| *** johnbelamaric1 has quit IRC | 00:03 | |
| *** johnbelamaric has joined #openstack-dns | 00:07 | |
| *** shakamunyi has joined #openstack-dns | 01:07 | |
| *** shakamunyi_ has joined #openstack-dns | 01:07 | |
| *** johnbelamaric has quit IRC | 01:14 | |
| *** johnbelamaric has joined #openstack-dns | 01:31 | |
| *** stanzgy has joined #openstack-dns | 01:39 | |
| *** johnbelamaric has quit IRC | 01:46 | |
| *** johnbelamaric has joined #openstack-dns | 02:06 | |
| *** johnbelamaric has quit IRC | 02:26 | |
| *** stanzgy has quit IRC | 02:32 | |
| *** stanzgy has joined #openstack-dns | 02:32 | |
| *** stanzgy has quit IRC | 02:50 | |
| *** stanzgy has joined #openstack-dns | 02:50 | |
| *** shakamunyi__ has joined #openstack-dns | 02:51 | |
| *** shakamunyi has quit IRC | 02:54 | |
| *** shakamunyi_ has quit IRC | 02:54 | |
| *** shakamunyi has joined #openstack-dns | 02:54 | |
| *** richm has quit IRC | 04:38 | |
| *** shakamunyi_ has joined #openstack-dns | 04:39 | |
| *** shakamunyi__ has quit IRC | 04:41 | |
| *** shakamunyi has quit IRC | 04:41 | |
| *** shakamunyi has joined #openstack-dns | 04:41 | |
| *** shakamunyi_ has quit IRC | 05:04 | |
| *** shakamunyi has quit IRC | 05:18 | |
| *** nkinder has quit IRC | 05:22 | |
| *** shakamunyi has joined #openstack-dns | 05:22 | |
| *** nkinder has joined #openstack-dns | 05:29 | |
| *** shakamunyi has quit IRC | 06:25 | |
| *** GonZo2K has quit IRC | 06:28 | |
| *** km has joined #openstack-dns | 07:31 | |
| *** jordanP has joined #openstack-dns | 07:38 | |
| *** GonZo2K has joined #openstack-dns | 09:07 | |
| *** GonZo2K has quit IRC | 09:23 | |
| *** untriaged-bot has joined #openstack-dns | 10:01 | |
| untriaged-bot | Untriaged bugs so far: | 10:01 |
|---|---|---|
| untriaged-bot | https://bugs.launchpad.net/designate/+bug/1437699 | 10:01 |
| openstack | Launchpad bug 1437699 in Designate "mDNS should Handle Bad File Descriptor" [Medium,New] | 10:02 |
| *** untriaged-bot has quit IRC | 10:02 | |
| openstackgerrit | Graham Hayes proposed openstack/designate: Add _to_string() methods on RRData Objects https://review.openstack.org/167672 | 10:20 |
| openstackgerrit | Graham Hayes proposed openstack/designate: Replace API_v2 views with Adapters https://review.openstack.org/166983 | 10:20 |
| openstackgerrit | Graham Hayes proposed openstack/designate: Add Schema to RRData Objects https://review.openstack.org/167600 | 10:20 |
| openstackgerrit | Graham Hayes proposed openstack/designate: Add Decorator to log RPCAPI Calls https://review.openstack.org/167352 | 10:23 |
| openstackgerrit | Graham Hayes proposed openstack/designate: WIP: Logging Cleanup https://review.openstack.org/167353 | 10:23 |
| openstackgerrit | Graham Hayes proposed openstack/designate: Add __str__() methods to DesignateObjects https://review.openstack.org/142218 | 10:23 |
| openstackgerrit | Graham Hayes proposed openstack/designate: Add Enhanced Logging to APIs https://review.openstack.org/167351 | 10:23 |
| *** km has quit IRC | 10:32 | |
| *** km has joined #openstack-dns | 10:32 | |
| *** stanzgy has quit IRC | 10:52 | |
| *** mwagner_lap has quit IRC | 10:52 | |
| *** kei_yama has quit IRC | 11:13 | |
| *** km has quit IRC | 11:27 | |
| *** MentalRay has joined #openstack-dns | 11:29 | |
| *** johnbelamaric has joined #openstack-dns | 11:40 | |
| openstackgerrit | Ihar Hrachyshka proposed openstack/designate: Fixed rootwrap file installation https://review.openstack.org/168402 | 11:50 |
| *** richm has joined #openstack-dns | 12:11 | |
| *** johnbelamaric has quit IRC | 12:33 | |
| *** mwagner_lap has joined #openstack-dns | 12:53 | |
| *** shakamunyi has joined #openstack-dns | 12:57 | |
| *** shakamunyi has quit IRC | 13:13 | |
| *** nkinder has quit IRC | 13:16 | |
| *** johnbelamaric has joined #openstack-dns | 13:25 | |
| *** chlong has joined #openstack-dns | 13:36 | |
| *** andrewbogott_afk is now known as andrewbogott | 13:54 | |
| *** nkinder has joined #openstack-dns | 14:06 | |
| *** paul_glass has joined #openstack-dns | 14:14 | |
| *** MentalRay has quit IRC | 14:15 | |
| *** ericpeterson has joined #openstack-dns | 14:18 | |
| *** vinod1 has joined #openstack-dns | 14:27 | |
| *** jmcbride has joined #openstack-dns | 14:31 | |
| *** OS-PV has quit IRC | 14:35 | |
| *** jmcbride has quit IRC | 14:46 | |
| openstackgerrit | Tim Simmons proposed openstack/designate: Reduce the # of SQL queries during AXFRs https://review.openstack.org/167809 | 14:59 |
| *** jmcbride has joined #openstack-dns | 15:00 | |
| *** kfox1111 has quit IRC | 15:03 | |
| openstackgerrit | Merged openstack/designate: Fixed rootwrap file installation https://review.openstack.org/168402 | 15:09 |
| *** MentalRay has joined #openstack-dns | 15:09 | |
| *** shakamunyi has joined #openstack-dns | 15:26 | |
| *** chlong has quit IRC | 15:26 | |
| mugsie | timsim: care to re-add your +2 to https://review.openstack.org/#/c/166983/11 ? | 15:29 |
| timsim | Nah | 15:29 |
| mugsie | :D | 15:30 |
| mugsie | and then there is a chain from https://review.openstack.org/#/c/167600/ | 15:30 |
| timsim | ;) | 15:30 |
| mugsie | if you have time | 15:30 |
| timsim | Working my way through right now. I'm just finishing up checking Kiall's mdns ones. | 15:30 |
| mugsie | cool | 15:30 |
| *** kfox1111 has joined #openstack-dns | 15:36 | |
| kfox1111 | morning. :) | 15:36 |
| kfox1111 | So, with the v2 client, how do I enable it? | 15:37 |
| mugsie | kfox1111: it is about to be broken ;) | 15:46 |
| mugsie | when https://review.openstack.org/#/c/166983/11 merges, it will be broken for master | 15:47 |
| mugsie | to enable it - i am not sure - ekarlso wrote it to plugin to openstack-client | 15:47 |
| timsim | probably pull down the change and pip install . --upgrade or something like that | 15:48 |
| openstackgerrit | Endre Karlson proposed openstack/designate: Sync up FIP behaviour to match latest PoolManager https://review.openstack.org/168357 | 15:49 |
| *** MentalRay has quit IRC | 15:49 | |
| openstackgerrit | Endre Karlson proposed openstack/designate: Fix ownership issue with Zone's created on set fip https://review.openstack.org/168310 | 15:52 |
| *** shakamunyi has quit IRC | 15:52 | |
| openstackgerrit | Merged openstack/designate: Replace API_v2 views with Adapters https://review.openstack.org/166983 | 15:55 |
| mugsie | woo \o/ | 15:55 |
| openstackgerrit | Merged openstack/designate: Unwind mDNS XFRMixin from NotifyEndpoint and Handler https://review.openstack.org/166627 | 15:56 |
| *** untriaged-bot has joined #openstack-dns | 16:01 | |
| untriaged-bot | Untriaged bugs so far: | 16:01 |
| untriaged-bot | https://bugs.launchpad.net/designate/+bug/1437699 | 16:01 |
| openstack | Launchpad bug 1437699 in Designate "mDNS should Handle Bad File Descriptor" [Medium,New] | 16:01 |
| *** untriaged-bot has quit IRC | 16:02 | |
| dvorak | we're running the juno version of designate in a dev environment, and we see database deadlocks that cause API failures when trying to do parallel creation of records. Is there some way to reduce the chance of this happening in Juno, or do we just need to tell clients they have to retry? | 16:03 |
| kfox1111 | timsim: did that. I see the code installed. just not sure how to activate it. | 16:06 |
| *** shakamunyi has joined #openstack-dns | 16:10 | |
| ekarlso | dvorak: I wonder if a dude from blizzard had a fix for it no Kiall ? | 16:25 |
| dvorak | we'd be glad to try a patch if there is one you can point us towards | 16:25 |
| openstackgerrit | Merged openstack/designate: Add Schema to RRData Objects https://review.openstack.org/167600 | 16:26 |
| openstackgerrit | Merged openstack/designate: Add _to_string() methods on RRData Objects https://review.openstack.org/167672 | 16:27 |
| openstackgerrit | Endre Karlson proposed openstack/designate: Add code to allow triggering of AXFR from API https://review.openstack.org/165331 | 16:29 |
| openstackgerrit | Endre Karlson proposed openstack/designate: Add code to allow triggering of AXFR from API https://review.openstack.org/165331 | 16:29 |
| dvorak | we're also seeing records that get dropped out of the designate db, but not out of the powerdns db, which is pretty horrible, but not real surprising since we're seeing the deadlocks | 16:37 |
| mugsie | dvorak: Kiall did some pretty extensive digging into that bug | 16:38 |
| mugsie | not sure if he is online right now | 16:39 |
| Kiall | dvorak: I have a branch in the office with a fix for that backported, but it's failing tests and I haven't had a chance to go back to it | 16:42 |
| dvorak | think you could put that up somewhere? I'd like to at least take a look if you don't mind | 16:42 |
| *** crc32 has joined #openstack-dns | 16:44 | |
| kfox1111 | ekarlso: so, how do you enable the v2 api plugin? | 16:47 |
| ekarlso | kfox1111: :P | 16:48 |
| ekarlso | I need to fix it first to work with the latest changes to the api | 16:48 |
| kfox1111 | I see the code in /srv/designate/venv/lib/python2.7/site-packages/designateclient/osc/plugin.py but I don't know how to load it. | 16:48 |
| kfox1111 | I've got a slightly older server, so I think it will still work. | 16:49 |
| kfox1111 | or should. | 16:49 |
| kfox1111 | hmm.... though I'm not sure it will load at all.. one of the first lines is: from openstackclient.common import utils as oscutils | 16:50 |
| Kiall | dvorak: Humm.. Can't find it :/ But here's the master branch commit - https://github.com/openstack/designate/commit/f8e5161569802db64d1d7d1d4864f8a77c4fe2dd | 16:51 |
| Kiall | The issue with a straight backport is, the various methods in central/service.py which are decorated need to be split up so that only the storage code is retried.. The retry ensures we succeed even with a deadlock, and the split means that 1) we only retry the storage parts, and 2) the transaction window is reduced - makeing deadlocks less likely. | 16:51 |
| kfox1111 | and it doesn't look like it was in requirements, so I don't have openstackclient. | 16:51 |
| kfox1111 | maybe the plugin is failing to load silently because of that? | 16:51 |
| kfox1111 | heh. its odd to pip install python-openstackclient... | 16:53 |
| kfox1111 | no, that didn't get the plugin to load. :/ | 16:57 |
| ekarlso | kfox1111: so, you need to install python-openstackclient | 16:57 |
| ekarlso | and the v2-osc branch | 16:58 |
| ekarlso | then you should be getting commands like "zone create" etc | 16:58 |
| *** MentalRay has joined #openstack-dns | 17:00 | |
| dvorak | are there any existing tools for syncing the powerdns and designate databases? | 17:05 |
| dvorak | is that what the sync api calls are supposed to be for? | 17:05 |
| *** einarr has quit IRC | 17:06 | |
| ekarlso | dvorak: how you mean sync ? | 17:06 |
| ekarlso | you mean sync records in dns vs designate db ? | 17:06 |
| ekarlso | s/dns/pdns | 17:06 |
| dvorak | powerdns database has records in it that aren't in the designate db | 17:06 |
| *** MentalRay_ has joined #openstack-dns | 17:14 | |
| *** MentalRay has quit IRC | 17:17 | |
| *** pk__ has joined #openstack-dns | 17:25 | |
| *** crc32 has quit IRC | 17:36 | |
| kfox1111 | v2-osc? Is that different then the patches in the review? | 17:38 |
| kfox1111 | I installed python-openstackclient, but still no zone stuff in the cli. | 17:42 |
| kfox1111 | its either failing to load the plugin, or just not trying. not sure which. | 17:44 |
| dvorak | actually, the situation we seem to have now is that the entry in records has been deleted, but the rows in the recordsets table and the powerdns records table are still hanging around | 17:44 |
| *** jmcbride has quit IRC | 17:46 | |
| ekarlso | kfox1111: you need designateclietn with a specific branc halso! | 17:47 |
| ekarlso | https://review.openstack.org/#/c/133676/ | 17:47 |
| openstackgerrit | Merged openstack/designate: Ensure mDNS originated queries are non-blocking https://review.openstack.org/168627 | 17:53 |
| *** jordanP has quit IRC | 17:54 | |
| *** jmcbride has joined #openstack-dns | 18:02 | |
| *** johnbelamaric1 has joined #openstack-dns | 18:04 | |
| *** johnbelamaric has quit IRC | 18:07 | |
| *** MentalRay_ has quit IRC | 18:12 | |
| *** MentalRay_ has joined #openstack-dns | 18:14 | |
| dvorak | in kilo, does minidns completely replace the existing record handling behavior? | 18:19 |
| mugsie | dvorak: yes - we no longer us the DB for record operations | 18:19 |
| dvorak | heh, the kilo backend for powerdns certainly looks simpler than the juno one | 18:20 |
| mugsie | yeah :) | 18:21 |
| ahu | btw, https://doc.powerdns.com/md/httpapi/README/ | 18:22 |
| *** crc32 has joined #openstack-dns | 18:26 | |
| ekarlso | ahu: patches welcome :P | 18:38 |
| mugsie | ahu: yeah. the problem for us is making a standard interface for all DNS servers | 18:41 |
| mugsie | also, marked as experimental, so more difficult to maintain. that said we could use that for provisioning the zones themselves | 18:42 |
| mugsie | s/maintain/ship code dependant on it/ | 18:43 |
| openstackgerrit | Kiall Mac Innes proposed openstack/designate: Implement Pool Targets https://review.openstack.org/167678 | 18:45 |
| openstackgerrit | Kiall Mac Innes proposed openstack/designate: DynECT: Convert to a Pool Backend https://review.openstack.org/168588 | 18:45 |
| openstackgerrit | Kiall Mac Innes proposed openstack/designate: Add Akamai EDNS Backend https://review.openstack.org/168638 | 18:45 |
| openstackgerrit | Kiall Mac Innes proposed openstack/designate: Add also-notifies support to Pool Manager https://review.openstack.org/168600 | 18:45 |
| openstackgerrit | Graham Hayes proposed openstack/designate: Enable Record Data Validation in v2 API https://review.openstack.org/169044 | 18:45 |
| openstackgerrit | Graham Hayes proposed openstack/designate: Enable Record Data Validation in v2 API https://review.openstack.org/169044 | 18:48 |
| *** andrewbogott is now known as andrewbogott_afk | 18:52 | |
| dvorak | Kiall: around? | 18:58 |
| dvorak | at the bottom of this patch set, it looks like there is some unrelated pool test changes - https://github.com/openstack/designate/commit/f8e5161569802db64d1d7d1d4864f8a77c4fe2dd | 18:58 |
| dvorak | if I backport that to juno, I assume I can ignore that stuff | 18:58 |
| Kiall | Yep, but there may be similar stuff needed.. the commit message explains why those changes were necessary | 18:59 |
| Kiall | Also .. A backport will need to break up central/service.py's create/update/delete_domain/record/etc methods into 2 each | 18:59 |
| ahu | mugsie: it will be stable soon, just making sure you know about it | 19:00 |
| ahu | ekarlso: I already gave powerdns to you, ok? ;-) | 19:00 |
| Kiall | dvorak: Otherwise, the retry will do things like retry the entire create - including provisioning on the backend, emitting notifications etc etc | 19:00 |
| Kiall | ahu: :D In all likelyhood, once that API is stable, we'll create a second pdns driver that uses it (have to support older powerdns for quite some time..) | 19:01 |
| dvorak | I'll take a look. Getting it passing tests on stable/juno was easy, not sure what you're describing will be :) | 19:01 |
| Kiall | Yep, that's where I stumbled too.. I think I might have rm -rf'd my designate clone since I started it :( | 19:02 |
| dvorak | heh, I'm guessing that's why this is at the top of create_domain: "# TODO(kiall): Refactor this method into *MUCH* smaller chunks." | 19:03 |
| ekarlso | ahu: hah! :p | 19:03 |
| Kiall | dvorak: Lol, I think that's the longest standing TODO in the codebase! pre-exists this particular issue :) | 19:05 |
| dvorak | did that break up happen for kilo? Since this patch landed | 19:05 |
| Kiall | Yea, but it landed as part of a much bigger change that happened to also need it | 19:05 |
| Kiall | https://github.com/openstack/designate/blob/master/designate/central/service.py#L847 | 19:06 |
| Kiall | Those <bla>_in_storage() methods are what get wrapped now, rather than the whole <bla> method | 19:06 |
| dvorak | makes sense | 19:06 |
| dvorak | would I be a crazy person if I was thinking of just using master for production use? I know I asked the same question about a month or two ago and was told yes :) | 19:07 |
| dvorak | where production = very limited beta | 19:07 |
| Kiall | master has got much better over the last little bit, but we're still working through issues and changes, but it's probably in a good enough state righ now... rc1 is 3 days after all | 19:08 |
| Kiall | is in 3 days* | 19:08 |
| Kiall | That said.. master in production is asking fort trouble ;) | 19:09 |
| dvorak | well, it seems like a lot of things are fixed on master | 19:09 |
| dvorak | Kiall: yes, but juno in production is known trouble :) | 19:09 |
| dvorak | if we spin up 5 threads creating records, we have a 100% failure rate on runs of 100 records | 19:09 |
| dvorak | rather, 100% chance of failure | 19:10 |
| openstackgerrit | Eric Larson proposed openstack/designate: Add the Hook Point API https://review.openstack.org/164748 | 19:10 |
| Kiall | yea, I was sent a small script that causes it to bail with less :( | 19:10 |
| Kiall | And - we now do this kinda testing in the gate so we don't let it happen again :) | 19:11 |
| dvorak | we're also ending up with orphaned records in the recordsets table and in the powerdns database | 19:11 |
| dvorak | when it fails | 19:11 |
| dvorak | so I can definitely write something to clean that up, but it's pretty gross, and hopefully not a problem we'd have with kilo | 19:11 |
| dvorak | sounds like I should probably take a day or two and see how much work moving to kilo would be | 19:12 |
| dvorak | we'd also like to be on a newer oslo.messaging anyway, and I assume that works with master | 19:12 |
| Kiall | In theory, the `designate sync-domain` command should help, but I think rjrjr found a bug with it.. | 19:12 |
| dvorak | Kiall: it's not implemented for the powerdns backend | 19:12 |
| mfisch | yeah it throws 404 currently | 19:12 |
| Kiall | There's a default implementation for all drivers, re the 404, enable the extension :) | 19:13 |
| Kiall | sync-domain was intended as a dev time tool to bascially delete and recreate the zone on the backend .. The default naieve implementation will just go "delete everything, recreate everything" | 19:13 |
| Kiall | i.e. it's not downtime free | 19:13 |
| mfisch | so for our issue we have orphans in the designate recordset table too | 19:14 |
| mfisch | will it solve that? | 19:14 |
| dvorak | we're not serving records directly from powerdns, so that's not an issue for us | 19:14 |
| openstackgerrit | Endre Karlson proposed openstack/designate: Sync up FIP behaviour to match latest PoolManager https://review.openstack.org/168357 | 19:14 |
| dvorak | we use infoblox for our auth dns, and we have it slaving zones from powerdns | 19:14 |
| Kiall | Ooo, really? The designate DB is filled with FK's to prevent orphans? | 19:14 |
| mfisch | data in the recordsets without a matching record, yes | 19:15 |
| mfisch | dvorak can show you the query | 19:15 |
| dvorak | select count(rs.id) from recordsets as rs left join records as r on r.recordset_id = rs.id where r.id is null; returns 19 | 19:15 |
| Kiall | Ah, Okay.. I get you | 19:15 |
| dvorak | I assume that's broken | 19:15 |
| mfisch | I work with dvorak btw - so Hi | 19:15 |
| Kiall | Let me dig through the commit logs, pretty sure that was a simple fix | 19:15 |
| mfisch | thx | 19:16 |
| Kiall | https://github.com/openstack/designate/commit/1dba323cd529654268e8f48bf673e6b0b60b0500 | 19:16 |
| dvorak | also, mfisch and I are giving the talk opposite the ebay guys about running designate in prod, except we're not yet. Our talk was optimistically titled :) | 19:16 |
| mfisch | we will be | 19:16 |
| Kiall | I don't particular like that fix, but the disconnect between the V1 API and V2 API models sucjs | 19:16 |
| Kiall | sucks* | 19:16 |
| mfisch | Is a recordset considered empty when there's no record pointing at it? | 19:17 |
| Kiall | Yea - It is.. which reminds me, before we release Kilo we have to fix the bug in that commit! (Check+Set race condition) | 19:18 |
| * Kiall looks forward to getting some bugfix only time | 19:19 | |
| mfisch | Kiall: that fix is already in juno | 19:20 |
| mfisch | its from July | 19:20 |
| Kiall | Oh, humm.. Then I wonder where the orphans are coming from? (are they orphans or parents without children? heh) | 19:21 |
| mfisch | Kiall: do you want some data? | 19:21 |
| Kiall | So - ebay, right? That means you're using designate-sink? | 19:21 |
| dvorak | no, time warner cable | 19:21 |
| mfisch | no we're not Ebay we're TWC | 19:21 |
| dvorak | we're planning to use sink though | 19:22 |
| mfisch | just happen to both be giving the same talk | 19:22 |
| Kiall | Ah, okay :) | 19:22 |
| Kiall | Since your not using that yet, sink can be ruled out as a possibility ... | 19:22 |
| dvorak | just not there yet. that's actually our biggest concern around the api failures, is that at some point we'll have records being created automatically, and probably not slowly like people will | 19:22 |
| dvorak | Kiall: the orphaned records are created during a stress test that mfisch wrote | 19:22 |
| mfisch | Well most are. Some are 2 weeks old | 19:22 |
| mfisch | Kiall: I create and the resolve N records in a loop. I ran the test 4 times in parallel | 19:23 |
| Kiall | I'm guessing the root cause is the DBDeadlocks... But it's hard to tell from IRC alone ;) If you guys don't have success doing that backport (the retry + the split), I can try tackle it again after rc1 ships... | 19:24 |
| mfisch | We certainly had deadlock messages in the logs | 19:24 |
| *** penick has joined #openstack-dns | 19:24 | |
| dvorak | Kiall: I tend to think it'd be a better use of our time to try kilo, and then if we find (and maybe fix) bugs then we're at least not working against a dead branch of code | 19:25 |
| dvorak | and if we can move to kilo now, we can avoid having to migrate to minidns later | 19:25 |
| Kiall | makes sense - Just have to remember that master is still baking :) e.g. we're aware of things like performance issues where things simply fall over, and silly mistakes imposiing artificial limits .. e.g. if zone ends up over 65k bytes when rendered to wire format it won't AXFR | 19:29 |
| dvorak | yeah, I saw something about that. Are you doing AXFR over UDP or something? | 19:29 |
| Kiall | TCP (or UDP..) but, the way an AXFR is rendered to wire format at 65k DNS wirefomat is somewhat odd... in that you send the first 65k (Header+First Part of Payload), then send (Header again+Second Part of Payload) | 19:32 |
| dvorak | ah weird. I've never done any DNS over TCP stuff, but I guess that makes sense. I'm guessing the packet size is hard coded to a 16bit value? | 19:33 |
| *** andrewbogott_afk is now known as andrewbogott | 19:34 | |
| Kiall | Anyway! It's 8pm and I still have wayy too much to get finished today :( Gotta run! | 19:35 |
| dvorak | np, thanks for all the help | 19:36 |
| openstackgerrit | Merged openstack/designate: RFC1996 Requires NOTIFY packets have the AA flag set https://review.openstack.org/168618 | 19:40 |
| mugsie | vinod1: timsim https://review.openstack.org/#/c/166983/ is a bit .... wierd :) ping me if you have any questions | 19:44 |
| timsim | Wrong link? | 19:45 |
| timsim | Or just that whole patch in general :P ? | 19:45 |
| mugsie | the whole patch ;) | 19:46 |
| mugsie | https://review.openstack.org/#/c/169044/2 | 19:46 |
| mugsie | ah | 19:46 |
| mugsie | damn tabs | 19:46 |
| mugsie | the core of the weirdness is here https://review.openstack.org/#/c/169044/2/designate/objects/recordset.py | 19:46 |
| *** jmcbride has quit IRC | 19:47 | |
| ericpeterson | hey all, I notice that the designate horizon stuff does not support policy files (so far as I read). Is there a bug / bp around this? I'd be willing to do the work for this / I need it. | 19:49 |
| openstackgerrit | Endre Karlson proposed openstack/designate: Add code to allow triggering of AXFR from API https://review.openstack.org/165331 | 19:51 |
| *** jmcbride has joined #openstack-dns | 19:52 | |
| Kiall | ericpeterson: the code was written before the horizon policy stuff existed (I think anyway!) .. Would be happy to see it implemented :) | 19:53 |
| ericpeterson | thanks, just wanted to check before I started on it. :D | 19:53 |
| openstackgerrit | Endre Karlson proposed openstack/designate: Add code to allow triggering of AXFR from API https://review.openstack.org/165331 | 19:55 |
| ekarlso | ok, addressed all in ^ | 19:55 |
| ekarlso | can we get it merged Kiall ? :p | 19:55 |
| Kiall | ekarlso: will take a look in a few mins :) | 19:55 |
| ekarlso | Kiall: and my fip patch as well please :P | 19:56 |
| Kiall | Nope! You can only have one, pick! :P | 19:56 |
| openstackgerrit | Kiall Mac Innes proposed openstack/designate: Implement Pool Targets https://review.openstack.org/167678 | 20:06 |
| openstackgerrit | Kiall Mac Innes proposed openstack/designate: DynECT: Convert to a Pool Backend https://review.openstack.org/168588 | 20:06 |
| openstackgerrit | Kiall Mac Innes proposed openstack/designate: Add Akamai EDNS Backend https://review.openstack.org/168638 | 20:06 |
| openstackgerrit | Kiall Mac Innes proposed openstack/designate: Add also-notifies support to Pool Manager https://review.openstack.org/168600 | 20:06 |
| Kiall | timsim / vinod1 - if you get time - https://review.openstack.org/167678 would be much appreciated, as I said in one of the review comments - it really looks larger than it is;) | 20:07 |
| Kiall | Also - you might have noticed gate-designate-dsvm-{dynect,akamai} ... Those are A) WIP .. and B) probably something we'll have to not leave public, trying to figure out how to pretent the credentials when you're effectivly running Remote Code Execution as a Service :( | 20:10 |
| Kiall | protect the credentials* | 20:10 |
| timsim | Does gerrit not have an API where you can just leave a +1, -1, no logs? | 20:10 |
| dvorak | is SSH an api? ;) | 20:11 |
| vinod1 | kiall: for your patchset, I noticed that there were a few py27 failures too - but I did not see any changes to address them | 20:11 |
| Kiall | vinod1: I did fix them, it was a misplaced ) | 20:11 |
| Kiall | timsim: it does, but that would be agaianst the 3rd Party Testing rules ;) | 20:11 |
| timsim | Ah. I've never looked at that :P | 20:12 |
| Kiall | we could try something simple .. like a sed 's/password/********/g' over everything... But then all you have to do is submit a patch with "echo $DYN_PASSWORD | rot13" and you get the keys :( | 20:12 |
| kfox1111 | ekarlso: I have /srv/designate/venv/lib/python2.7/site-packages/designateclient/osc/plugin.py on the file system. that means I have the right branch, right? | 20:15 |
| kfox1111 | I see the code from both reviews under designateclient. | 20:17 |
| kfox1111 | I'm just not seeing the plugin code activate. | 20:18 |
| ekarlso | kfox1111: you wanna do a tmate session ? | 20:18 |
| kfox1111 | tmate? | 20:18 |
| kfox1111 | oh. | 20:19 |
| kfox1111 | yeah, won't work due to it constraints. :/ | 20:19 |
| ekarlso | kfox1111: bad firewallS ?:p | 20:22 |
| kfox1111 | yeah. corperate firewall blocks a lot of things. | 20:22 |
| kfox1111 | policy tends to block much of the rest. ;) | 20:22 |
| ekarlso | :p | 20:22 |
| kfox1111 | so, if I grep for plugin in the code, I only see the osc/plugin.py file. | 20:23 |
| kfox1111 | through what mechanism does the plugin get loaded? | 20:23 |
| ekarlso | kfox1111: stevedore | 20:23 |
| kfox1111 | heh. could be worse though. I'm at least dealing with the unclassified stuff.... | 20:24 |
| ekarlso | kfox1111: if you type "openstack help" it doesn't like "zone create" etc ? | 20:24 |
| kfox1111 | AH... the answer is, I'm using it wrong. :) | 20:25 |
| kfox1111 | it doesn't show up in 'designate -h' | 20:25 |
| kfox1111 | it does show up in 'openstack -h' :) | 20:25 |
| dvorak | huh, how do I get the plugin? Someone said something about it being in a special branch? | 20:26 |
| ekarlso | dvorak: yeah, but it doesn't work with latest master though | 20:26 |
| dvorak | ah, ok | 20:26 |
| dvorak | is that adding v1 or v2 api support or both? | 20:26 |
| openstackgerrit | Tim Simmons proposed openstack/designate: Agent Pool Manager Backend https://review.openstack.org/169073 | 20:27 |
| kfox1111 | I did a git clone of the python-designateclient, then: git fetch ssh://kfox1111@review.openstack.org:29418/openstack/python-designateclient refs/changes/75/133675/9 && git checkout FETCH_HEAD | 20:27 |
| kfox1111 | v2. | 20:27 |
| timsim | Kiall: ^^ Managed to get Agent working with the targets. | 20:27 |
| Kiall | timsim: excellent, hit any issues? | 20:27 |
| timsim | I had to put host, port into the options part of the config. Are they supposed to be on the same level as options, or contained in there? | 20:28 |
| *** ericpete_ has joined #openstack-dns | 20:28 | |
| ekarlso | dvorak: v2 to openstack client | 20:28 |
| dvorak | nice | 20:28 |
| Kiall | timsim: yea, the config format is a tad weird for that.. try "designate-manage pool show_config" - you'll recognise the format from the mid-cycle ;) | 20:29 |
| openstackgerrit | Graham Hayes proposed openstack/designate: Enable Record Data Validation in v2 API https://review.openstack.org/169044 | 20:30 |
| kfox1111 | openstack zone list: ERROR: openstack __init__() got an unexpected keyword argument 'allow' | 20:30 |
| kfox1111 | so close... | 20:30 |
| timsim | Kiall: Yeah, here's my question: http://paste.openstack.org/show/197615/ | 20:31 |
| *** ericpeterson has quit IRC | 20:32 | |
| Kiall | timsim: no, I don't think so? host/port aren't common to all backends .. e.g. for powerdns it's a DB URL, for Dyn, a user and password etc | 20:32 |
| timsim | Ok cool, I didn't remember if that was what we had decided or not. Totally fine. | 20:32 |
| Kiall | :) | 20:33 |
| *** vinod1 has quit IRC | 20:35 | |
| openstackgerrit | Tim Simmons proposed openstack/designate: Agent Pool Manager Backend https://review.openstack.org/169073 | 20:38 |
| *** mwagner_lap has quit IRC | 20:40 | |
| ekarlso | kfox1111: yeah ,that's the perk of the v2 thing not being stable yet and keystone discovery not supporting anything but "stable" .. | 20:42 |
| kfox1111 | ok. :/ | 20:44 |
| kfox1111 | thanks. :) | 20:44 |
| kfox1111 | would pulling grunk keystoneclient fix it then, or do I need trunk lots of things? | 20:45 |
| kfox1111 | s/grunk/trunk/ | 20:45 |
| *** paul_glass1 has joined #openstack-dns | 20:47 | |
| *** paul_glass has quit IRC | 20:47 | |
| ekarlso | kfox1111: not a good patch but https://review.openstack.org/130159 | 20:47 |
| *** vinod1 has joined #openstack-dns | 20:49 | |
| kfox1111 | oh. ok. thanks. | 20:49 |
| *** vinod1 has quit IRC | 21:09 | |
| *** vinod1 has joined #openstack-dns | 21:09 | |
| openstackgerrit | Kiall Mac Innes proposed openstack/designate: Implement Pool Targets https://review.openstack.org/167678 | 21:13 |
| openstackgerrit | Kiall Mac Innes proposed openstack/designate: DynECT: Convert to a Pool Backend https://review.openstack.org/168588 | 21:13 |
| openstackgerrit | Kiall Mac Innes proposed openstack/designate: Add Akamai EDNS Backend https://review.openstack.org/168638 | 21:13 |
| openstackgerrit | Kiall Mac Innes proposed openstack/designate: Add also-notifies support to Pool Manager https://review.openstack.org/168600 | 21:13 |
| Kiall | better be the last time -_- | 21:13 |
| *** vinod1 has quit IRC | 21:15 | |
| *** vinod1 has joined #openstack-dns | 21:17 | |
| *** boris-42 has quit IRC | 21:18 | |
| *** boris-42 has joined #openstack-dns | 21:30 | |
| timsim | Kiall: What's with the the powerdns gate? | 21:35 |
| Kiall | I made the change graham asked for and It's broke ;) | 21:35 |
| timsim | Figures. :) | 21:36 |
| Kiall | Was working from home today, so didn't bother running devstack locally on my laptop ;) | 21:36 |
| Kiall | serves me right... | 21:36 |
| Kiall | Fingers crossed -_- | 21:38 |
| openstackgerrit | Kiall Mac Innes proposed openstack/designate: Implement Pool Targets https://review.openstack.org/167678 | 21:39 |
| openstackgerrit | Kiall Mac Innes proposed openstack/designate: DynECT: Convert to a Pool Backend https://review.openstack.org/168588 | 21:39 |
| openstackgerrit | Kiall Mac Innes proposed openstack/designate: Add Akamai EDNS Backend https://review.openstack.org/168638 | 21:39 |
| openstackgerrit | Kiall Mac Innes proposed openstack/designate: Add also-notifies support to Pool Manager https://review.openstack.org/168600 | 21:39 |
| Kiall | Battery in laptop nearly dead... Not even enough juice for a devsack run ;) will be my last attempt today ;) | 21:39 |
| *** vinod1 has quit IRC | 21:42 | |
| timsim | Vinod just showed me that you can select a "reference version" in gerrit. http://courseware.codeschool.com/images/blog/mind-blown.gif | 21:43 |
| Kiall | timsim: it's nearly useless IMO ;) Half the time it'll show the differeneces between PS1 and PS2, with all changes megred to master between the two aswell :( | 21:44 |
| timsim | Better than trying to find the two lines that changed in a 2000 line patchset :P | 21:44 |
| kfox1111 | ok. so, I have a domain. pilot.cloud.pnnl.gov. its owned by the mgmt tenant. | 21:45 |
| kfox1111 | I want to make a tenant for a specific tenant, 'kfox'. called kfox.pilot.cloud.pnnl.gov and give it to that thenant, so it can manage it, itself. | 21:46 |
| kfox1111 | how do I go about doing that? | 21:46 |
| kfox1111 | I can't create it as kfox. says its parent is owned by another tenant. which is true. | 21:46 |
| kfox1111 | I created it ok in the mgmt tenant. how do I transfer it to the user tenant? | 21:46 |
| timsim | kfox1111: Are you on Juno or master? | 21:47 |
| kfox1111 | master | 21:47 |
| kfox1111 | minuts a few days. | 21:47 |
| *** jmcbride has quit IRC | 21:47 | |
| timsim | kfox1111: http://docs-draft.openstack.org/09/167809/5/check/gate-designate-docs/4ae5f17//doc/build/html/rest/v2/zones.html#transfer-zone | 21:48 |
| kfox1111 | .... ok. | 21:48 |
| *** jmcbride has joined #openstack-dns | 21:49 | |
| timsim | Weird link because the docs have changed in the last few days and I don't know how to get older docs. | 21:49 |
| * kfox1111 loves how openstack projects change terms often from the things they wrap. | 21:49 | |
| timsim | Does that make sense? Basically you're going to initialize a transfer to your new tenant, go accept it, and the zone will transfer tenants. | 21:49 |
| kfox1111 | well, I guess the order is changed... zone transfer vs transfer zone... | 21:49 |
| kfox1111 | yeah. it does. | 21:49 |
| kfox1111 | thanks. :) | 21:50 |
| * kfox1111 breaks out curl. :/ | 21:50 | |
| *** johnbelamaric1 has left #openstack-dns | 21:54 | |
| *** johnbelamaric has joined #openstack-dns | 21:55 | |
| *** pk__ has quit IRC | 21:57 | |
| *** johnbelamaric has quit IRC | 21:59 | |
| *** pk has joined #openstack-dns | 21:59 | |
| *** jmcbride has quit IRC | 21:59 | |
| *** nkinder has quit IRC | 22:13 | |
| openstackgerrit | Eric Peterson proposed openstack/designate: Add policy support to designatedashboard https://review.openstack.org/169106 | 22:14 |
| *** pk has quit IRC | 22:15 | |
| *** ericpete_ has quit IRC | 22:26 | |
| *** paul_glass1 has quit IRC | 22:35 | |
| *** MentalRay_ has quit IRC | 22:48 | |
| *** andrewbogott has quit IRC | 23:10 | |
| *** andrewbogott has joined #openstack-dns | 23:11 | |
| *** andrewbogott has joined #openstack-dns | 23:11 | |
| *** pk has joined #openstack-dns | 23:13 | |
| *** pk__ has joined #openstack-dns | 23:15 | |
| *** pk has quit IRC | 23:15 | |
| *** penick has quit IRC | 23:16 | |
| *** jmcbride has joined #openstack-dns | 23:24 | |
| *** logan2 has quit IRC | 23:25 | |
| *** logan2 has joined #openstack-dns | 23:33 | |
| *** pk__ has quit IRC | 23:45 | |
| *** pk has joined #openstack-dns | 23:47 | |
| *** km has joined #openstack-dns | 23:50 | |
| *** shakamunyi has quit IRC | 23:55 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
