| *** rrecio_ has joined #openstack-mistral | 00:04 | |
| *** toddjohn has joined #openstack-mistral | 00:15 | |
| *** rrecio_ has quit IRC | 00:16 | |
| *** toddjohn has quit IRC | 00:20 | |
| *** toddjohn has joined #openstack-mistral | 01:47 | |
| *** toddjohn has quit IRC | 01:51 | |
| *** toddjohn has joined #openstack-mistral | 02:48 | |
| rakhmerov | hparekh, kong, ddeja, akuznetsova: guys, please review https://review.openstack.org/#/c/352075, it's pretty urgent because it blocks client development | 02:48 |
|---|---|---|
| *** toddjohn has quit IRC | 02:52 | |
| *** FL1SK has quit IRC | 03:03 | |
| *** chlong has quit IRC | 03:25 | |
| *** toddjohn has joined #openstack-mistral | 03:50 | |
| *** toddjohn has quit IRC | 03:54 | |
| openstackgerrit | Renat Akhmerov proposed openstack/mistral: Fix GET /executions/<id> to init 'output' attribute explicitly https://review.openstack.org/352207 | 04:05 |
| *** jaosorior has joined #openstack-mistral | 04:14 | |
| rakhmerov | and https://review.openstack.org/#/c/352207/ too | 04:16 |
| openstackgerrit | Renat Akhmerov proposed openstack/mistral: Towards non-locking model: adapt 'join' tasks to work w/o locks https://review.openstack.org/351727 | 04:16 |
| rakhmerov | hparekh: hi, can you please also look at https://review.openstack.org/#/c/352207/ | 04:17 |
| rakhmerov | devstack tests are failing because of that too | 04:17 |
| hparekh | rakhmerov, sure | 04:17 |
| rakhmerov | thanks, it's super simple | 04:17 |
| openstackgerrit | Merged openstack/mistral: fix for get action executions fails with "has no property 'type" https://review.openstack.org/352075 | 04:27 |
| *** Ravikiran_K has joined #openstack-mistral | 04:29 | |
| hparekh | rakhmerov, Do we need database transaction ? I think it is ok if we don't use database transaction | 04:31 |
| rakhmerov | hparekh: no-no, we do need it | 04:31 |
| rakhmerov | because this field is loaded lazily | 04:31 |
| rakhmerov | both operations must happen within the same SQLAlchemy sessions | 04:32 |
| rakhmerov | I also just filed another bug that's related to this: https://bugs.launchpad.net/mistral/+bug/1610817 | 04:32 |
| openstack | Launchpad bug 1610817 in Mistral "GET /action_executions should not include 'output' field" [High,New] - Assigned to Michal Gershenzon (michal-gershenzon) | 04:32 |
| rakhmerov | and it has a similar problem | 04:32 |
| hparekh | rakhmerov, Oh ok. Then I think we need same in mistral/api/controllers/v2/action_execution.py | 04:33 |
| rakhmerov | yeah | 04:33 |
| rakhmerov | that bug has info about it :) | 04:33 |
| rakhmerov | hparekh: so, we always need to use transactions of we have any kind of lazy loading, whether it's a collection or a deferred field | 04:34 |
| rakhmerov | in practice though it works but not always | 04:34 |
| rakhmerov | I don't know why | 04:34 |
| rakhmerov | I observe both behaviors | 04:34 |
| rakhmerov | but in theory it must never work if we load a collection or deferred field in a separate transaction, seems like we have another bug somewhere that we are not aware of yet | 04:35 |
| rakhmerov | because sometimes it actually works | 04:35 |
| hparekh | rakhmerov, hmmm.. Thanks | 04:36 |
| rakhmerov | you're still in doubts? | 04:37 |
| rakhmerov | :) | 04:37 |
| hparekh | rakhmerov, No doubts, I understand your point. :) | 04:37 |
| rakhmerov | ok | 04:37 |
| rakhmerov | yeah, I just don't know yet why it works sometimes | 04:37 |
| rakhmerov | as if a transaction (and hence SQLAlchemy session) was ended after "with db_api.transaction():" block | 04:38 |
| rakhmerov | have no idea how it is possible | 04:38 |
| *** tonytan4ever has quit IRC | 04:39 | |
| *** toddjohn has joined #openstack-mistral | 04:50 | |
| hparekh | ddeja, kong , akuznetsova If you have free time please review https://review.openstack.org/#/c/346814/ | 04:51 |
| kong | hparekh: ok | 04:52 |
| * kong is looking | 04:52 | |
| hparekh | thanks | 04:54 |
| *** toddjohn has quit IRC | 04:55 | |
| kong | hparekh: left my comment after a quick review | 04:57 |
| *** FL1SK has joined #openstack-mistral | 05:03 | |
| *** janki has joined #openstack-mistral | 05:06 | |
| openstackgerrit | Renat Akhmerov proposed openstack/mistral: Towards non-locking model: adapt 'join' tasks to work w/o locks https://review.openstack.org/351727 | 05:10 |
| openstackgerrit | Merged openstack/mistral: Fix GET /executions/<id> to init 'output' attribute explicitly https://review.openstack.org/352207 | 05:10 |
| *** chlong has joined #openstack-mistral | 05:59 | |
| hparekh | kong, rakhmerov, Please look at https://bugs.launchpad.net/mistral/+bug/1609335 and https://blueprints.launchpad.net/mistral/+spec/mistral-retry-continue-on | 06:21 |
| openstack | Launchpad bug 1609335 in Mistral "Retry get executed even if action is success" [Undecided,New] - Assigned to Lucky samadhiya (lucky-samadhiya) | 06:21 |
| hparekh | We may need some discussion regarding this | 06:21 |
| hparekh | this feature is little bit confusing | 06:22 |
| *** toddjohn has joined #openstack-mistral | 06:52 | |
| *** toddjohn has quit IRC | 06:56 | |
| rakhmerov | yes, ok | 07:14 |
| *** sshnaidm has joined #openstack-mistral | 07:15 | |
| sshnaidm | hi, I'd appreciate your attention to bug (in case it's missed) https://bugs.launchpad.net/mistral/+bug/1610269 (db missing the 'unique_key' column for delayed_calls_v2 table) which is affecting TripleO periodic jobs now, thanks | 07:18 |
| openstack | Launchpad bug 1610269 in Mistral "db missing the 'unique_key' column for delayed_calls_v2 table" [Undecided,New] | 07:18 |
| *** FL1SK has quit IRC | 07:21 | |
| *** jpich has joined #openstack-mistral | 07:36 | |
| *** nmakhotkin has joined #openstack-mistral | 08:02 | |
| *** jtomasek has quit IRC | 08:07 | |
| *** jistr|mtg is now known as jistr | 08:50 | |
| *** toddjohn has joined #openstack-mistral | 08:53 | |
| *** toddjohn has quit IRC | 08:58 | |
| rakhmerov | sshnaidm: I already told jaosorior that it's currently a work-in-progress thing and DB migration is now missing for this column | 09:11 |
| rakhmerov | should be made today | 09:11 |
| rakhmerov | so please be careful when updating from master | 09:11 |
| sshnaidm | rakhmerov, thanks | 09:12 |
| *** FL1SK has joined #openstack-mistral | 09:19 | |
| d0ugal | rakhmerov: Is there anything I can do to help with it? | 10:08 |
| rakhmerov | d0ugal: thanks, it's actually already in progress | 10:09 |
| d0ugal | okay, cool :) | 10:09 |
| d0ugal | Thanks! | 10:10 |
| d0ugal | rakhmerov: Which patch is it? | 10:11 |
| rakhmerov | no patch yet, but there will be soon | 10:11 |
| d0ugal | k | 10:11 |
| rakhmerov | Michal is working on it | 10:12 |
| rakhmerov | I mean mgershen | 10:12 |
| d0ugal | okay, cool | 10:12 |
| rakhmerov | yep | 10:12 |
| d0ugal | I'll look out for it :) | 10:12 |
| *** sshnaidm has quit IRC | 10:20 | |
| *** sshnaidm has joined #openstack-mistral | 10:40 | |
| *** sshnaidm is now known as sshnaidm|afk | 10:40 | |
| *** Ravikiran_K has quit IRC | 10:40 | |
| openstackgerrit | Merged openstack/mistral: Fix past migration scripts discrepancies https://review.openstack.org/352091 | 10:43 |
| *** toddjohn has joined #openstack-mistral | 10:54 | |
| *** toddjohn has quit IRC | 10:59 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/mistral: Updated from global requirements https://review.openstack.org/352349 | 11:03 |
| openstackgerrit | prince_katiyar proposed openstack/python-mistralclient: Add error message when OS_USERNAME or OS_PASSWORD not provided https://review.openstack.org/348142 | 11:13 |
| *** jaosorior has quit IRC | 11:27 | |
| *** sshnaidm|afk is now known as sshnaidm | 11:28 | |
| *** jaosorior has joined #openstack-mistral | 11:28 | |
| *** toddjohn has joined #openstack-mistral | 11:48 | |
| openstackgerrit | Michal Gershenzon proposed openstack/mistral: Add unique keys for non locking model https://review.openstack.org/352376 | 11:57 |
| mgershen | d0ugal: here it is ^ | 11:59 |
| mgershen | sorry for keeping you waiting | 11:59 |
| *** dprince has joined #openstack-mistral | 12:00 | |
| d0ugal | mgershen: no problem, thanks - I'll take a look! | 12:03 |
| *** toddjohn has quit IRC | 12:06 | |
| *** toddjohn has joined #openstack-mistral | 12:07 | |
| *** toddjohn has quit IRC | 12:11 | |
| openstackgerrit | Renat Akhmerov proposed openstack/mistral: Towards non-locking model: adapt 'join' tasks to work w/o locks https://review.openstack.org/351727 | 12:19 |
| openstackgerrit | Renat Akhmerov proposed openstack/mistral: Towards non-locking model: make 'with-items' work w/o locks https://review.openstack.org/352386 | 12:19 |
| openstackgerrit | Renat Akhmerov proposed openstack/mistral: Towards non-locking model: make 'with-items' work w/o locks https://review.openstack.org/352386 | 12:25 |
| openstackgerrit | Renat Akhmerov proposed openstack/mistral: Towards non-locking model: adapt 'join' tasks to work w/o locks https://review.openstack.org/351727 | 12:25 |
| openstackgerrit | Merged openstack/mistral: Updated from global requirements https://review.openstack.org/352349 | 12:27 |
| rakhmerov | ddeja: hi Dawid, I addressed your comments in https://review.openstack.org/#/c/351727/ | 12:27 |
| ddeja | hi rakhmerov | 12:36 |
| ddeja | looking | 12:36 |
| rakhmerov | ok ) | 12:36 |
| rakhmerov | you can also ask me here if needed | 12:37 |
| rakhmerov | the changes are not trivial | 12:37 |
| ddeja | about the key = 'th_c_t_s_a-%s' % task_ex.id | 12:37 |
| rakhmerov | yes | 12:37 |
| rakhmerov | it's purely internal data | 12:37 |
| rakhmerov | it will never be exposed to users | 12:38 |
| ddeja | I'm OK with reducing space in the DB | 12:38 |
| rakhmerov | ok | 12:38 |
| ddeja | and I understand it is in only visible from the codebase or database | 12:38 |
| ddeja | can you just add comment for this? | 12:38 |
| rakhmerov | yeah, it just should be enough for debugging | 12:38 |
| *** Ravikiran_K has joined #openstack-mistral | 12:38 | |
| rakhmerov | I can, yes | 12:38 |
| rakhmerov | ok in a separate patch? ;) | 12:39 |
| ddeja | sure | 12:39 |
| rakhmerov | ok | 12:39 |
| rakhmerov | I'll make a note | 12:39 |
| ddeja | someone just might go there after a year and spend some time wondering 'what kind of language is this' ;) | 12:39 |
| rakhmerov | :))) | 12:39 |
| ddeja | oh, and since we are talking | 12:40 |
| ddeja | I have some questions about the insert_or_ignore method | 12:40 |
| rakhmerov | yes | 12:40 |
| rakhmerov | go ahead | 12:41 |
| ddeja | I don't like that it hack a little the sql alchemy | 12:41 |
| ddeja | and depends on the dialect value | 12:41 |
| ddeja | isn't there any better way of achiving this? | 12:42 |
| ddeja | I mean, some internal mechanism of SQLAlchemy | 12:42 |
| rakhmerov | I didn't find such a way yet | 12:42 |
| rakhmerov | unfortunately | 12:42 |
| rakhmerov | I don't like this hack too but I spent several days trying to find a better solution | 12:42 |
| rakhmerov | nothing | 12:43 |
| ddeja | OK then | 12:43 |
| rakhmerov | the key problem here is that we need this kind of conflict check at DB level | 12:43 |
| rakhmerov | it must be atomic | 12:43 |
| rakhmerov | we could make it based on just ignoring exceptions but this approach doesn't work with Postgres | 12:44 |
| rakhmerov | because in Postgres if you get ANY sql query error it switches TX to "rollback-only" mode | 12:44 |
| rakhmerov | so it's a different behavior across databases | 12:44 |
| rakhmerov | and there are some nuances two | 12:45 |
| ddeja | oh, I see | 12:45 |
| rakhmerov | yeah, it's all painful | 12:45 |
| rakhmerov | ddeja: as far as supporting different DBs, we're lucky that we can do it even for two DB dialects :) | 12:46 |
| rakhmerov | even Postgres and Mysql work completely differently in many cases | 12:46 |
| ddeja | yup | 12:46 |
| rakhmerov | e.g. transactional support | 12:46 |
| rakhmerov | in fact, Mysql doesn't really support real 'repeatable read' :) | 12:47 |
| rakhmerov | which we tried to use in the past | 12:47 |
| rakhmerov | if they were supported normally, I wouldn't have to do this kind of hacks | 12:47 |
| * ddeja just give +2 for the patch | 12:47 | |
| rakhmerov | :))) | 12:47 |
| rakhmerov | thanks | 12:47 |
| ddeja | no problem | 12:48 |
| rakhmerov | anyway, let me know if you want to know more, I spent last several weeks learning about this kind of tiny details in DB behavior | 12:48 |
| rakhmerov | I can tell | 12:48 |
| rakhmerov | ddeja: but the good news is that Mistral now works ~100 times faster | 12:49 |
| rakhmerov | may be even more than 100 times | 12:49 |
| ddeja | wow | 12:49 |
| rakhmerov | yes | 12:49 |
| ddeja | that's good | 12:49 |
| rakhmerov | without locks | 12:49 |
| rakhmerov | I test these changes w/o locks locally with multiple Mistral instances | 12:50 |
| rakhmerov | it works like a rocket | 12:50 |
| rakhmerov | so my plan is to make all needed changes and then remove locks completely | 12:50 |
| ddeja | 'It's not a rocket science... well, it is kind of' :) | 12:50 |
| rakhmerov | yeah, it's kind of :)) | 12:50 |
| ddeja | rakhmerov: great! | 12:51 |
| ddeja | rakhmerov: Do we have a meeting today? There is no mail on opensktack-dev... | 12:54 |
| rakhmerov | :) | 12:54 |
| rakhmerov | yes, we do | 12:54 |
| rakhmerov | I'll send an email now | 12:54 |
| openstackgerrit | Renat Akhmerov proposed openstack/mistral: Towards non-locking model: make 'with-items' work w/o locks https://review.openstack.org/352386 | 12:57 |
| *** toddjohn has joined #openstack-mistral | 12:57 | |
| nmakhotkin | hi guys | 13:04 |
| nmakhotkin | hi rakhmerov! | 13:04 |
| rakhmerov | hi ) | 13:05 |
| nmakhotkin | great news! If it is indeed that Mistral now works about 100 times faster it is amazing | 13:06 |
| nmakhotkin | I really want to review these patches | 13:06 |
| rakhmerov | yeah, I believe I'm very close to finish these changes | 13:07 |
| nmakhotkin | I'll do it asap | 13:07 |
| nmakhotkin | tomorrow, I think :) | 13:07 |
| *** nmakhotkin has quit IRC | 13:08 | |
| rakhmerov | ok | 13:08 |
| *** tonytan4ever has joined #openstack-mistral | 13:12 | |
| openstackgerrit | Renat Akhmerov proposed openstack/mistral: Towards non-locking model: make 'with-items' work w/o locks https://review.openstack.org/352386 | 13:26 |
| *** bobh has joined #openstack-mistral | 13:38 | |
| *** bobh has quit IRC | 13:44 | |
| *** bobh has joined #openstack-mistral | 13:45 | |
| openstackgerrit | Merged openstack/mistral: Add unique keys for non locking model https://review.openstack.org/352376 | 13:55 |
| openstackgerrit | Ryan Brady proposed openstack/mistral-lib: [WIP] Porting Security Related Code to Mistral-Lib https://review.openstack.org/352435 | 14:00 |
| *** rrecio has joined #openstack-mistral | 14:08 | |
| *** rrecio_ has joined #openstack-mistral | 14:09 | |
| *** rrecio has quit IRC | 14:12 | |
| *** janki has quit IRC | 14:14 | |
| *** rrecio_ has quit IRC | 14:35 | |
| *** rrecio has joined #openstack-mistral | 14:45 | |
| *** jaosorior has quit IRC | 14:56 | |
| *** mgershen has quit IRC | 15:08 | |
| *** dprince has quit IRC | 15:14 | |
| *** toddjohn has quit IRC | 15:16 | |
| *** toddjohn has joined #openstack-mistral | 15:20 | |
| openstackgerrit | Dawid Deja proposed openstack/mistral: Make mistral work with amqp and zmq backends https://review.openstack.org/352480 | 15:23 |
| *** toddjohn has quit IRC | 15:25 | |
| *** brian_price has joined #openstack-mistral | 15:32 | |
| *** toddjohn has joined #openstack-mistral | 15:35 | |
| *** rbrady-afk is now known as rbrady | 15:38 | |
| *** Ravikiran_K has quit IRC | 15:45 | |
| *** sshnaidm has quit IRC | 16:01 | |
| *** Ravikiran_K has joined #openstack-mistral | 16:11 | |
| *** sshnaidm has joined #openstack-mistral | 16:13 | |
| *** toddjohn has quit IRC | 16:26 | |
| *** catintheroof has joined #openstack-mistral | 16:28 | |
| *** toddjohn has joined #openstack-mistral | 16:29 | |
| *** toddjohn_ has joined #openstack-mistral | 16:31 | |
| *** toddjoh__ has joined #openstack-mistral | 16:32 | |
| *** tonytan4ever has quit IRC | 16:34 | |
| *** toddjohn has quit IRC | 16:34 | |
| *** tonytan4ever has joined #openstack-mistral | 16:34 | |
| *** toddjohn_ has quit IRC | 16:35 | |
| *** toddjoh__ has quit IRC | 16:36 | |
| *** tonytan_brb has joined #openstack-mistral | 16:37 | |
| *** tonytan4ever has quit IRC | 16:38 | |
| openstackgerrit | Renat Akhmerov proposed openstack/mistral: Towards non-locking model: make 'with-items' work w/o locks https://review.openstack.org/352386 | 16:40 |
| *** jpich has quit IRC | 16:51 | |
| *** bobh has quit IRC | 17:34 | |
| *** dprince has joined #openstack-mistral | 18:08 | |
| *** toddjohn has joined #openstack-mistral | 18:28 | |
| *** toddjohn has quit IRC | 18:33 | |
| *** bobh has joined #openstack-mistral | 18:34 | |
| *** toddjohn has joined #openstack-mistral | 18:38 | |
| *** bobh has quit IRC | 18:39 | |
| *** Ephur has quit IRC | 18:40 | |
| *** bobh has joined #openstack-mistral | 18:42 | |
| *** dulek_ has joined #openstack-mistral | 18:50 | |
| *** dulek has quit IRC | 18:50 | |
| *** dulek_ is now known as dulek | 18:50 | |
| *** Ravikiran_K has quit IRC | 18:51 | |
| *** dprince has quit IRC | 18:54 | |
| openstackgerrit | Ryan Brady proposed openstack/mistral-lib: [WIP] Porting Security Related Code to Mistral-Lib https://review.openstack.org/352435 | 18:58 |
| *** bobh has quit IRC | 19:03 | |
| *** Ephur has joined #openstack-mistral | 19:22 | |
| *** jistr has quit IRC | 19:44 | |
| *** bobh has joined #openstack-mistral | 19:45 | |
| *** jistr has joined #openstack-mistral | 19:48 | |
| *** jistr has quit IRC | 19:53 | |
| *** jistr has joined #openstack-mistral | 19:56 | |
| *** dprince has joined #openstack-mistral | 20:11 | |
| *** tonytan_brb has quit IRC | 20:16 | |
| *** dprince has quit IRC | 20:33 | |
| *** dprince has joined #openstack-mistral | 20:35 | |
| *** dprince has quit IRC | 21:02 | |
| *** tonytan4ever has joined #openstack-mistral | 21:17 | |
| *** tonytan4ever has quit IRC | 21:22 | |
| *** bobh has quit IRC | 21:23 | |
| *** bobh has joined #openstack-mistral | 21:33 | |
| *** tonytan4ever has joined #openstack-mistral | 21:35 | |
| *** catintheroof has quit IRC | 21:43 | |
| *** bobh has quit IRC | 22:10 | |
| *** chlong has quit IRC | 22:35 | |
| *** sshnaidm is now known as sshnaidm|afk | 22:38 | |
| *** vishwanathj has quit IRC | 23:20 | |
| *** toddjohn has quit IRC | 23:25 | |
| *** rrecio has quit IRC | 23:50 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!