| *** dshulyak_ has joined #openstack-solar | 07:56 | |
| *** salmon_ has joined #openstack-solar | 08:33 | |
| openstackgerrit | Dmitry Shulyak proposed openstack/solar: Add init script for custom worker https://review.openstack.org/270665 | 08:44 |
|---|---|---|
| openstackgerrit | Lukasz Oles proposed openstack/solar: Use exported env name, change default name https://review.openstack.org/270679 | 09:17 |
| pigmej | sup guys? | 09:22 |
| bogdando | folks, where is the "gate-solar-docs" script? | 09:26 |
| bogdando | it calls "tox -evenv -- python setup.py build_sphinx" I need it as tox -edocs instead | 09:27 |
| salmon_ | bogdando: here https://github.com/openstack-infra/project-config/blob/master/jenkins/jobs/python-jobs.yaml#L250 :) | 09:28 |
| salmon_ | it's not defined by us | 09:28 |
| openstackgerrit | Lukasz Oles proposed openstack/solar: Use exported env name, change default name https://review.openstack.org/270679 | 09:43 |
| salmon_ | guys, please take a look at this https://review.fuel-infra.org/#/c/16320/2 it adds jenkins job to run example test | 09:45 |
| pigmej | I looked | 09:55 |
| *** openstackgerrit has quit IRC | 10:02 | |
| *** openstackgerrit has joined #openstack-solar | 10:02 | |
| bogdando | salmon_, https://review.openstack.org/#/c/270698 | 10:05 |
| salmon_ | thx | 10:05 |
| openstackgerrit | Bogdan Dobrelya proposed openstack/solar: Add tox checkniceness,checksyntax for docs https://review.openstack.org/269559 | 10:06 |
| pigmej | https://review.openstack.org/#/c/269653/ + https://review.openstack.org/#/c/270247/ review it guys | 10:07 |
| salmon_ | pigmej: does this two depends on each other? | 10:20 |
| pigmej | yup | 10:20 |
| pigmej | 208 => 247 | 10:20 |
| pigmej | 247 depends on 208 | 10:20 |
| pigmej | 208 is simple it just adds one exception in 'else' | 10:21 |
| pigmej | 247 uses this exception ;) | 10:21 |
| *** mihgen has quit IRC | 10:29 | |
| *** mihgen has joined #openstack-solar | 10:39 | |
| openstackgerrit | Jedrzej Nowak proposed openstack/solar: Unified exception in repo create https://review.openstack.org/270208 | 11:06 |
| pigmej | dshulyak_: check this again, salmon wanted to change exception string :) | 11:06 |
| openstackgerrit | Jedrzej Nowak proposed openstack/solar: Unified exception in repo create https://review.openstack.org/270208 | 11:28 |
| pigmej | wrrr, pep8 dshulyak_ salmon_ I fixed that pep8.... | 11:28 |
| bogdando | salmon_, tha infra patch throws Job gate-solar-tox-checkniceness not defined | 11:55 |
| bogdando | Job gate-solar-tox-checksyntax not defined | 11:55 |
| bogdando | I'm totally confused now, which places are to define gates | 11:55 |
| salmon_ | bogdando: no idea, it's better to ask them on IRC | 11:55 |
| bogdando | let's at least accept my patch and later think about making a check gate for it as well | 11:55 |
| dshulyak_ | catched siblings error | 11:59 |
| dshulyak_ | SiblingsError: No empty object for resolution not sure what to do with siblings: ['{"status": "INPROGRESS", "task_type": "solar_resource", "target": "", "args": ["data_resource_2_6", "run"], "timelimit": 0, "childs": ["system_log:a377dc6b-f3df-4002-b1a4-0409d7eac130~data_resource_3_6.run"], "parents": ["system_log:a377dc6b-f3df-4002-b1a4-0409d7eac130~data_resource_1_6.run"], "execution": "system_log:a377dc6b-f3df-4002-b1a4-0409d7eac130", "errmsg": "" | 11:59 |
| dshulyak_ | "name": "data_resource_2_6.run"}', '{"status": "PENDING", "task_type": "solar_resource", "target": "", "args": ["data_resource_2_6", "run"], "timelimit": 0, "childs": ["system_log:a377dc6b-f3df-4002-b1a4-0409d7eac130~data_resource_3_6.run"], "parents": ["system_log:a377dc6b-f3df-4002-b1a4-0409d7eac130~data_resource_1_6.run"], "execution": "system_log:a377dc6b-f3df-4002-b1a4-0409d7eac130", "errmsg": "", "name": "data_resource_2_6.run"}'] | 11:59 |
| pigmej | dshulyak_: on what ? | 11:59 |
| dshulyak_ | i have test | 11:59 |
| dshulyak_ | https://review.openstack.org/#/c/269166/10/solar/test/functional/test_complete_solar_workflow.py | 12:00 |
| dshulyak_ | but i also changed zeromq puller implementation | 12:00 |
| pigmej | does you have concurrency there ? | 12:01 |
| pigmej | do you have* | 12:01 |
| dshulyak_ | yeah, scale in that tests represents number of concurrent chains | 12:01 |
| dshulyak_ | and in puller as well | 12:01 |
| dshulyak_ | lock spams like crazy :) | 12:02 |
| pigmej | it would mean that lock is not working properly OR some other issue ... | 12:02 |
| pigmej | https://bpaste.net/show/d9cd2b8ff933 | 12:03 |
| pigmej | dshulyak_: ^ | 12:03 |
| pigmej | it should basically print always list with the same value on every index | 12:04 |
| pigmej | https://github.com/Mirantis/solar-resources/pull/5 | 12:08 |
| openstackgerrit | Dmitry Shulyak proposed openstack/solar: Add implementation of puller with pool of threads and tracking of tasks https://review.openstack.org/270765 | 12:25 |
| openstackgerrit | Dmitry Shulyak proposed openstack/solar: Implement timelimit mechanism based on tracking of tasks https://review.openstack.org/270766 | 12:25 |
| pigmej | dshulyak_: can we focus on release today ? please ? | 12:26 |
| pigmej | :) | 12:26 |
| dshulyak_ | y, i’m not asking to debug it | 12:33 |
| pigmej | dshulyak_: yeah I know, but we need to document stuff still | 12:36 |
| dshulyak_ | which stuff? | 12:36 |
| pigmej | configuration options for example | 12:36 |
| dshulyak_ | for celery :) ? | 12:37 |
| pigmej | and we have 4 pending change sets | 12:37 |
| pigmej | dshulyak_: yeah | 12:37 |
| dshulyak_ | oh, nobody cares about that | 12:37 |
| salmon_ | about what? | 12:38 |
| pigmej | we have also undocummented "how to run solar without vagrant" | 12:38 |
| pigmej | ( salmon_ has it in TODO though, isn't it ?) | 12:38 |
| salmon_ | pigmej: which part? | 12:39 |
| pigmej | solar without vagrant | 12:39 |
| salmon_ | nope | 12:39 |
| pigmej | you said that you will test it and add to readmes :) | 12:39 |
| salmon_ | ah, I will test it :) | 12:39 |
| pigmej | you said that you will add it to readme :P | 12:39 |
| salmon_ | ok, I remember ;) | 12:40 |
| openstackgerrit | Merged openstack/solar: Use isolated filebased sqlite for each test process https://review.openstack.org/270181 | 12:40 |
| pigmej | https://github.com/Mirantis/solar-resources/pull/6 | 12:42 |
| pigmej | we need to improve and note things that we can change / adjust etc, because otherwise there will be flame war, why the fuck do you use sqlite for celery not redis, etc. | 12:45 |
| salmon_ | pigmej, please add such notes in this etherpads | 12:45 |
| salmon_ | https://etherpad.openstack.org/p/os-dev-annoucement-solar-0.0.1 | 12:45 |
| salmon_ | https://etherpad.openstack.org/p/mos-dev-annoucement-solar-0.0.1 | 12:45 |
| pigmej | what do you mean? | 12:46 |
| pigmej | I'm now adding simple FAQ to our documentation (about key/password fuckup) | 12:46 |
| salmon_ | ah, ok. FAQ is better | 12:46 |
| salmon_ | can someone merge it https://review.openstack.org/#/c/270062/ ? | 12:52 |
| pigmej | I add +2 yesterday :P | 12:55 |
| pigmej | added* | 12:55 |
| openstackgerrit | Merged openstack/solar: Unified exception in repo create https://review.openstack.org/270208 | 13:00 |
| dshulyak_ | pigmej: i’m seeing this error with hosts_file example - http://paste.openstack.org/show/484562/ | 13:01 |
| dshulyak_ | it is with unified results | 13:02 |
| pigmej | yeah because you wanted None there | 13:02 |
| pigmej | isn't it ? | 13:03 |
| dshulyak_ | aha, but rsync returns None None None even if everything is ok | 13:03 |
| pigmej | ah, ok I will check it asap then, because it shouldn't | 13:03 |
| pigmej | weird | 13:04 |
| openstackgerrit | Lukasz Oles proposed openstack/solar: Set correct classifier https://review.openstack.org/270785 | 13:05 |
| salmon_ | required for pypi^ | 13:05 |
| salmon_ | afk,lunch | 13:05 |
| pigmej | ah ok I see the fuckup dshulyak_ :) | 13:05 |
| openstackgerrit | Jedrzej Nowak proposed openstack/solar: Unify returned transport results https://review.openstack.org/269653 | 13:12 |
| pigmej | dshulyak_: ^ | 13:12 |
| openstackgerrit | Jedrzej Nowak proposed openstack/solar: Added EGroup click group which can handle errors https://review.openstack.org/270247 | 13:13 |
| pigmej | ^ rebased | 13:13 |
| openstackgerrit | Jedrzej Nowak proposed openstack/solar: Unify returned transport results https://review.openstack.org/269653 | 13:17 |
| dshulyak_ | pigmej: you have prints there :) https://review.openstack.org/#/c/269653/6/solar/core/transports/base.py | 13:20 |
| pigmej | dshulyak_: removed already :D | 13:20 |
| pigmej | in next changeset :) | 13:20 |
| dshulyak_ | now error is nicer | 13:20 |
| dshulyak_ | hosts_file2.run -> ERROR :: ssh: connect to host 10.0.0.4 port 22: No route to host | 13:20 |
| dshulyak_ | rsync: connection unexpectedly closed (0 bytes received so far) [sender] | 13:20 |
| dshulyak_ | rsync error: unexplained error (code 255) at io.c(226) [sender=3.1.0] | 13:21 |
| dshulyak_ | hosts_file1.run -> ERROR :: ssh: connect to host 10.0.0.3 port 22: No route to host | 13:21 |
| dshulyak_ | rsync: connection unexpectedly closed (0 bytes received so far) [sender] | 13:21 |
| dshulyak_ | rsync error: unexplained error (code 255) at io.c(226) [sender=3.1.0] | 13:21 |
| dshulyak_ | oh | 13:21 |
| pigmej | :) | 13:21 |
| openstackgerrit | Jedrzej Nowak proposed openstack/solar: Added simple FAQ to our docs https://review.openstack.org/270791 | 13:24 |
| pigmej | brb dinner is there :) | 13:43 |
| openstackgerrit | Dmitry Shulyak proposed openstack/solar: Add documentation about celery configuration options https://review.openstack.org/270806 | 13:48 |
| dshulyak_ | hm, i will abandon my patch, that faq is more than enough | 13:48 |
| salmon_ | dshulyak_: workflow +1 ;) https://review.openstack.org/#/c/270062/2 | 14:12 |
| openstackgerrit | Merged openstack/solar: Move all text from README to documentation https://review.openstack.org/270062 | 14:15 |
| salmon_ | thx | 14:16 |
| openstackgerrit | Lukasz Oles proposed openstack/solar: Remove last `/` from urls https://review.openstack.org/270825 | 14:23 |
| openstackgerrit | Jedrzej Nowak proposed openstack/solar: Added simple FAQ to our docs https://review.openstack.org/270791 | 14:26 |
| pigmej | salmon_: changed what you wanted | 14:26 |
| pigmej | salmon_: https://review.openstack.org/#/c/269653/ workflow please | 14:27 |
| pigmej | dshulyak_ check this please https://review.openstack.org/#/c/270247/ | 14:27 |
| openstackgerrit | Merged openstack/solar: Added EGroup click group which can handle errors https://review.openstack.org/270247 | 14:34 |
| salmon_ | pigmej: https://review.openstack.org/#/c/270679/ | 14:39 |
| salmon_ | and https://review.openstack.org/#/c/270785/ | 14:39 |
| salmon_ | and all https://review.openstack.org/#/c/270825/ | 14:40 |
| pigmej | salmon_: yeah doing it, we already discussed some of them ;P | 14:40 |
| pigmej | salmon_: should we link to 'latest' ? | 14:41 |
| pigmej | I missclicked that one :D | 14:42 |
| salmon_ | for now we don't have any other | 14:44 |
| pigmej | awesome! | 14:46 |
| pigmej | I wonder wtf https://review.openstack.org/#/c/267453/ | 14:46 |
| pigmej | salmon_: you're moving all stuf on etherpad ? | 14:47 |
| salmon_ | yup | 14:48 |
| openstackgerrit | Dmitry Shulyak proposed openstack/solar: Zerorpc worker for orchestration modules https://review.openstack.org/269166 | 14:48 |
| openstackgerrit | Dmitry Shulyak proposed openstack/solar: Add implementation of puller with pool of threads and tracking of tasks https://review.openstack.org/270765 | 14:51 |
| openstackgerrit | Merged openstack/solar: Use exported env name, change default name https://review.openstack.org/270679 | 14:51 |
| openstackgerrit | Merged openstack/solar: Set correct classifier https://review.openstack.org/270785 | 14:52 |
| openstackgerrit | Merged openstack/solar: Unify returned transport results https://review.openstack.org/269653 | 14:54 |
| openstackgerrit | Merged openstack/solar: Add tox checkniceness,checksyntax for docs https://review.openstack.org/269559 | 14:54 |
| pigmej | dshulyak_: https://github.com/Mirantis/solar-resources/pulls | 14:54 |
| pigmej | salmon_: dshulyak_ I cannot fix / improve "parse puppet log from STDOUT" because I lack puppet knowlegde now, or do you have some resources where I can check what can I expect on stdout ? | 14:57 |
| salmon_ | hmm, what do we need to do here? I thought each puppet resource has logs in separate file | 14:58 |
| dshulyak_ | and also in stdout | 14:58 |
| dshulyak_ | but idont remember what it writes to stdout and what to stderr | 14:58 |
| salmon_ | btw. if we are requiring sqlalchemy now can we drop peewee? | 15:01 |
| dshulyak_ | maybe someone have solar log with puppet task? | 15:01 |
| pigmej | salmon_: sqlalchemy is "BIG" and we require it only for celery ' | 15:02 |
| pigmej | so the answer is no :) | 15:02 |
| salmon_ | ah, ok | 15:02 |
| salmon_ | dshulyak_: https://review.openstack.org/#/c/270825/ https://review.openstack.org/#/c/270791/ | 15:03 |
| salmon_ | last two commits required for release | 15:03 |
| dshulyak_ | i think its ok to merge small patches wihtout 2 reviews | 15:04 |
| pigmej | I think the same | 15:04 |
| salmon_ | rules breakers! | 15:04 |
| pigmej | because we're overflowed otherwise | 15:04 |
| salmon_ | but I agree :) | 15:04 |
| pigmej | and now define 'small' :) | 15:04 |
| salmon_ | hmm, where is jenkins? | 15:04 |
| pigmej | salmon_: it has a lag now | 15:04 |
| salmon_ | ;( | 15:04 |
| pigmej | but hmm | 15:05 |
| pigmej | https://review.openstack.org/#/c/270791/ this looks weird | 15:05 |
| salmon_ | yup | 15:05 |
| openstackgerrit | Lukasz Oles proposed openstack/solar: Added simple FAQ to our docs https://review.openstack.org/270791 | 15:05 |
| salmon_ | I rebased it | 15:05 |
| salmon_ | do we need anything else to commit/merge? Except maybe some docs fixes? | 15:06 |
| pigmej | probably not | 15:07 |
| pigmej | but I will check all thigns again | 15:07 |
| salmon_ | ok, so let's test it :) | 15:07 |
| salmon_ | I will try wordpress example without vagrant | 15:08 |
| pigmej | remember about running celery ;) | 15:08 |
| salmon_ | yeah | 15:08 |
| salmon_ | and please update this https://etherpad.openstack.org/p/solar-release-0.1.0 | 15:10 |
| openstackgerrit | Merged openstack/solar: Remove last `/` from urls https://review.openstack.org/270825 | 15:14 |
| pigmej | salmon_: it is updated | 15:14 |
| salmon_ | good :) | 15:15 |
| pigmej | salmon_: https://review.openstack.org/#/c/270791/ wtf is there ? | 15:22 |
| pigmej | I will rebase in on my local machine | 15:23 |
| openstackgerrit | Jedrzej Nowak proposed openstack/solar: Added simple FAQ to our docs https://review.openstack.org/270791 | 15:23 |
| pigmej | rebased salmon_ | 15:24 |
| salmon_ | it's in a queue | 15:24 |
| salmon_ | http://status.openstack.org/zuul/ | 15:24 |
| pigmej | now it it | 15:24 |
| pigmej | it is | 15:24 |
| pigmej | salmon_: just remember to run celery from the same user as cli commands | 15:26 |
| salmon_ | I know | 15:30 |
| dshulyak_ | altered test a bit - http://paste.openstack.org/show/484585/ | 15:31 |
| pigmej | dshulyak_: it's incorrect | 15:32 |
| dshulyak_ | why? | 15:32 |
| pigmej | all items should be equal | 15:32 |
| pigmej | ah you have retries... | 15:33 |
| dshulyak_ | no | 15:33 |
| pigmej | but I'm not sure if you will be able to get all locks | 15:33 |
| pigmej | within time | 15:33 |
| dshulyak_ | if im not able to get lock - then lock should fail, isnt it? | 15:33 |
| pigmej | yeah, but it will be not equal range(100) then | 15:33 |
| dshulyak_ | yes, but i what i see is that sometimes there is 2 lockers with 1 | 15:34 |
| pigmej | perfect | 15:34 |
| dshulyak_ | no :) | 15:34 |
| pigmej | why? | 15:34 |
| pigmej | it's perfectly fine if you see multiple items | 15:34 |
| dshulyak_ | 18, 20, 21, 21, | 15:34 |
| dshulyak_ | 21 and 19 | 15:34 |
| pigmej | and ? | 15:34 |
| pigmej | dshulyak_: you coded that lock in a bit different way.. :) | 15:34 |
| pigmej | who_is_locking determines who owns lock | 15:35 |
| dshulyak_ | and 19 thought that it acquired lock | 15:35 |
| pigmej | hmm, how ? | 15:35 |
| dshulyak_ | but in fact it was acquired by 21 | 15:35 |
| pigmej | do you have who_is_locking() == 19 ? | 15:35 |
| dshulyak_ | thats exactly the problem | 15:36 |
| pigmej | your acquire doesn't break when acquire fails, it returns `lk` | 15:36 |
| pigmej | and at the end you need to check if lk.who_is_locking() == me | 15:36 |
| dshulyak_ | but i shouldnt enter context manager at all | 15:36 |
| dshulyak_ | it should fail with exception or acquire lock | 15:36 |
| dshulyak_ | in __enter__ | 15:36 |
| pigmej | hmm | 15:36 |
| pigmej | ah | 15:37 |
| pigmej | why clear_cache ? | 15:37 |
| dshulyak_ | on range of 100 - there 3 cases like this | 15:37 |
| dshulyak_ | but still my worker fails even if i have concurrency > 2 :( | 15:37 |
| pigmej | dshulyak_: you shouldn't clean cache between __enter__ and __exit__ that's for sure | 15:38 |
| pigmej | (it may be noop though for this situation but still) | 15:38 |
| dshulyak_ | but that just means that 19 and 21 will simutaneously acquire a lock, no? | 15:41 |
| dshulyak_ | if they entered | 15:41 |
| pigmej | generally who_is_locking should determine what's going on | 15:41 |
| pigmej | but for me they don't enter | 15:41 |
| pigmej | or hmm | 15:42 |
| pigmej | wait it's hard to intepret the results;D | 15:42 |
| pigmej | hmm | 15:43 |
| pigmej | ok, BUT, why you lock again inside ? | 15:43 |
| pigmej | yeah, you're breaking it | 15:44 |
| pigmej | dshulyak_: return lk in __enter__ | 15:45 |
| pigmej | and with RiakLock(...) as l: | 15:45 |
| pigmej | and remove your additional acquire | 15:45 |
| pigmej | https://bpaste.net/show/f0b5bddc1014 | 15:45 |
| pigmej | works for me | 15:45 |
| dshulyak_ | but thats doesnt tell you if there is 2 concurrent locks | 15:46 |
| pigmej | it does | 15:46 |
| pigmej | p.map() is parallel | 15:46 |
| pigmej | you're acquiring one lock twice within one loop, and you're breaking release method probably, | 15:46 |
| pigmej | aka: untested behaviour | 15:46 |
| dshulyak_ | but i shouldnt even enter this block :) | 15:48 |
| pigmej | it enters it, and it's correct | 15:48 |
| pigmej | but THEN you do additional acquire | 15:48 |
| pigmej | which breaks everything | 15:48 |
| openstackgerrit | Merged openstack/solar: Added simple FAQ to our docs https://review.openstack.org/270791 | 15:48 |
| pigmej | (at least it seems so dshulyak_ ) | 15:49 |
| pigmej | and because you use different stamp, you're changing lock ownership | 15:49 |
| dshulyak_ | i changed it a bit, result is the same - http://paste.openstack.org/show/484589/ | 15:50 |
| dshulyak_ | ok, lets leave it | 15:51 |
| dshulyak_ | i will do some additional debuging | 15:51 |
| pigmej | you cannot do it dshulyak_ like that I think | 15:51 |
| pigmej | you're destroying lock ;D | 15:51 |
| pigmej | remove that .get() | 15:51 |
| pigmej | and add return in __enter__ | 15:51 |
| dshulyak_ | ok, i will use separate object for tracking collision | 15:51 |
| pigmej | ? | 15:52 |
| pigmej | https://bpaste.net/show/6a5dc73f3906 | 15:52 |
| dshulyak_ | but afaiu there should be no 2 lockers with 1, and no -1 | 15:52 |
| pigmej | ? | 15:52 |
| pigmej | with that modification that I did, I have equality in res == range(100) | 15:53 |
| pigmej | I executed it 50 times and 50 times equality | 15:53 |
| pigmej | dshulyak_: what probably happens, when you _acquire lock again (or get it), you alter behaviour of __release__ | 15:54 |
| pigmej | because different state will be saved now | 15:54 |
| pigmej | and weird things happens :D | 15:54 |
| pigmej | at least I think it happens... | 15:56 |
| salmon_ | /usr/local/bin/celery multi stopwait 2 :( | 16:11 |
| pigmej | salmon_: you can use service solar-celery start|stop | 16:12 |
| salmon_ | does it require full path? | 16:12 |
| pigmej | ah fuck | 16:12 |
| pigmej | yeah it should say 'celery' instead :D | 16:12 |
| pigmej | /usr/bin/env celery ;) | 16:13 |
| pigmej | and we're fine | 16:13 |
| salmon_ | ImportError: No module named gevent :( | 16:13 |
| salmon_ | I think we require gevent now :) | 16:13 |
| pigmej | no | 16:13 |
| pigmej | you need to execute worker correctly | 16:13 |
| salmon_ | ./utils/solar-celery start is not correct? :P | 16:14 |
| pigmej | no because it requires gevent :D | 16:14 |
| pigmej | because it has -P gevent | 16:14 |
| salmon_ | so we also require a gevent :P | 16:14 |
| pigmej | salmon_: no, you can prefork | 16:14 |
| pigmej | salmon_: /usr/bin/env celery start 1 -A solar.orchestration.runner -P prefork -c 2 -Q system_log,celery,scheduler --pidfile=/var/run/celery/%N.pid --logfile=/var/run/celery/%N.log | 16:16 |
| pigmej | or something like that | 16:16 |
| salmon_ | pfff | 16:16 |
| salmon_ | you recommend solar-celery scrip in FAQ! | 16:16 |
| pigmej | salmon_: I'm extending that entry then | 16:18 |
| salmon_ | ok | 16:19 |
| *** salmon_ has quit IRC | 16:37 | |
| openstackgerrit | Jedrzej Nowak proposed openstack/solar: Improved celery worker section in FAQ https://review.openstack.org/270916 | 16:49 |
| openstackgerrit | Jedrzej Nowak proposed openstack/solar: use /usr/bin/env in celery script https://review.openstack.org/270917 | 16:50 |
| *** salmon_ has joined #openstack-solar | 17:28 | |
| dshulyak_ | pigmej: looks like my test was catching intermediate state | 17:53 |
| pigmej | dshulyak_: you mean? | 19:41 |
| pigmej | when you _acquire three is local state involved, which *could* be wrong (you used different stamp) | 19:41 |
| dshulyak_ | this one log.debug("I was not locking, so removing me %s" % identity) | 19:42 |
| pigmej | this is ok, in fact | 19:42 |
| pigmej | I mean it's not bad that it reaches that stage | 19:42 |
| pigmej | but that additional _acquire for sure changes semantic, I don't know if it's good or bad though ;D | 19:43 |
| pigmej | btw you're still working? | 19:43 |
| dshulyak_ | i think that this Lock.get happened before this write, thats why i was able to get this illusion of conflict | 19:43 |
| pigmej | posssible | 19:47 |
| *** salmon_ has quit IRC | 19:51 | |
| *** salmon_ has joined #openstack-solar | 20:42 | |
| openstackgerrit | Merged openstack/solar: Improved celery worker section in FAQ https://review.openstack.org/270916 | 20:51 |
| openstackgerrit | Merged openstack/solar: use /usr/bin/env in celery script https://review.openstack.org/270917 | 20:51 |
| *** dshulyak_ has quit IRC | 21:18 | |
| *** dshulyak_ has joined #openstack-solar | 21:26 | |
| openstackgerrit | Lukasz Oles proposed openstack/solar: Fix celery start script https://review.openstack.org/271059 | 22:20 |
| *** dshulyak_ has quit IRC | 22:22 | |
| openstackgerrit | Merged openstack/solar: Fix celery start script https://review.openstack.org/271059 | 22:41 |
| *** salmon_ has quit IRC | 23:02 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!