| *** ryanpetrello has quit IRC | 00:00 | |
| *** ryanpetrello has joined #openstack-dns | 00:11 | |
| *** jmcbride has joined #openstack-dns | 00:26 | |
| *** vinod1 has quit IRC | 00:31 | |
| *** ryanpetrello has quit IRC | 00:46 | |
| *** EricGonczer_ has joined #openstack-dns | 00:46 | |
| *** vinod1 has joined #openstack-dns | 00:47 | |
| *** ryanpetrello has joined #openstack-dns | 00:51 | |
| *** ryanpetrello has quit IRC | 00:56 | |
| *** rjrjr has joined #openstack-dns | 00:59 | |
| *** rmoe has quit IRC | 01:04 | |
| *** EricGonczer_ has quit IRC | 01:05 | |
| *** ryanpetrello has joined #openstack-dns | 01:18 | |
| *** ToMiles has quit IRC | 01:26 | |
| *** ToMiles has joined #openstack-dns | 01:29 | |
| *** rmoe has joined #openstack-dns | 01:30 | |
| *** ryanpetrello_ has joined #openstack-dns | 01:33 | |
| *** ryanpetrello has quit IRC | 01:34 | |
| *** ryanpetrello_ is now known as ryanpetrello | 01:34 | |
| *** nosnos has joined #openstack-dns | 01:52 | |
| *** ryanpetrello has quit IRC | 01:55 | |
| *** EricGonczer_ has joined #openstack-dns | 01:55 | |
| *** linstatsdr_ has joined #openstack-dns | 02:31 | |
| *** jmcbride has quit IRC | 02:33 | |
| *** LinStatSDR has quit IRC | 02:34 | |
| *** linstatsdr__ has joined #openstack-dns | 02:34 | |
| *** ryanpetrello has joined #openstack-dns | 02:37 | |
| *** linstatsdr_ has quit IRC | 02:38 | |
| *** LinStatSDR has joined #openstack-dns | 02:38 | |
| *** linstatsdr__ has quit IRC | 02:40 | |
| *** EricGonczer_ has quit IRC | 03:10 | |
| *** EricGonczer_ has joined #openstack-dns | 03:15 | |
| *** EricGonczer_ has quit IRC | 03:17 | |
| *** nosnos has quit IRC | 03:23 | |
| *** richm has quit IRC | 03:55 | |
| *** ToMiles_ has joined #openstack-dns | 04:02 | |
| *** ToMiles has quit IRC | 04:03 | |
| *** nosnos has joined #openstack-dns | 04:16 | |
| *** GonZo2K has quit IRC | 04:21 | |
| *** vipul has quit IRC | 05:09 | |
| *** vipul has joined #openstack-dns | 05:10 | |
| *** LinStatSDR has quit IRC | 05:13 | |
| *** ryanpetrello has quit IRC | 06:33 | |
| *** ToMiles_ has quit IRC | 06:53 | |
| *** k4n0 has joined #openstack-dns | 06:54 | |
| *** nihilifer has joined #openstack-dns | 07:47 | |
| *** jordanP has joined #openstack-dns | 08:53 | |
| *** jordanP has quit IRC | 09:05 | |
| *** jordanP has joined #openstack-dns | 09:17 | |
| *** alokj has joined #openstack-dns | 09:28 | |
| *** pino has joined #openstack-dns | 09:59 | |
| *** jmcbride has joined #openstack-dns | 10:55 | |
| *** jmcbride has quit IRC | 11:13 | |
| *** jmcbride has joined #openstack-dns | 11:14 | |
| *** kbyrne has joined #openstack-dns | 11:26 | |
| *** mwagner_lap has quit IRC | 11:39 | |
| *** jmcbride1 has joined #openstack-dns | 12:08 | |
| *** jmcbride has quit IRC | 12:12 | |
| openstackgerrit | Kiall Mac Innes proposed openstack/designate: Handle multi-packet TCP queries https://review.openstack.org/138885 | 13:03 |
|---|---|---|
| openstackgerrit | Kiall Mac Innes proposed openstack/designate: Add DynECT DevStack plugin https://review.openstack.org/138866 | 13:03 |
| *** vinod1 has joined #openstack-dns | 13:11 | |
| *** jmcbride1 has quit IRC | 13:15 | |
| *** mwagner_lap has joined #openstack-dns | 13:17 | |
| *** nosnos has quit IRC | 13:27 | |
| *** GonZo2K has joined #openstack-dns | 13:52 | |
| rjrjr | just 13 more FAILed unit tests to fix. | 13:58 |
| *** vinod1 has quit IRC | 14:02 | |
| Kiall | See, now you scared vinod off ;) | 14:04 |
| *** jordanP has quit IRC | 14:07 | |
| *** pino has quit IRC | 14:14 | |
| *** pino has joined #openstack-dns | 14:14 | |
| *** richm has joined #openstack-dns | 14:16 | |
| *** rediskin has joined #openstack-dns | 14:16 | |
| *** jordanP has joined #openstack-dns | 14:20 | |
| *** nihilifer has quit IRC | 14:31 | |
| *** k4n0 has quit IRC | 14:33 | |
| *** jmcbride has joined #openstack-dns | 14:36 | |
| *** nkinder has quit IRC | 14:45 | |
| *** ryanpetrello has joined #openstack-dns | 14:45 | |
| *** rediskin has quit IRC | 14:52 | |
| *** GonZo2K has quit IRC | 14:57 | |
| *** rmoe has quit IRC | 15:20 | |
| openstackgerrit | Merged openstack/designate: Handle multi-packet TCP queries https://review.openstack.org/138885 | 15:28 |
| *** rediskin has joined #openstack-dns | 15:30 | |
| *** timsim has joined #openstack-dns | 15:30 | |
| *** EricGonczer_ has joined #openstack-dns | 15:31 | |
| rjrjr | kiall: any reason why the deleting of an empty recordset is in the API code and not central? | 15:32 |
| Kiall | Ehh, I can't remember to be honest... Probably! | 15:39 |
| *** vinod1 has joined #openstack-dns | 15:43 | |
| rjrjr | okay. i'm going to try to move this code to storage and see if it breaks any additional unit tests. | 15:47 |
| rjrjr | problem with it in the API code, we don't actually delete the record when we call delete_record method anymore. so, checking to see if the recordset has no more records will always fail. | 15:48 |
| rjrjr | the deletion of a record has been moved to update_status instead. | 15:48 |
| rjrjr | someday, we should go through this code and pick a style: record['recordset'] or record.recordset - i prefer the later personally (less typing and easier on the eyes.) | 15:50 |
| vinod1 | The former one was the original one - when we did not have objects. Once we had objects we started using the latter format | 15:53 |
| vinod1 | Of course now we have to convert the dict syntax to the object syntax | 15:53 |
| *** betsy has joined #openstack-dns | 15:58 | |
| *** paul_glass has joined #openstack-dns | 15:58 | |
| *** LinStatSDR has joined #openstack-dns | 16:00 | |
| *** EricGonc_ has joined #openstack-dns | 16:03 | |
| *** rjrjr has quit IRC | 16:05 | |
| *** EricGonczer_ has quit IRC | 16:07 | |
| *** arn_ has joined #openstack-dns | 16:16 | |
| *** rjrjr has joined #openstack-dns | 16:16 | |
| rjrjr | moving the logical to central or storage breaks other unit tests. <sigh. | 16:17 |
| rjrjr | so, how do we want to handle this? the problem, we have code in the API when a record is deleted to also delete any empty recordset. | 16:17 |
| rjrjr | we cannot leave it in the API, because we don't actually delete the record when we call central.delete_record method. | 16:18 |
| *** d34dh0r53 has quit IRC | 16:18 | |
| *** jordanP has quit IRC | 16:18 | |
| *** arn has quit IRC | 16:18 | |
| rjrjr | if deleting an empty recordset when deleting a record is desireable, then we will want to implement it in another layer. i think storage makes the most sense. | 16:19 |
| *** jordanP has joined #openstack-dns | 16:19 | |
| rjrjr | the ramifications, there are other places where we delete a record that we do not expect the recordset to also be deleted. i would have to fix all those too. | 16:19 |
| *** d34dh0r53 has joined #openstack-dns | 16:20 | |
| rjrjr | another solution, remove this functionality (deleting of an empty recordset) but we reintroduce the bug where we cannot reuse names (for example, reuse a CNAME in the same domain.) | 16:20 |
| rjrjr | thoughts? | 16:21 |
| vinod1 | Currently we have that kind of logic (deleting empty recordsets) in central with the storage doing what is asked of it. | 16:22 |
| rjrjr | no, we don't. | 16:22 |
| rjrjr | sorry, i thought you implied that logic exists. you mean that logic should go in central. my bad. 8^) | 16:22 |
| vinod1 | For this case, we could possibly have a flag in storage to specify if we want to delete empty recordsets when records are deleted | 16:23 |
| rjrjr | so, keep storage dump (and single focused.) | 16:23 |
| vinod1 | Yes | 16:23 |
| rjrjr | makes sense. so implement this in central and fix what it breaks? | 16:23 |
| vinod1 | Yes IMHO | 16:23 |
| vinod1 | Kiall/mugsie/betsy/timsim: Thoughts? | 16:24 |
| rjrjr | for example, another unit test it breaks, we delete a record. the record is deleted, but so it the recordset. then we go to fetch the record (you can see where this is going) but instead of getting the RecordNotFound exception, we get the RecordsetNotFound exception. | 16:24 |
| betsy | vinod1: wasn’t reading let me catch up | 16:24 |
| rjrjr | vinod, you are correct. this is business logic and it should go in central. | 16:25 |
| betsy | rjrjr: I’ve just about finished my patchset that gets rid of server and replaces it with nameserver | 16:25 |
| betsy | This is part of the pool storage spec. I shoud have mine up by this afternoon. | 16:26 |
| rjrjr | betsy: cool. | 16:26 |
| rjrjr | i figured i'm going to be in merge heaven regardless. 8^( | 16:26 |
| betsy | I’ve redone central to basically get rid of servers and use nameservers instead. I’ve changed the v1 api to reflect that, as well as all the tets | 16:26 |
| betsy | too true | 16:26 |
| betsy | But I think we’ve worked on some of the same parts. For instance, I’m eliminating the create_server call from central | 16:27 |
| timsim | I'm having a hard time finding it in the API code... | 16:27 |
| rjrjr | my stuff is blocked waiting for a pdns backend. | 16:27 |
| rjrjr | timsim: ? | 16:27 |
| rjrjr | v1/records.py | 16:27 |
| timsim | ah. v1. | 16:28 |
| rjrjr | almost the last line in the file. | 16:28 |
| *** mikedillion has joined #openstack-dns | 16:28 | |
| vinod1 | https://github.com/openstack/designate/blob/master/designate/api/v1/records.py#L251 | 16:28 |
| timsim | Kind of seems like that logic should be in central.delete_record | 16:28 |
| rjrjr | it can't go there either. central.delete_record just updates the record atttributes and calls the pool manager update_domain method. | 16:29 |
| rjrjr | it will need to go in the update_status where we actually delete records and "delete" domains. | 16:29 |
| rjrjr | i think we are in agreement it goes into central. | 16:30 |
| rjrjr | i'll have to track down all the stuff it breaks by moving it there. hope the list isn't long... | 16:30 |
| vinod1 | central.delete_record can check if this is the last record and call central.delete_recordset. That way both are marked as PENDING | 16:31 |
| timsim | That's what I was thinking ^ | 16:31 |
| vinod1 | and when update_status returns it does the right thing | 16:31 |
| rjrjr | so, you can imagine, unit tests at the API level handle this. unit tests at the central level do not. | 16:31 |
| rjrjr | central.delete_recordset does not set a PENDING status. there is no status for recordsets. | 16:32 |
| *** ryanpetrello has quit IRC | 16:32 | |
| timsim | Wait wut | 16:33 |
| rjrjr | we have status on records, not recordsets. | 16:33 |
| timsim | Ah | 16:33 |
| rjrjr | so, when you update a recordset, it changes the status field of a record. | 16:33 |
| rjrjr | remember, the backend could care less about recordsets. (at least BIND9 doesn't care.) | 16:33 |
| vinod1 | How about if you change the ttl of a recordset - how do you mark it as PENDING? | 16:33 |
| rjrjr | the records are marked as PENDING when you do that. | 16:34 |
| timsim | All of them? | 16:34 |
| rjrjr | correct. | 16:34 |
| rjrjr | all of them in the recordset. | 16:34 |
| timsim | What if there is 100000 of them? Do you poll for each individual one? | 16:34 |
| rjrjr | ? | 16:34 |
| rjrjr | one call to update_domain. | 16:34 |
| timsim | Fixes the records to ACTIVE once the serial for the zone is right? | 16:35 |
| rjrjr | the code no longer handles records one at a time. it does them all at once. (NS, SOA, etc.) | 16:35 |
| rjrjr | timsim: correct. | 16:35 |
| timsim | ok | 16:35 |
| rjrjr | the code is much more efficient than it was at calling the backend. | 16:36 |
| rjrjr | sorry, performing an AXFR. | 16:36 |
| rjrjr | so, unless i want to introduce status and action to recordsets, i'll handle this in update_status. | 16:37 |
| rjrjr | sorry, meant "we". | 16:37 |
| vinod1 | Handling it in update_status, then seems logical | 16:38 |
| timsim | I guess I'd have to see what it looks like. | 16:38 |
| rjrjr | and fix what it breaks. okay, that is the way i'll go. if it looks like it breaks too much or i run into a road block, we can look at adding status and action to recordset. | 16:38 |
| rjrjr | funny, the code is working and passes all the tests we have. it's these unit tests that i'm spending all my time on... | 16:39 |
| rjrjr | all the tests = exercises/designate.sh | 16:39 |
| rjrjr | and rally | 16:40 |
| vinod1 | If you need help fixing the unit tests - let me know | 16:40 |
| rjrjr | been spending my time simulating stuff happening on the backend and calling update_status for most of the fixes. | 16:40 |
| rjrjr | i was down to 13. with the change needed for delete_record, the number has jumped a bit. | 16:41 |
| rjrjr | i want to get these existing tests done today and the code up for review. | 16:42 |
| rjrjr | have other unit tests to write for pool manager. | 16:42 |
| vinod1 | Can the delete_record be done in a different patchset? | 16:42 |
| rjrjr | and i suspect a fun merge in the near future. 8^) | 16:42 |
| rjrjr | not if we want the unit tests to pass. | 16:42 |
| rjrjr | i'm just trying to get them to pass so the code i submitted can be reviewed and accepted. | 16:43 |
| rjrjr | https://review.openstack.org/#/c/138406/ | 16:44 |
| rjrjr | also waiting for a powerdns pool backend driver. | 16:44 |
| *** paul_glass has quit IRC | 16:46 | |
| *** pino has quit IRC | 17:02 | |
| *** untriaged-bot has joined #openstack-dns | 17:03 | |
| untriaged-bot | Untriaged bugs so far: | 17:03 |
| untriaged-bot | https://bugs.launchpad.net/designate/+bug/1390357 | 17:03 |
| untriaged-bot | https://bugs.launchpad.net/designate/+bug/1289444 | 17:03 |
| *** pino has joined #openstack-dns | 17:03 | |
| untriaged-bot | https://bugs.launchpad.net/designate/+bug/1338256 | 17:03 |
| uvirtbot | Launchpad bug 1390357 in designate "Designate sync creates duplicated records in powerdns backend" [Medium,New] | 17:03 |
| untriaged-bot | https://bugs.launchpad.net/designate/+bug/1398989 | 17:03 |
| uvirtbot | Launchpad bug 1289444 in designate "Designate with postgres backend is having issues" [Undecided,New] | 17:03 |
| *** untriaged-bot has quit IRC | 17:03 | |
| uvirtbot | Launchpad bug 1338256 in designate "There's no record validation in v2" [Undecided,New] | 17:03 |
| uvirtbot | Launchpad bug 1398989 in designate "DynECT backend fails to catch certain failures during zone creation" [Undecided,New] | 17:03 |
| *** jmcbride has quit IRC | 17:06 | |
| *** jmcbride has joined #openstack-dns | 17:07 | |
| *** jmcbride has quit IRC | 17:07 | |
| *** jmcbride has joined #openstack-dns | 17:08 | |
| *** LinStatSDR has quit IRC | 17:16 | |
| *** boris-42 has joined #openstack-dns | 17:29 | |
| *** alokj has quit IRC | 17:29 | |
| *** mwagner_lap has quit IRC | 17:30 | |
| *** pino has quit IRC | 17:34 | |
| *** mikedillion has quit IRC | 17:40 | |
| *** jordanP has quit IRC | 17:44 | |
| *** jmcbride1 has joined #openstack-dns | 17:51 | |
| *** jmcbride has quit IRC | 17:51 | |
| *** ryanpetrello has joined #openstack-dns | 17:57 | |
| *** jmcbride1 has quit IRC | 18:01 | |
| *** jmcbride has joined #openstack-dns | 18:01 | |
| *** jmcbride has quit IRC | 18:06 | |
| *** jmcbride has joined #openstack-dns | 18:08 | |
| *** LinStatSDR has joined #openstack-dns | 18:15 | |
| *** shakamunyi has joined #openstack-dns | 18:18 | |
| *** mwagner_lap has joined #openstack-dns | 18:18 | |
| rjrjr | seeing some strange behavior. the test for record paging is failing. | 18:31 |
| rjrjr | the created_at timestamp for the SOA record is strange. | 18:32 |
| rjrjr | that record has a timestamp greater than all other records but the last record. | 18:32 |
| rjrjr | why would the created_at timestamp change at all? from the logic of the code, it should be right after the NS records. | 18:32 |
| *** mikedillion has joined #openstack-dns | 18:33 | |
| rjrjr | it seems like that was accounted for in the unit test, as the unit test expects that record to be created last, after all records. | 18:33 |
| rjrjr | which makes no sense to me.... | 18:33 |
| rjrjr | okay, figured it out. increment_serial in central is also updating the created_at value of a domain. | 18:43 |
| rjrjr | sorry, meant create_at value of the SOA record. | 18:48 |
| rjrjr | basically, the _update_soa method is creating a new SOA record everytime. | 18:48 |
| *** openstackgerrit has quit IRC | 18:50 | |
| *** openstackgerrit has joined #openstack-dns | 18:50 | |
| *** ChanServ sets mode: +v openstackgerrit | 18:50 | |
| vinod1 | _update_soa is calling update_recordset_in_storage and not create_recordset? Is update_recordset updating the create_time? | 18:50 |
| rjrjr | look at the code. it is creating a new record everytime. | 18:53 |
| rjrjr | i'm fixing it. | 18:53 |
| *** jmcbride has joined #openstack-dns | 18:54 | |
| rjrjr | the SOA record will be created first for a domain. it will always be the first 'created_at' record. | 18:56 |
| *** amcrn has joined #openstack-dns | 19:12 | |
| *** GonZo2K has joined #openstack-dns | 19:13 | |
| *** amcrn has quit IRC | 19:34 | |
| *** jmcbride has quit IRC | 19:57 | |
| *** jmcbride has joined #openstack-dns | 20:02 | |
| *** jmcbride has quit IRC | 20:02 | |
| *** jmcbride has joined #openstack-dns | 20:02 | |
| *** openstackgerrit has quit IRC | 20:04 | |
| *** openstackgerrit has joined #openstack-dns | 20:04 | |
| *** ChanServ sets mode: +v openstackgerrit | 20:04 | |
| *** shakamunyi has quit IRC | 20:30 | |
| *** shakamunyi has joined #openstack-dns | 20:32 | |
| *** jmcbride has quit IRC | 21:02 | |
| *** mikedillion has quit IRC | 21:08 | |
| openstackgerrit | Betsy Luzader proposed openstack/designate: Pool_Attribute API https://review.openstack.org/133549 | 21:12 |
| openstackgerrit | Betsy Luzader proposed openstack/designate: Migrate Server table https://review.openstack.org/136440 | 21:12 |
| *** vinod1 has quit IRC | 21:21 | |
| *** jmcbride has joined #openstack-dns | 21:25 | |
| *** vinod1 has joined #openstack-dns | 21:35 | |
| *** bitblt has joined #openstack-dns | 21:37 | |
| *** rediskin has quit IRC | 21:37 | |
| *** pino has joined #openstack-dns | 21:51 | |
| *** pino has quit IRC | 21:51 | |
| *** jmcbride has quit IRC | 21:52 | |
| *** jmcbride has joined #openstack-dns | 21:57 | |
| *** pino has joined #openstack-dns | 21:59 | |
| *** ryanpetrello has quit IRC | 22:18 | |
| *** jmcbride has quit IRC | 22:33 | |
| *** jmcbride has joined #openstack-dns | 22:35 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/designate: Updated from global requirements https://review.openstack.org/139227 | 22:36 |
| *** jmcbride1 has joined #openstack-dns | 22:52 | |
| *** jmcbride has quit IRC | 22:53 | |
| *** betsy has quit IRC | 22:57 | |
| *** jmcbride1 has quit IRC | 23:17 | |
| *** bitblt has quit IRC | 23:18 | |
| *** shakamunyi has quit IRC | 23:19 | |
| *** shakamunyi has joined #openstack-dns | 23:20 | |
| *** timsim has quit IRC | 23:28 | |
| *** ryanpetrello has joined #openstack-dns | 23:32 | |
| *** pino has quit IRC | 23:34 | |
| *** EricGonc_ has quit IRC | 23:48 | |
| *** vinod1 has quit IRC | 23:56 | |
| *** EricGonczer_ has joined #openstack-dns | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!