| *** cleverdevil has quit IRC | 00:11 | |
| *** puranamr has quit IRC | 00:15 | |
| *** sarob has quit IRC | 00:20 | |
| *** puranamr has joined #akanda | 04:58 | |
| *** puranamr has quit IRC | 05:07 | |
| *** rods has joined #akanda | 09:38 | |
| *** openstackgerrit has quit IRC | 11:13 | |
| *** openstackgerrit has joined #akanda | 11:14 | |
| openstackgerrit | David Lenwell proposed stackforge/akanda-rug: WIP: sketch of lbaas driver https://review.openstack.org/181545 | 16:52 |
|---|---|---|
| *** sarob has joined #akanda | 16:53 | |
| openstackgerrit | David Lenwell proposed stackforge/akanda-rug: WIP: sketch of lbaas driver https://review.openstack.org/181545 | 16:57 |
| *** puranamr has joined #akanda | 17:06 | |
| adam_g | sarob, some comments on https://review.openstack.org/#/c/190312/ | 17:12 |
| adam_g | oh, cool https://review.openstack.org/#/c/188909/ merged, so your other doc patch will be a good test to make sure they update | 17:14 |
| openstackgerrit | Adam Gandelman proposed stackforge/akanda-appliance: Test broken experimental devstack image build (DO NOT MERGE) https://review.openstack.org/189380 | 18:05 |
| adam_g | ryanpetrello, do you guys actually use the RUG's ceilometer notifier thing? | 18:12 |
| ryanpetrello | nope | 18:12 |
| adam_g | hmm | 18:12 |
| adam_g | i wonder if it works | 18:12 |
| markmcclain | it did at one time | 18:18 |
| openstackgerrit | Merged stackforge/akanda-appliance: Remove arping from the list of base packages. https://review.openstack.org/189419 | 18:21 |
| sarob | adam_g: why keep the akanda-rug/doc(s) director? | 18:22 |
| sarob | directy | 18:22 |
| sarob | arg | 18:22 |
| adam_g | sarob, so that we can have a pointer to the main docs for people who aren't aware of projet layout | 18:23 |
| sarob | ah | 18:23 |
| sarob | so along that line | 18:23 |
| adam_g | renaming it ./doc -> ./docs means readthedocs.org can pick it up | 18:23 |
| adam_g | and by extension, google | 18:23 |
| sarob | we should have docs dir in all the projects | 18:23 |
| sarob | repos | 18:24 |
| sarob | that is | 18:24 |
| sarob | i get it | 18:24 |
| adam_g | i know ive asked before but i guess i dont remember the reasons, but why shouldn't we move the things in akanda-appliance-builder into akanda-appliance? | 18:28 |
| adam_g | im looking at adding the post-job to publish build and publish images, but it doesn't make sense to do post-commit on 2 repos | 18:29 |
| adam_g | markmcclain, i remember you had some reasons for the split | 18:30 |
| markmcclain | they were originally split for speed, but you know dib better | 18:32 |
| markmcclain | mainly because I couldn't find the right hook to make dib use my local tree for Akanda | 18:33 |
| markmcclain | but you've worked around it | 18:33 |
| markmcclain | for the devstack stuff, so I'd be find rolling the two together | 18:33 |
| adam_g | ok cool | 18:35 |
| adam_g | yeah, it seems awkward to have a single image be generated for commits to two repositories | 18:35 |
| openstackgerrit | Sean Roberts proposed stackforge/akanda-rug: Update the docs directory https://review.openstack.org/190312 | 18:35 |
| adam_g | having the build recipe coupled with the code its building seems to make sense, and we can have a make target or build script in the akanda-appliance repo that the bot can use to build | 18:36 |
| adam_g | sarob, any objection to dropping the akanda-appliance-builder project/repo? | 18:36 |
| sarob | adam_g: nope | 18:37 |
| sarob | adam_g: less is more | 18:37 |
| markmcclain | yeah.. simplifying is definitely the way to go | 18:38 |
| adam_g | cool. filed https://bugs.launchpad.net/akanda/+bug/1464359 to track, will take a look at merging later today | 18:42 |
| openstack | Launchpad bug 1464359 in akanda "akanda-appliance-builder should be merged into akanda-appliance" [Undecided,New] | 18:42 |
| adam_g | markmcclain, re ceilometer stuff, what ends up causing the rug workers to call the bandwidth callback? | 18:43 |
| sarob | adam_g: you are picking up https://bugs.launchpad.net/akanda/+bug/1464359 as part of your ci work? | 18:57 |
| openstack | Launchpad bug 1464359 in akanda "akanda-appliance-builder should be merged into akanda-appliance" [High,Confirmed] | 18:57 |
| adam_g | sarob, yesw | 18:57 |
| sarob | okey | 19:00 |
| *** puranamr has quit IRC | 19:29 | |
| openstackgerrit | Merged stackforge/akanda-rug: Migrate to oslo.utils https://review.openstack.org/183156 | 20:07 |
| *** puranamr has joined #akanda | 20:54 | |
| adam_g | davidlenwell, you've been deep into state machine, any idea how to trigger a transition through the ReadStats state? | 21:24 |
| *** puranamr has quit IRC | 21:26 | |
| davidlenwell | off the top of my head.. no .. right now im trying to accurately map its behavior in a flow chart.. the current docs aren't exactly accurate or easy to follow | 21:30 |
| *** puranamr has joined #akanda | 21:35 | |
| *** puranamr has quit IRC | 21:38 | |
| *** puranamr has joined #akanda | 21:47 | |
| adam_g | seemed to b e able to get it to execute that transition by hacking a READ command into rug-ctl | 21:51 |
| *** puranamr has quit IRC | 21:54 | |
| *** rods has quit IRC | 21:54 | |
| *** puranamr has joined #akanda | 22:00 | |
| markmcclain | adam_g: yeah.. looks like read support hook was removed | 22:13 |
| adam_g | markmcclain, ive got it running through there now to discover the vm manager doesn't support what the state machine/ReadStats state expects | 22:14 |
| adam_g | im going to fix it up to the point where its putting some form of bandwidth notification messages on the wire /w oslo.messaging and we can worry about fixing up the interface, defining a message format, etc later | 22:15 |
| adam_g | i dont really know what ceilometer expects to see, or how we'd get bandwidth stats to begin with | 22:15 |
| markmcclain | adam_g: that makes sense just to get something working | 22:24 |
| markmcclain | as for format we could try to match the metering messages from the neutron ref impl | 22:24 |
| adam_g | markmcclain, what was the original idea behind the read state? would that be an additional part of the polling phase? | 22:24 |
| markmcclain | otherwise we'd ahve to write a plugin for ceilometer to understand the message | 22:25 |
| adam_g | or something that was triggered via rug-ctl or w/e | 22:25 |
| adam_g | markmcclain, following the neutron spec seems like the best idea | 22:25 |
| markmcclain | originally we'd just spawn a sub process to post the event for each router periodically | 22:28 |
| markmcclain | or it's also possible to change the healthmon | 22:28 |
| markmcclain | to create this event vs the poll which requests the stats, but throws them away | 22:29 |
| openstackgerrit | Sean Roberts proposed stackforge/akanda-rug: Update the docs directory https://review.openstack.org/190312 | 22:30 |
| adam_g | ah | 22:33 |
| *** puranamr has quit IRC | 22:40 | |
| openstackgerrit | Adam Gandelman proposed stackforge/akanda-rug: Rework messaging layer for oslo.messaging (WIP) https://review.openstack.org/190401 | 22:52 |
| openstackgerrit | Adam Gandelman proposed stackforge/akanda-rug: Add a log shim to avoid a dependency on openstack.common.log https://review.openstack.org/190400 | 22:52 |
| openstackgerrit | Adam Gandelman proposed stackforge/akanda-rug: Repair the semi-functional notification publisher https://review.openstack.org/190854 | 22:52 |
| *** puranamr has joined #akanda | 23:01 | |
| *** puranamr has quit IRC | 23:06 | |
| *** puranamr has joined #akanda | 23:11 | |
| openstackgerrit | Adam Gandelman proposed stackforge/akanda-appliance: Check in diskimage-builder elements https://review.openstack.org/190859 | 23:15 |
| openstackgerrit | Adam Gandelman proposed stackforge/akanda-rug: Build appliance with DIB elements from akanda-appliance https://review.openstack.org/190860 | 23:16 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!