| *** shakamunyi has joined #openstack-dns | 00:24 | |
| *** HenryG_afk is now known as HenryG | 00:57 | |
| *** shakamunyi_ has joined #openstack-dns | 01:03 | |
| *** EricGonczer_ has joined #openstack-dns | 01:29 | |
| *** EricGonczer_ has quit IRC | 01:50 | |
| *** balles has quit IRC | 02:20 | |
| *** balles has joined #openstack-dns | 02:22 | |
| *** vinod1 has joined #openstack-dns | 02:40 | |
| *** vinod1 has quit IRC | 03:11 | |
| *** balles has quit IRC | 03:54 | |
| *** balles has joined #openstack-dns | 03:56 | |
| *** EricGonczer_ has joined #openstack-dns | 04:04 | |
| *** EricGonczer_ has quit IRC | 04:11 | |
| *** shakamunyi_ has quit IRC | 04:47 | |
| *** shakamunyi has quit IRC | 04:47 | |
| *** balles has quit IRC | 07:12 | |
| *** balles has joined #openstack-dns | 07:13 | |
| *** balles has quit IRC | 09:02 | |
| *** balles has joined #openstack-dns | 09:04 | |
| *** balles has quit IRC | 10:30 | |
| *** balles has joined #openstack-dns | 10:31 | |
| *** EricGonczer_ has joined #openstack-dns | 11:44 | |
| *** EricGonczer_ has quit IRC | 11:58 | |
| *** EricGonczer_ has joined #openstack-dns | 12:03 | |
| *** EricGonczer_ has quit IRC | 12:04 | |
| *** EricGonczer_ has joined #openstack-dns | 12:17 | |
| *** vinod1 has joined #openstack-dns | 12:34 | |
| *** vinod1 has quit IRC | 12:59 | |
| *** vinod1 has joined #openstack-dns | 12:59 | |
| *** vinod1 has quit IRC | 13:30 | |
| *** vinod1 has joined #openstack-dns | 13:41 | |
| openstackgerrit | Kiall Mac Innes proposed a change to openstack/designate: Move DNS packet serialization into mDNS service https://review.openstack.org/118046 | 14:09 |
|---|---|---|
| openstackgerrit | Kiall Mac Innes proposed a change to openstack/designate: Implement mDNS Middleware https://review.openstack.org/118047 | 14:09 |
| Kiall | vinod1: about? Started work on getting the Middleware/TSIG suff in place ^ | 14:09 |
| Kiall | (It's layered on top of betsy + your changes - so no conflicts ;) | 14:10 |
| vinod1 | Kiall - I will take a look at it shortly. | 14:10 |
| Kiall | Sure, The actual TSIG bits are coming next.. Only been at it a few mins ;) | 14:11 |
| vinod1 | Now that you are here - I added checking AXFR in the devstack exercises | 14:11 |
| vinod1 | but looking at the log http://logs.openstack.org/31/109731/10/check/gate-designate-devstack-dsvm/25dc762/console.html | 14:12 |
| vinod1 | designate-mdns is not started | 14:12 |
| vinod1 | Relevant lines from the log | 14:12 |
| vinod1 | http://logs.openstack.org/31/109731/10/check/gate-designate-devstack-dsvm/25dc762/console.html#_2014-08-30_20_46_13_340 | 2014-08-30 20:46:13.275 | +++ [[ -z designate,designate-api,designate-central,designate-sink ]] | 14:12 |
| vinod1 | http://logs.openstack.org/31/109731/10/check/gate-designate-devstack-dsvm/25dc762/console.html#_2014-08-30_20_46_13_385 | 2014-08-30 20:46:13.332 | ++++ ENABLED_SERVICES=c-api,c-bak,c-sch,c-vol,ceilometer-acentral,ceilometer-acompute,ceilometer-alarm-evaluator,ceilometer-alarm-notifier,ceilometer-anotification,ceilometer-api,ceilometer-collector,cinder,dstat,g-api,g-reg,h-api,h-api-cfn,h-api-cw,h-eng,heat,horizon,key,mysql,n-api,n- | 14:12 |
| vinod1 | I added designate-mdns to localrc - https://review.openstack.org/#/c/109731/10/contrib/devstack/localrc | 14:13 |
| vinod1 | Also in localrc I don't actually see designate-sink enabled - so is there a different place that it needs to be enabled? | 14:14 |
| Kiall | Sorry, was AFK a minute | 14:19 |
| Kiall | mdns starts up for be locally with devstack+the vagrant VM | 14:19 |
| Kiall | I *think* the gate job is configured to start just the core services | 14:19 |
| Kiall | Yep | 14:20 |
| Kiall | https://github.com/openstack-infra/config/blob/master/modules/openstack_project/files/jenkins_job_builder/config/designate.yaml#L18 | 14:20 |
| Kiall | If you add this to the gate_hook.sh file, next to the export KEEP_LOCALRC=1, it should start it up: | 14:21 |
| Kiall | export ENABLED_SERVICES=designate,designate-api,designate-central,designate-sink,designate-mdns | 14:21 |
| Kiall | then we can remove the hardcode fom the gate job itself | 14:21 |
| Kiall | And - the localrc file is only used for the vagrant devstack VM, devstack-gate will generate a localrc during the job run | 14:22 |
| Kiall | (The vagrant stuff should probably be pulled out of the contrib/devstack folder, to make ^ more obvious) | 14:22 |
| openstackgerrit | Vinod Mangalpally proposed a change to openstack/designate: Add support for AXFR requests https://review.openstack.org/109731 | 14:28 |
| *** vinod1 has quit IRC | 14:34 | |
| openstackgerrit | Kiall Mac Innes proposed a change to openstack/designate: Move DNS packet serialization into mDNS service https://review.openstack.org/118046 | 14:37 |
| openstackgerrit | Kiall Mac Innes proposed a change to openstack/designate: Implement mDNS Middleware https://review.openstack.org/118047 | 14:37 |
| openstackgerrit | Kiall Mac Innes proposed a change to openstack/designate: Implement mDNS Middleware https://review.openstack.org/118047 | 14:42 |
| *** vinod1 has joined #openstack-dns | 14:47 | |
| *** EricGonczer_ has quit IRC | 15:00 | |
| openstackgerrit | Kiall Mac Innes proposed a change to openstack/designate: Implement mDNS Middleware https://review.openstack.org/118047 | 15:09 |
| Kiall | Annoyingly, DNSPython won't let me implement tsig the way I want :'( | 15:28 |
| ekarlso | :P | 15:43 |
| Kiall | Yea.. It's being a... Pain :/ | 15:50 |
| openstackgerrit | Vinod Mangalpally proposed a change to openstack/designate: Add support for AXFR requests https://review.openstack.org/109731 | 15:50 |
| Kiall | It won't decode a TSIG signed packet without also validating it.. Which kinda forces the location of TSIG code -_- ... And, once validated, it doesn't actually seem to offer any way for us to know what key was used to validate, so, we have no clue what key to use in the response. | 15:51 |
| *** EricGonczer_ has joined #openstack-dns | 15:54 | |
| Kiall | It remembers the algorithm and that it was signed, just not the keyname | 15:54 |
| Kiall | (even though it says it should be :() | 15:54 |
| *** vinod1 has quit IRC | 15:59 | |
| Kiall | OMG. And, DNSPython strips out the TSIG fields -_- | 16:15 |
| Kiall | So, I can't even parse them myself to get the key name -_- | 16:15 |
| *** EricGonczer_ has quit IRC | 16:18 | |
| Kiall | lol | 16:23 |
| Kiall | https://github.com/rthalley/dnspython/commit/2db54ba975ed5be22f92af105bb2bd6b0ccbbef1 | 16:23 |
| Kiall | i.e. it's broke, and they haven't done a release -_- | 16:24 |
| Kiall | Well, that's that then. | 16:24 |
| Kiall | It's actually not possible to implement due to the bug, and we can't depend on an unreleased version. | 16:25 |
| *** vinod1 has joined #openstack-dns | 16:40 | |
| openstackgerrit | Vinod Mangalpally proposed a change to openstack/designate: Add support for AXFR requests https://review.openstack.org/109731 | 16:43 |
| Kiall | vinod1: BTW, I've pushed the tsig stuff to Kilo.. A dnspython bug makes implementing it impossible :( | 16:44 |
| vinod1 | ok. can we try to get it fixed in dnspython? | 16:45 |
| Kiall | https://github.com/rthalley/dnspython/commit/2db54ba975ed5be22f92af105bb2bd6b0ccbbef1 is the fix, but we would need them to do a release etc in the next couple of days before the j3 cutoff | 16:45 |
| Kiall | And, then we would need to get openstack/requirements updated to accept a min of that new version | 16:46 |
| Kiall | etc etc | 16:46 |
| Kiall | I'm going to email them tomorrow in the hopes of getting it done.. We'll see though! | 16:46 |
| vinod1 | agreed - a lot of other dependencies to get done - i am fine with moving it to kilo | 16:47 |
| *** balles has quit IRC | 16:50 | |
| *** balles has joined #openstack-dns | 16:52 | |
| *** vinod1 has quit IRC | 17:18 | |
| *** balles has quit IRC | 17:52 | |
| *** balles has joined #openstack-dns | 17:54 | |
| harmw | Kiall: are you around to help in some troubleshooting? | 18:40 |
| *** vinod1 has joined #openstack-dns | 19:06 | |
| Kiall | harmw: kinda, what's up? | 19:37 |
| harmw | I'm having trouble hooking up designate with nova | 19:39 |
| harmw | ive got designate listening in on notifications topic, in the nova_fixed handler | 19:40 |
| harmw | but sink is tripping on almost every message | 19:40 |
| harmw | TypeError: __init__() got an unexpected keyword argument 'instance_lock_checked' | 19:40 |
| harmw | https://bugs.launchpad.net/designate/+bug/1360696 | 19:41 |
| uvirtbot | Launchpad bug 1360696 in designate "oslo notify dispatcher, unexpected keyword argument" [Undecided,New] | 19:41 |
| harmw | I can add that keyword to init() but that ofcourse isn't the way to go - I'd rather know what I'm doing wrong in my setup :) | 19:42 |
| Kiall | Looking now | 19:44 |
| harmw | ok, cool | 19:46 |
| Kiall | I think I might see what's happening, 1 sec | 19:46 |
| harmw | ah nice | 19:50 |
| Kiall | Humm... So, we're trying to take a JSON serialized context from nova/neutron, and deserialize it into a Designate Context object. The issue is that they aren't actually the same. | 19:50 |
| harmw | ok, sad story :p | 19:51 |
| Kiall | There is a separate code path for deserializing contexts from notifications, but we're just assuming Nova/Neutron's match ours in it | 19:51 |
| harmw | ok, does the bugreport contain enough info to brew up a fix for this? | 19:52 |
| Kiall | Yea - It does. | 19:53 |
| harmw | great | 19:53 |
| Kiall | There is likely an easy fix, just looking through oslo.messaging now to see if I need to write the 5 lines of code myself or not ;) | 19:53 |
| harmw | oh lol | 19:53 |
| harmw | nice | 19:53 |
| harmw | btw, the docs mention notification_topics = monitor but that yields nothing - atleast with the current codebase | 19:54 |
| harmw | switching to notification_topics = notifications brings me a lot of magic, but is that the way to go? | 19:54 |
| *** vinod1 has quit IRC | 19:57 | |
| Kiall | The value depends on what you have set at the nova/neutron side.. if both sides are set to "abcdef", then "abcdef" is the right value.. That said, everyones defaults bar ours have recently-ish changed to notifications ... we should update our docs | 20:00 |
| openstackgerrit | Kiall Mac Innes proposed a change to openstack/designate: Received notifications context is not a DesignateContext https://review.openstack.org/118060 | 20:03 |
| Kiall | harmw: untested, but I believe ^ should work. | 20:03 |
| harmw | Kiall: thanks a lot, Ill take a look | 20:03 |
| Kiall | Could you try it out and let me know? | 20:04 |
| harmw | but ofcourse :) | 20:04 |
| * Kiall should have ran the tests before uploading.. Looks like loads are broken now ;) | 20:05 | |
| Kiall | Oh.. nvm | 20:05 |
| Kiall | dodgy local env | 20:05 |
| openstackgerrit | Kiall Mac Innes proposed a change to openstack/designate: Received notifications context is not a DesignateContext https://review.openstack.org/118060 | 20:07 |
| Kiall | harmw: patchset 2 shouldn't, ya know, break everything.. | 20:07 |
| harmw | hehe | 20:08 |
| harmw | ok, so, what's the easy way of getting the patch without a full git checkout? | 20:08 |
| Kiall | If you have a local clone of designate, you can: | 20:09 |
| Kiall | git fetch https://review.openstack.org/openstack/designate refs/changes/60/118060/2 && git format-patch -1 --stdout FETCH_HEAD | 20:09 |
| Kiall | and .diff will be output.. | 20:09 |
| Kiall | you can apply that to your install to test | 20:09 |
| harmw | I know I can do that, but I don't have the local checkout ;) | 20:09 |
| harmw | https://review.openstack.org/#/c/118060/2/designate/rpc.py,unified | 20:10 |
| harmw | from there I can download a zip, as it seems | 20:10 |
| Kiall | harmw: http://paste.openstack.org/show/103823/ | 20:10 |
| Kiall | or.. http://paste.openstack.org/raw/103823/ | 20:11 |
| Kiall | for less HTML | 20:11 |
| harmw | yea, that'll do fine as well :) | 20:11 |
| harmw | thanks | 20:11 |
| harmw | ok, creating new instance now | 20:14 |
| harmw | ouch, new error | 20:14 |
| harmw | http://paste.openstack.org/show/103824/ | 20:14 |
| Kiall | Oh, Humm.. Okay, even less code is needed. oslo.messaging already did the JSON decode.. heh | 20:16 |
| harmw | :p | 20:16 |
| Kiall | If you reverse the previous patch, then try this: http://paste.openstack.org/show/103826/ | 20:17 |
| openstackgerrit | Kiall Mac Innes proposed a change to openstack/designate: Received notifications context is not a DesignateContext https://review.openstack.org/118060 | 20:17 |
| * Kiall needs to get designate-sink tested in the gate. | 20:18 | |
| harmw | Kiall: great, I know get a DomainNotFound exception, so it's finally atleast trying to do somthing :) | 20:24 |
| harmw | thanks! | 20:24 |
| Kiall | Okay, it's got past that issue anyway :) excellent! | 20:24 |
| harmw | indeed | 20:24 |
| harmw | ah, my config contains some other id :| | 20:26 |
| *** EricGonczer_ has joined #openstack-dns | 20:26 | |
| openstackgerrit | Kiall Mac Innes proposed a change to openstack/designate: Update default notification-topics for nova/neutron https://review.openstack.org/118063 | 20:27 |
| Kiall | And, defaults updated from monitor -> notifications ... | 20:27 |
| harmw | ah great :) | 20:27 |
| harmw | Kiall: it's working now, nice! | 20:27 |
| Kiall | Excellent, all that's left to do is get designate-sink tested as part of the gate so it doesn't break again :D | 20:28 |
| harmw | it just created the A and AAAA records for this new instance | 20:28 |
| harmw | haha yea | 20:28 |
| harmw | anyway, thanks a lot Kiall :) | 20:32 |
| Kiall | No problem | 20:32 |
| *** EricGonczer_ has quit IRC | 20:48 | |
| *** vinod1 has joined #openstack-dns | 22:17 | |
| *** vinod1 has quit IRC | 23:32 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!