| openstackgerrit | Adam Gandelman proposed openstack/astara: WIP Add clustering capabilities to instance manager https://review.openstack.org/267263 | 00:42 |
|---|---|---|
| openstackgerrit | Adam Gandelman proposed openstack/astara-appliance: Ensure interface cache up to date https://review.openstack.org/273706 | 00:48 |
| *** xiayu has joined #openstack-astara | 01:12 | |
| openstackgerrit | Adam Gandelman proposed openstack/astara: Fix docs path in setup.cfg https://review.openstack.org/273887 | 05:32 |
| *** Liuqing has joined #openstack-astara | 07:38 | |
| *** Liuqing has quit IRC | 09:34 | |
| *** Liuqing has joined #openstack-astara | 10:48 | |
| *** Liuqing has quit IRC | 11:15 | |
| *** xiayu has left #openstack-astara | 12:24 | |
| *** prithiv has joined #openstack-astara | 17:55 | |
| *** prithiv has quit IRC | 18:04 | |
| *** pcaruana has joined #openstack-astara | 18:57 | |
| openstackgerrit | Adam Gandelman proposed openstack/astara-appliance: Adjust build script to not use rev-parse https://review.openstack.org/274206 | 19:21 |
| openstackgerrit | Adam Gandelman proposed openstack/astara-appliance: Adjust build script to not use rev-parse https://review.openstack.org/274206 | 19:22 |
| *** pcaruana has quit IRC | 19:24 | |
| *** pcaruana has joined #openstack-astara | 19:37 | |
| *** shashank_hegde has joined #openstack-astara | 20:01 | |
| *** pcaruana has quit IRC | 20:06 | |
| openstackgerrit | Merged openstack/astara: devstack: Fix image build variable check https://review.openstack.org/273726 | 20:08 |
| *** pcaruana has joined #openstack-astara | 20:09 | |
| *** pcaruana has quit IRC | 20:19 | |
| openstackgerrit | mark mcclain proposed openstack/astara-appliance: Add vim package back to image https://review.openstack.org/274224 | 20:37 |
| markmcclain | adam_g: I think I know why we're setting random freezes during configuration | 20:42 |
| adam_g | markmcclain, oh? | 20:45 |
| markmcclain | seems there's a subtle behavior difference when we're sending oversized packets | 20:46 |
| markmcclain | lowering the MTU on the mgt interface seems to solve the issue in some local testing | 20:46 |
| adam_g | hmm | 20:47 |
| cleverdevil | fzylogic ^^ | 20:48 |
| cleverdevil | this may be relevant to your interests :) | 20:48 |
| cleverdevil | (and drwahl) | 20:48 |
| adam_g | markmcclain, lowering the MTU on the mgt interface in the appliance, or on the orchestrator node? | 20:48 |
| markmcclain | the orchestrator node usually gets the right value | 20:48 |
| markmcclain | within the appliance | 20:49 |
| markmcclain | it's low vol mgt traffic | 20:49 |
| adam_g | so, maybe your idea of defaulting its MTU to 1450 is a good one? | 20:49 |
| markmcclain | yeah.. that's my plan | 20:49 |
| markmcclain | and then I we can push a coordinated config update allows the orchestrator to tell the appliance to correct size | 20:50 |
| markmcclain | do you have a logstash query handy? | 20:50 |
| markmcclain | or have we basically gotten rid of the freeze within the gate? | 20:50 |
| * markmcclain wants to check some other test variables | 20:51 | |
| adam_g | markmcclain, no its not possible to query really | 20:52 |
| adam_g | i have seen it just a couple times today on stable/liberty (*where we still run the old test suite) | 20:52 |
| adam_g | one sec | 20:52 |
| adam_g | markmcclain, http://logs.openstack.org/78/273778/1/check/gate-functional-dsvm-astara/b3e8da7/ | 20:53 |
| adam_g | but im not certain those failures we saw were really caused by a freeze | 20:53 |
| adam_g | when it passed, the job finished in the same state, so.. | 20:53 |
| markmcclain | this might work | 20:54 |
| * adam_g needs to run some errands. back in a bit | 21:09 | |
| drwahl | yo, akanda. just had an idea... | 22:10 |
| drwahl | we're talking about using flavors (and network QoS) to throttle tenants. this would also effect astara | 22:11 |
| drwahl | however, it's concievable to have some tenants who need/want/get more internets than some other tenant | 22:11 |
| markmcclain | it could depending on how the flavors are created | 22:11 |
| drwahl | which would mean that we'd need different flavors for routers | 22:11 |
| drwahl | thoughts? | 22:11 |
| markmcclain | right.. out plan is to add flavor support this cycle | 22:12 |
| drwahl | i don't know of a way for the rug to apply different flavors right now | 22:12 |
| drwahl | ah, kk | 22:12 |
| drwahl | you're ahead of me, then :) | 22:12 |
| drwahl | in that case... as you were | 22:12 |
| openstackgerrit | mark mcclain proposed openstack/astara-appliance: Make the management interface MTU configurable https://review.openstack.org/274281 | 22:46 |
| openstackgerrit | mark mcclain proposed openstack/astara: Add variable MTU support to the orchestrator. https://review.openstack.org/274293 | 23:54 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!