| *** reed has joined #openstack-zaqar | 00:04 | |
| openstackgerrit | OpenStack Proposal Bot proposed a change to openstack/python-zaqarclient: Updated from global requirements https://review.openstack.org/114068 | 00:31 |
|---|---|---|
| openstackgerrit | OpenStack Proposal Bot proposed a change to openstack/zaqar: Updated from global requirements https://review.openstack.org/120330 | 00:32 |
| *** kgriffs|afk is now known as kgriffs | 01:21 | |
| *** amitgandhinz has quit IRC | 01:42 | |
| *** amitgandhinz has joined #openstack-zaqar | 01:43 | |
| *** kgriffs is now known as kgriffs|afk | 01:44 | |
| *** amitgandhinz has quit IRC | 01:49 | |
| *** jeffrey4l has joined #openstack-zaqar | 02:32 | |
| *** vkmc has quit IRC | 02:40 | |
| *** jeffrey4l has quit IRC | 03:17 | |
| *** jeffrey4l has joined #openstack-zaqar | 03:29 | |
| *** sgotliv has joined #openstack-zaqar | 04:10 | |
| *** jeffrey4l has quit IRC | 04:13 | |
| *** echevemaster has quit IRC | 04:22 | |
| *** jeffrey4l has joined #openstack-zaqar | 04:29 | |
| *** prashanthr_ has joined #openstack-zaqar | 05:02 | |
| *** prashanthr_ has quit IRC | 05:22 | |
| *** reed has quit IRC | 05:22 | |
| *** jeffrey4l has quit IRC | 05:35 | |
| *** jeffrey4l has joined #openstack-zaqar | 05:52 | |
| *** flaper87|afk is now known as flaper87 | 05:58 | |
| flaper87 | o/ | 06:05 |
| wpf | flwang, flaper87 : all, would you please help to check https://review.openstack.org/#/c/117706/ again. thanks | 06:08 |
| flaper87 | wpf: sure | 06:08 |
| wpf | flaper87: thanks , :P | 06:09 |
| *** flwang1 has quit IRC | 06:11 | |
| flaper87 | wpf: a tiny nit that I think is worth it | 06:15 |
| jeffrey4l | flaper87, pls help to check https://review.openstack.org/#/c/118269/ again. You added -1 due to the pass of j-3. | 06:18 |
| flaper87 | jeffrey4l: sure | 06:18 |
| flaper87 | :) | 06:18 |
| jeffrey4l | thx | 06:18 |
| * flaper87 so happy to have more patches coming in in zaqar | 06:18 | |
| wpf | flaper87: I will check it now | 06:20 |
| *** jeffrey4l has quit IRC | 06:26 | |
| *** jeffrey4l has joined #openstack-zaqar | 06:37 | |
| *** jeffrey4l has quit IRC | 07:14 | |
| openstackgerrit | Peng Fei Wang proposed a change to openstack/zaqar: Not permitted to delete pools which are used by flavor https://review.openstack.org/117706 | 07:20 |
| *** jeffrey4l has joined #openstack-zaqar | 07:27 | |
| *** AAzza_afk is now known as AAzza | 07:51 | |
| wpf | flaper87: https://review.openstack.org/#/c/117706/ , again , :P | 07:53 |
| *** jeffrey4l has quit IRC | 08:55 | |
| *** jeffrey4l has joined #openstack-zaqar | 09:11 | |
| openstackgerrit | Jeffrey Zhang proposed a change to openstack/python-zaqarclient: Fix the help string for queue in cli https://review.openstack.org/120735 | 09:34 |
| *** jeffrey4l has quit IRC | 09:38 | |
| *** jeffrey4l has joined #openstack-zaqar | 09:50 | |
| *** jeffrey4l has quit IRC | 10:08 | |
| *** prashanthr_ has joined #openstack-zaqar | 10:12 | |
| *** jeffrey4l has joined #openstack-zaqar | 10:20 | |
| *** flwang1 has joined #openstack-zaqar | 10:27 | |
| *** vkmc has joined #openstack-zaqar | 10:36 | |
| vkmc | morning :) | 10:38 |
| openstackgerrit | Jeffrey Zhang proposed a change to openstack/python-zaqarclient: Fix the help string for queue in cli https://review.openstack.org/120735 | 10:51 |
| flaper87 | vkmc: :) | 11:03 |
| vkmc | flaper87, hi o/ | 11:05 |
| flaper87 | vkmc: I just sent another email on that thread. Mind reading? | 11:14 |
| flaper87 | I'd like to know what you think | 11:14 |
| vkmc | flaper87, sure thing | 11:15 |
| flaper87 | vkmc: thank you :) | 11:16 |
| flaper87 | vkmc: how are you doing? btw | 11:16 |
| flaper87 | (didn't mean to be rude) | 11:16 |
| vkmc | flaper87, haha np | 11:19 |
| vkmc | flaper87, I'm doing ok, bit sleepy though | 11:20 |
| vkmc | I was going to reply to that thread, or the first one... but TBH I didn't know who to reply first | 11:21 |
| flaper87 | I'd really prefer the discussion to focus on the second one | 11:22 |
| vkmc | flwang1, around? :) | 11:22 |
| flaper87 | The current concern is suppose to be discussed there | 11:22 |
| vkmc | flaper87, yeah me too... the first one is quite confusing right now | 11:22 |
| flaper87 | just ignore it | 11:23 |
| vkmc | already did | 11:25 |
| flaper87 | So, since January, SQS has dead letter queue: http://aws.amazon.com/blogs/aws/amazon-sqs-new-dead-letter-queue/ | 11:25 |
| vkmc | great email flaper87 | 11:26 |
| vkmc | I think it shed some light over most of the concerns | 11:26 |
| * flaper87 is becoming really paranoid about this emailing process | 11:27 | |
| vkmc | just one thing that let me thinking yesterday | 11:27 |
| vkmc | and now I saw you brought it up in your email | 11:27 |
| vkmc | what are the Zaqar features that keep us away from being considered queue? | 11:27 |
| flaper87 | vkmc: get-message-by-id is one of them | 11:28 |
| vkmc | yeah, that is clear | 11:28 |
| flaper87 | TBH, I kinda think it's the only one | 11:28 |
| flaper87 | :P | 11:28 |
| vkmc | but besides of that | 11:28 |
| vkmc | lol | 11:28 |
| flaper87 | As gordon said, there's no real difference | 11:28 |
| flaper87 | messaging is just a bit more generic | 11:28 |
| vkmc | there is where I wanted to arrive | 11:28 |
| vkmc | they don't like the fact that Zaqar behave as a queue but we are defending that we are a messages service | 11:29 |
| flaper87 | right, which doesn't mean we should remove all the other things | 11:29 |
| vkmc | everyone is like 'oh but you mentioned prod/cons, then you are a queue!' | 11:29 |
| vkmc | about SQS dead letter queue, good stuff! | 11:57 |
| *** flwang2 has joined #openstack-zaqar | 12:01 | |
| *** flwang1 has quit IRC | 12:02 | |
| *** cpallares has joined #openstack-zaqar | 12:02 | |
| flaper87 | vkmc: btw, I responded to your comments in gerrit | 12:17 |
| vkmc | flaper87, thanks | 12:18 |
| vkmc | flaper87, btw could you reproduce https://review.openstack.org/#/c/119900/? | 12:18 |
| vkmc | AAzza couldn't yesterday | 12:18 |
| vkmc | I'm debugging the server | 12:19 |
| vkmc | flaper87, oh and I replied too | 12:20 |
| flaper87 | ah, erm... | 12:20 |
| * flaper87 silently goes away and reads | 12:20 | |
| *** jeffrey4l has quit IRC | 12:30 | |
| *** jeffrey4l has joined #openstack-zaqar | 12:44 | |
| vkmc | flaper87, do you know if someone in Glance is interested in mentoring for the next OPW round? | 12:56 |
| flaper87 | vkmc: I'm not sure :/ | 13:00 |
| flaper87 | I'd like to mentor for Zaqar | 13:00 |
| flaper87 | I think I've a project idea already | 13:01 |
| vkmc | that's so great :) | 13:01 |
| *** flwang2 has quit IRC | 13:03 | |
| vkmc | well, when you have a moment please update the wiki with the project you have in mind | 13:04 |
| vkmc | so applicants start making up their minds | 13:05 |
| vkmc | right now we have two interested applicants, one in Keystone and one in Glance | 13:05 |
| vkmc | and we don't have mentors for those projects :/ | 13:05 |
| vkmc | I started reaching out former mentors | 13:05 |
| flaper87 | vkmc: what was the link again? | 13:05 |
| vkmc | https://wiki.openstack.org/wiki/OutreachProgramForWomen | 13:14 |
| vkmc | flaper87, ^ | 13:14 |
| *** sriram has joined #openstack-zaqar | 13:14 | |
| vkmc | more precisely, https://wiki.openstack.org/wiki/OutreachProgramForWomen/Ideas | 13:14 |
| * flaper87 adds something | 13:20 | |
| *** jchai has joined #openstack-zaqar | 13:21 | |
| vkmc | :) | 13:21 |
| vkmc | you rock | 13:21 |
| vkmc | iff I get a job | 13:21 |
| vkmc | I'll propose myself as mentor for Zaqar as well | 13:21 |
| flaper87 | vkmc: AWESOME! | 13:22 |
| flaper87 | vkmc: btw, https://review.openstack.org/#/c/120155/ :D | 13:22 |
| * flaper87 needs to get kgriffs|afk to review client patches | 13:22 | |
| vkmc | flaper87, I'm on it | 13:22 |
| vkmc | first I want to debug the claim thing | 13:23 |
| *** mpanetta has joined #openstack-zaqar | 13:31 | |
| *** earnThis has joined #openstack-zaqar | 13:52 | |
| *** amitgandhinz has joined #openstack-zaqar | 14:01 | |
| *** kgriffs|afk is now known as kgriffs | 14:08 | |
| *** prashanthr_1 has joined #openstack-zaqar | 14:14 | |
| *** prashanthr_ has quit IRC | 14:14 | |
| vkmc | flaper87, I'm revisiting the claims_id bug in the client | 14:22 |
| flaper87 | vkmc: what's up with that? | 14:22 |
| flaper87 | :D | 14:22 |
| vkmc | it's happening because claimed messages has a claim_id | 14:22 |
| vkmc | and that is fine | 14:22 |
| flaper87 | it's fine for v1.1 not v1 | 14:23 |
| flaper87 | Are you testing it in your laptop? | 14:23 |
| vkmc | the odd thing is that the client is trying to create a claimed message as it were a not claimed message | 14:23 |
| vkmc | yeah | 14:23 |
| vkmc | check line 131 https://review.openstack.org/#/c/119347/3/zaqar/queues/transport/wsgi/v1_0/claims.py | 14:23 |
| vkmc | here we add the claimed_id | 14:23 |
| vkmc | param | 14:24 |
| vkmc | it's in the URL as well | 14:24 |
| vkmc | so.. we should remove that param in there | 14:24 |
| vkmc | in v1 the claim should be in the URL only | 14:24 |
| flaper87 | mmh, but the `message_url` function adds it to the URL | 14:25 |
| flaper87 | ah well, yeah | 14:26 |
| flaper87 | no wait | 14:26 |
| flaper87 | nevermind, you're right | 14:26 |
| flaper87 | it needs to be removed manually in the claim code | 14:26 |
| flaper87 | as we remove the id | 14:26 |
| vkmc | oh yes, message_url adds it to the url | 14:27 |
| flaper87 | to be fair, we could use the _format_message in wsgi/v1/messages.py | 14:27 |
| flaper87 | s/use/reuse/ | 14:27 |
| vkmc | oh no! I was right in the first time --> path += falcon.to_query_str({'claim_id': claim_id}) | 14:27 |
| vkmc | it's adding the field, not changing the url | 14:27 |
| *** ametts has quit IRC | 14:27 | |
| vkmc | pfff | 14:28 |
| vkmc | yeah we could | 14:28 |
| flaper87 | cool | 14:28 |
| vkmc | or remove that param | 14:28 |
| flaper87 | easy fix, remove the param :) | 14:28 |
| flaper87 | re-using _format_message reduces the chance of duplicated bugs | 14:28 |
| vkmc | ok, I'll take the red pill and do the second one | 14:29 |
| AAzza | flaper87: vkmc: i didn't get what is the problem... I want to understand since I was touching the code | 14:30 |
| vkmc | AAzza, hey :) | 14:32 |
| AAzza | there is no claim_id in the message dict in that file, i checked it. the message-url adds it only to the url, and it was adding before me | 14:32 |
| AAzza | vkmc: hi) | 14:32 |
| vkmc | AAzza, this is the output I got when running the bench mark with one consumer worker http://paste.openstack.org/show/110176/ | 14:32 |
| vkmc | AAzza, the first message shows that id has a claim_id param | 14:32 |
| flaper87 | AAzza: but the message controller is adding the claim_id | 14:33 |
| vkmc | I guess it's because of line 132 here https://review.openstack.org/#/c/119347/3/zaqar/queues/transport/wsgi/v1_0/claims.py | 14:33 |
| flaper87 | and it's returning it through the claim controller back to the API | 14:33 |
| vkmc | ^ | 14:33 |
| vkmc | :) | 14:33 |
| AAzza | flaper87: but i tried to delete it in that point. i checked - claim_controller format the messages it returns and do not add the claim_id, let me find that piece of code. | 14:36 |
| AAzza | orrr... maye some controllers do not format the messages... | 14:36 |
| AAzza | vkmc: do you test on mongo? | 14:36 |
| *** alcabrera|afk is now known as alcabrera | 14:36 | |
| vkmc | AAzza, yep, mongo :) | 14:37 |
| AAzza | vkmc: i do testing on sqla | 14:37 |
| vkmc | oh :o | 14:37 |
| vkmc | probably it's that | 14:37 |
| vkmc | this sounds like kgriffs fault | 14:37 |
| prashanthr_1 | Good morning all :) | 14:38 |
| kgriffs | vkmc: probably | 14:38 |
| flaper87 | mongo's claim driver doesn't format the message | 14:38 |
| kgriffs | prashanthr_1: good evening! | 14:38 |
| flaper87 | it gets it right from the message controller | 14:38 |
| flaper87 | whereas the sql driver does this https://github.com/openstack/zaqar/blob/master/zaqar/queues/storage/sqlalchemy/claims.py#L45 | 14:38 |
| prashanthr_1 | Thanks kgriffs. | 14:38 |
| vkmc | huh.. what about redis? | 14:40 |
| *** jeffrey4l_ has joined #openstack-zaqar | 14:41 | |
| flaper87 | https://github.com/openstack/zaqar/blob/master/zaqar/queues/storage/redis/models.py#L165 | 14:41 |
| flaper87 | same thing, it returns the claim_id | 14:42 |
| openstackgerrit | Jeffrey Zhang proposed a change to openstack/python-zaqarclient: CLI support for queue existence and metadata operations https://review.openstack.org/120797 | 14:42 |
| flaper87 | kgriffs: whenever you have time, pleeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeease take a look at the client reviews :D | 14:42 |
| AAzza | flaper87: i think right fix will be to explicity format message in transport layer, the same we do in messages | 14:43 |
| AAzza | flaper87: to be sure, that we are not dependent on drivers implementations | 14:44 |
| vkmc | +1 AAzza | 14:44 |
| flaper87 | AAzza: +!, that's what I was proposing. We should re-use `_format_message` | 14:44 |
| vkmc | AAzza, want to add this or should I finish it? | 14:45 |
| *** jeffrey4l has quit IRC | 14:45 | |
| flaper87 | kgriffs: do we still need that `test_correct_has` test | 14:47 |
| flaper87 | ? | 14:47 |
| flaper87 | I'm kind of tempted to remove it | 14:47 |
| flaper87 | if you run tests with changes that have not been commited it normally fails | 14:47 |
| kgriffs | oh. | 14:47 |
| flaper87 | and if you run it outside a `git` repo it fails too | 14:48 |
| flaper87 | it's probably pbr's fault | 14:48 |
| flaper87 | dunno | 14:48 |
| kgriffs | yeah, sometimes I have to recreate when switching branches | 14:48 |
| kgriffs | anyway | 14:48 |
| kgriffs | um | 14:48 |
| kgriffs | considering that code is stable and probably won't change for the forseeable future | 14:48 |
| kgriffs | i'm fine with removing the test | 14:48 |
| flaper87 | cool | 14:49 |
| flaper87 | wait for it | 14:52 |
| openstackgerrit | Flavio Percoco proposed a change to openstack/zaqar: Don't import client from the version package https://review.openstack.org/120802 | 14:52 |
| openstackgerrit | Flavio Percoco proposed a change to openstack/zaqar: Remove test_correct_has from our test suite https://review.openstack.org/120803 | 14:52 |
| flaper87 | BANG! | 14:52 |
| prashanthr_1 | kgriffs: I have finished with | 14:54 |
| prashanthr_1 | "* Add the ability to configure Sentinel support in zaqar.conf | 14:54 |
| prashanthr_1 | * Read that config and use it to construct a Sentinel instance and discover the master in lieu of instantiating StrictRedis directly." | 14:54 |
| prashanthr_1 | can you explain me a bit about "* Test master-slave failover with the above and see if the retry decorator needs to do anything to find the new slave, or if redis-py does it automagically." | 14:55 |
| prashanthr_1 | I have found out that redis-py does not provide default fallback | 14:55 |
| prashanthr_1 | :( | 14:55 |
| flaper87 | malini: around? | 14:56 |
| flaper87 | ah no, malini is on vacation | 14:56 |
| flaper87 | nevermind | 14:56 |
| flaper87 | AFAIK | 14:56 |
| flaper87 | malini: stay where you are, don't pay attention to me. Trust me, it's better like this | 14:56 |
| kgriffs | prashanthr_1: ok, so what error does redis-py raise when the master fails over to a slave? | 14:56 |
| prashanthr_1 | ConnectionError | 14:56 |
| prashanthr_1 | as with a default standalone redis | 14:57 |
| kgriffs | and I am assuming the decorator retries? | 14:57 |
| vkmc | AAzza, let me know about the above later :) | 14:57 |
| prashanthr_1 | Yes it retries | 14:57 |
| kgriffs | ok. TBH I haven't looked at that decorator code very closely yet | 14:57 |
| kgriffs | so, i suppose what it needs to do is rediscover the new master. which means it will need access to the sentinel class instance | 14:58 |
| kgriffs | what do you think? | 14:59 |
| prashanthr_1 | kgriffs: precisely | 14:59 |
| AAzza | vkmc: if you want, you can do it) i would be glad to do it, but I should do many projects for the university - deadlinea are in few weeks although semester is just started :) | 14:59 |
| prashanthr_1 | so we need to store a state variable in the driver can say if redis is operating in sentinel or stand alone mode | 15:00 |
| prashanthr_1 | if yes we might need to use the sentinel object to retrieve the next slave | 15:00 |
| vkmc | AAzza, ok, no worries! I'm on it :) thanks for helping to debug | 15:00 |
| prashanthr_1 | i also had a small question here, in the failover case we need to check for any other slaves and then create a master-slave replication hierarchy in the decorator | 15:01 |
| prashanthr_1 | Am i correct here ? | 15:01 |
| AAzza | kgriffs: flaper87: could we discuss later today - probably in few hours - what are left from api-v1.1-response changes blueprint? So I could finish it tomorrow/on the weekends) | 15:02 |
| kgriffs | kk | 15:02 |
| flaper87 | AAzza: totally | 15:05 |
| flaper87 | thanks :) | 15:05 |
| prashanthr_1 | kgriffs: Should we re-establish the master slave hierarchy when we do a failover | 15:10 |
| prashanthr_1 | ? | 15:10 |
| prashanthr_1 | also most slaves operate read-only | 15:10 |
| prashanthr_1 | so we must try to change the config in the decorator i guess | 15:10 |
| *** AAzza is now known as AAzza_afk | 15:11 | |
| *** jeffrey4l_ has quit IRC | 15:13 | |
| kgriffs | back | 15:15 |
| kgriffs | prashanthr_1: not sure I understand the question | 15:16 |
| kgriffs | let me pull up sentinel docs | 15:16 |
| kgriffs | ok, found this "If a master is not working as expected, Sentinel can start a failover process where a slave is promoted to master, the other additional slaves are reconfigured to use the new master, and the applications using the Redis server informed about the new address to use when connecting." | 15:17 |
| kgriffs | so, if sentinel is enabled | 15:18 |
| kgriffs | isn't it as simple as calling sentinel.discover_master in the decorator before retrying the operation? | 15:18 |
| kgriffs | oops | 15:19 |
| kgriffs | I mean sentinel.master_for | 15:19 |
| kgriffs | btw, if you aren't already, it would be good log a warning when this happens | 15:19 |
| kgriffs | flaper87: btw, saw your note about epel serving redis 2.4 | 15:21 |
| prashanthr_1 | Ohh okay. I will look into this kgriffs. | 15:21 |
| kgriffs | flaper87: were you going to follow up with the package maintainer? | 15:21 |
| flaper87 | kgriffs: look who's the maintainer now: https://admin.fedoraproject.org/pkgdb/ | 15:21 |
| flaper87 | erm https://admin.fedoraproject.org/pkgdb/package/redis/ | 15:22 |
| * kgriffs looks | 15:22 | |
| kgriffs | crap | 15:22 |
| flaper87 | ;) | 15:22 |
| kgriffs | now it's *never* going to get done | 15:22 |
| flaper87 | that's why I asked someone else to do it :P | 15:22 |
| flaper87 | btw, el7 will get 2.8 | 15:23 |
| flaper87 | el6 ain't going to be updated | 15:23 |
| kgriffs | gtk | 15:23 |
| kgriffs | no ZSCAN love for el6 then | 15:24 |
| flaper87 | I wouldn't worry much about el, TBH. | 15:24 |
| flaper87 | is ZSCAN in 2.6 ? | 15:25 |
| flaper87 | el6* | 15:25 |
| kgriffs | 2.8 | 15:25 |
| flaper87 | doh, ok | 15:25 |
| kgriffs | not sure what else landed in 2.8 - that was just first thing I thought of | 15:26 |
| *** reed has joined #openstack-zaqar | 15:26 | |
| kgriffs | I heard it on this radio show. "Random techie trivia to win friends and influence people" | 15:26 |
| * kgriffs winks | 15:27 | |
| *** prashanthr_1 has quit IRC | 15:36 | |
| *** prashanthr_ has joined #openstack-zaqar | 15:36 | |
| openstackgerrit | A change was merged to openstack/zaqar: Updated from global requirements https://review.openstack.org/120330 | 15:41 |
| *** jimhoagland has joined #openstack-zaqar | 15:42 | |
| flaper87 | LOL | 15:44 |
| flaper87 | kgriffs: "Any resemblance to reality is purely a coincidence" | 15:44 |
| *** jimhoagland has left #openstack-zaqar | 15:45 | |
| prashanthr_ | kgriffs: We can use discover_master on the sentinel to obtain the next master | 15:50 |
| prashanthr_ | :) | 15:50 |
| kgriffs | kewl | 15:52 |
| earnThis | hey guys, any idea why im getting this error when i switch branches to my previos patch and run tox? All im going is git review -d "117574" to pick up where i left off then running tox, the branch previously passed the tests http://paste.openstack.org/show/110196/ | 15:57 |
| earnThis | im sure im missing somethign obvious | 15:57 |
| flaper87 | earnThis: it's not your fault | 16:00 |
| flaper87 | earnThis: this fill fix that error: https://review.openstack.org/#/c/120803/ | 16:00 |
| flaper87 | it's a test we have that requires you to have a clean git workspace | 16:00 |
| flaper87 | that thing is used for production environments | 16:00 |
| vkmc | the problem is that he has tox 1.7.2 | 16:01 |
| flaper87 | ? | 16:02 |
| earnThis | vkmc: i have 1.6.1 now | 16:02 |
| flaper87 | vkmc: open the paste :P | 16:02 |
| vkmc | and probably in the other branch he has .tox env with that version | 16:02 |
| vkmc | flaper87, yeah Fla, I already look at it | 16:02 |
| vkmc | it's an old bug heh | 16:02 |
| flaper87 | mmh, wait because this paste is different from the one I saw the other day | 16:03 |
| vkmc | this -> File "tests/unit/test_version.py", line 34, in test_correct_hash | 16:03 |
| flaper87 | that test that is failing is not because of tox | 16:03 |
| vkmc | is tox fault | 16:03 |
| flaper87 | mmh, nope | 16:04 |
| flaper87 | I can make that test fail without tos | 16:04 |
| flaper87 | tox | 16:04 |
| flaper87 | the reason it is failing is because it requires a clean working copy | 16:04 |
| earnThis | that sounds righ | 16:05 |
| earnThis | right* | 16:05 |
| vkmc | it does yeap | 16:05 |
| earnThis | what can i do | 16:06 |
| flaper87 | earnThis: review the patch :D | 16:06 |
| flaper87 | earnThis: ignore the failure | 16:06 |
| flaper87 | earnThis: or commit everything locally and re-run the test | 16:06 |
| *** prashanthr_ has quit IRC | 16:08 | |
| *** prashanthr_1 has joined #openstack-zaqar | 16:08 | |
| vkmc | flaper87, http://paste.openstack.org/show/110201/ | 16:08 |
| flaper87 | vkmc: ohhhhhhhhhhhhhh now I understand what you meant | 16:08 |
| flaper87 | I'm sorry | 16:09 |
| flaper87 | although, that bug is not really related to tox, it's pbr version discovery thingy failing misserably :( | 16:09 |
| vkmc | earnThis, I think that rm -rf .tox/ should do the trick | 16:10 |
| *** prashanthr_ has joined #openstack-zaqar | 16:10 | |
| vkmc | flaper87, yeah I didn't asked what it was about that time.. so I blame tox ;p | 16:10 |
| vkmc | so it's pbr messing with us | 16:10 |
| *** prashanthr_1 has quit IRC | 16:12 | |
| vkmc | flaper87, you owe me a gummybear | 16:12 |
| * vkmc hides | 16:12 | |
| earnThis | http://paste.openstack.org/show/110203/ | 16:15 |
| flaper87 | vkmc: you should've asked 10 mins ago, I just finished this week's bag :( | 16:15 |
| flaper87 | I'm sorry | 16:15 |
| flaper87 | and I still have friday ahead | 16:15 |
| flaper87 | damnit | 16:15 |
| flaper87 | earnThis: you seriously can't blame zaqar for that :P | 16:15 |
| earnThis | hahah wasn't at all | 16:16 |
| *** Guest96101 is now known as sebasmagri | 16:18 | |
| flaper87 | vkmc: if you review those 2 patches that I submitted like 30mins ago we can get rid of that issue forever | 16:18 |
| *** sebasmagri has quit IRC | 16:18 | |
| *** sebasmagri has joined #openstack-zaqar | 16:18 | |
| *** sebasmagri has joined #openstack-zaqar | 16:18 | |
| flaper87 | and I'll bring gummy bears to paris | 16:18 |
| vkmc | ok... I'm updating my bet to TWO gummybears | 16:19 |
| vkmc | or I won't review those patches | 16:19 |
| flaper87 | vkmc: 1 full bag because you rock | 16:19 |
| vkmc | :D | 16:20 |
| vkmc | reviewed | 16:20 |
| * kgriffs observes the blatant corruption of the OS community as demonstrated by rampant bribery | 16:21 | |
| flaper87 | vkmc: you'll have to review https://review.openstack.org/#/c/120802/1 to get the other one merged | 16:21 |
| flaper87 | 1 bag + 1 gummy bear for that | 16:21 |
| flaper87 | and don't tell kgriffs | 16:21 |
| vkmc | done | 16:21 |
| vkmc | kgriffs, you didn't see nothing | 16:21 |
| * kgriffs didn't see nothin' | 16:22 | |
| vkmc | you know nothing John Snow | 16:22 |
| flaper87 | kgriffs: client client client | 16:22 |
| flaper87 | there's a client release freeze on the 18th | 16:22 |
| flaper87 | :D | 16:22 |
| kgriffs | oic | 16:23 |
| vkmc | flaper87, btw, using format_messages in the claim_id bug didn't work | 16:45 |
| vkmc | I'm uploading a possible fix, but it's not fancy | 16:45 |
| prashanthr_ | kgriffs: I am almost done with the patch will submit it early tmrw :) | 16:46 |
| kgriffs | sounds good, thanks! | 16:46 |
| *** prashanthr_ has quit IRC | 16:46 | |
| *** alcabrera is now known as alcabrera|afk | 16:47 | |
| flaper87 | vkmc: mmh, why didn't it work? | 16:50 |
| flaper87 | that's weird | 16:50 |
| openstackgerrit | A change was merged to openstack/zaqar: Don't import client from the version package https://review.openstack.org/120802 | 16:50 |
| openstackgerrit | A change was merged to openstack/zaqar: Remove test_correct_has from our test suite https://review.openstack.org/120803 | 16:51 |
| flaper87 | vkmc: I think there are 2 format_messages, 1 in the storage and 1 in the transport | 16:51 |
| vkmc | flaper87, I was using the one in transport | 16:51 |
| vkmc | the claim_id field is still there because it only formats the style | 16:51 |
| vkmc | nothing else | 16:51 |
| vkmc | so... I had to delete it | 16:51 |
| flaper87 | mmmh | 16:52 |
| vkmc | I'll think in another solution though | 16:52 |
| flaper87 | what do you mean it it only formats the style? | 16:52 |
| vkmc | forget that phrase haha | 16:53 |
| vkmc | the format_message only modifies some of the fields, but claim_id is still there | 16:54 |
| vkmc | brb | 16:54 |
| vkmc | I'll expland better soon | 16:54 |
| flaper87 | ok | 16:54 |
| flaper87 | I'm not understanding :P | 16:54 |
| flaper87 | https://github.com/openstack/zaqar/blob/master/zaqar/queues/transport/wsgi/v1_0/messages.py#L313 | 16:54 |
| flaper87 | I mean, that should return a dict w/o claim_id | 16:55 |
| *** jchai is now known as jchai_afk | 17:01 | |
| *** jchai_afk is now known as jchai | 17:04 | |
| vkmc | back | 17:37 |
| vkmc | flaper87, yeah sorry for the confusion | 17:37 |
| vkmc | I not sure what happens precisely, I'm trying to understand | 17:38 |
| *** openstackgerrit has quit IRC | 17:46 | |
| flaper87 | vkmc: did you change both methods ? | 17:47 |
| flaper87 | vkmc: I mean `on_get` for the claim resource and `on_post` for the collection ? | 17:47 |
| flaper87 | both methods return messages | 17:47 |
| *** openstackgerrit has joined #openstack-zaqar | 17:47 | |
| flaper87 | and both messages need to be normalized | 17:47 |
| *** alcabrera|afk is now known as alcabrera | 17:47 | |
| vkmc | flaper87, yeah | 17:49 |
| vkmc | now it's failing the client | 17:49 |
| vkmc | because apparently there is no href field | 17:49 |
| vkmc | that is what confused me in the first place | 17:49 |
| vkmc | http://paste.openstack.org/show/110256/ | 17:50 |
| flaper87 | that's weird, may I see the changes you did? | 17:50 |
| flaper87 | It's not that I don't trust what you're doing, btw. I just want to wrap my head around the problem | 17:51 |
| flaper87 | git diff should be enough | 17:51 |
| vkmc | k | 17:53 |
| vkmc | I could upload it as WIP | 17:54 |
| vkmc | so it's easier for you | 17:54 |
| flaper87 | sure, that works too | 17:54 |
| vkmc | http://paste.openstack.org/show/110257/ | 17:54 |
| vkmc | ok :) | 17:54 |
| flaper87 | vkmc: question: Shouldn't the claim_id still be in the URL ? | 17:57 |
| vkmc | it should | 17:58 |
| vkmc | but besides of that, the href field should be there | 17:59 |
| flaper87 | right | 17:59 |
| flaper87 | sorry, I just happened to notice that | 17:59 |
| * flaper87 got confused | 17:59 | |
| * flaper87 STFU and keeps reading | 17:59 | |
| vkmc | its ok :) | 17:59 |
| flaper87 | wait, I know what's going on | 18:00 |
| flaper87 | that for loop | 18:00 |
| flaper87 | you're not modifying the `msg` dict in `resp_msgs` | 18:00 |
| flaper87 | you're just re-assigning `msg` | 18:00 |
| flaper87 | you may want to use a generator or a list comprehension there | 18:01 |
| flaper87 | or msg.update(..format_message()) | 18:01 |
| flaper87 | and then del msg['claim_id'] | 18:01 |
| flaper87 | but that would be silly | 18:01 |
| flaper87 | so, I'd probably go with the list comprehension | 18:02 |
| flaper87 | vkmc: https://github.com/openstack/zaqar/blob/master/zaqar/queues/transport/wsgi/v1_0/messages.py#L69 | 18:02 |
| openstackgerrit | Victoria MartÃnez de la Cruz proposed a change to openstack/zaqar: Removes claim_id param in v1.0 https://review.openstack.org/120867 | 18:02 |
| vkmc | oh | 18:02 |
| flaper87 | that's a python gotcha, FWIW | 18:03 |
| * flaper87 has been there | 18:03 | |
| vkmc | lol | 18:04 |
| vkmc | so... | 18:04 |
| vkmc | (recalculando) | 18:04 |
| flaper87 | LOL | 18:05 |
| vkmc | which for loop were you talking about? | 18:06 |
| flaper87 | both, I'll comment on the review | 18:06 |
| vkmc | lol k | 18:06 |
| vkmc | I think I know | 18:06 |
| flaper87 | vkmc: done | 18:08 |
| flaper87 | (and remember the cid :P ) | 18:08 |
| flaper87 | ok, brb | 18:09 |
| flaper87 | hope that helps | 18:09 |
| vkmc | :D | 18:09 |
| *** kgriffs is now known as kgriffs|afk | 18:18 | |
| *** echevemaster has joined #openstack-zaqar | 18:19 | |
| openstackgerrit | Griffin Ashe proposed a change to openstack/zaqar: Add _ prefix to local variables https://review.openstack.org/117574 | 18:20 |
| openstackgerrit | Victoria MartÃnez de la Cruz proposed a change to openstack/zaqar: Removes claim_id param in v1.0 https://review.openstack.org/120867 | 18:24 |
| vkmc | flaper87, yay it worked! | 18:26 |
| vkmc | python gotchas are new for me :x | 18:26 |
| vkmc | although that loops seemed suspicious | 18:27 |
| *** mpanetta has quit IRC | 18:34 | |
| *** sgotliv has quit IRC | 18:48 | |
| *** mpanetta has joined #openstack-zaqar | 18:51 | |
| *** kgriffs|afk is now known as kgriffs | 19:06 | |
| vkmc | bbl | 19:07 |
| *** kgriffs is now known as kgriffs|afk | 19:17 | |
| *** flwang1 has joined #openstack-zaqar | 19:23 | |
| *** amitgandhinz has quit IRC | 19:26 | |
| *** mpanetta has quit IRC | 19:26 | |
| *** echevemaster has quit IRC | 19:50 | |
| *** earnThis is now known as earnThis|afk | 19:51 | |
| *** yograterol has joined #openstack-zaqar | 20:03 | |
| *** kgriffs|afk is now known as kgriffs | 20:04 | |
| *** echevemaster has joined #openstack-zaqar | 20:06 | |
| *** flwang1 has quit IRC | 20:08 | |
| *** kgriffs is now known as kgriffs|afk | 20:13 | |
| *** echevemaster has quit IRC | 20:14 | |
| *** flaper87 is now known as flaper87|afk | 20:24 | |
| openstackgerrit | Clark Boylan proposed a change to openstack/zaqar: Workaround pypy bug on trusty when running tests https://review.openstack.org/120906 | 20:40 |
| *** alcabrera is now known as alcabrera|afk | 21:05 | |
| *** kgriffs|afk is now known as kgriffs | 21:05 | |
| *** yograterol has quit IRC | 21:49 | |
| *** alcabrera|afk is now known as alcabrera | 21:49 | |
| *** jchai has quit IRC | 21:52 | |
| *** sriram has quit IRC | 21:55 | |
| *** yograterol has joined #openstack-zaqar | 22:03 | |
| *** earnThis|afk has quit IRC | 22:08 | |
| *** sgotliv has joined #openstack-zaqar | 22:12 | |
| *** openstackgerrit has quit IRC | 22:38 | |
| *** openstackgerrit_ has joined #openstack-zaqar | 22:38 | |
| *** openstackgerrit_ is now known as openstackgerrit | 22:40 | |
| *** cpallares has quit IRC | 23:04 | |
| *** boris-42 has joined #openstack-zaqar | 23:38 | |
| *** sriram1 has joined #openstack-zaqar | 23:41 | |
| *** flwang1 has joined #openstack-zaqar | 23:43 | |
| *** jeffrey4l_ has joined #openstack-zaqar | 23:55 | |
| *** kgriffs is now known as kgriffs|afk | 23:55 | |
| *** yograterol has quit IRC | 23:57 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!