*** rmoe has quit IRC | 00:10 | |
*** rmoe has joined #openstack-dns | 00:24 | |
*** jaycaz has quit IRC | 00:40 | |
*** jaycaz has joined #openstack-dns | 00:42 | |
*** eankutse has joined #openstack-dns | 01:01 | |
*** vinod1 has joined #openstack-dns | 01:05 | |
*** eankutse has quit IRC | 01:09 | |
*** vinod1 has quit IRC | 01:11 | |
*** eankutse has joined #openstack-dns | 01:30 | |
*** eankutse has quit IRC | 01:30 | |
*** eankutse has joined #openstack-dns | 01:30 | |
*** eankutse1 has joined #openstack-dns | 01:39 | |
*** eankutse has quit IRC | 01:41 | |
*** eankutse1 has quit IRC | 01:42 | |
*** eankutse has joined #openstack-dns | 01:43 | |
*** eankutse has quit IRC | 01:47 | |
*** nosnos has joined #openstack-dns | 01:49 | |
*** jaycaz has quit IRC | 01:56 | |
*** richm has quit IRC | 01:59 | |
*** darshan104 has quit IRC | 02:05 | |
*** nosnos has quit IRC | 03:06 | |
*** vinod1 has joined #openstack-dns | 03:20 | |
*** vinod1 has quit IRC | 03:29 | |
*** nosnos has joined #openstack-dns | 03:33 | |
*** crc32 has quit IRC | 03:41 | |
*** msisk has quit IRC | 03:42 | |
*** rjrjr has quit IRC | 03:46 | |
*** darshan104 has joined #openstack-dns | 04:42 | |
*** k4n0 has joined #openstack-dns | 06:22 | |
*** sballe_ has joined #openstack-dns | 06:28 | |
*** sballe has quit IRC | 06:34 | |
*** rektide_ has joined #openstack-dns | 06:34 | |
*** rektide has quit IRC | 06:34 | |
*** k4n0 has quit IRC | 06:34 | |
*** k4n0 has joined #openstack-dns | 06:37 | |
*** darshan104 has quit IRC | 08:23 | |
*** darshan104 has joined #openstack-dns | 08:23 | |
*** darshan104 has quit IRC | 08:28 | |
*** Dafna has quit IRC | 09:43 | |
*** nosnos has quit IRC | 09:54 | |
*** nosnos has joined #openstack-dns | 10:01 | |
*** nosnos has quit IRC | 10:10 | |
*** nosnos has joined #openstack-dns | 10:12 | |
*** nosnos has quit IRC | 10:38 | |
openstackgerrit | Kiall Mac Innes proposed a change to openstack/designate: Ensure each greenthread uses it's own Session instance https://review.openstack.org/107644 | 10:53 |
---|---|---|
*** Dafna has joined #openstack-dns | 11:00 | |
*** Dafna has quit IRC | 11:29 | |
*** Dafna has joined #openstack-dns | 11:43 | |
*** mwagner_lap has quit IRC | 12:17 | |
*** richm has joined #openstack-dns | 12:46 | |
*** k4n0 has quit IRC | 13:01 | |
*** vinod1 has joined #openstack-dns | 13:04 | |
*** vinod1 has quit IRC | 13:07 | |
Kiall | FYI .. https://review.openstack.org/#/c/107688/1 | 13:22 |
Kiall | Gerrit ACL changes | 13:22 |
*** vinod1 has joined #openstack-dns | 13:26 | |
*** vinod1 has quit IRC | 13:27 | |
mugsie | FYI - links for designate on docs.openstack.org https://review.openstack.org/#/c/107691/ | 13:33 |
*** jaycaz has joined #openstack-dns | 13:35 | |
*** eankutse has joined #openstack-dns | 13:36 | |
*** eankutse has quit IRC | 13:38 | |
*** eankutse has joined #openstack-dns | 13:38 | |
*** jaycaz has quit IRC | 13:41 | |
*** msisk has joined #openstack-dns | 13:48 | |
*** mwagner_lap has joined #openstack-dns | 13:53 | |
*** msisk has quit IRC | 14:10 | |
openstackgerrit | Kiall Mac Innes proposed a change to openstack/designate: Attach records to RecordSet objects https://review.openstack.org/105887 | 14:15 |
*** timsim has joined #openstack-dns | 14:15 | |
*** crc32 has joined #openstack-dns | 14:18 | |
*** msisk has joined #openstack-dns | 14:18 | |
*** betsy has quit IRC | 14:20 | |
openstackgerrit | Kiall Mac Innes proposed a change to openstack/designate: Attach records to RecordSet objects https://review.openstack.org/105887 | 14:39 |
*** crc32_ has joined #openstack-dns | 14:53 | |
*** eankutse1 has joined #openstack-dns | 14:56 | |
*** eankutse1 has quit IRC | 14:56 | |
*** eankutse1 has joined #openstack-dns | 14:56 | |
*** eankutse has quit IRC | 14:57 | |
*** jaycaz has joined #openstack-dns | 14:58 | |
*** jaycaz has quit IRC | 14:58 | |
*** crc32 has quit IRC | 15:02 | |
*** rektide_ has quit IRC | 15:02 | |
*** Dafna has quit IRC | 15:02 | |
*** rektide has joined #openstack-dns | 15:03 | |
*** Dafna has joined #openstack-dns | 15:07 | |
*** eankutse1 has quit IRC | 15:15 | |
*** eankutse has joined #openstack-dns | 15:16 | |
*** jaycaz has joined #openstack-dns | 15:17 | |
*** crc32_ has quit IRC | 15:20 | |
*** betsy has joined #openstack-dns | 15:20 | |
*** eankutse1 has joined #openstack-dns | 15:22 | |
*** msisk has quit IRC | 15:24 | |
*** msisk_ has joined #openstack-dns | 15:24 | |
*** jaycaz1 has joined #openstack-dns | 15:25 | |
*** eankutse has quit IRC | 15:25 | |
*** jaycaz has quit IRC | 15:26 | |
*** timsim has quit IRC | 15:26 | |
*** vinod1 has joined #openstack-dns | 15:27 | |
*** rmoe has quit IRC | 15:29 | |
*** eankutse has joined #openstack-dns | 15:31 | |
*** eankutse1 has quit IRC | 15:31 | |
*** eankutse1 has joined #openstack-dns | 15:47 | |
*** eankutse has quit IRC | 15:47 | |
*** eankutse1 has quit IRC | 15:47 | |
*** eankutse has joined #openstack-dns | 15:47 | |
*** eankutse has quit IRC | 15:48 | |
*** eankutse has joined #openstack-dns | 15:48 | |
*** rmoe has joined #openstack-dns | 15:49 | |
*** eankutse1 has joined #openstack-dns | 15:54 | |
*** eankutse has quit IRC | 15:55 | |
*** eankutse1 has quit IRC | 15:56 | |
*** eankutse has joined #openstack-dns | 15:56 | |
openstackgerrit | A change was merged to openstack/designate: Attach records to RecordSet objects https://review.openstack.org/105887 | 16:00 |
*** eankutse1 has joined #openstack-dns | 16:03 | |
*** eankutse has quit IRC | 16:05 | |
*** eankutse2 has joined #openstack-dns | 16:06 | |
*** eankutse1 has quit IRC | 16:07 | |
*** eankutse2 has quit IRC | 16:07 | |
*** eankutse1 has joined #openstack-dns | 16:08 | |
Kiall | woo - 1 step closer :) | 16:10 |
vinod1 | kiall: regarding your change - https://review.openstack.org/#/c/107644/ | 16:11 |
Kiall | vinod1: heya | 16:12 |
vinod1 | did you hit some issue with this? | 16:12 |
Kiall | kinda ;) | 16:12 |
vinod1 | did youget a chance to look through at other places in the code where we might need thread locals? | 16:13 |
Kiall | I don't think they are any other places we need them, and if there are, then those are bugs which should be fixed without the threadlocal. Ideally this wouldn't use it either, but avoiding it would be a large change to how our Storage layer works and interacts with the rest of the system :( | 16:15 |
Kiall | (But - Since our storage API is the way it is - it's OK in this case to use them, it's actually even how SQLA recommends you do it..) | 16:16 |
vinod1 | Thanks Kiall for the explanations | 16:17 |
Kiall | np | 16:17 |
*** crc32 has joined #openstack-dns | 16:22 | |
*** eankutse has joined #openstack-dns | 16:23 | |
*** eankutse1 has quit IRC | 16:24 | |
*** eankutse has quit IRC | 16:25 | |
*** jaycaz1 has quit IRC | 16:25 | |
*** eankutse has joined #openstack-dns | 16:25 | |
*** jaycaz has joined #openstack-dns | 16:26 | |
*** jaycaz1 has joined #openstack-dns | 16:28 | |
*** jaycaz has quit IRC | 16:31 | |
*** eankutse has quit IRC | 16:34 | |
*** JensOfSweden has quit IRC | 16:34 | |
*** eankutse has joined #openstack-dns | 16:34 | |
openstackgerrit | A change was merged to openstack/designate: Ensure each greenthread uses it's own Session instance https://review.openstack.org/107644 | 16:37 |
*** eankutse has quit IRC | 16:39 | |
*** eankutse has joined #openstack-dns | 16:40 | |
*** rossk has joined #openstack-dns | 16:43 | |
*** eankutse1 has joined #openstack-dns | 16:53 | |
*** eankutse has quit IRC | 16:56 | |
*** eankutse1 has quit IRC | 16:57 | |
*** eankutse has joined #openstack-dns | 16:57 | |
jaycaz1 | Hey, has anyone ran into issues with oslo.messaging when they try to send api requests from multiple processes? | 17:11 |
jaycaz1 | That's what I'm running into. vinod mentioned that it's related to this bug? https://launchpad.net/mistral/+bug/1337705 | 17:11 |
uvirtbot | Launchpad bug 1337705 in mistral "Exception during message handling: _oslo_messaging_localcontext" [Undecided,New] | 17:11 |
jaycaz1 | The situation is that I have a python script to create n zones, one at a time, and I am trying to run that simultaneously on multiple processes | 17:12 |
vinod1 | kiall:/ekarlso: Jordan ran into this issue when trying to test out zone creates | 17:12 |
vinod1 | Do you know if there is already a fix for this is oslo.messaging? | 17:12 |
*** eankutse1 has joined #openstack-dns | 17:15 | |
*** Dafna has quit IRC | 17:16 | |
*** eankutse has quit IRC | 17:18 | |
*** msisk_ has quit IRC | 17:34 | |
*** msisk has joined #openstack-dns | 17:35 | |
*** eankutse1 has quit IRC | 17:36 | |
*** eankutse has joined #openstack-dns | 17:36 | |
*** msisk has quit IRC | 17:41 | |
*** msisk has joined #openstack-dns | 17:41 | |
vinod1 | kiall: i had a question on mdns | 18:06 |
vinod1 | the payload received on tcp seems different from that received on udp. the tcp one seems to have about 4 extra bytes in the beginning. | 18:17 |
vinod1 | but i could not see why that should be | 18:17 |
vinod1 | wondering if you could spot something in the code there | 18:17 |
Kiall | vinod1: could there be 4 extra bytes going over the wire? (e.g. have you used wireshark or something to look at the packets on-the-wire?) | 18:25 |
vinod1 | no not yet | 18:29 |
vinod1 | i used dig to send a tcp request to mdns | 18:29 |
vinod1 | with udp there is no issue, but tcp has something extra before the payload | 18:30 |
*** darshan104 has joined #openstack-dns | 18:40 | |
*** JensOfSweden has joined #openstack-dns | 18:53 | |
Kiall | vinod1: humm - if you strip the leading 4 bytes of a TCP payload before passing to the handler, does it work? | 18:57 |
Kiall | jaycaz1: for some reason, I think that might be an out of date dependency. No clue why I'm thinking that though | 18:57 |
ekarlso | good q vinod1 ;) | 18:58 |
Kiall | jaycaz1: related bug - https://bugs.launchpad.net/oslo.messaging/+bug/1288878 | 18:58 |
uvirtbot | Launchpad bug 1288878 in ceilometer "localcontext store is not monkeypatched by eventlet" [Undecided,Fix released] | 18:58 |
ekarlso | or what Kiall said :D | 18:58 |
*** timsim has joined #openstack-dns | 18:59 | |
Kiall | jaycaz1: Is this a script something that talks to the REST API, or RabbitMQ BTW? | 19:00 |
jaycaz1 | Kiall: it's through the REST API | 19:01 |
Kiall | Humm - I wonder if we've introduced an issue recently (or if oslo.messaging has..) | 19:01 |
ekarlso | jaycaz1: u care to share your script ? | 19:02 |
jaycaz1 | sure thing! | 19:03 |
jaycaz1 | you can clone it here: https://github.com/jaycaz/designate-scripts | 19:03 |
Kiall | Just standing up a fresh copy of designate.. Let's see if I can reproduce | 19:04 |
Kiall | Humm.. Should have turned keystone off before starting :( | 19:06 |
ekarlso | :P | 19:07 |
Kiall | OOOHHH.. jaycaz1 .. I may have an idea. | 19:08 |
ekarlso | Kiall: ? | 19:08 |
Kiall | looking for the commit | 19:09 |
Kiall | jaycaz1: if you try revision 0a2406739a588a52c8555bcd7ca8695323cfa0f1 of designate, does it work? | 19:09 |
jaycaz1 | Kiall: I'll try that and get back to you! | 19:10 |
Kiall | actually, you can scratch that | 19:12 |
Kiall | https://github.com/openstack/designate/blob/master/designate/__init__.py | 19:12 |
Kiall | imports oslo.messaging, and it does it before https://github.com/openstack/designate/blob/master/designate/cmd/__init__.py issues the eventlet.monkey_patch(os=False) | 19:13 |
Kiall | Looks like ekarlso broke it ;) | 19:13 |
ekarlso | PFFFT | 19:13 |
Kiall | https://github.com/openstack/designate/commit/0d3a4427d9c9111cd0229d1b05d8517d9a2945b6 <-- That's the first commit that is going to fail ;) | 19:14 |
ekarlso | :D | 19:14 |
* Kiall is VERY glad we haven't deployed that at HP Cloud yet | 19:14 | |
ekarlso | why does it need to patch before o.m ? | 19:14 |
Kiall | because otherwise oslo-messaging get's + holds an unpatched reference to the threading module | 19:15 |
*** timsim has quit IRC | 19:16 | |
*** timsim has joined #openstack-dns | 19:17 | |
ekarlso | :) | 19:18 |
openstackgerrit | Kiall Mac Innes proposed a change to openstack/designate: Ensure eventlet monkey patching happens as early as possible https://review.openstack.org/107807 | 19:21 |
Kiall | jaycaz1: ^ should sort it | 19:21 |
Kiall | also filed bug 1288878 for it | 19:21 |
uvirtbot | Launchpad bug 1288878 in ceilometer "localcontext store is not monkeypatched by eventlet" [Undecided,Fix released] https://launchpad.net/bugs/1288878 | 19:21 |
Kiall | ehh | 19:21 |
Kiall | I mean 1343526 | 19:21 |
jaycaz1 | Kiall: great, thanks! I'll give it a shot! | 19:21 |
Kiall | bug 1343526 | 19:21 |
uvirtbot | Launchpad bug 1343526 in designate "Eventlet patching is occurring too late during service startup" [Critical,In progress] https://launchpad.net/bugs/1343526 | 19:21 |
Kiall | there we go :) | 19:21 |
jaycaz1 | Kiall: success! | 19:23 |
Kiall | Cool :D | 19:23 |
* Kiall is very glad that was found before we released with it -_- | 19:24 | |
*** JensOfSweden has quit IRC | 19:30 | |
*** JensOfSweden has joined #openstack-dns | 19:32 | |
*** vinod1 has quit IRC | 19:38 | |
*** JensOfSweden has quit IRC | 19:50 | |
timsim | Kiall: We're seeing an error "Backend: Unknown backend failure: DBError(u'Session is already flushing',)" Is that related to the bug you filed this morning? | 19:52 |
Kiall | timsim: not that I know of, any more info to hand? | 19:56 |
jaycaz1 | it happens when we are sending requests from two processes at once | 19:56 |
timsim | We're just playing around, but yeah that ^ We're not pushing the requests through a web server or anything, and it's quite a lot of requests. | 19:57 |
jaycaz1 | when the requests are slowed down (i.e. 1 per second) there seem to be no issues | 19:58 |
timsim | http://paste.openstack.org/show/87010/ <- stacktrace | 19:58 |
Kiall | Just managed to reproduce with jaycaz1's change | 19:58 |
Kiall | jaycaz1's script& | 19:58 |
Kiall | jaycaz1's script* | 19:58 |
*** JensOfSweden has joined #openstack-dns | 20:01 | |
Kiall | Without my fix from earlier today - the SQLA Session one - It's a different trace: | 20:02 |
Kiall | 2014-07-17 20:01:36.474 TRACE oslo.messaging.rpc.dispatcher InvalidRequestError: This Session's transaction has been rolled back by a nested rollback() call. To begin a new transaction, issue Session.rollback() first. | 20:02 |
Kiall | ^ is what we saw that prompted the fix in the first place | 20:02 |
Kiall | https://bugs.launchpad.net/ceilometer/+bug/1183996 seems similar, and their fix was basically what we did to the session today. | 20:04 |
uvirtbot | Launchpad bug 1183996 in ceilometer "session flushing error" [Undecided,Fix released] | 20:04 |
Kiall | timsim: can you file a bug? Looking into it now | 20:06 |
Kiall | It looks related, but not 100% sure yet | 20:07 |
*** JensOfSweden has quit IRC | 20:07 | |
timsim | Sure. | 20:09 |
timsim | bug 1343561 | 20:12 |
uvirtbot | Launchpad bug 1343561 in designate "DB Error ('Sesssion is already flushing')" [Undecided,New] https://launchpad.net/bugs/1343561 | 20:12 |
Kiall | timsim: Just in time :) Got it | 20:14 |
Kiall | Exact same issue, different place. Can't believe I didn't realize this when vinod actually asked me earlier -_- | 20:14 |
Kiall | 10 copies of jaycaz1 script going creating 1000 zones each, no stack traces :) | 20:14 |
timsim | Nice. | 20:15 |
jaycaz1 | awesome! | 20:15 |
* Kiall schedules another deploy for tomorrow :( | 20:16 | |
*** vinod1 has joined #openstack-dns | 20:16 | |
openstackgerrit | Kiall Mac Innes proposed a change to openstack/designate: PowerDNS: Ensure each greenthread uses it's own Session instance https://review.openstack.org/107817 | 20:18 |
Kiall | vinod1: ^ you were totally right earlier, and even with your question, I didn't realize it ;) | 20:18 |
*** msisk has quit IRC | 20:22 | |
*** JensOfSweden has joined #openstack-dns | 20:24 | |
*** msisk has joined #openstack-dns | 20:24 | |
vinod1 | today seems to be the day to identify issues | 20:25 |
Kiall | :) | 20:27 |
Kiall | jaycaz1: is that working for you too? | 20:28 |
jaycaz1 | we're making a branch with that change and the earlier change combined | 20:32 |
jaycaz1 | once we do that, we'll let you know | 20:32 |
timsim | Kiall: It's working great | 20:34 |
Kiall | cool.. and on that note... I'm going to get out of here before you find something else! | 20:34 |
openstackgerrit | Graham Hayes proposed a change to openstack/designate: Added functionality to allow for zone ownership transfers https://review.openstack.org/107822 | 20:38 |
* mugsie runs | 20:38 | |
timsim | mugsie: oooh interesting. | 20:39 |
jaycaz1 | Kiall: in case you're still here, it all works! thanks for all your help! | 20:40 |
*** darshan104 has quit IRC | 20:49 | |
*** darshan104 has joined #openstack-dns | 20:56 | |
openstackgerrit | Betsy Luzader proposed a change to openstack/designate: Modify Recordset API https://review.openstack.org/102892 | 20:57 |
*** jaycaz1 has quit IRC | 21:07 | |
*** timsim has quit IRC | 21:12 | |
*** mwagner_lap has quit IRC | 21:43 | |
openstackgerrit | A change was merged to openstack/designate: PowerDNS: Ensure each greenthread uses it's own Session instance https://review.openstack.org/107817 | 22:32 |
*** mwagner_lap has joined #openstack-dns | 22:35 | |
*** betsy has quit IRC | 22:38 | |
openstackgerrit | A change was merged to openstack/designate: Ensure eventlet monkey patching happens as early as possible https://review.openstack.org/107807 | 22:39 |
*** darshan104 has quit IRC | 22:41 | |
*** mwagner_lap has quit IRC | 22:47 | |
*** JensOfSweden has quit IRC | 22:51 | |
*** ekarlso has quit IRC | 22:51 | |
*** mugsie has quit IRC | 22:51 | |
*** retr0h has quit IRC | 22:51 | |
*** shufflebot has quit IRC | 22:51 | |
*** uvirtbot has quit IRC | 22:52 | |
*** vinod1 has quit IRC | 22:53 | |
*** uvirtbot has joined #openstack-dns | 22:53 | |
*** JensOfSweden has joined #openstack-dns | 22:53 | |
*** ekarlso has joined #openstack-dns | 22:53 | |
*** mugsie has joined #openstack-dns | 22:53 | |
*** retr0h has joined #openstack-dns | 22:53 | |
*** shufflebot has joined #openstack-dns | 22:53 | |
*** mikal has quit IRC | 22:54 | |
*** mikal has joined #openstack-dns | 22:54 | |
*** soren has quit IRC | 22:54 | |
*** zigo has quit IRC | 22:54 | |
*** mgagne has quit IRC | 22:54 | |
*** simonmcc has quit IRC | 22:54 | |
*** baffle has quit IRC | 22:55 | |
*** bauruine has quit IRC | 22:55 | |
*** soren has joined #openstack-dns | 22:55 | |
*** zigo has joined #openstack-dns | 22:55 | |
*** mgagne has joined #openstack-dns | 22:55 | |
*** simonmcc has joined #openstack-dns | 22:55 | |
*** mikal has left #openstack-dns | 22:55 | |
*** mikal has joined #openstack-dns | 22:55 | |
*** baffle has joined #openstack-dns | 22:56 | |
*** bauruine has joined #openstack-dns | 22:56 | |
*** darshan104 has joined #openstack-dns | 23:11 | |
*** darshan104 has quit IRC | 23:16 | |
*** darshan104 has joined #openstack-dns | 23:21 | |
*** eankutse1 has joined #openstack-dns | 23:51 | |
*** eankutse has quit IRC | 23:52 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!