:54 | |
| *** _amrith_ is now known as amrith | 01:58 | |
| *** arnaud____ has joined #openstack-oslo | 02:04 | |
| tchaypo | and I think I’m wrong. The format we came up with yesterday is lovely and clear and completely different to everything else in setup.cfg | 02:07 |
| *** achanda has quit IRC | 02:07 | |
| tchaypo | the code for parsing it is going to be completely disjoint from everything else | 02:08 |
| lifeless | tchaypo: I couldn't tell if they are looking for the right data because I can't tell what the inputs are, not easily | 02:08 |
| *** bknudson has quit IRC | 02:08 | |
| *** arnaud____ has quit IRC | 02:09 | |
| *** achanda has joined #openstack-oslo | 02:09 | |
| tchaypo | I think you couldn’t tell what the inputs are because I’d left out the call to the function I wanted to test | 02:09 |
| tchaypo | I think just adding the line where I call setup_cfg_to_setup_kwargs(self.config) would fix that | 02:10 |
| *** vilobhmm1 has quit IRC | 02:10 | |
| lifeless | ok, let me go look | 02:10 |
| *** browne has joined #openstack-oslo | 02:13 | |
| lifeless | tchaypo: ok, not convinced. I've put a few questions on the first test :) | 02:14 |
| tchaypo | just a few :) | 02:16 |
| *** amrith is now known as _amrith_ | 02:17 | |
| openstackgerrit | Merged openstack/oslotest: Updated from global requirements https://review.openstack.org/181228 | 02:19 |
| tchaypo | To step back though - I think we need to change the format | 02:19 |
| tchaypo | the parsing code we already have is very geared towards one arg in setup.cfg <-> one kwarg to setup.py | 02:20 |
| tchaypo | this thing we’ve proposed of having all the values in a section get mapped to values inside a dict inside one kwarg to setup.py is very different. It’s doable, but not inside the existing " for arg in D1_D2_SETUP_ARGS” loop in setup_cfg_to_setup_kwargs loop. | 02:21 |
| lifeless | tchaypo: that seems like a very weak reason to change the format | 02:21 |
| *** sigmavirus24_awa is now known as sigmavirus24 | 02:21 | |
| lifeless | tchaypo: sure it is | 02:21 |
| lifeless | tchaypo: by which I mean, it needs tweaks, not a wholesale rewrite | 02:22 |
| tchaypo | the existing loop looks for a (section, option) tuple which it can map to an arg. we have the arg (extras_require) and a section (extras) but there’s no single option we’re looking for - we want to iterate over all the options in that section | 02:24 |
| lifeless | I think you'll find you can write some code on line 369 that will work just dandy | 02:25 |
| lifeless | I'm not trying to be obtuse btw | 02:26 |
| lifeless | feel free to say 'wtf sketch it out' | 02:26 |
| tchaypo | no, I agree with you, we’ve both said the same thing | 02:27 |
| tchaypo | assuming you’re looking at http://git.openstack.org/cgit/openstack-dev/pbr/tree/pbr/util.py#n369 | 02:27 |
| tchaypo | we can certainly put something on line 369, but at that point it’s outside the main "for arg in D1_D2_SETUP_ARGS:” that handles every other argument | 02:28 |
| tchaypo | unless I’ve misread where that loop ends *checks* | 02:28 |
| lifeless | its outside the loop | 02:28 |
| lifeless | but you said we'd have to change the format because the function wouldn't fit it | 02:28 |
| tchaypo | well. Yes, I did. | 02:29 |
| tchaypo | What I meant was “inside the main loop" | 02:30 |
| tchaypo | This function is huge and could really use some refactoring, but I’m not going to tackle that right now. | 02:30 |
| *** yamahata has quit IRC | 02:31 | |
| lifeless | jd__: mordred: https://review.openstack.org/#/c/181785/ +2's please. | 02:43 |
| *** ashleighfarnham has quit IRC | 02:44 | |
| *** dims has quit IRC | 02:45 | |
| *** sputnik13 has joined #openstack-oslo | 02:46 | |
| openstackgerrit | Joshua Harlow proposed openstack/taskflow: Use hash path lookup vs path finding https://review.openstack.org/182132 | 03:04 |
| openstackgerrit | Merged openstack/oslo.concurrency: Updated from global requirements https://review.openstack.org/181900 | 03:14 |
| *** sputnik13 has quit IRC | 03:14 | |
| *** sputnik13 has joined #openstack-oslo | 03:16 | |
| *** dstanek has quit IRC | 03:16 | |
| *** peterstac has quit IRC | 03:16 | |
| openstackgerrit | Merged openstack/oslo.middleware: Updated from global requirements https://review.openstack.org/181223 | 03:16 |
| *** dstanek has joined #openstack-oslo | 03:19 | |
| *** peterstac has joined #openstack-oslo | 03:19 | |
| *** sputnik13 has quit IRC | 03:20 | |
| openstackgerrit | Joshua Harlow proposed openstack/tooz: No longer need kazoo lock custom retry code https://review.openstack.org/182146 | 03:24 |
| openstackgerrit | Joshua Harlow proposed openstack/tooz: No longer need kazoo lock custom retry code https://review.openstack.org/182146 | 03:25 |
| *** amotoki_ has joined #openstack-oslo | 03:30 | |
| *** stevemar has quit IRC | 03:31 | |
| *** jecarey_ has joined #openstack-oslo | 03:36 | |
| lifeless | we gots the lightning | 03:36 |
| openstackgerrit | Joshua Harlow proposed openstack/tooz: No longer need kazoo lock custom retry code https://review.openstack.org/182146 | 03:38 |
| lifeless | and the hail | 03:40 |
| *** yamahata has joined #openstack-oslo | 03:40 | |
| sigmavirus24 | lifeless: stay safe | 03:40 |
| sigmavirus24 | lifeless: in spite of size, do not use the hail to go golfing | 03:40 |
| *** ozamiatin has quit IRC | 03:41 | |
| lifeless | :) | 03:44 |
| lifeless | http://www.metservice.com/maps-radar/rain-radar/canterbury | 03:44 |
| tchaypo | I live in canterbury! | 03:45 |
| lifeless | I'm about a mouse pointer up and slightly to the west of the label christchurch | 03:45 |
| tchaypo | There’s meant to be an archbishop around here somewhere but I haven’t been able to find him | 03:45 |
| lifeless | tchaypo: you do? | 03:45 |
| *** arnaud____ has joined #openstack-oslo | 03:50 | |
| tchaypo | I do! | 03:52 |
| tchaypo | http://www.bom.gov.au/nsw/forecasts/canterbury.shtml is my local weather. no radar though | 03:52 |
| lifeless | hows the patch? | 03:52 |
| tchaypo | http://www.weatherzone.com.au/nsw/sydney/canterbury offers a tiny little radar that looks like it can click through to something else | 03:52 |
| tchaypo | oh, the patch is very sad right now. I smacked together some code and ran the tests over it and the tests tell me I did something very wrong. I had to take a break to help Andre with something, and now I’m getting lunch before I poke at it again. I think I’ve just used the wrong variable when creating the keys for the dict, so it shouldn’t be hard to | 03:54 |
| tchaypo | fix. | 03:54 |
| *** arnaud____ has quit IRC | 03:55 | |
| tchaypo | okay, I got distracted and fixed it so the tests pass. Now to get lunch and then tackle the “but what are we even testing” question | 04:04 |
| *** jamespage has quit IRC | 04:05 | |
| tchaypo | in summary, based on your questions, my thinking there is that rather than reading in the whole setup.cfg, I could pass in a simpler dict (constructed in the test) which has specific things I’m testing for | 04:05 |
| *** sigmavirus24 is now known as sigmavirus24_awa | 04:05 | |
| openstackgerrit | Joshua Harlow proposed openstack/taskflow: Remove listener stack and replace with exit stack https://review.openstack.org/182148 | 04:05 |
| openstackgerrit | Joshua Harlow proposed openstack/taskflow: Remove listener stack and replace with exit stack https://review.openstack.org/182148 | 04:06 |
| *** jamespage has joined #openstack-oslo | 04:06 | |
| lifeless | SafeConfigParser object, yes. | 04:09 |
| *** dims has joined #openstack-oslo | 04:33 | |
| *** sdake_ has quit IRC | 04:37 | |
| *** dims has quit IRC | 04:38 | |
| *** jaypipes has quit IRC | 04:40 | |
| *** stevemar has joined #openstack-oslo | 04:54 | |
| *** ashleighfarnham has joined #openstack-oslo | 04:54 | |
| *** jaypipes has joined #openstack-oslo | 04:57 | |
| *** dulek has joined #openstack-oslo | 05:08 | |
| *** nkrinner has joined #openstack-oslo | 05:20 | |
| openstackgerrit | Joshua Harlow proposed openstack/taskflow: Ensure empty paths raise a value error https://review.openstack.org/182157 | 05:33 |
| tchaypo | I don’t think I care about the SafeConfigParser object. I’m trying to test the code I just wrote, I assume SafeConfigParser is going to work correctly. | 05:35 |
| tchaypo | did I mention this method being a tad large? It’s really not possible for me to test just the code I wrote | 05:35 |
| *** arnaud____ has joined #openstack-oslo | 05:36 | |
| *** arnaud____ has quit IRC | 05:41 | |
| openstackgerrit | Joshua Harlow proposed openstack/taskflow: Use newer versions of futures that adds exception tracebacks https://review.openstack.org/179972 | 05:42 |
| *** arnaud____ has joined #openstack-oslo | 05:57 | |
| *** tapesh_ has joined #openstack-oslo | 05:57 | |
| tapesh_ | hi everyone :) | 05:57 |
| tapesh_ | i am a newbie :) plz help me make my first contribution | 05:58 |
| *** tapesh_ has left #openstack-oslo | 06:01 | |
| openstackgerrit | Joshua Harlow proposed openstack/taskflow: Executors come in via options config, not keyword arguments https://review.openstack.org/182159 | 06:01 |
| *** dstanek has quit IRC | 06:08 | |
| *** dstanek has joined #openstack-oslo | 06:09 | |
| *** sdake_ has joined #openstack-oslo | 06:38 | |
| openstackgerrit | Ethan Lynn proposed openstack/oslo.utils: Add mac validate function in netutils https://review.openstack.org/182167 | 06:39 |
| openstackgerrit | Merged openstack/taskflow: Just let the future executors handle the max workers https://review.openstack.org/164352 | 06:40 |
| *** ashleighfarnham has quit IRC | 06:40 | |
| openstackgerrit | Merged openstack/taskflow: Avoid duplicating exception message https://review.openstack.org/180815 | 06:40 |
| openstackgerrit | Merged openstack/taskflow: Avoid re-normalizing paths when following links https://review.openstack.org/181265 | 06:45 |
| openstackgerrit | Merged openstack/taskflow: Expose in memory backend split staticmethod https://review.openstack.org/181585 | 06:45 |
| *** arnaud____ has quit IRC | 06:47 | |
| ttx | dhellmann: added to agenda (as well as the oslo release process one | 06:54 |
| *** stevemar has quit IRC | 06:57 | |
| *** haypo has joined #openstack-oslo | 07:02 | |
| *** browne has quit IRC | 07:15 | |
| *** sdake_ has quit IRC | 07:23 | |
| *** exploreshaifali has joined #openstack-oslo | 07:28 | |
| *** inc0 has joined #openstack-oslo | 07:29 | |
| *** jaosorior has joined #openstack-oslo | 07:38 | |
| *** abhishekk has joined #openstack-oslo | 07:49 | |
| *** shardy_z is now known as shardy | 07:50 | |
| *** openstackstatus has quit IRC | 07:52 | |
| *** openstack has joined #openstack-oslo | 07:54 | |
| *** openstackstatus has joined #openstack-oslo | 07:54 | |
| *** ChanServ sets mode: +v openstackstatus | 07:54 | |
| *** exploreshaifali has quit IRC | 07:54 | |
| *** exploreshaifali has joined #openstack-oslo | 08:04 | |
| *** e0ne has joined #openstack-oslo | 08:17 | |
| *** e0ne is now known as e0ne_ | 08:17 | |
| *** achanda has quit IRC | 08:25 | |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/oslo.messaging: rabbit: smarter declaration of the notif. queue https://review.openstack.org/179270 | 08:27 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/oslo.messaging: rabbit: Set timeout on the underlying socket https://review.openstack.org/179357 | 08:27 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/oslo.messaging: Bump kombu and amqp requirements https://review.openstack.org/179356 | 08:27 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/oslo.messaging: TEST: enable heartbeat https://review.openstack.org/179359 | 08:27 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/oslo.messaging: rabbit: smart timeout on missing exchange https://review.openstack.org/180905 | 08:27 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/oslo.messaging: rabbit: redeclare consumers when ack/requeue fail https://review.openstack.org/180059 | 08:27 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/oslo.messaging: rabbit: remove publisher classes https://review.openstack.org/180129 | 08:27 |
| *** _amrith_ is now known as amrith | 08:38 | |
| *** e0ne_ is now known as e0ne | 08:46 | |
| *** e0ne is now known as e0ne_ | 08:46 | |
| *** e0ne_ is now known as e0ne | 08:55 | |
| *** eezhova has joined #openstack-oslo | 09:02 | |
| openstackgerrit | Oleksii Zamiatin proposed openstack/oslo.messaging: ZeroMQ driver Req/Rep implementation https://review.openstack.org/182201 | 09:08 |
| *** exploreshaifali has quit IRC | 09:23 | |
| tchaypo | lifeless: i spent the afternoon working semi-offline, but I’ve simplified the tests down to something closer to what I actually care about, I think - although perhaps I can get them even simpler | 09:45 |
| *** ihrachyshka has joined #openstack-oslo | 09:46 | |
| tchaypo | I’m not at a state where I have one test that shows that configparser parses the data in setup.cfg correctly, and another showing that the code I added to setup_cfg_to_setup_kwargs does what I think it does | 09:46 |
| lifeless | tchaypo: k | 09:46 |
| lifeless | tchaypo: thats excellent | 09:46 |
| tchaypo | once I have the output from the former the same as the input to the latter we’ll be somewhere useful :p | 09:46 |
| tchaypo | it was when I created the test to make sure configparser was reading in the data I expected that I realised the mismatch. | 09:47 |
| *** ozamiatin has joined #openstack-oslo | 09:50 | |
| *** cdent has joined #openstack-oslo | 09:55 | |
| *** ozamiatin has quit IRC | 10:09 | |
| *** e0ne is now known as e0ne_ | 10:21 | |
| *** dims has joined #openstack-oslo | 10:33 | |
| *** yamahata has quit IRC | 10:35 | |
| *** e0ne_ is now known as e0ne | 10:39 | |
| *** ozamiatin has joined #openstack-oslo | 10:56 | |
| *** links has joined #openstack-oslo | 11:04 | |
| *** amrith is now known as _amrith_ | 11:12 | |
| *** _amrith_ is now known as amrith | 11:15 | |
| *** dguitarbite has joined #openstack-oslo | 11:15 | |
| openstackgerrit | Merged openstack/stevedore: Add pypi download + version badges https://review.openstack.org/171275 | 11:28 |
| openstackgerrit | Merged openstack/oslo.messaging: Bump kombu and amqp requirements https://review.openstack.org/179356 | 11:29 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/oslo.messaging: Remove usage of contentmanager for executors https://review.openstack.org/140985 | 11:33 |
| openstackgerrit | Mehdi Abaakouk (sileht) proposed openstack/oslo.messaging: Remove usage of contentmanager for executors https://review.openstack.org/140985 | 11:38 |
| *** stpierre has joined #openstack-oslo | 11:43 | |
| *** viktors has quit IRC | 11:44 | |
| *** mfedosin has joined #openstack-oslo | 11:54 | |
| *** exploreshaifali has joined #openstack-oslo | 11:57 | |
| *** kgiusti has joined #openstack-oslo | 11:58 | |
| *** e0ne is now known as e0ne_ | 12:00 | |
| openstackgerrit | Grzegorz Grasza (xek) proposed openstack/oslo.versionedobjects: Add field name to error messages in object type checking https://review.openstack.org/168650 | 12:00 |
| *** dims_ has joined #openstack-oslo | 12:03 | |
| *** dims has quit IRC | 12:06 | |
| *** zul has quit IRC | 12:06 | |
| *** gordc has joined #openstack-oslo | 12:18 | |
| *** zul has joined #openstack-oslo | 12:18 | |
| *** dulek_ has joined #openstack-oslo | 12:27 | |
| *** dulek has quit IRC | 12:27 | |
| *** dulek_ has quit IRC | 12:28 | |
| *** dulek_ has joined #openstack-oslo | 12:28 | |
| *** e0ne_ is now known as e0ne | 12:34 | |
| *** dims_ has quit IRC | 12:40 | |
| haypo | jd__, do you really want to play with timezones? :) | 12:42 |
| haypo | jd__, (i commented https://review.openstack.org/#/c/148500/ & https://review.openstack.org/#/c/166861/) | 12:42 |
| *** dims has joined #openstack-oslo | 12:42 | |
| jd__ | haypo: yeah, I replied | 12:47 |
| *** viktors has joined #openstack-oslo | 12:48 | |
| openstackgerrit | Victor Sergeyev proposed openstack/oslo.messaging: WIP: use channel pool https://review.openstack.org/182272 | 12:49 |
| dhellmann | ttx: added to the agenda? | 12:49 |
| ttx | <dhellmann> 22:14:25> bnemec: we should encourage ttx and sdague to attend your session on optional requirements: https://etherpad.openstack.org/p/YVR-oslo-optional-dependencies | 12:51 |
| ttx | I added it to my agenda | 12:51 |
| dhellmann | ttx: ah, right | 12:51 |
| dhellmann | ETOOMUCHSCROLLBACK | 12:51 |
| ihrachyshka | dhellmann, any updates on the plan to solve juno failures due to missing juno branches for some oslo libraries? | 12:51 |
| haypo | jd__, IMO you should replace strtime(t) with t.isoformat() to reduce the risk of backward compatibility issue | 12:52 |
| haypo | jd__, oh, strtime() simply ignores the timezone, too bad | 12:52 |
| jd__ | haypo: that's not what I'm trying to do | 12:52 |
| jd__ | I'm trying to fix openstack | 12:52 |
| dhellmann | ihrachyshka: do you have a list? | 12:52 |
| jd__ | :-D | 12:52 |
| ihrachyshka | dhellmann, well, pycadf and oslo.utils for the start? | 12:53 |
| dhellmann | ihrachyshka: we're mostly waiting to sort that out at the summit | 12:53 |
| haypo | jd__, in this case, you should start by creating datetime objects with the timezone information | 12:53 |
| ihrachyshka | dhellmann, ok, so we leave it broken for time being? if that's the plan, I will just ignore those failures. I'm lazy and fine with it. :) | 12:53 |
| jd__ | haypo: what's the point since strtime() does not output the timezone? | 12:53 |
| dhellmann | ihrachyshka: ok, I'll make sure those are in my notes | 12:53 |
| haypo | jd__, isotime(datetime.datetime.now()) returns '2015-05-12T14:53:32Z', it's wrong: it's not 14:53 but 12:53 in UTC | 12:53 |
| ihrachyshka | dhellmann, btw should I now ping dims instead? :) | 12:54 |
| dhellmann | ihrachyshka: yeah, it's just another week or two, and we all want time to focus on the problem to stop playing whack-a-mole | 12:54 |
| dhellmann | ihrachyshka: yes :-) | 12:54 |
| ihrachyshka | it will take time before I become used to it! | 12:54 |
| haypo | jd__, i agree to deprecate strtime(), but i don't like isotime() (it's wrong for unaware datetime objects) | 12:54 |
| ihrachyshka | ok, I then relax on those failures | 12:54 |
| haypo | jd__, announcing a wrong datetime may be worse than omitting the timezone | 12:55 |
| haypo | oops | 12:55 |
| jd__ | haypo: we can fix isotime() | 12:55 |
| haypo | jd__, announcing a wrong timezone may be worse than omitting the timezone | 12:55 |
| jd__ | haypo: agreed let's fix isotime() then | 12:55 |
| *** pasquier-s has quit IRC | 12:55 | |
| haypo | jd__, hum, we should first fix utcnow() to use datetime.datetime.now(tz=datetime.timezone.utc) | 12:59 |
| *** zul has quit IRC | 12:59 | |
| dhellmann | ihrachyshka: are you talking about the periodic jobs failing? (I want to grab a link to a failure log) | 12:59 |
| jd__ | haypo: does that work with Py2? | 13:00 |
| haypo | jd__, nope, python2 sucks | 13:00 |
| haypo | jd__, let's drop py2 too! | 13:00 |
| jd__ | :( | 13:00 |
| jd__ | haypo: you sound as unhelpful as me sometimes | 13:01 |
| haypo | jd__, lol | 13:01 |
| jd__ | :) | 13:01 |
| ihrachyshka | dhellmann, yes. like http://logs.openstack.org/periodic-stable/periodic-glance-python27-juno/9057881/ | 13:01 |
| haypo | jd__, well, it's possible to hack a UTC timezone on python 2 | 13:01 |
| haypo | jd__, i did that in the paste | 13:01 |
| haypo | past | 13:01 |
| jd__ | haypo: tell me how :-) | 13:01 |
| *** sdake has joined #openstack-oslo | 13:01 | |
| *** dulek_ has quit IRC | 13:01 | |
| dhellmann | ihrachyshka: cool, thanks | 13:02 |
| *** dulek_ has joined #openstack-oslo | 13:02 | |
| *** kgiusti has quit IRC | 13:02 | |
| haypo | jd__, https://bitbucket.org/haypo/hachoir/src/f9c997b6fcd3abcd0b265f1307d7aea572827667/hachoir-metadata/hachoir_metadata/timezone.py?at=default#cl-3 | 13:02 |
| *** sdake_ has joined #openstack-oslo | 13:03 | |
| *** zul has joined #openstack-oslo | 13:03 | |
| jd__ | haypo: lol this is so going to break OpenStack | 13:04 |
| jd__ | this all code is completely broken and everything relies on it | 13:04 |
| haypo | jd__, why? | 13:04 |
| jd__ | haypo: because of things like TypeError: can't compare offset-naive and offset-aware datetimes | 13:04 |
| jd__ | but let's try :) | 13:04 |
| haypo | jd__, yeah, it's a problem very close to bytes versus unicode | 13:05 |
| jd__ | agreed | 13:05 |
| haypo | bytes is "unaware" of the encoding :) | 13:05 |
| jd__ | I'm working on that right now | 13:05 |
| haypo | jd__, "now" in which timezone? | 13:05 |
| jd__ | haypo: CEST | 13:06 |
| haypo | jd__, if you don't want to break the world at once, maybe we can start by only changing the behaviour on Python 3? | 13:06 |
| jd__ | haypo: brrr | 13:06 |
| jd__ | haypo: I'll write the patch and we'll discuss after | 13:06 |
| jd__ | haypo: we can also create a new library :p | 13:06 |
| jd__ | oslo.utilsnotbroken | 13:06 |
| haypo | jd__, i suggest that because python 3 has a better support for aware datetime objects (as python 3 has a better support of unicode in general) | 13:06 |
| *** sdake has quit IRC | 13:06 | |
| haypo | dhellmann will love the idea of a new library | 13:07 |
| haypo | jd__, unit tests are fun: "self.skynet_self_aware_time = datetime.datetime(1997, 8, 29, 6, 14, 0)", this object is *not* aware | 13:08 |
| jd__ | haypo: that the line I'm changing | 13:08 |
| jd__ | haypo: are you reading my screen? | 13:08 |
| haypo | only JCVD is aware | 13:08 |
| haypo | jd__, i'm probably hacking the same code | 13:08 |
| jd__ | it's pretty messy and ugly | 13:11 |
| jd__ | no timezone can mean UTC or local time zone | 13:11 |
| *** BrianShang has quit IRC | 13:14 | |
| haypo | jd__, hum, it looks like you understood the issue | 13:15 |
| jd__ | I'm smart, right? | 13:16 |
| haypo | jd__, with encoding, at least it's possible to check if a string is encoded to UTF-8 | 13:16 |
| haypo | jd__, with a timestamp, well, good luck :) it depends on the context | 13:16 |
| *** sdake_ is now known as sdake | 13:18 | |
| *** jecarey_ has quit IRC | 13:22 | |
| dhellmann | haypo, jd__ : we could just make a new function with the new behavior, and then deprecate the old one | 13:27 |
| dhellmann | ETOOMANYLIBS | 13:27 |
| jd__ | dhellmann: lol I don't know | 13:28 |
| jd__ | I'll write a patch and we'll discuss I guess | 13:28 |
| haypo | dhellmann, something like timetime.reallyisotime() ? | 13:28 |
| dhellmann | jd__: doing something that lets the app developer decide when to make a potentially breaking change gives them the flexibility and control they want | 13:28 |
| dhellmann | haypo: I'm bad at naming things, but we can probably do better than that :-) | 13:29 |
| haypo | "almostnow()" | 13:29 |
| dhellmann | haha | 13:29 |
| *** prad has joined #openstack-oslo | 13:29 | |
| *** samueldmq has joined #openstack-oslo | 13:30 | |
| dhellmann | samueldmq: hi! | 13:30 |
| samueldmq | dhellmann, hi, I'd like to talk about graduating oslo-quota :) | 13:30 |
| dhellmann | yes, we've had a couple of discussions about that in the past | 13:31 |
| dhellmann | salv-orlando brought it up for kilo, and we had a summit session in paris | 13:31 |
| dhellmann | during that discussion I think we mostly agreed that because the quota thing will have a database, and therefore database migrations, we probably want it to be a service rather than just a library | 13:31 |
| dhellmann | and there is just such a service, but it hasn't seen much development work lately | 13:32 |
| dhellmann | let me find the etherpad from that session... | 13:32 |
| dhellmann | https://etherpad.openstack.org/p/kilo-oslo-common-quota-library | 13:32 |
| *** amuller has joined #openstack-oslo | 13:32 | |
| salv-orlando | that was my understanding as well, also I and dhellmann verified that oslo-quota was not used by any project, how could we graduate that in this conditions? | 13:32 |
| dhellmann | and the current spec is https://review.openstack.org/#/c/132127/ | 13:33 |
| samueldmq | dhellmann, cool ... I would be able to work on that front for L if you want ... | 13:33 |
| salv-orlando | mind me, I do not no understand anything about oslo processes, but I expect a library to be widely adopted before graduating it | 13:33 |
| samueldmq | dhellmann, I am also working on the cross-project policy changes as well | 13:33 |
| dhellmann | salv-orlando: oh, thanks, I'd forgotten about the fact that the existing quota code isn't used | 13:33 |
| samueldmq | dhellmann, looking at those links | 13:33 |
| amuller | sileht: I have a question about oslo messaging 'cast' if you're up for it :) | 13:33 |
| sileht | amuller, sure | 13:34 |
| samueldmq | salv-orlando, ++ I agree with you | 13:34 |
| dhellmann | samueldmq: as a service, that work would be done outside of oslo, so we'll have to hook you up with the other folks interested in it | 13:34 |
| salv-orlando | samueldmq: you're more than welcome to take ownership of that specification for oslo. My impression however was that the consensus was to reject it, indeed I moved on looking at a github project called boson as a basis for a quota service. | 13:34 |
| dhellmann | salv-orlando: it depends, but in most cases with incubated code, yes, we want it to be used before graduation | 13:35 |
| amuller | sileht: call blocks until the other side finishes processing and you get a return value or an exception. Does cast wait for anything at all? Does it block until you get a confirmation from the messaging server (Assuming rabbit backend?) | 13:35 |
| amuller | Does it block until the intended recipient gets the message? (If direct or topic) | 13:35 |
| amuller | Or is it just fire and forget? | 13:35 |
| samueldmq | dhellmann, salv-orlando great .. I will take a look at that spec | 13:36 |
| samueldmq | but at a glance it looks like it is just proposing the oslo library , and not the quota service | 13:36 |
| sileht | amuller, it blocks until rabbit take the msg | 13:37 |
| amuller | sileht: ok that confirms my assumption | 13:37 |
| sileht | amuller, it does wait for reply or anything else | 13:37 |
| sileht | *doesn't* | 13:37 |
| amuller | sileht: do you think there is value in another kind of message, that waits until the other side picks up the message? (But doesn't block until it finishes processing) | 13:38 |
| amuller | does the rabbit backend even support that functionality? | 13:38 |
| amuller | sileht: and do you know what does oslo notify do in this context? | 13:38 |
| sileht | amuller, no application have got a use-case that need this | 13:39 |
| dhellmann | samueldmq: yes, the spec is for the old library idea, I was just sharing history with you | 13:39 |
| salv-orlando | samueldmq: yes, the spec proposes a library. You can look at the etherpad for the bits that were deemed "awkward" at the Paris summit concerning the approach. | 13:39 |
| samueldmq | dhellmann, great .. the first question coming is, what would be the overhead on having all the quotas in a separate service ? | 13:40 |
| amuller | sileht: What makes you determine that? I'd love that for Neutron in several places, unless I'm way off base? | 13:40 |
| sileht | amuller, sending notification is like a cast except we create we enforce default queue | 13:40 |
| samueldmq | dhellmann, I am asking because you guys possible already have thought about htis | 13:40 |
| sileht | amuller, you means retrieves all replies from all workers that got the cast ? | 13:41 |
| dhellmann | samueldmq: rest calls are more expensive than database lookups, I would think, but I don't know | 13:41 |
| amuller | sileht: let's start with direct I suppose | 13:41 |
| amuller | sileht: I'd like something like cast, but that blocks until the other side got the message, not just the messaging server | 13:41 |
| amuller | sileht: this would be used as a (more) reliabe means of notifications | 13:42 |
| amuller | reliable* | 13:42 |
| samueldmq | dhellmann, salv-orlando k got it .. I will take a deeper look and come back later | 13:42 |
| samueldmq | dstanek, salv-orlando you guys will be at the summit ? | 13:42 |
| sileht | amuller, how will you known how many replies you should wait ? | 13:42 |
| samueldmq | dstanek, not you ... I meant dhellmann :) | 13:42 |
| amuller | sileht: in a direct queue it's just 1 right | 13:42 |
| amuller | sileht: and in topic 1 as well | 13:43 |
| amuller | this wouldn't be relevant for fanout | 13:43 |
| *** links has quit IRC | 13:44 | |
| dhellmann | samueldmq: yes | 13:44 |
| samueldmq | dhellmann, great we can talk a bit more about this there, thanks | 13:45 |
| *** mriedem_away is now known as mriedem | 13:46 | |
| *** stpierre has quit IRC | 13:47 | |
| *** exploreshaifali has quit IRC | 13:49 | |
| *** browne has joined #openstack-oslo | 13:50 | |
| *** xek is now known as x3k | 13:52 | |
| dims | dansmith: looks like we have 3 reviews in flight in o.vo - can you please +W if appropriate? https://review.openstack.org/#/q/project:openstack/oslo.versionedobjects+status:open,n,z | 13:53 |
| *** e0ne is now known as e0ne_ | 13:53 | |
| sileht | amuller, I don't understand what the different with a rpc.call if you want to wait a return from the server | 13:54 |
| sileht | (even you don't care of the value returned) | 13:54 |
| *** stpierre has joined #openstack-oslo | 13:54 | |
| amuller | sileht: I want to wait until the other *receives* the message but not until it finishes processing the message | 13:54 |
| sileht | amuller, got it | 13:54 |
| dansmith | dims: two of them were co-authored by me, or already +2 by me.. you okay with me slamming those in? | 13:55 |
| *** e0ne_ is now known as e0ne | 13:55 | |
| dims | dansmith: yep | 13:55 |
| sileht | amuller, you can just do eventlet.spawn() in the application that will return directly | 13:56 |
| dansmith | okay done | 13:56 |
| dims | thanks dansmith | 13:56 |
| amuller | sileht: you mean eventlet.spawn with a call inside? | 13:56 |
| sileht | amuller, in the rpc.endpoint that receive the call | 13:57 |
| *** bknudson has joined #openstack-oslo | 13:57 | |
| sileht | amuller, nova already does this kind of thing | 13:57 |
| amuller | hmm | 13:57 |
| * amuller is thinking | 13:57 | |
| *** abhishekk has quit IRC | 13:57 | |
| *** jecarey has joined #openstack-oslo | 13:59 | |
| sileht | amuller, for the performance of replies itself, we have two idea to improve it https://bugs.launchpad.net/oslo.messaging/+bug/1437951 and https://bugs.launchpad.net/oslo.messaging/+bug/1437951 | 14:04 |
| openstack | Launchpad bug 1437951 in oslo.messaging "rpc call leads to unnecerry exchange declaration" [Wishlist,In progress] - Assigned to Mehdi Abaakouk (sileht) | 14:04 |
| sileht | oups seconds:https://review.openstack.org/#/c/181010/ | 14:05 |
| sileht | the first looks tricky to not break rolling upgrade | 14:05 |
| sileht | for the second one, we should have the enhancement soon | 14:06 |
| *** inc0 has quit IRC | 14:11 | |
| amuller | sileht: about the receipient processing in an eventlet thread. I think it would help in a lot of cases, but, while I'm far from an eventlet expert... A receipient might have dozens+ of green threads already up and running when it receives the message. If I'm not mistaken, it might take a non-trivial amount of time to return back to the code that returns the RPC call | 14:13 |
| sileht | amuller, oslo.messaging limits the number of messages handled in parallel with rpc_thread_pool_size | 14:15 |
| sileht | amuller, if you use eventlet.spawn, that will return a oslo.msg threads in the pool | 14:15 |
| *** zzzeek has joined #openstack-oslo | 14:15 | |
| dhellmann | ihrachyshka: can you lend me a hand with getting the neutron team to prioritize work on the namespace patch in the client lib? https://review.openstack.org/#/c/180700/ I just asked in #openstack-neutron, too, fwiw | 14:15 |
| amuller | sileht: I'm thinking of a server, it would have green threads running unrelated to oslo messaging | 14:16 |
| jd__ | haypo: did you just start an entire new thread to troll on the same mysql subject lol? | 14:16 |
| haypo | jd__, zigo suggested me to open a new thread | 14:17 |
| *** zul has quit IRC | 14:17 | |
| haypo | jd__, the previous one ends with discussing rust and go, i don't understand how it's related to python 3 | 14:17 |
| sileht | amuller, yes I don't see a problem | 14:17 |
| jd__ | haypo + zigo = #trollteam ;) | 14:17 |
| ihrachyshka | dhellmann, ack. we may need a hacking check similar to what is in neutron | 14:17 |
| *** stevemar has joined #openstack-oslo | 14:17 | |
| amuller | sileht: if the server has tons of green threads going, it might take a while to get back to the code that calls eventlet.spawn when it receives the RPC call | 14:17 |
| jd__ | haypo: but I admit the last thread kinda died weirdly | 14:18 |
| sileht | amuller, you perhaps need to have a greenpool in your app | 14:18 |
| haypo | jd__, the discussion doesn't produce anything concrete | 14:18 |
| jd__ | haypo: that's the point of most discussions you know | 14:18 |
| sileht | amuller, to define how many times you want this processing in parallel | 14:18 |
| jd__ | that's why I write more patches than email :) | 14:18 |
| haypo | jd__, except that monty voted -1 on my patch because of it :-p | 14:18 |
| dhellmann | ihrachyshka: if you like. I would prefer to just move ahead with deleting things, but that's up to the neutron team | 14:18 |
| sileht | "eventlet.spawn_n" returns imediality and the greenthread in scheduled later | 14:19 |
| sileht | amuller, ^ | 14:19 |
| amuller | sileht: ok | 14:19 |
| amuller | sileht: I can probably work with that | 14:19 |
| sileht | amuller, note the '_n' | 14:19 |
| amuller | sileht: yeah | 14:19 |
| *** zul has joined #openstack-oslo | 14:19 | |
| ihrachyshka | dhellmann, W+1 | 14:19 |
| amuller | sileht: It's a useful pattern, do you think it belongs in oslo messaging, or in the different apps? | 14:19 |
| amuller | sileht: in neutron I could just define a decorator for RPC endpoints that processes the RPC message in eventlet.spawn_n | 14:20 |
| sileht | amuller, for example, nova do some processing sync, then spawn a thread and reply | 14:21 |
| sileht | amuller, but oslo.messaging can provide the helper to do that | 14:21 |
| amuller | sileht: you can look at this from the recipient's point of view or the sender | 14:23 |
| amuller | so for example oslo messaging (Or the different apps) could define a decorator that you'd put around an RPC endpoint, and it'd just call eventlet.spawn_n(f, args) | 14:24 |
| amuller | sileht: or, if you see it from the client's point of view, it'd be a flag to call, or a new type of message | 14:24 |
| sileht | amuller, yeah all works | 14:25 |
| *** sigmavirus24_awa is now known as sigmavirus24 | 14:25 | |
| *** dulek_ has quit IRC | 14:25 | |
| sileht | but I think that the application should create a thread pool to do that, to not exhaute resources | 14:26 |
| sileht | amuller, if oslo.messaging do that we have to implement that for all executors | 14:27 |
| amuller | sileht: yeah I see your point | 14:27 |
| sileht | for me how the application want to handle sync and async thing is its business, oslo.messaging just dipatcher message in a sync way with the select executor | 14:28 |
| sileht | amuller, I still wonder from the caller POV, why is it need to have a special handling between 'the message haven't been delivred' and 'it have been delivred but the processing failed and I don't care to known how/why' | 14:30 |
| amuller | sileht: I just see the distinction between getting an ACK from the messaging server (cast) and getting an ACK from the intended recipient | 14:30 |
| amuller | it's one step more reliable | 14:31 |
| sileht | yeah but why is usefull if the caller doesn't care if the processing succeed | 14:31 |
| amuller | so you're saying why not use cast in these cases | 14:32 |
| sileht | yes :p | 14:32 |
| dhellmann | dims, gordc: can we do an oslo.middleware release today or tomorrow? I need to get a version that doesn't use the oslo namespace package when importing other libs | 14:34 |
| dims | dhellmann: i am good pulling trigger, gordc, can you confirm it's ok? | 14:35 |
| dhellmann | dims: I can run the script, if you both confirm it's ok to do so | 14:35 |
| amuller | sileht: the one distinction is if the recipient is not up at all | 14:36 |
| amuller | sileht: with cast the client wouldn't know | 14:36 |
| amuller | sileht: so the client can handle that type of error | 14:37 |
| sileht | amuller, ok so you care of the return :) | 14:37 |
| dims | dhellmann: +2W from me :) | 14:37 |
| sileht | amuller, clearly rpc.call + spawn_n, should fit your use-case | 14:37 |
| amuller | sileht: I don't think it's ideal but I can work with it | 14:37 |
| *** Guest20306 is now known as redrobot | 14:39 | |
| dhellmann | dims: ok, I'll track down gordc | 14:39 |
| jd__ | dhellmann: can we remove old tests directory quickly? Just lost 30 mins not understanding why my tests change were not passsing locally because I was modifying only one of the file /me crying | 14:40 |
| dhellmann | jd__: https://review.openstack.org/#/q/topic:bp/remove-namespace-packages,n,z | 14:40 |
| dhellmann | jd__: I have wip patches up to remove all of that | 14:41 |
| jd__ | dhellmann: thanks! | 14:41 |
| dhellmann | I have about 24 libraries that need patches landed and new releases | 14:41 |
| sileht | amuller, we can try to add the feature to oslo.msg, but I need to dig how | 14:41 |
| dhellmann | then we can make progress on the patches in our libs and the applications | 14:41 |
| dhellmann | jd__: you could look at the patches related to ceilometer, ceilometermiddleware, and ceilometerclient :-) | 14:42 |
| sileht | amuller, I'm thinking to something like 'a post dispatcher callback' or 'a post reply callback' | 14:42 |
| openstackgerrit | Merged openstack/oslo.versionedobjects: Add enum fieldtype field https://review.openstack.org/168649 | 14:42 |
| openstackgerrit | Merged openstack/oslo.versionedobjects: Add field name to error messages in object type checking https://review.openstack.org/168650 | 14:42 |
| amuller | sileht: ack | 14:42 |
| sileht | amuller, that could be a good idea, that will still hold the oslo.msg thread from the pool, and avoid resource exhauted | 14:43 |
| sileht | amuller, application can do some processing before the reply and some other after | 14:44 |
| amuller | sileht: sounds good | 14:44 |
| *** mtanino has joined #openstack-oslo | 14:45 | |
| dhellmann | dims: I thought https://review.openstack.org/#/c/182090/1 had landed, but hasn't, so I need that approved before we release | 14:45 |
| sileht | that should be just some changes into the rpc dispatcher | 14:45 |
| dhellmann | ETOOMANYPATCHES | 14:45 |
| sileht | dhellmann, done | 14:46 |
| dhellmann | sileht: thanks! | 14:46 |
| *** ashleighfarnham has joined #openstack-oslo | 14:46 | |
| *** sdake_ has joined #openstack-oslo | 14:49 | |
| *** e0ne is now known as e0ne_ | 14:51 | |
| *** achanda has joined #openstack-oslo | 14:52 | |
| *** sdake has quit IRC | 14:52 | |
| dims | sileht: thanks! | 14:53 |
| *** jaypipes has quit IRC | 14:54 | |
| openstackgerrit | Victor Sergeyev proposed openstack/oslo.messaging: WIP: use channel pool https://review.openstack.org/182272 | 14:55 |
| *** nkrinner has quit IRC | 14:55 | |
| *** e0ne_ is now known as e0ne | 14:58 | |
| *** browne has quit IRC | 15:00 | |
| *** achanda has quit IRC | 15:01 | |
| *** alexpilotti has joined #openstack-oslo | 15:01 | |
| *** sputnik13 has joined #openstack-oslo | 15:03 | |
| *** zz_jgrimm is now known as jgrimm | 15:06 | |
| *** ozamiatin has quit IRC | 15:11 | |
| *** cdent has quit IRC | 15:12 | |
| *** mfedosin has quit IRC | 15:16 | |
| *** shakamunyi has joined #openstack-oslo | 15:22 | |
| *** barra204 has joined #openstack-oslo | 15:22 | |
| ihrachyshka | dhellmann, https://review.openstack.org/#/c/176333/ neutron would like to see it | 15:24 |
| dhellmann | ihrachyshka: are names unique? I'm not sure we want the default to be to log non-unique values | 15:29 |
| ihrachyshka | dhellmann, as per http://docs.openstack.org/cli-reference/content/keystoneclient_commands.html , yes | 15:31 |
| *** dguitarbite has quit IRC | 15:31 | |
| ihrachyshka | see user-create and tenant-create | 15:31 |
| dhellmann | ihrachyshka: ok. it's also a backwards-incompatible change for anyone currently parsing the logs we generate, so we need to think about that | 15:32 |
| *** sputnik13 has quit IRC | 15:32 | |
| dhellmann | ihrachyshka: I'm not opposed to the change, per se, but we need to take care. I'll leave some comments on the review | 15:32 |
| *** sputnik1_ has joined #openstack-oslo | 15:32 | |
| *** yamahata has joined #openstack-oslo | 15:34 | |
| *** yamahata has quit IRC | 15:36 | |
| *** yamahata has joined #openstack-oslo | 15:37 | |
| openstackgerrit | Joshua Harlow proposed openstack/tooz: Use a lua script(s) instead of transactions (WIP) https://review.openstack.org/180800 | 15:40 |
| *** ccrouch has quit IRC | 15:42 | |
| openstackgerrit | Ihar Hrachyshka proposed openstack/oslo.policy: Expose base check classes as part of public API https://review.openstack.org/176683 | 15:45 |
| *** jaypipes has joined #openstack-oslo | 15:45 | |
| *** links has joined #openstack-oslo | 15:50 | |
| *** arnaud____ has joined #openstack-oslo | 15:51 | |
| *** haypo has quit IRC | 15:52 | |
| *** mtreinish has quit IRC | 15:58 | |
| *** mtreinish has joined #openstack-oslo | 16:00 | |
| *** kgiusti has joined #openstack-oslo | 16:05 | |
| *** ozamiatin has joined #openstack-oslo | 16:08 | |
| openstackgerrit | Merged openstack/oslo.middleware: Drop use of 'oslo' namespace package https://review.openstack.org/182090 | 16:08 |
| *** exploreshaifali has joined #openstack-oslo | 16:11 | |
| *** openstackgerrit_ has quit IRC | 16:11 | |
| bknudson | does anyone know off the top of their head what changed to fix the doubling-up of [cors] section in keystone.conf? | 16:16 |
| *** ozamiatin has quit IRC | 16:28 | |
| *** sputnik13 has joined #openstack-oslo | 16:33 | |
| *** browne has joined #openstack-oslo | 16:33 | |
| *** yamahata has quit IRC | 16:35 | |
| dhellmann | dims, bnemec, jd__ : do any of you know about "hidden" configuration options? I'm not seeing anything in oslo.config about hiding option definitions | 16:36 |
| *** sputnik1_ has quit IRC | 16:36 | |
| *** ashleighfarnham has left #openstack-oslo | 16:36 | |
| bnemec | dhellmann: I'm not aware of anything. What would be the purpose? | 16:37 |
| sileht | dhellmann, perhaps oslo.config ConfigFilter ? | 16:37 |
| dhellmann | sileht: the filter doesn't hide them from deployers | 16:37 |
| dhellmann | sileht: I'm reviewing https://review.openstack.org/#/c/181010/3/specs/liberty/oslo.messaging-remove-double-reply.rst,cm | 16:37 |
| dhellmann | bnemec: ^^ | 16:37 |
| sileht | dhellmann, oh, the bp need to be corrected so | 16:38 |
| *** arnaud____ has quit IRC | 16:44 | |
| *** amuller has left #openstack-oslo | 16:52 | |
| openstackgerrit | Elena Ezhova proposed openstack/oslo-incubator: Add documentation for launcher classes in service.py https://review.openstack.org/182386 | 16:56 |
| openstackgerrit | Joshua Harlow proposed openstack/taskflow: Use hash path lookup vs path finding https://review.openstack.org/182132 | 16:59 |
| *** mtreinish has quit IRC | 17:02 | |
| *** tteggel has quit IRC | 17:02 | |
| *** AAzza has quit IRC | 17:02 | |
| *** mtreinish_ has joined #openstack-oslo | 17:02 | |
| *** ashleighfarnham has joined #openstack-oslo | 17:02 | |
| *** mtreinish_ is now known as mtreinish | 17:02 | |
| *** yamahata has joined #openstack-oslo | 17:03 | |
| *** tteggel has joined #openstack-oslo | 17:03 | |
| *** flaper87 has quit IRC | 17:03 | |
| *** flaper87 has joined #openstack-oslo | 17:05 | |
| *** flaper87 has quit IRC | 17:06 | |
| *** flaper87 has joined #openstack-oslo | 17:06 | |
| *** AAzza_afk has joined #openstack-oslo | 17:06 | |
| *** AAzza_afk is now known as AAzza | 17:06 | |
| *** sdake has joined #openstack-oslo | 17:11 | |
| *** links has quit IRC | 17:14 | |
| *** sdake_ has quit IRC | 17:14 | |
| *** harlowja has quit IRC | 17:22 | |
| *** harlowja has joined #openstack-oslo | 17:22 | |
| *** achanda has joined #openstack-oslo | 17:30 | |
| *** exploreshaifali has quit IRC | 17:41 | |
| harlowja | dims ^ that might interest u also | 17:42 |
| *** vilobhmm11 has joined #openstack-oslo | 17:46 | |
| *** e0ne is now known as e0ne_ | 17:47 | |
| *** ihrachyshka has quit IRC | 17:51 | |
| *** amuller has joined #openstack-oslo | 17:51 | |
| *** e0ne_ is now known as e0ne | 17:56 | |
| harlowja | damn, seems like my links never showed up | 17:56 |
| harlowja | lol | 17:56 |
| harlowja | reposting | 17:57 |
| harlowja | samueldmq poke me on quota stuff to | 17:57 |
| harlowja | i've been working on https://etherpad.openstack.org/p/fix-the-quota-please which hopefully has some of the history /archalogy and ideas and ... | 17:57 |
| harlowja | salv-orlando dims ^ | 17:57 |
| harlowja | ok, hopefully thats not broken this time (for some reason not showing up in eavesdrop) | 17:57 |
| dims | harlowja: ack. wrapping something up | 17:57 |
| harlowja | thx | 17:57 |
| harlowja | maybe eavesdrop broken, idk | 17:57 |
| harlowja | haha | 17:57 |
| harlowja | anyway, double posted, oh ya | 17:58 |
| samueldmq | harlowja, hi | 17:58 |
| harlowja | hi there | 17:58 |
| harlowja | https://etherpad.openstack.org/p/fix-the-quota-please might interest u | 17:58 |
| harlowja | been working on that semi-internally, gathering some kind of basis for something to be fixed | 17:58 |
| samueldmq | dhellmann, salv-orlando so the biggest cons against quota as a separate service ... | 17:58 |
| samueldmq | a service needs to get involved for every action | 17:58 |
| samueldmq | so another single point of failure/bottleneck | 17:58 |
| samueldmq | .... that's the point people don't like I think | 17:59 |
| samueldmq | harlowja, yeah I will take a look | 17:59 |
| harlowja | cool | 17:59 |
| harlowja | samueldmq i would agree, this has to be done carefully :) | 17:59 |
| *** e0ne has quit IRC | 17:59 | |
| harlowja | and without trying to take over other projects stuff (to much) | 17:59 |
| harlowja | at least not initially | 17:59 |
| samueldmq | harlowja, oh, your first point on that pad says that, great | 17:59 |
| harlowja | ;) | 17:59 |
| openstackgerrit | garyk proposed openstack/oslo.vmware: Add support for dynamicProperty https://review.openstack.org/182398 | 17:59 |
| *** mfedosin has joined #openstack-oslo | 17:59 | |
| harlowja | samueldmq hopefully some other useful stuff on that etherpad also | 18:00 |
| samueldmq | harlowja, great ... my team is developing the hierarchical multitenancy stuff in keystone | 18:00 |
| harlowja | ya, i and others have tried to capture something that hopefully will succed this time :) | 18:01 |
| samueldmq | harlowja, also we are grabbing some work on the hierarchical quotas implementation | 18:01 |
| samueldmq | harlowja, so it will be great to come up with all this together | 18:01 |
| harlowja | ya, there is some info about that there to | 18:01 |
| harlowja | agreed | 18:01 |
| harlowja | vilobhmm11 and others are involved from y! and pretty kin on getting something to happen | 18:01 |
| *** ericksonsantos has joined #openstack-oslo | 18:01 | |
| harlowja | *and me* | 18:01 |
| samueldmq | ericksonsantos, cc ^ read up :) | 18:02 |
| harlowja | feel free to tweak that etherpad with any ideas u have | 18:02 |
| samueldmq | ericksonsantos is working on the hierarchical driver on nova side | 18:02 |
| samueldmq | harlowja, k thanks | 18:02 |
| harlowja | cool | 18:02 |
| harlowja | np, let's get something to happen | 18:03 |
| harlowja | the 20th time around is the one that makes it right, lol | 18:03 |
| harlowja | :) | 18:03 |
| harlowja | samueldmq as soon as we have some decent etherpad, either at the summit we need to figure out the strategy to make something happen | 18:05 |
| harlowja | and ensure we are very very much in touch with the various projects | 18:05 |
| harlowja | ^ so that this plan doesn't fail again .... | 18:05 |
| samueldmq | harlowja, great ! I'll be at the summit, we can talk more about all this there :) | 18:07 |
| harlowja | cools | 18:07 |
| samueldmq | harlowja, however I still don't have a great knowledge on the quota system :p | 18:07 |
| samueldmq | harlowja, your initial plan is to get soemthing similar to oslo.quota right ? | 18:08 |
| samueldmq | harlowja, since each service would still own its resources and use that common library | 18:08 |
| *** tsekiyama has joined #openstack-oslo | 18:10 | |
| harlowja | samueldmq did u see my braindump @ bottom of https://review.openstack.org/#/c/132127/ :-P | 18:10 |
| harlowja | samueldmq ya, we need to be careful here, i'm not 100% on the right approach yet either | 18:10 |
| harlowja | something simple to make starting integration easy | 18:10 |
| ericksonsantos | harlowja, nice to know that you're working with quota management. :) | 18:11 |
| harlowja | vs a service, which seems harder to start | 18:11 |
| harlowja | ericksonsantos trying to wrangle it a little :-P | 18:11 |
| harlowja | its failed to many times, and its a pain point for yahoo | 18:11 |
| harlowja | *failed getting centralized/addressed/other alot | 18:12 |
| harlowja | *well like 4+ times i think, lol | 18:12 |
| harlowja | with 'Known hacks:' (other that etherpad) also make me sad | 18:12 |
| harlowja | *on that etherpad | 18:12 |
| samueldmq | harlowja, nice .. I think a library containing only the duplicated enforcement code is the first step | 18:13 |
| ericksonsantos | harlowja, I see. I'll read the etherpad. | 18:13 |
| samueldmq | for sure, since we will be changing nothing than putting common code together (oh I said common, it's oslo :)) | 18:13 |
| harlowja | cool, let's see how this goes | 18:13 |
| samueldmq | k | 18:14 |
| harlowja | samueldmq well we do need to think how to handle schema, eventually | 18:14 |
| harlowja | if we want to go there | 18:14 |
| harlowja | but thats imho sorta far off, there's enough work to get to that stage to keep people busy for a while | 18:14 |
| harlowja | *and u have to imho have a better foundation before that's even a concern... | 18:15 |
| harlowja | don't try to fly before u can walk... | 18:15 |
| *** amotoki_ has quit IRC | 18:15 | |
| samueldmq | harlowja, hehe yep :p I neeed to deeply understand how different services implement their quota management | 18:15 |
| harlowja | ya | 18:15 |
| harlowja | we seem to have some of the people that know this (not neutron though, anyone know that) | 18:16 |
| samueldmq | and see what's common/being duplicated | 18:16 |
| harlowja | vilobhmm11 i think understands the cinder quota, some nova folks | 18:16 |
| harlowja | documenting and understanding that as much as can be, would be a great first step (to writeup and understand that) | 18:16 |
| harlowja | i have some level of knowledge of nova, but not alot | 18:16 |
| harlowja | dims u da man for that right :-P | 18:17 |
| samueldmq | harlowja, great, do you have another etherpad ? (in which you write info on how quota works for different services) | 18:17 |
| harlowja | don't | 18:17 |
| harlowja | maybe a new section on that one? | 18:17 |
| harlowja | or a linked one? | 18:17 |
| dims | harlowja: i can help find things, yes. nova is vast :) | 18:17 |
| harlowja | ;) | 18:18 |
| harlowja | captain dims sailing the nova sea | 18:18 |
| harlowja | lol | 18:18 |
| samueldmq | harlowja, we can put in a section, and if we see it gets huge, we split and link to another | 18:18 |
| samueldmq | harlowja, we change as we go :) | 18:18 |
| harlowja | samueldmq on that etherpad there is some info scattered about how its done, maybe move that around also? | 18:18 |
| harlowja | some links and such, probably should/could reorganize that | 18:18 |
| harlowja | under 'See:' (at top) | 18:19 |
| samueldmq | yeah, I will be taking a deeper look on that with ericksonsantos :) | 18:19 |
| harlowja | kk | 18:19 |
| harlowja | sounds good | 18:19 |
| ericksonsantos | sure | 18:19 |
| ericksonsantos | :) | 18:19 |
| dims | :) | 18:19 |
| vilobhmm11 | harlowja, dims : started working on https://review.openstack.org/#/c/173141/ ..will update finding on cinder-quota as and when i find more here https://etherpad.openstack.org/p/fix-the-quota-please…more than happy to help | 18:20 |
| harlowja | cool | 18:20 |
| harlowja | thx vilobhmm11 | 18:20 |
| dims | vilobhmm11: nice | 18:21 |
| *** cdent has joined #openstack-oslo | 18:29 | |
| *** exploreshaifali has joined #openstack-oslo | 18:29 | |
| lifeless | mordred: where's my +2+A? https://review.openstack.org/#/c/181785/ | 18:32 |
| lifeless | mordred: I'd like that in 1.0.0 if we can (to salve the rpm heads) | 18:32 |
| lifeless | harlowja: hows rrepete?s issue | 18:32 |
| harlowja | let me find him | 18:33 |
| mordred | lifeless: done | 18:33 |
| harlowja | lifeless poked him on YIM (will see if he gets on) | 18:34 |
| harlowja | and then he can tell u, ha | 18:34 |
| harlowja | :-P | 18:34 |
| harlowja | he's our main CI guy, so might be busy though, let's see | 18:35 |
| lifeless | harlowja: kk | 18:36 |
| lifeless | harlowja: no skin of my nose :) | 18:36 |
| harlowja | :) | 18:36 |
| openstackgerrit | Merged openstack/oslo.vmware: Remove run_cross_tests.sh https://review.openstack.org/180282 | 18:41 |
| openstackgerrit | Merged openstack/oslosphinx: Advertise support for Python3.4 / Remove support for Python 3.3 https://review.openstack.org/181527 | 18:42 |
| openstackgerrit | Merged openstack/oslotest: Advertise support for Python3.4 / Remove support for Python 3.3 https://review.openstack.org/181698 | 18:42 |
| *** rrepaka has joined #openstack-oslo | 18:43 | |
| rrepaka | lifeless: Hi | 18:44 |
| harlowja | rrepaka has arrived! | 18:44 |
| harlowja | :) | 18:44 |
| rrepaka | lifeless: Yesterday i tried the patch http://paste.ubuntu.com/11086963/ | 18:44 |
| lifeless | rrepaka: and what did it output? | 18:46 |
| rrepaka | lifeless: i was unable to see the log output | 18:46 |
| rrepaka | lifeless: How to i configure the log file for pbr ? | 18:46 |
| lifeless | rrepaka: the stdout is the log | 18:47 |
| lifeless | rrepaka: how did you test it? | 18:47 |
| rrepaka | lifeless: I checked out the pbr code from git hub, applied your patch | 18:48 |
| rrepaka | lifeless: did "python setup.py install" in pbr module | 18:48 |
| lifeless | rrepaka: then went to the olso.config dir and checked the version? | 18:49 |
| rrepaka | lifeless: after that i went to oslo-config modle and ran "python setup.py --version | 18:49 |
| lifeless | rrepaka: ok, try http://paste.ubuntu.com/11100035/ instead | 18:49 |
| lifeless | that should drop you into a debugger | 18:49 |
| lifeless | have you used pdb before? | 18:49 |
| rrepaka | lifeless: sure | 18:49 |
| harlowja | dhellmann added comment on https://review.openstack.org/#/c/181688/ for why not removing just yet, soon is the answer to that one :) | 18:51 |
| *** e0ne has joined #openstack-oslo | 18:51 | |
| harlowja | soon like probably in next few days (hopefully) | 18:51 |
| *** Rockyg has joined #openstack-oslo | 18:53 | |
| lifeless | rrepaka: I put the breakpoint right before the routine used to calculate versions | 18:54 |
| lifeless | rrepaka: you'll want to print out the parameters | 18:54 |
| lifeless | and then step in | 18:54 |
| lifeless | and at each call within that routine print the function inputs and outputs | 18:55 |
| lifeless | e.g. the first one is | 18:55 |
| lifeless | print tag, distance, git_dir | 18:55 |
| lifeless | you'll drop into pdb when installing pbr too, just continue out of it there | 18:55 |
| lifeless | since its oslo.config we care about | 18:55 |
| rrepaka | lifeless: no i didnt use pdb | 18:57 |
| lifeless | rrepaka: ok, quick primer | 18:57 |
| rrepaka | lifeless: I now have latest pbr code patched | 18:58 |
| lifeless | c<ENTER> to continue | 18:58 |
| lifeless | print to print stuff, just like in python | 18:58 |
| lifeless | s<enter> to step into a function | 18:58 |
| lifeless | n<enter> to step over it | 18:58 |
| rrepaka | lifeless: how to i configure the python to pick the code from source | 18:58 |
| lifeless | rrepaka: I'd use a virtualenv for testing | 18:58 |
| lifeless | inside the virtualenv do | 18:59 |
| lifeless | pip install -U pip setuptools | 18:59 |
| lifeless | then pip install -e <path-to-pbr> | 18:59 |
| rrepaka | rrepaka: ok | 18:59 |
| harlowja | no talking to yourself | 18:59 |
| rrepaka | lifeless: ok got it let me do pip install -e first | 19:00 |
| *** exploreshaifali has quit IRC | 19:01 | |
| rrepaka | lifeless: i am able to start the debug run, i see the following "/home/rrepaka/community/pbr/pbr/packaging.py(582)_get_version_from_git()->'1.4.0.0a5.dev5'" | 19:02 |
| lifeless | rrepaka: ok so that indicates you're at the end of the function we want to dig through | 19:02 |
| lifeless | rrepaka: continue then run it again | 19:03 |
| rrepaka | lifeless: i will run and send you pastebin output | 19:03 |
| lifeless | this time step into the _get_version_from_git_target call | 19:03 |
| rrepaka | lifeless: sure doing it | 19:04 |
| openstackgerrit | Joshua Harlow proposed openstack/taskflow: Remove hack + fix for issue no longer needed https://review.openstack.org/182416 | 19:07 |
| sdague | so... question - https://github.com/openstack/oslo.messaging/blob/master/oslo_messaging/notify/notifier.py#L188 | 19:07 |
| sdague | is audit in the notification path supposed to be deprecated now in the same way as for logging? | 19:07 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/stevedore: Fix Python versions supported https://review.openstack.org/181524 | 19:08 |
| rrepaka | lifeless: http://paste.ubuntu.com/11100215/ | 19:08 |
| rrepaka | lifeless:http://paste.ubuntu.com/11100333/ | 19:14 |
| *** ccrouch has joined #openstack-oslo | 19:15 | |
| *** openstackgerrit_ has joined #openstack-oslo | 19:16 | |
| openstackgerrit | Joshua Harlow proposed openstack/taskflow: Remove custom py26/py27 and py33 environment + fix requirements https://review.openstack.org/182423 | 19:24 |
| harlowja | dhellmann ^ should be that final thing that goes in | 19:24 |
| *** openstackgerrit_ has quit IRC | 19:25 | |
| openstackgerrit | Sean McGinnis proposed openstack/oslo-incubator: Clean up logging to conform to guidelines https://review.openstack.org/182424 | 19:28 |
| *** jaosorior has quit IRC | 19:32 | |
| rrepaka | lifeless: more depth of debug run http://paste.ubuntu.com/11100657/ | 19:32 |
| *** achanda has quit IRC | 19:33 | |
| *** shakamunyi has quit IRC | 19:35 | |
| dhellmann | harlowja: +2 | 19:35 |
| harlowja | thx | 19:35 |
| *** barra204 has quit IRC | 19:36 | |
| harlowja | let me just merge https://review.openstack.org/#/c/181688/ into that | 19:37 |
| harlowja | and all will be merry | 19:37 |
| openstackgerrit | Doug Hellmann proposed openstack/oslo-incubator: Remove versionutils https://review.openstack.org/182428 | 19:37 |
| openstackgerrit | Joshua Harlow proposed openstack/taskflow: Remove custom py26/py27 and py33 environment + fix requirements https://review.openstack.org/182423 | 19:40 |
| harlowja | ok dokie, all now merry | 19:41 |
| openstackgerrit | Merged openstack/oslo.context: Remove support for Python 3.3 https://review.openstack.org/181764 | 19:43 |
| openstackgerrit | Merged openstack/oslo.db: Remove support for Python 3.3 https://review.openstack.org/181765 | 19:43 |
| openstackgerrit | Merged openstack/oslo.log: Advertise support for Python3.4 / Remove support for Python 3.3 https://review.openstack.org/181535 | 19:43 |
| openstackgerrit | Merged openstack/oslo.messaging: Remove support for Python 3.3 https://review.openstack.org/181766 | 19:43 |
| openstackgerrit | Merged openstack/oslo.policy: Remove support for Python 3.3 https://review.openstack.org/181767 | 19:44 |
| openstackgerrit | Merged openstack/oslo.serialization: Advertise support for Python3.4 / Remove support for Python 3.3 https://review.openstack.org/181532 | 19:44 |
| openstackgerrit | Merged openstack/oslo.utils: Advertise support for Python3.4 / Remove support for Python 3.3 https://review.openstack.org/181531 | 19:44 |
| openstackgerrit | Merged openstack/oslo.vmware: Remove support for Python 3.3 https://review.openstack.org/181529 | 19:44 |
| openstackgerrit | Merged openstack/taskflow: Remove tox py33 environment no longer used https://review.openstack.org/181686 | 19:44 |
| harlowja | death to py3.3 | 19:45 |
| harlowja | may u rest in peace | 19:45 |
| *** shakamunyi has joined #openstack-oslo | 19:49 | |
| *** barra204 has joined #openstack-oslo | 19:49 | |
| openstackgerrit | Doug Hellmann proposed openstack/oslo.log: Remove is_compatible from versionutils https://review.openstack.org/170471 | 19:50 |
| openstackgerrit | Doug Hellmann proposed openstack/oslo.log: Add versionutils options to list_opts https://review.openstack.org/167769 | 19:50 |
| openstackgerrit | Doug Hellmann proposed openstack/oslo.log: Add versionutils to API documentation https://review.openstack.org/167768 | 19:50 |
| lifeless | rrepaka: thanks, looking | 19:55 |
| openstackgerrit | Merged openstack/debtcollector: Ensure that the incoming 'new_class' is actually a class https://review.openstack.org/172613 | 19:57 |
| openstackgerrit | Merged openstack/oslo.config: Fix logging of deprecated opts with dest override https://review.openstack.org/169392 | 19:57 |
| openstackgerrit | Merged openstack/oslo.rootwrap: Generate a oslo-rootwrap console script https://review.openstack.org/174349 | 19:57 |
| openstackgerrit | Merged openstack/oslo.config: Allow loading custom config files into Config Fixture. https://review.openstack.org/172566 | 20:02 |
| *** achanda has joined #openstack-oslo | 20:04 | |
| *** ashleighfarnham has quit IRC | 20:11 | |
| openstackgerrit | Merged openstack/oslo-incubator: Remove timeutils.strtime() usage https://review.openstack.org/165340 | 20:12 |
| openstackgerrit | Merged openstack/oslo-incubator: Add gerrit dashboard files https://review.openstack.org/167272 | 20:12 |
| openstackgerrit | Merged openstack/oslosphinx: Remove the unneeded horizontal scrollbar https://review.openstack.org/172073 | 20:12 |
| *** flaper87 has quit IRC | 20:20 | |
| *** flaper87 has joined #openstack-oslo | 20:20 | |
| *** achanda has quit IRC | 20:21 | |
| *** ashleighfarnham has joined #openstack-oslo | 20:26 | |
| openstackgerrit | James Polley proposed openstack-dev/pbr: WIP: Teach pbr to read extras from setup.cfg https://review.openstack.org/181795 | 20:26 |
| tchaypo | lifeless: ^^ | 20:27 |
| lifeless | tchaypo: https://review.openstack.org/182433 <- a sample use case | 20:32 |
| tchaypo | which should be on master, according to the comments | 20:35 |
| amuller | sileht: there? | 20:35 |
| tchaypo | Is there somewhere we should be documenting this? | 20:35 |
| *** vilobhmm11 has quit IRC | 20:35 | |
| *** achanda has joined #openstack-oslo | 20:36 | |
| dims | tchaypo: etherpads for oslo are open, there's a session on optional dependencies... | 20:36 |
| *** vilobhmm1 has joined #openstack-oslo | 20:37 | |
| *** openstackgerrit has quit IRC | 20:37 | |
| *** openstackgerrit has joined #openstack-oslo | 20:37 | |
| tchaypo | dims: do you have a link handy? | 20:37 |
| dims | https://etherpad.openstack.org/p/YVR-oslo-optional-dependencies | 20:38 |
| dims | tchaypo: ^^ | 20:38 |
| tchaypo | which is probably linked from https://etherpad.openstack.org/p/liberty-oslo-summit-planning, which I just found in my email.. | 20:38 |
| *** vilobhmm1 has quit IRC | 20:38 | |
| tchaypo | actually it doesn’t seem to be | 20:40 |
| *** ashleighfarnham has quit IRC | 20:41 | |
| dims | tchaypo: https://wiki.openstack.org/wiki/Summit/Liberty/Etherpads is the main landing page, oslo etherpads are listed there by time/date | 20:42 |
| *** mriedem has quit IRC | 20:44 | |
| tchaypo | thanks | 20:45 |
| dims | harlowja: just saw this float by - https://review.openstack.org/#/c/182445/1/specs/liberty/approved/quotas-reimagined.rst,cm | 20:47 |
| *** vilobhmm1 has joined #openstack-oslo | 20:49 | |
| *** vilobhmm1 has quit IRC | 20:50 | |
| *** vilobhmm1 has joined #openstack-oslo | 20:50 | |
| tchaypo | I’ve added that to my sched | 20:51 |
| tchaypo | lifeless: Is there anything we can do for 182433 right now? Maybe clone it to master, put the extra in setup.cfg, and add a dependency link to the pbr review? | 20:52 |
| *** mdbooth has quit IRC | 20:52 | |
| *** e0ne has quit IRC | 20:52 | |
| *** e0ne has joined #openstack-oslo | 20:53 | |
| *** kgiusti has left #openstack-oslo | 20:54 | |
| lifeless | tchaypo: we need to unbreak neutron immediately | 20:57 |
| lifeless | tchaypo: so no. Once we've got markers in requirements.txt working in a pbr release we can circle back and cleanup neutron. | 20:58 |
| *** mdbooth has joined #openstack-oslo | 20:58 | |
| *** ozamiatin has joined #openstack-oslo | 21:00 | |
| tchaypo | markers in requirements.txt? I’ve been aiming for markers in setup.cfg | 21:00 |
| *** vilobhmm1 has quit IRC | 21:01 | |
| *** vilobhmm1 has joined #openstack-oslo | 21:02 | |
| *** shardy is now known as shardy_z | 21:02 | |
| *** e0ne has quit IRC | 21:03 | |
| tchaypo | In that case, I’m going to go have breakfast and coffee | 21:03 |
| *** vilobhmm1 has quit IRC | 21:03 | |
| tchaypo | and then come back and document pbr’s new support for extras and env markers | 21:04 |
| *** vilobhmm1 has joined #openstack-oslo | 21:04 | |
| *** mriedem has joined #openstack-oslo | 21:07 | |
| *** stevemar has quit IRC | 21:08 | |
| *** amuller is now known as amuller_afk | 21:11 | |
| *** ashleighfarnham has joined #openstack-oslo | 21:12 | |
| lifeless | tchaypo: ok so | 21:19 |
| lifeless | tchaypo: same same, one big arc :) | 21:19 |
| lifeless | tchaypo: what you're doing is v closely related, and may even fit in the same patch | 21:19 |
| lifeless | tchaypo: I'm going to look at your current stuff soon | 21:20 |
| *** rrepaka has quit IRC | 21:22 | |
| *** ashleighfarnham has quit IRC | 21:26 | |
| *** ozamiatin has quit IRC | 21:35 | |
| *** salv-orlando has quit IRC | 21:42 | |
| *** cdent has quit IRC | 21:47 | |
| dims | sigmavirus24: i am seeing lots of hits for "Connection pool is full, discarding connection" in logstash, seems to be from connectionpool.py. am i reading this right? | 21:54 |
| sigmavirus24 | dims: in urllib3, correct, but I'm curious where it's coming from | 21:58 |
| sigmavirus24 | I had gone around attempting to squash the cause of that | 21:58 |
| *** vilobhmm1 has quit IRC | 22:00 | |
| *** stpierre has quit IRC | 22:03 | |
| *** vilobhmm1 has joined #openstack-oslo | 22:04 | |
| *** vilobhmm1 has quit IRC | 22:05 | |
| *** vilobhmm1 has joined #openstack-oslo | 22:05 | |
| *** vilobhmm1 has quit IRC | 22:07 | |
| *** vilobhmm11 has joined #openstack-oslo | 22:10 | |
| *** bogdando has quit IRC | 22:10 | |
| harlowja | dims thx | 22:10 |
| harlowja | intersting | 22:10 |
| *** jecarey has quit IRC | 22:13 | |
| *** jaypipes has quit IRC | 22:13 | |
| lifeless | tchaypo: ok pop() back to you | 22:13 |
| tchaypo | huh | 22:13 |
| tchaypo | how did you manage to do that seconds after i sat down and opened the window? | 22:14 |
| openstackgerrit | Victor Stinner proposed openstack/oslo.versionedobjects: VersionedObjectDictCompat inherits MutableMapping https://review.openstack.org/182477 | 22:16 |
| lifeless | tchaypo: its a skill | 22:17 |
| *** bknudson has quit IRC | 22:18 | |
| tchaypo | how far is canterbury from you? | 22:18 |
| tchaypo | *looks at maps* Oh, you’re *in* canterbury. | 22:18 |
| lifeless | I'm in waimakariri | 22:19 |
| tchaypo | according to google maps, waimakariri has some interesting across-the-river bits | 22:20 |
| lifeless | yup | 22:20 |
| tchaypo | it looks like someone was coloring in “this area is waimakariri” but their pen was a bit too wide and accidentally coloured in little bits on the other side of the river | 22:21 |
| tchaypo | I’m sure the real historical reason is different | 22:21 |
| openstackgerrit | Victor Stinner proposed openstack/oslo.versionedobjects: VersionedObjectDictCompat inherits MutableMapping https://review.openstack.org/182477 | 22:23 |
| tchaypo | so anyway, back to my patch | 22:23 |
| *** Rockyg has quit IRC | 22:24 | |
| tchaypo | http://logs.openstack.org/95/181795/8/check/check-pbr-installation-dsvm/bfc99a8/console.html#_2015-05-12_20_53_57_833 Is that related to the neutron issue you mentioned earlier? | 22:26 |
| *** dims_ has joined #openstack-oslo | 22:28 | |
| lifeless | well | 22:28 |
| lifeless | the river didn't stay will | 22:28 |
| lifeless | still | 22:28 |
| *** sdake_ has joined #openstack-oslo | 22:29 | |
| *** dims has quit IRC | 22:30 | |
| dims_ | sigmavirus24: http://logstash.openstack.org/#eyJzZWFyY2giOiJcIkNvbm5lY3Rpb24gcG9vbCBpcyBmdWxsLCBkaXNjYXJkaW5nIGNvbm5lY3Rpb25cIiIsImZpZWxkcyI6W10sIm9mZnNldCI6MCwidGltZWZyYW1lIjoiNDMyMDAiLCJncmFwaG1vZGUiOiJjb3VudCIsInRpbWUiOnsidXNlcl9pbnRlcnZhbCI6MH0sInN0YW1wIjoxNDMxNDY5ODI5MTYwfQ== | 22:30 |
| *** sdake has quit IRC | 22:33 | |
| sigmavirus24 | dims_: so previously glanceclient and swiftclient were the two big offenders | 22:35 |
| sigmavirus24 | Now all of these are coming out of devstack gates from screen-n-api | 22:35 |
| sigmavirus24 | (or at least the first bunch that I've looked at) | 22:35 |
| sigmavirus24 | Found one from glance and one other from n-cpu | 22:36 |
| dims_ | sigmavirus24: i feel i saw one where nova api was calling glance client | 22:36 |
| *** ashleighfarnham has joined #openstack-oslo | 22:36 | |
| sigmavirus24 | It looks like Nova is streaming responses and not properly handling them if they're not completely consumed | 22:37 |
| sigmavirus24 | dims_: see https://review.openstack.org/#/c/147678/ for an idea of how I squashed it in glanceclient (there's a similar patch in swiftclient) | 22:38 |
| lifeless | tchaypo: doing a detailed review now | 22:41 |
| tchaypo | Thanks | 22:41 |
| *** salv-orlando has joined #openstack-oslo | 22:43 | |
| *** sdake has joined #openstack-oslo | 22:43 | |
| *** ihrachyshka has joined #openstack-oslo | 22:44 | |
| *** salv-orlando has quit IRC | 22:46 | |
| *** sdake_ has quit IRC | 22:46 | |
| *** vilobhmm11 has quit IRC | 22:49 | |
| *** vilobhmm1 has joined #openstack-oslo | 22:49 | |
| dims_ | sigmavirus24: don't see any iter_content in nova | 23:03 |
| *** amuller_afk is now known as amuller | 23:05 | |
| lifeless | tchaypo: thats looking good | 23:07 |
| lifeless | tchaypo: thoughts and the next bit included | 23:07 |
| lifeless | now to write a blog post about this and stuff and stuff | 23:07 |
| sigmavirus24 | dims_: must be in something nova's using with stream=True then | 23:09 |
| sigmavirus24 | dims_: in general it's not a serious message, it's a debug level message from urllib3 | 23:09 |
| sigmavirus24 | It /may/ hint towards a potential socket leak | 23:10 |
| sigmavirus24 | But that's not necessarily the case | 23:10 |
| *** ozamiatin has joined #openstack-oslo | 23:19 | |
| lifeless | tchaypo: crickets? | 23:25 |
| tchaypo | I’m breathing deeply and trying to let go of the gas bill thing | 23:26 |
| harlowja | lifeless added comment on https://review.openstack.org/#/c/182423/ | 23:31 |
| harlowja | maybe i should just totally drop Mysql-python blah blah | 23:31 |
| harlowja | *thats the only test-requirements differnce, lol | 23:31 |
| harlowja | seems like that would be ok (and is the general direction of all the things happening anway) | 23:31 |
| openstackgerrit | Joshua Harlow proposed openstack/taskflow: Remove custom py26/py27 and py33 environment + fix requirements https://review.openstack.org/182423 | 23:33 |
| openstackgerrit | Joshua Harlow proposed openstack/taskflow: Remove custom py26/py27 and tweak requirements https://review.openstack.org/182423 | 23:34 |
| tchaypo | okay. Calm has been restored, approximately. | 23:34 |
| harlowja | vilobhmm1 u might be interested in https://review.openstack.org/#/c/182445 to | 23:38 |
| harlowja | intersting to see what will happen there | 23:38 |
| openstackgerrit | Morgan Fainberg proposed openstack/oslo.log: Make remove_in=0 (no removal) use a better syntax https://review.openstack.org/168186 | 23:40 |
| morganfainberg | ^^ that is a simple rebase, just needs someone to press +A again | 23:41 |
| vilobhmm1 | sure…thanks harlowja for the link | 23:44 |
| harlowja | np | 23:44 |
| *** ericksonfgds has joined #openstack-oslo | 23:45 | |
| dhellmann | morganfainberg: +W | 23:46 |
| morganfainberg | dhellmann: tyvm sir | 23:46 |
| dhellmann | morganfainberg: think nothing of it | 23:47 |
| lifeless | harlowja: do you depend on oslo.db ? | 23:48 |
| harlowja | nope, not right now, maybe someday | 23:48 |
| * harlowja trying to keep the additions of oslo.* deps to a minmum if i can | 23:50 | |
| harlowja | and only things like oslo.utils or oslo.serialization which are dinky | 23:50 |
| harlowja | although i should be able to easily switch to https://github.com/openstack/oslo.db/blob/master/oslo_db/sqlalchemy/session.py#L712 (and use that, vs just use sqlalchemy engine calls via the standard sqlalchemy stuffs...) | 23:51 |
| openstackgerrit | Dan Smith proposed openstack/oslo.versionedobjects: VersionedObjectDictCompat inherits MutableMapping https://review.openstack.org/182477 | 23:51 |
| openstackgerrit | Dan Smith proposed openstack/oslo.versionedobjects: Add delattr support https://review.openstack.org/182501 | 23:51 |
| harlowja | but i've always had this tension of making taskflow just that much more only openstack useable | 23:52 |
| harlowja | ^ which i don't really like :-/ | 23:52 |
| harlowja | lifeless whats a project that uses the environment markers in setup.py like u said | 23:55 |
| harlowja | can just do that until https://review.openstack.org/#/c/181795/ happens | 23:56 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!