| *** harlowja has quit IRC | 00:00 | |
| *** shakamunyi has quit IRC | 00:05 | |
| manishg | _cjones_ got to run. I don't have horizon setup. | 00:06 |
|---|---|---|
| *** manishg has quit IRC | 00:07 | |
| *** lykinsbd has quit IRC | 00:07 | |
| *** lykinsbd has joined #openstack-neutron | 00:07 | |
| *** dfarrell07 has joined #openstack-neutron | 00:08 | |
| _cjones_ | manishg: It seems that you need to specify --provider:network_type local. --provider:network_type local | 00:08 |
| _cjones_ | No worries. Thanks for the help. | 00:08 |
| *** prad_ has quit IRC | 00:09 | |
| *** lykinsbd has quit IRC | 00:09 | |
| *** xuhanp has joined #openstack-neutron | 00:09 | |
| *** lykinsbd has joined #openstack-neutron | 00:09 | |
| *** vivek-ebay has quit IRC | 00:11 | |
| openstackgerrit | Armando Migliaccio proposed a change to openstack/neutron: L2 Model additions to support DVR https://review.openstack.org/102101 | 00:14 |
| *** annegent_ has joined #openstack-neutron | 00:15 | |
| openstackgerrit | Armando Migliaccio proposed a change to openstack/neutron: RPC additions to support DVR https://review.openstack.org/102332 | 00:18 |
| *** annegent_ has quit IRC | 00:20 | |
| *** ranger81_ has quit IRC | 00:21 | |
| *** dsneddon has joined #openstack-neutron | 00:21 | |
| *** dsneddon has quit IRC | 00:23 | |
| *** german_ has quit IRC | 00:25 | |
| *** nlahouti has quit IRC | 00:25 | |
| *** dfarrell07 has quit IRC | 00:25 | |
| openstackgerrit | Armando Migliaccio proposed a change to openstack/neutron: ML2 additions to support DVR https://review.openstack.org/102398 | 00:27 |
| *** xuhanp has quit IRC | 00:28 | |
| *** karimb has joined #openstack-neutron | 00:31 | |
| *** Sukhdev has joined #openstack-neutron | 00:31 | |
| *** jaypipes has quit IRC | 00:35 | |
| *** matsuhashi has joined #openstack-neutron | 00:35 | |
| *** padkrish has quit IRC | 00:36 | |
| *** Sukhdev has quit IRC | 00:38 | |
| *** kong has joined #openstack-neutron | 00:41 | |
| *** coolsvap is now known as coolsvap|afk | 00:44 | |
| *** WackoRobie has joined #openstack-neutron | 00:52 | |
| *** karimb has quit IRC | 00:53 | |
| *** carl_baldwin has quit IRC | 00:54 | |
| *** morganfainberg is now known as morganfainberg_Z | 00:57 | |
| *** bandarji has quit IRC | 00:58 | |
| *** achanda has quit IRC | 01:01 | |
| *** _cjones_ has quit IRC | 01:03 | |
| *** _cjones_ has joined #openstack-neutron | 01:04 | |
| *** puck has quit IRC | 01:05 | |
| *** zhipeng has quit IRC | 01:05 | |
| *** zhipeng has joined #openstack-neutron | 01:06 | |
| *** _cjones_ has quit IRC | 01:08 | |
| *** ijw has joined #openstack-neutron | 01:09 | |
| *** rkukura_ has joined #openstack-neutron | 01:10 | |
| *** baoli has quit IRC | 01:10 | |
| *** ijw has quit IRC | 01:10 | |
| *** rkukura has quit IRC | 01:10 | |
| *** rkukura_ is now known as rkukura | 01:10 | |
| *** ijw has joined #openstack-neutron | 01:11 | |
| *** baoli has joined #openstack-neutron | 01:14 | |
| *** annegent_ has joined #openstack-neutron | 01:16 | |
| openstackgerrit | Armando Migliaccio proposed a change to openstack/neutron: L2 Agent-side additions to support DVR https://review.openstack.org/87730 | 01:16 |
| *** annegent_ has quit IRC | 01:21 | |
| *** diegows has quit IRC | 01:22 | |
| *** a_le has quit IRC | 01:24 | |
| *** a_le has joined #openstack-neutron | 01:24 | |
| *** sweston_ has quit IRC | 01:25 | |
| *** a_le has quit IRC | 01:25 | |
| *** iwamoto has joined #openstack-neutron | 01:26 | |
| *** oda-g has joined #openstack-neutron | 01:26 | |
| *** seizadi has joined #openstack-neutron | 01:26 | |
| *** gduan has quit IRC | 01:27 | |
| *** harlowja_ has quit IRC | 01:28 | |
| *** garyduan has joined #openstack-neutron | 01:31 | |
| *** yamamoto_ has joined #openstack-neutron | 01:32 | |
| *** harlowja has joined #openstack-neutron | 01:32 | |
| *** amcrn_ has quit IRC | 01:33 | |
| *** manishg has joined #openstack-neutron | 01:34 | |
| *** dfarrell07 has joined #openstack-neutron | 01:34 | |
| *** manishg_ has joined #openstack-neutron | 01:37 | |
| *** sbfox has joined #openstack-neutron | 01:37 | |
| *** geekinutah has quit IRC | 01:37 | |
| *** manishg has quit IRC | 01:38 | |
| *** manishg_ is now known as manishg | 01:38 | |
| *** nati_ueno has quit IRC | 01:40 | |
| manishg | rkukura, yt? | 01:44 |
| *** sbfox has quit IRC | 01:46 | |
| *** shakamunyi has joined #openstack-neutron | 01:47 | |
| amotoki | gus: thanks for reminding. I just reviewed again. | 01:47 |
| *** alexpilotti has quit IRC | 01:55 | |
| *** dfarrell07 has quit IRC | 01:57 | |
| *** sbfox has joined #openstack-neutron | 01:57 | |
| *** dfarrell07 has joined #openstack-neutron | 01:57 | |
| *** banix has joined #openstack-neutron | 01:59 | |
| *** ranger81 has joined #openstack-neutron | 02:00 | |
| *** manishg has quit IRC | 02:01 | |
| *** zhipeng has quit IRC | 02:02 | |
| *** dfarrell07 has quit IRC | 02:02 | |
| *** zhipeng has joined #openstack-neutron | 02:03 | |
| *** morganfainberg_Z is now known as morganfainberg | 02:06 | |
| *** seizadi has quit IRC | 02:10 | |
| *** TrevorV_ has joined #openstack-neutron | 02:12 | |
| openstackgerrit | IWAMOTO Toshihiro proposed a change to openstack/neutron: Make l3-agent be able to load advanced service drivers https://review.openstack.org/96691 | 02:22 |
| *** seizadi has joined #openstack-neutron | 02:24 | |
| *** sbfox has quit IRC | 02:26 | |
| *** dims__ has quit IRC | 02:28 | |
| *** VijayB has quit IRC | 02:31 | |
| *** openstackgerrit has quit IRC | 02:31 | |
| *** changbl has joined #openstack-neutron | 02:32 | |
| HenryG | sc68cal: ping | 02:33 |
| *** baojg has joined #openstack-neutron | 02:33 | |
| *** otherwiseguy has joined #openstack-neutron | 02:33 | |
| *** carl_baldwin has joined #openstack-neutron | 02:35 | |
| *** mitz_ has joined #openstack-neutron | 02:36 | |
| *** otherwiseguy has quit IRC | 02:40 | |
| *** vivek-ebay has joined #openstack-neutron | 02:41 | |
| *** otherwiseguy has joined #openstack-neutron | 02:41 | |
| *** annegent_ has joined #openstack-neutron | 02:42 | |
| *** TrevorV_ has quit IRC | 02:43 | |
| *** bandarji has joined #openstack-neutron | 02:44 | |
| *** annegent_ has quit IRC | 02:46 | |
| *** fifieldt_ is now known as fifieldt | 02:47 | |
| *** ranger81 has quit IRC | 02:48 | |
| *** sweston_ has joined #openstack-neutron | 02:54 | |
| *** dims__ has joined #openstack-neutron | 02:54 | |
| *** konglingxian has joined #openstack-neutron | 02:55 | |
| *** otherwiseguy has quit IRC | 02:55 | |
| *** baojg has quit IRC | 02:56 | |
| *** otherwiseguy has joined #openstack-neutron | 02:56 | |
| *** baojg has joined #openstack-neutron | 02:56 | |
| *** kong has quit IRC | 02:58 | |
| *** dims__ has quit IRC | 02:58 | |
| *** harlowja is now known as harlowja_away | 03:04 | |
| *** zhhuabj has joined #openstack-neutron | 03:05 | |
| *** armax has left #openstack-neutron | 03:07 | |
| *** essessv has joined #openstack-neutron | 03:11 | |
| *** sbfox has joined #openstack-neutron | 03:17 | |
| *** seizadi has quit IRC | 03:22 | |
| *** ranger81 has joined #openstack-neutron | 03:22 | |
| *** gildub has quit IRC | 03:24 | |
| *** achanda has joined #openstack-neutron | 03:25 | |
| *** nati_ueno has joined #openstack-neutron | 03:25 | |
| *** spandhe has quit IRC | 03:25 | |
| *** WackoRobie has quit IRC | 03:26 | |
| *** WackoRobie has joined #openstack-neutron | 03:26 | |
| *** WackoRobie has quit IRC | 03:26 | |
| *** WackoRobie has joined #openstack-neutron | 03:27 | |
| *** WackoRobie has quit IRC | 03:31 | |
| *** manishg has joined #openstack-neutron | 03:32 | |
| *** sbfox has quit IRC | 03:35 | |
| *** dfarrell07 has joined #openstack-neutron | 03:37 | |
| *** bandarji has quit IRC | 03:41 | |
| *** Xurong has quit IRC | 03:42 | |
| *** matsuhashi has quit IRC | 03:42 | |
| *** matsuhashi has joined #openstack-neutron | 03:43 | |
| *** german_ has joined #openstack-neutron | 03:44 | |
| *** ijw has quit IRC | 03:44 | |
| *** matsuhashi has quit IRC | 03:47 | |
| *** zhiyan_ is now known as zhiyan | 03:49 | |
| *** gongysh has joined #openstack-neutron | 03:51 | |
| *** dims__ has joined #openstack-neutron | 03:55 | |
| *** WackoRobie has joined #openstack-neutron | 03:57 | |
| *** WackoRobie has quit IRC | 03:58 | |
| *** dims__ has quit IRC | 03:59 | |
| *** ijw has joined #openstack-neutron | 04:01 | |
| *** dfarrell07 has quit IRC | 04:01 | |
| *** ijw has quit IRC | 04:02 | |
| *** ijw has joined #openstack-neutron | 04:02 | |
| *** WackoRobie has joined #openstack-neutron | 04:04 | |
| *** baojg has quit IRC | 04:04 | |
| *** carl_baldwin has quit IRC | 04:06 | |
| *** carl_baldwin has joined #openstack-neutron | 04:07 | |
| *** bandarji has joined #openstack-neutron | 04:08 | |
| *** nati_ueno has quit IRC | 04:08 | |
| *** nati_ueno has joined #openstack-neutron | 04:09 | |
| *** bandarji has quit IRC | 04:10 | |
| *** WackoRobie has quit IRC | 04:13 | |
| *** otherwiseguy has quit IRC | 04:13 | |
| *** otherwiseguy has joined #openstack-neutron | 04:14 | |
| *** gildub has joined #openstack-neutron | 04:15 | |
| *** shakamunyi has quit IRC | 04:16 | |
| *** annegent_ has joined #openstack-neutron | 04:16 | |
| *** shakamunyi has joined #openstack-neutron | 04:17 | |
| *** shakamunyi has quit IRC | 04:17 | |
| *** nlahouti has joined #openstack-neutron | 04:17 | |
| *** nlahouti has joined #openstack-neutron | 04:18 | |
| *** shakamunyi has joined #openstack-neutron | 04:20 | |
| *** annegent_ has quit IRC | 04:21 | |
| *** nplanel_ has quit IRC | 04:21 | |
| *** baojg has joined #openstack-neutron | 04:22 | |
| *** mlavalle has quit IRC | 04:25 | |
| *** banix has quit IRC | 04:25 | |
| *** yfried has quit IRC | 04:25 | |
| *** ramishra has joined #openstack-neutron | 04:26 | |
| *** carl_baldwin has quit IRC | 04:26 | |
| *** nati_ueno has quit IRC | 04:29 | |
| *** matsuhashi has joined #openstack-neutron | 04:31 | |
| *** garyk has quit IRC | 04:32 | |
| *** puck has joined #openstack-neutron | 04:33 | |
| *** Demitar has joined #openstack-neutron | 04:34 | |
| *** VenuMurthy has joined #openstack-neutron | 04:34 | |
| *** german_ has quit IRC | 04:36 | |
| *** Demitar has quit IRC | 04:38 | |
| *** lykinsbd has quit IRC | 04:38 | |
| *** otherwiseguy has quit IRC | 04:39 | |
| *** achanda has quit IRC | 04:39 | |
| *** lori|away is now known as lori | 04:42 | |
| *** Sukhdev has joined #openstack-neutron | 04:44 | |
| *** kopparam has joined #openstack-neutron | 04:46 | |
| *** nplanel_ has joined #openstack-neutron | 04:52 | |
| *** SridharG has joined #openstack-neutron | 04:52 | |
| *** seizadi has joined #openstack-neutron | 04:55 | |
| *** dims__ has joined #openstack-neutron | 04:56 | |
| *** ramishra has quit IRC | 04:59 | |
| *** achanda has joined #openstack-neutron | 04:59 | |
| *** ajc_ has joined #openstack-neutron | 04:59 | |
| *** yfried has joined #openstack-neutron | 04:59 | |
| *** shakamunyi has quit IRC | 05:00 | |
| *** dims__ has quit IRC | 05:00 | |
| *** baojg has quit IRC | 05:02 | |
| *** baojg has joined #openstack-neutron | 05:02 | |
| *** nplanel_ has quit IRC | 05:03 | |
| *** baojg has quit IRC | 05:04 | |
| *** baojg has joined #openstack-neutron | 05:04 | |
| *** vivek-ebay has quit IRC | 05:05 | |
| *** ramishra has joined #openstack-neutron | 05:05 | |
| *** zhipeng has quit IRC | 05:06 | |
| *** zhipeng has joined #openstack-neutron | 05:06 | |
| *** WackoRobie has joined #openstack-neutron | 05:08 | |
| *** blogan is now known as zz_blogan | 05:09 | |
| *** zhipeng has quit IRC | 05:11 | |
| *** a_le has joined #openstack-neutron | 05:11 | |
| *** WackoRobie has quit IRC | 05:13 | |
| *** morganfainberg is now known as morganfainberg_Z | 05:14 | |
| *** annegent_ has joined #openstack-neutron | 05:16 | |
| *** pradipta_away is now known as pradipta | 05:17 | |
| *** markmcclain has joined #openstack-neutron | 05:17 | |
| *** nati_ueno has joined #openstack-neutron | 05:17 | |
| *** markmcclain1 has joined #openstack-neutron | 05:18 | |
| *** coolsvap|afk is now known as coolsvap | 05:19 | |
| *** manishg has quit IRC | 05:20 | |
| *** annegent_ has quit IRC | 05:21 | |
| *** markmcclain has quit IRC | 05:22 | |
| *** rwsu has quit IRC | 05:24 | |
| *** nati_ueno has quit IRC | 05:26 | |
| *** ramishra has quit IRC | 05:31 | |
| *** amitpp has joined #openstack-neutron | 05:31 | |
| *** ramishra has joined #openstack-neutron | 05:37 | |
| *** markmcclain1 has quit IRC | 05:39 | |
| *** amitpp has quit IRC | 05:42 | |
| *** garyk has joined #openstack-neutron | 05:43 | |
| *** ihrachyshka has joined #openstack-neutron | 05:46 | |
| *** ihrachyshka has quit IRC | 05:46 | |
| *** sbfox has joined #openstack-neutron | 05:47 | |
| kashyap | rpm -qa | grep -i neutron | 05:47 |
| kashyap | Bah, sorry, wrong window | 05:47 |
| *** suresh12 has quit IRC | 05:50 | |
| *** suresh12 has joined #openstack-neutron | 05:51 | |
| *** suresh12 has quit IRC | 05:56 | |
| *** Sukhdev has quit IRC | 05:56 | |
| *** dims__ has joined #openstack-neutron | 05:56 | |
| *** chandan_kumar has quit IRC | 06:00 | |
| *** dims__ has quit IRC | 06:01 | |
| *** Xurong has joined #openstack-neutron | 06:03 | |
| *** vimal_ has joined #openstack-neutron | 06:07 | |
| kopparam | Hello! | 06:07 |
| *** ijw_ has joined #openstack-neutron | 06:07 | |
| *** zhipeng has joined #openstack-neutron | 06:08 | |
| *** ijw_ has quit IRC | 06:09 | |
| kopparam | I'm trying to install OVS on a compute node following this doc http://docs.openstack.org/havana/install-guide/install/yum/content/install-neutron.install-plugin-compute.ovs.html | 06:09 |
| *** seizadi has quit IRC | 06:09 | |
| kopparam | But in step 3, I cant seem to find the file neutron.conf | 06:09 |
| *** WackoRobie has joined #openstack-neutron | 06:09 | |
| *** sweston__ has joined #openstack-neutron | 06:10 | |
| *** ijw has quit IRC | 06:10 | |
| kopparam | Looks like this file is created when installing the openstack-neutron package, which is not mentioned to be installed on a compute node. | 06:10 |
| kopparam | Please help | 06:10 |
| *** achanda has quit IRC | 06:10 | |
| *** sweston_ has quit IRC | 06:10 | |
| *** sbfox has quit IRC | 06:11 | |
| *** afazekas_ has joined #openstack-neutron | 06:13 | |
| *** WackoRobie has quit IRC | 06:13 | |
| *** moha_hunt has joined #openstack-neutron | 06:14 | |
| konglingxian | kopparam, you mean step 4 | 06:15 |
| *** achanda has joined #openstack-neutron | 06:15 | |
| vimal_ | kopparam, that should be done on neutron server in order to use ovs plugin | 06:15 |
| *** annegent_ has joined #openstack-neutron | 06:16 | |
| *** sweston__ has quit IRC | 06:18 | |
| kopparam | Yes, step 4 | 06:19 |
| kopparam | Thanks vimal_ | 06:20 |
| *** annegent_ has quit IRC | 06:21 | |
| *** achanda has quit IRC | 06:22 | |
| *** a_le has quit IRC | 06:25 | |
| *** konglingxian has quit IRC | 06:25 | |
| *** seizadi has joined #openstack-neutron | 06:28 | |
| *** sweston_ has joined #openstack-neutron | 06:30 | |
| *** VenuMurthy has quit IRC | 06:30 | |
| *** VenuMurthy has joined #openstack-neutron | 06:30 | |
| *** sweston__ has joined #openstack-neutron | 06:31 | |
| *** essessv has quit IRC | 06:32 | |
| *** xuhanp has joined #openstack-neutron | 06:32 | |
| *** sbfox has joined #openstack-neutron | 06:34 | |
| *** a_le has joined #openstack-neutron | 06:34 | |
| *** sweston_ has quit IRC | 06:35 | |
| *** seizadi has quit IRC | 06:39 | |
| *** ijw has joined #openstack-neutron | 06:40 | |
| *** nati_ueno has joined #openstack-neutron | 06:42 | |
| *** catohornet has quit IRC | 06:42 | |
| *** ijw_ has joined #openstack-neutron | 06:43 | |
| *** kashyap has quit IRC | 06:45 | |
| *** ijw has quit IRC | 06:45 | |
| *** SridharG has quit IRC | 06:46 | |
| *** ijw_ has quit IRC | 06:48 | |
| *** kashyap has joined #openstack-neutron | 06:48 | |
| *** bloodypenguin_ has joined #openstack-neutron | 06:51 | |
| *** bloodypenguin_ has quit IRC | 06:51 | |
| *** kashyap` has joined #openstack-neutron | 06:52 | |
| *** kashyap has quit IRC | 06:56 | |
| *** gongysh has quit IRC | 06:57 | |
| *** dims__ has joined #openstack-neutron | 06:57 | |
| *** suresh12 has joined #openstack-neutron | 07:02 | |
| kopparam | Hi vimal_, according to this doc http://docs.openstack.org/havana/install-guide/install/yum/content/install-neutron.dedicated-compute-node.html. I need to edit the configs for neutron.conf on compute node? These instructions are repeated if they are meant to be on network node. | 07:02 |
| *** dims__ has quit IRC | 07:04 | |
| *** nlahouti has quit IRC | 07:04 | |
| *** kashyap` is now known as kashyap | 07:05 | |
| *** kickinz1|away is now known as kickinz1 | 07:05 | |
| VenuMurthy | Hey there, i see that the veth pair between the br-int and br-ex doesn’t exist on my packstack all-in-one install. am unable to assign floating ips to the VMs | 07:06 |
| *** suresh12 has quit IRC | 07:06 | |
| *** kopparam has quit IRC | 07:07 | |
| *** gongysh has joined #openstack-neutron | 07:07 | |
| *** kopparam has joined #openstack-neutron | 07:08 | |
| *** a_le has quit IRC | 07:08 | |
| *** _cjones_ has joined #openstack-neutron | 07:09 | |
| *** shausy has joined #openstack-neutron | 07:09 | |
| *** roeyc has joined #openstack-neutron | 07:10 | |
| *** WackoRobie has joined #openstack-neutron | 07:10 | |
| *** _cjones_ has quit IRC | 07:10 | |
| *** _cjones_ has joined #openstack-neutron | 07:11 | |
| *** kopparam has quit IRC | 07:12 | |
| *** gongysh has quit IRC | 07:13 | |
| *** rotbeard has joined #openstack-neutron | 07:13 | |
| *** WackoRobie has quit IRC | 07:14 | |
| *** annegent_ has joined #openstack-neutron | 07:16 | |
| *** gongysh has joined #openstack-neutron | 07:18 | |
| *** annegent_ has quit IRC | 07:23 | |
| *** SridharG has joined #openstack-neutron | 07:34 | |
| *** nlahouti_ has joined #openstack-neutron | 07:35 | |
| *** Longgeek has joined #openstack-neutron | 07:37 | |
| *** Longgeek has quit IRC | 07:37 | |
| *** Longgeek has joined #openstack-neutron | 07:37 | |
| *** ranger81 has quit IRC | 07:37 | |
| *** sbfox has quit IRC | 07:38 | |
| *** ajo|mac has joined #openstack-neutron | 07:39 | |
| *** gildub has quit IRC | 07:41 | |
| *** ijw has joined #openstack-neutron | 07:41 | |
| *** nlahouti_ has quit IRC | 07:43 | |
| *** Demitar has joined #openstack-neutron | 07:46 | |
| *** ijw has quit IRC | 07:47 | |
| *** kopparam has joined #openstack-neutron | 07:47 | |
| *** sweston_ has joined #openstack-neutron | 07:48 | |
| *** baojg has quit IRC | 07:48 | |
| *** baojg has joined #openstack-neutron | 07:49 | |
| *** baojg has quit IRC | 07:49 | |
| *** baojg has joined #openstack-neutron | 07:49 | |
| *** jpich has joined #openstack-neutron | 07:50 | |
| *** sweston__ has quit IRC | 07:51 | |
| *** kopparam has quit IRC | 07:52 | |
| *** yfauser has joined #openstack-neutron | 07:52 | |
| *** zzelle has joined #openstack-neutron | 07:52 | |
| *** yfauser has left #openstack-neutron | 07:53 | |
| *** VenuMurthy has quit IRC | 07:55 | |
| *** ajo|mac has quit IRC | 07:56 | |
| *** ajo|mac has joined #openstack-neutron | 07:57 | |
| *** matsuhashi has quit IRC | 07:57 | |
| *** matsuhashi has joined #openstack-neutron | 07:58 | |
| *** ajo|mac has quit IRC | 07:58 | |
| *** _cjones__ has joined #openstack-neutron | 08:00 | |
| *** evgenyf has joined #openstack-neutron | 08:00 | |
| *** doude has joined #openstack-neutron | 08:02 | |
| *** ihrachyshka has joined #openstack-neutron | 08:03 | |
| *** _cjones_ has quit IRC | 08:03 | |
| *** baojg has quit IRC | 08:04 | |
| *** kopparam has joined #openstack-neutron | 08:05 | |
| *** matsuhashi has quit IRC | 08:05 | |
| *** jlibosva has joined #openstack-neutron | 08:05 | |
| *** kashyap is now known as kashyap_lunch | 08:07 | |
| *** kopparam has quit IRC | 08:08 | |
| *** kopparam has joined #openstack-neutron | 08:09 | |
| *** nlahouti has joined #openstack-neutron | 08:10 | |
| *** matsuhashi has joined #openstack-neutron | 08:10 | |
| yfried | Hi, who's on LBaaS team | 08:10 |
| *** amaretskiy has joined #openstack-neutron | 08:13 | |
| *** nlahouti has quit IRC | 08:14 | |
| *** baojg has joined #openstack-neutron | 08:14 | |
| *** annegent_ has joined #openstack-neutron | 08:16 | |
| *** roeyc has quit IRC | 08:17 | |
| *** matsuhashi has quit IRC | 08:17 | |
| *** matsuhashi has joined #openstack-neutron | 08:18 | |
| *** amarao has joined #openstack-neutron | 08:18 | |
| *** Guest99196 is now known as atmark | 08:19 | |
| *** atmark has joined #openstack-neutron | 08:20 | |
| *** annegent_ has quit IRC | 08:21 | |
| *** nijaba has joined #openstack-neutron | 08:21 | |
| *** nijaba has quit IRC | 08:22 | |
| *** nijaba has joined #openstack-neutron | 08:22 | |
| *** Demitar has quit IRC | 08:23 | |
| *** _cjones__ has quit IRC | 08:23 | |
| *** _cjones_ has joined #openstack-neutron | 08:23 | |
| *** ramishra has quit IRC | 08:23 | |
| *** VenuMurthy has joined #openstack-neutron | 08:28 | |
| *** matsuhashi has quit IRC | 08:31 | |
| *** Ju_ has quit IRC | 08:34 | |
| *** maishsk has quit IRC | 08:35 | |
| *** yamamoto_ has quit IRC | 08:39 | |
| *** _cjones_ has quit IRC | 08:40 | |
| *** _cjones_ has joined #openstack-neutron | 08:41 | |
| *** nlahouti has joined #openstack-neutron | 08:41 | |
| *** ijw has joined #openstack-neutron | 08:41 | |
| *** _cjones_ has quit IRC | 08:45 | |
| *** igordcard has joined #openstack-neutron | 08:45 | |
| *** nlahouti has quit IRC | 08:46 | |
| *** eezhova has quit IRC | 08:46 | |
| *** ijw has quit IRC | 08:47 | |
| *** openstackgerrit has joined #openstack-neutron | 08:53 | |
| *** nijaba has quit IRC | 08:56 | |
| *** eezhova has joined #openstack-neutron | 08:57 | |
| openstackgerrit | Sumit Naiksatam proposed a change to openstack/neutron: Group Policy API-1: EP, EPG, L2 Policy, L3 Policy https://review.openstack.org/95900 | 09:00 |
| *** dims has joined #openstack-neutron | 09:02 | |
| *** jp_at_hp has joined #openstack-neutron | 09:06 | |
| *** ygbo has quit IRC | 09:07 | |
| *** jp_at_hp has quit IRC | 09:07 | |
| *** jp_at_hp has joined #openstack-neutron | 09:07 | |
| *** dims has quit IRC | 09:08 | |
| *** kashyap_lunch is now known as kashyap | 09:10 | |
| *** Xurong has quit IRC | 09:10 | |
| *** nlahouti has joined #openstack-neutron | 09:11 | |
| openstackgerrit | Cedric Brandily proposed a change to openstack/neutron-specs: DHCP agent customization https://review.openstack.org/99356 | 09:12 |
| *** WackoRobie has joined #openstack-neutron | 09:13 | |
| *** Xurong has joined #openstack-neutron | 09:13 | |
| *** nlahouti has quit IRC | 09:16 | |
| *** annegent_ has joined #openstack-neutron | 09:16 | |
| *** WackoRobie has quit IRC | 09:18 | |
| *** ygbo has joined #openstack-neutron | 09:20 | |
| *** annegent_ has quit IRC | 09:21 | |
| *** amotoki has quit IRC | 09:24 | |
| *** zhipeng has quit IRC | 09:25 | |
| *** zhipeng has joined #openstack-neutron | 09:29 | |
| *** iwamoto has quit IRC | 09:29 | |
| *** evgenyf has quit IRC | 09:30 | |
| *** nati_ueno has quit IRC | 09:31 | |
| *** safchain has joined #openstack-neutron | 09:32 | |
| *** yfauser has joined #openstack-neutron | 09:34 | |
| *** yfauser has left #openstack-neutron | 09:34 | |
| *** devvesa has joined #openstack-neutron | 09:36 | |
| *** matsuhashi has joined #openstack-neutron | 09:36 | |
| *** nijaba has joined #openstack-neutron | 09:37 | |
| *** nijaba has quit IRC | 09:37 | |
| *** nijaba has joined #openstack-neutron | 09:37 | |
| enikanorov | kevinbenton: hi | 09:40 |
| enikanorov | kevinbenton: why get rid of autodeletion? | 09:40 |
| *** ijw has joined #openstack-neutron | 09:41 | |
| *** nlahouti has joined #openstack-neutron | 09:42 | |
| *** zhipeng has quit IRC | 09:44 | |
| *** ihrachyshka has quit IRC | 09:44 | |
| *** evgenyf has joined #openstack-neutron | 09:45 | |
| *** zhipeng has joined #openstack-neutron | 09:46 | |
| *** Demitar has joined #openstack-neutron | 09:47 | |
| *** ijw has quit IRC | 09:47 | |
| *** nlahouti has quit IRC | 09:47 | |
| *** matsuhashi has quit IRC | 09:54 | |
| *** matsuhashi has joined #openstack-neutron | 09:55 | |
| *** Demitar has quit IRC | 09:57 | |
| *** matsuhas_ has joined #openstack-neutron | 09:57 | |
| *** matsuhashi has quit IRC | 09:57 | |
| kevinbenton | enikanorov: I briefly mentioned it in the commit, but it’s part of the process to improve the performance of the unit tests | 09:57 |
| kevinbenton | enikanorov: right now auto-deletion adds a ton of extra API calls | 09:58 |
| enikanorov | but that's defeat the purpose of using context managers to create objects | 09:58 |
| kevinbenton | enikanorov: yes, the context managers will eventually be removed completely | 09:59 |
| enikanorov | also, you still need to delete objects, will not it be the same amount of API calls? | 09:59 |
| kevinbenton | enikanorov: no, in most cases you don’t need to delete the objects. it will just be a DB reset | 09:59 |
| openstackgerrit | Evgeny Fedoruk proposed a change to openstack/neutron-specs: LBaaS TLS termination capability specification https://review.openstack.org/98640 | 09:59 |
| enikanorov | i'm not sure agree with this approach as it relies on external logic (external relative to the test) | 10:00 |
| kevinbenton | enikanorov: what external logic? | 10:00 |
| enikanorov | that DB is reset | 10:00 |
| enikanorov | i don't like the idea that test will leave objects behind | 10:01 |
| kevinbenton | enikanorov: many already do it | 10:01 |
| kevinbenton | enikanorov: the db is reset between tests | 10:01 |
| enikanorov | yes, i know, but that's just because some tests are not clean enough to do their stuff gracefully | 10:02 |
| zzelle | enikanorov: hi | 10:02 |
| enikanorov | what I would really like to see is separation between plugin tests and API tests | 10:02 |
| enikanorov | that is what really could speed things up | 10:02 |
| kevinbenton | enikanorov: the point is that a unit test shouldn’t have to clean up objects unless that’s specifically what it’s testing | 10:03 |
| enikanorov | zzelle: hi | 10:04 |
| zzelle | enikanorov: i did some perf tests on https://review.openstack.org/101982 ... it reduces vxlan sync allocations time by about 40 to 50 % on sqlite/myysql and postgresql (when only reducing ranges) | 10:04 |
| enikanorov | zzelle: cool | 10:05 |
| kevinbenton | enikanorov: yes, separation would be another improvement as well but it will require a significant refactor of the existing tests | 10:05 |
| zzelle | enikanorov: this afternoon i will share the results on the ML with the bench module | 10:06 |
| kevinbenton | enikanorov: take this test for example. https://github.com/bigswitch/neutron/blob/master/neutron/tests/unit/nec/test_nec_plugin.py#L893 | 10:06 |
| kevinbenton | enikanorov: it doesn’t delete the port | 10:06 |
| enikanorov | right | 10:06 |
| kevinbenton | enikanorov: and there really is no reason it should need to since everything is recreated for each test | 10:06 |
| enikanorov | but what's the problem with that test? | 10:07 |
| kevinbenton | enikanorov: there isn’t one, but you are suggesting that it is not clean enough to do it gracefully | 10:07 |
| enikanorov | context manager can be used with or without autodeletion | 10:07 |
| enikanorov | sometimes autodeletion catch issues | 10:08 |
| *** yamahata has quit IRC | 10:08 | |
| kevinbenton | enikanorov: yes, definitely | 10:08 |
| enikanorov | kevinbenton: without context manager you'll need to do extrawork to do things gracefully | 10:08 |
| enikanorov | as you do in the patch | 10:08 |
| openstackgerrit | Koteswara Rao Kelam proposed a change to openstack/neutron: Add -s option for neutron metering rules https://review.openstack.org/95526 | 10:09 |
| kevinbenton | enikanorov: yes, and we will need some extra tests to guaruntee certain objects are blocked from deletion, etc | 10:10 |
| enikanorov | ok, personally i don't like this approach | 10:11 |
| enikanorov | (as i would not like any extra coding) | 10:11 |
| kevinbenton | enikanorov: you shouldn’t encounter it very often | 10:12 |
| kevinbenton | enikanorov: I believe marun will be proposing a blueprint for this work when he returns | 10:13 |
| *** nlahouti has joined #openstack-neutron | 10:13 | |
| kevinbenton | enikanorov: so you can provide your feedback there | 10:13 |
| *** WackoRobie has joined #openstack-neutron | 10:14 | |
| *** gildub has joined #openstack-neutron | 10:14 | |
| *** busterswt has quit IRC | 10:15 | |
| *** annegent_ has joined #openstack-neutron | 10:16 | |
| *** ihrachyshka has joined #openstack-neutron | 10:17 | |
| *** nlahouti has quit IRC | 10:17 | |
| *** igordcard has quit IRC | 10:18 | |
| *** WackoRobie has quit IRC | 10:19 | |
| *** jpich has quit IRC | 10:20 | |
| *** jpich has joined #openstack-neutron | 10:20 | |
| openstackgerrit | Koteswara Rao Kelam proposed a change to openstack/neutron: Proper validation for inserting firewall rule https://review.openstack.org/100560 | 10:21 |
| *** annegent_ has quit IRC | 10:21 | |
| *** ihrachyshka has quit IRC | 10:21 | |
| *** ihrachyshka has joined #openstack-neutron | 10:22 | |
| *** yfauser has joined #openstack-neutron | 10:22 | |
| *** yfauser has left #openstack-neutron | 10:22 | |
| *** amarao has quit IRC | 10:22 | |
| *** rm_work is now known as rm_work|away | 10:23 | |
| *** matsuhas_ has quit IRC | 10:28 | |
| *** matsuhashi has joined #openstack-neutron | 10:28 | |
| *** amitpp has joined #openstack-neutron | 10:28 | |
| openstackgerrit | Sumit Naiksatam proposed a change to openstack/neutron: Group Policy DB-1: EP, EPG, L2 Policy, L3 Policy https://review.openstack.org/96050 | 10:30 |
| openstackgerrit | Koteswara Rao Kelam proposed a change to openstack/neutron: Add -s option for neutron metering rules https://review.openstack.org/95526 | 10:32 |
| *** moha_hunt has quit IRC | 10:36 | |
| *** matsuhashi has quit IRC | 10:37 | |
| *** matsuhashi has joined #openstack-neutron | 10:37 | |
| *** xuhanp has quit IRC | 10:38 | |
| *** ramishra has joined #openstack-neutron | 10:38 | |
| *** ijw has joined #openstack-neutron | 10:41 | |
| *** nlahouti has joined #openstack-neutron | 10:44 | |
| *** ihrachyshka has quit IRC | 10:47 | |
| *** ijw has quit IRC | 10:47 | |
| *** ihrachyshka has joined #openstack-neutron | 10:47 | |
| *** nlahouti has quit IRC | 10:48 | |
| openstackgerrit | Sumit Naiksatam proposed a change to openstack/neutron: Group Policy API-1: EP, EPG, L2 Policy, L3 Policy https://review.openstack.org/95900 | 10:50 |
| *** roeyc has joined #openstack-neutron | 10:51 | |
| openstackgerrit | Sumit Naiksatam proposed a change to openstack/neutron: Group Policy DB-1: EP, EPG, L2 Policy, L3 Policy https://review.openstack.org/96050 | 10:53 |
| *** dims has joined #openstack-neutron | 11:03 | |
| openstackgerrit | Kevin Benton proposed a change to openstack/neutron: UTs: Disable auto deletion of ports/subnets/nets https://review.openstack.org/102465 | 11:06 |
| rossella_s | ihrachyshka: ping | 11:06 |
| ihrachyshka | rossella_s: pong | 11:07 |
| openstackgerrit | Sumit Naiksatam proposed a change to openstack/neutron: Group Policy Plugin-1: EP, EPG, L2 Policy, L3 Policy https://review.openstack.org/96393 | 11:08 |
| rossella_s | ihrachyshka just a quick question, I guess you are the right person to ask since you did the porting to oslo.messaging. I am working on my patch https://review.openstack.org/#/c/66899/ ...I rebased and unit tests are failing | 11:08 |
| ihrachyshka | rossella_s: looking | 11:08 |
| *** dims has quit IRC | 11:08 | |
| rossella_s | ihrachyshka: I see during the port you removed for the rpc unit test the line where the version is checked | 11:09 |
| rossella_s | ihrachyshka: just to be clear for example here https://review.openstack.org/#/c/97291/20/neutron/tests/unit/openvswitch/test_ovs_rpcapi.py line 37 | 11:09 |
| ihrachyshka | rossella_s: yes, because it's not passed with message now, it belongs to client used | 11:09 |
| rossella_s | ihrachyshka: so here comes my question...I am adding a new method, should I pass the version or not? | 11:10 |
| *** baojg has quit IRC | 11:10 | |
| ihrachyshka | rossella_s: specific place in the code? | 11:10 |
| *** baojg has joined #openstack-neutron | 11:11 | |
| *** baojg has quit IRC | 11:11 | |
| rossella_s | ihrachyshka: see https://review.openstack.org/#/c/66899/36/neutron/agent/rpc.py | 11:11 |
| *** coolsvap is now known as coolsvap|afk | 11:11 | |
| *** baojg has joined #openstack-neutron | 11:12 | |
| rossella_s | ihrachyshka: the version is there because this method is not supported by the base rpc api version | 11:12 |
| ihrachyshka | rossella_s: if you needed to pass it in previous version of code, you need to do it now also | 11:14 |
| ihrachyshka | version will be set in __call_rpc_method from rpc.py | 11:14 |
| ihrachyshka | as: callee = self._client.prepare(**options) | 11:14 |
| ihrachyshka | options will contain your version, among other things | 11:14 |
| rossella_s | ihrachyshka: so only if the version is the base rpc version that param is not passed anymore? | 11:14 |
| *** WackoRobie has joined #openstack-neutron | 11:14 | |
| ihrachyshka | the test change you referred is just because msg object doesn't contain it. but it's still set, thru .prepare() | 11:15 |
| *** yamahata has joined #openstack-neutron | 11:15 | |
| *** nlahouti has joined #openstack-neutron | 11:15 | |
| ihrachyshka | rossella_s: honestly, I don't know what are the rules of setting versions when introducing new methods. this should probably be documented in oslo.messaging though. let me check | 11:15 |
| *** baojg has quit IRC | 11:16 | |
| rossella_s | ihrachyshka: thanks! | 11:16 |
| *** annegent_ has joined #openstack-neutron | 11:16 | |
| openstackgerrit | Sumit Naiksatam proposed a change to openstack/neutron: Group Policy API-2: Classifiers, Actions, and Rules https://review.openstack.org/101816 | 11:16 |
| *** nlahouti has quit IRC | 11:19 | |
| *** WackoRobie has quit IRC | 11:19 | |
| ihrachyshka | shouldn't you just bump client version (aka BASE_..._VERSION) to 1.2? in that way, client will be able to send both methods without setting version explicitly | 11:19 |
| ihrachyshka | and since 1.1 is compatible with 1.2, it probably will work just fine | 11:19 |
| *** dims_ has joined #openstack-neutron | 11:19 | |
| ihrachyshka | rossella_s: ^^ | 11:20 |
| *** annegent_ has quit IRC | 11:21 | |
| *** dims__ has joined #openstack-neutron | 11:22 | |
| *** WackoRobie has joined #openstack-neutron | 11:22 | |
| ihrachyshka | hm, I guess if you don't pass version_cap to the client (and we don't), no checks will be applied to incoming message, it will be sent with whatever version is passed | 11:23 |
| *** moha_hunt has joined #openstack-neutron | 11:23 | |
| *** yfauser has joined #openstack-neutron | 11:24 | |
| *** yfauser has left #openstack-neutron | 11:24 | |
| rossella_s | ihrachyshka: so you recommend to bump the client version? | 11:25 |
| *** dims_ has quit IRC | 11:25 | |
| rossella_s | ihrachyshka: doing that...the client will require a server that support that new version | 11:27 |
| *** gongysh has quit IRC | 11:28 | |
| *** banix has joined #openstack-neutron | 11:29 | |
| *** kopparam has quit IRC | 11:30 | |
| ihrachyshka | rossella_s: probably you're right. so we bump client version only after bumping callbacks, right? | 11:30 |
| *** kopparam has joined #openstack-neutron | 11:30 | |
| rossella_s | ihrachyshka: what do you mean by bumping callbacks? | 11:30 |
| ihrachyshka | rossella_s: their versions | 11:31 |
| ihrachyshka | set via RPC_API_VERSION in corresponding RpcCallback-inherited class | 11:31 |
| rossella_s | ihrachyshka: yes | 11:31 |
| rossella_s | ihrachyshka: anyway I guess there's no other solution if we don't use the version cap by oslo messaging | 11:32 |
| ihrachyshka | rossella_s: ok, so you want agent to continue work with servers of older versions, so you don't bump the version. now it's clear you're correct and I'm wrong :) | 11:32 |
| rossella_s | ihrachyshka: yes that was my goal...that's why the version field was handy | 11:32 |
| rossella_s | ihrachyshka: I don't know anything about oslo.messaging...can you see a way to support that? you are the expert | 11:33 |
| *** banix has quit IRC | 11:33 | |
| *** yfried_ has joined #openstack-neutron | 11:35 | |
| *** kopparam has quit IRC | 11:35 | |
| *** yfried has quit IRC | 11:35 | |
| *** SridharG has quit IRC | 11:36 | |
| ihrachyshka | rossella_s: haha, no, I'm not, I just made migration effort :D | 11:36 |
| ihrachyshka | rossella_s: I guess you code is correct, at least in that file. I'm looking thru the patch atm. | 11:36 |
| rossella_s | ihrachyshka: well I will send another one shortly...I need to fix the tests | 11:37 |
| rossella_s | ihrachyshka: for now I will leave it the way it is and fix the ut accordingly | 11:37 |
| ihrachyshka | wait a bit, I have some other minor nits to point out | 11:38 |
| *** ramishra has quit IRC | 11:38 | |
| *** ramishra has joined #openstack-neutron | 11:39 | |
| rossella_s | ok | 11:40 |
| *** ijw has joined #openstack-neutron | 11:41 | |
| *** ramishra has quit IRC | 11:43 | |
| *** erecio has joined #openstack-neutron | 11:44 | |
| *** diegows has joined #openstack-neutron | 11:46 | |
| *** nlahouti has joined #openstack-neutron | 11:46 | |
| *** ijw has quit IRC | 11:47 | |
| *** baoli has quit IRC | 11:48 | |
| *** devvesa has quit IRC | 11:50 | |
| *** nlahouti has quit IRC | 11:51 | |
| *** Xurong has quit IRC | 11:51 | |
| *** Xurong has joined #openstack-neutron | 11:52 | |
| *** mwagner_lap has quit IRC | 11:53 | |
| *** ramishra has joined #openstack-neutron | 11:54 | |
| ihrachyshka | rossella_s: posted comments | 11:54 |
| rossella_s | ihrachyshka: thanks! | 11:54 |
| *** VenuMurthy has quit IRC | 11:55 | |
| *** amarao has joined #openstack-neutron | 11:55 | |
| *** VenuMurthy has joined #openstack-neutron | 12:02 | |
| *** moha_hunt has quit IRC | 12:05 | |
| *** rkukura has quit IRC | 12:05 | |
| rossella_s | ihrachyshka: thanks for your comment. What do you mean by "If you want to check version here, you need to catch arguments to .prepare()" ? | 12:06 |
| ihrachyshka | rossella_s: I mean that message returned by self.make_msg(), doesn't have 'version' set | 12:07 |
| ihrachyshka | but I have another idea - you may check msg['args']['version'], this will probably work | 12:08 |
| rossella_s | ihrachyshka: it's not clear to me | 12:09 |
| ihrachyshka | rossella_s: have you checked make_msg from neutron/common/rpc.py? that's what you call thru self.make_msg | 12:10 |
| *** gildub has quit IRC | 12:10 | |
| rossella_s | ihrachyshka: yes, version is not set there | 12:10 |
| ihrachyshka | ah, wait checking msg['args']['version'] is a bad idea, I mixed expected and actual :) | 12:10 |
| ihrachyshka | so yes, if you want to check version, you need to catch self.client.prepare() and check that it's called with version=expected | 12:11 |
| *** erecio has quit IRC | 12:11 | |
| *** erecio has joined #openstack-neutron | 12:12 | |
| *** moha_hunt has joined #openstack-neutron | 12:12 | |
| *** erecio has quit IRC | 12:13 | |
| ihrachyshka | rossella_s: is it clear now? | 12:14 |
| *** erecio has joined #openstack-neutron | 12:14 | |
| rossella_s | ihrachyshka: thanks for your help I think I can go on from here | 12:15 |
| ihrachyshka | rossella_s: good, feel free to ping in case of questions. I'm not an expert in that thing, but open for mutual learning :) | 12:15 |
| *** alexpilotti has joined #openstack-neutron | 12:15 | |
| rossella_s | ihrachyshka: I will, thanks again! | 12:16 |
| *** annegent_ has joined #openstack-neutron | 12:16 | |
| *** nlahouti has joined #openstack-neutron | 12:17 | |
| *** trad511 has joined #openstack-neutron | 12:17 | |
| *** kopparam has joined #openstack-neutron | 12:19 | |
| *** amitpp has quit IRC | 12:19 | |
| *** amitpp has joined #openstack-neutron | 12:19 | |
| *** baoli has joined #openstack-neutron | 12:19 | |
| *** baoli has quit IRC | 12:19 | |
| *** baoli has joined #openstack-neutron | 12:20 | |
| *** erecio has quit IRC | 12:20 | |
| *** annegent_ has quit IRC | 12:21 | |
| *** erecio has joined #openstack-neutron | 12:21 | |
| *** nlahouti has quit IRC | 12:21 | |
| *** amitpp has quit IRC | 12:22 | |
| *** ajc_ has quit IRC | 12:23 | |
| *** shausy has quit IRC | 12:24 | |
| *** shausy has joined #openstack-neutron | 12:24 | |
| *** shausy has quit IRC | 12:25 | |
| *** shausy has joined #openstack-neutron | 12:25 | |
| *** moha_hunt has quit IRC | 12:28 | |
| *** yfried_ has quit IRC | 12:29 | |
| *** shausy has quit IRC | 12:31 | |
| *** dims__ has quit IRC | 12:31 | |
| *** shausy has joined #openstack-neutron | 12:31 | |
| *** rotbeard has quit IRC | 12:31 | |
| *** dims__ has joined #openstack-neutron | 12:31 | |
| *** shausy has left #openstack-neutron | 12:32 | |
| openstackgerrit | Ravindra Joshi proposed a change to openstack/neutron: description for "default" security group changed https://review.openstack.org/102509 | 12:32 |
| *** ramishra has quit IRC | 12:33 | |
| *** radez_g0n3 is now known as radez | 12:36 | |
| *** kopparam has quit IRC | 12:38 | |
| *** ramishra has joined #openstack-neutron | 12:40 | |
| *** ijw has joined #openstack-neutron | 12:41 | |
| *** yfried has joined #openstack-neutron | 12:43 | |
| *** sweston_ has quit IRC | 12:44 | |
| openstackgerrit | Murali Birru proposed a change to openstack/neutron: Don't schedule a router with no interfaces attached to it. https://review.openstack.org/102512 | 12:44 |
| *** annegent_ has joined #openstack-neutron | 12:45 | |
| *** baoli has quit IRC | 12:46 | |
| *** ijw has quit IRC | 12:47 | |
| *** nlahouti has joined #openstack-neutron | 12:48 | |
| *** SridharG has joined #openstack-neutron | 12:50 | |
| HenryG | sc68cal: ping | 12:51 |
| *** nlahouti has quit IRC | 12:52 | |
| *** jecarey has joined #openstack-neutron | 12:54 | |
| *** annegent_ has quit IRC | 12:54 | |
| openstackgerrit | Xu Chen proposed a change to openstack/neutron: call security_groups_member_updated in port_update https://review.openstack.org/101293 | 12:55 |
| *** pradipta is now known as pradipta_away | 12:55 | |
| *** prad_ has joined #openstack-neutron | 12:56 | |
| *** amitpp has joined #openstack-neutron | 12:58 | |
| *** matsuhashi has quit IRC | 12:59 | |
| *** matsuhashi has joined #openstack-neutron | 12:59 | |
| *** enikanorov_ has quit IRC | 13:00 | |
| *** garyk has quit IRC | 13:00 | |
| *** enikanorov_ has joined #openstack-neutron | 13:00 | |
| *** vimal_ has quit IRC | 13:02 | |
| *** julim has joined #openstack-neutron | 13:03 | |
| *** yfried has quit IRC | 13:04 | |
| *** matsuhashi has quit IRC | 13:04 | |
| *** ramishra has quit IRC | 13:05 | |
| *** ramishra has joined #openstack-neutron | 13:05 | |
| *** safchain has quit IRC | 13:06 | |
| *** igordcard has joined #openstack-neutron | 13:06 | |
| *** safchain has joined #openstack-neutron | 13:06 | |
| *** shakamunyi has joined #openstack-neutron | 13:08 | |
| *** matsuhashi has joined #openstack-neutron | 13:10 | |
| openstackgerrit | Jakub Libosvar proposed a change to openstack/neutron: Exit rpc_loop when SIGTERM is recieved in ovs-agent https://review.openstack.org/102519 | 13:10 |
| *** trad511 has quit IRC | 13:13 | |
| *** yamamoto has joined #openstack-neutron | 13:16 | |
| *** mwagner_lap has joined #openstack-neutron | 13:16 | |
| *** salv-orlando_ has joined #openstack-neutron | 13:18 | |
| *** nlahouti has joined #openstack-neutron | 13:18 | |
| *** salv-orlando has quit IRC | 13:20 | |
| *** salv-orlando_ is now known as salv-orlando | 13:20 | |
| *** nlahouti has quit IRC | 13:23 | |
| *** nplanel_ has joined #openstack-neutron | 13:23 | |
| openstackgerrit | Jakub Libosvar proposed a change to openstack/neutron: Exit rpc_loop when SIGTERM is recieved in ovs-agent https://review.openstack.org/102519 | 13:24 |
| *** busterswt has joined #openstack-neutron | 13:25 | |
| *** jecarey has quit IRC | 13:25 | |
| *** jecarey has joined #openstack-neutron | 13:25 | |
| *** annegent_ has joined #openstack-neutron | 13:26 | |
| openstackgerrit | vinay yadhav proposed a change to openstack/neutron-specs: Introducing Tap-as-a-Service https://review.openstack.org/96149 | 13:28 |
| HenryG | mestery: ping re: devstack patch https://review.openstack.org/102284 | 13:29 |
| *** amitpp has quit IRC | 13:30 | |
| mestery | HenryG: Whatsu p? | 13:31 |
| *** baojg has joined #openstack-neutron | 13:31 | |
| HenryG | mestery: Looks like when no *-dkms package is needed, then we don't need the | 13:32 |
| HenryG | dkms package itself, nor do we need the linux headers. If it's OK with | 13:32 |
| HenryG | you I will take over this patch and do the proper cleanup. | 13:32 |
| HenryG | mestery: I should have done the job properly in my first patch. | 13:33 |
| *** matsuhashi has quit IRC | 13:33 | |
| *** matsuhashi has joined #openstack-neutron | 13:33 | |
| mestery | HenryG: I am fine either way, I could also push a new one, I see what you mean there. :) | 13:33 |
| *** ramishra has quit IRC | 13:34 | |
| HenryG | mestery: Also, the cloned code needs to go in a common function. | 13:34 |
| mestery | HenryG: Likely the case, yes. :) | 13:35 |
| *** markmcclain has joined #openstack-neutron | 13:36 | |
| *** radez is now known as radez_g0n3 | 13:36 | |
| HenryG | mestery: I am hacking it now. I can push a change for you to look at soon. | 13:37 |
| mestery | HenryG: Thanks! I'll review once it's there. It would be good to test it with ODL as well, I can PM you the details for localrc if you want. | 13:37 |
| *** markmcclain1 has joined #openstack-neutron | 13:37 | |
| *** matsuhashi has quit IRC | 13:38 | |
| HenryG | mestery: sounds good | 13:38 |
| *** markmcclain1 has quit IRC | 13:38 | |
| *** markmcclain has quit IRC | 13:40 | |
| *** ijw has joined #openstack-neutron | 13:41 | |
| *** coolsvap|afk is now known as coolsvap | 13:43 | |
| *** matsuhashi has joined #openstack-neutron | 13:44 | |
| *** matsuhashi has quit IRC | 13:44 | |
| *** garyk has joined #openstack-neutron | 13:45 | |
| *** matsuhashi has joined #openstack-neutron | 13:45 | |
| *** ijw has quit IRC | 13:47 | |
| *** nlahouti has joined #openstack-neutron | 13:49 | |
| *** matsuhashi has quit IRC | 13:49 | |
| openstackgerrit | Chengli Xu proposed a change to openstack/neutron: ovs agent: re-sync tunnels if openvswitch restarts https://review.openstack.org/102532 | 13:50 |
| *** esco-franc has joined #openstack-neutron | 13:50 | |
| esco-franc | hi | 13:51 |
| *** VenuMurthy has quit IRC | 13:51 | |
| *** jmsoares has joined #openstack-neutron | 13:53 | |
| *** yfried has joined #openstack-neutron | 13:53 | |
| *** kopparam has joined #openstack-neutron | 13:53 | |
| *** nlahouti has quit IRC | 13:53 | |
| *** yfried has quit IRC | 13:54 | |
| *** yfried has joined #openstack-neutron | 13:54 | |
| ihrachyshka | mestery: hey. can we briefly discuss https://blueprints.launchpad.net/neutron/+spec/schedule-actions-for-db-transaction-closed ? | 13:54 |
| openstackgerrit | Cedric Brandily proposed a change to openstack/neutron: OVS flows apply concurrently using a deferred OVSBridge https://review.openstack.org/77578 | 13:55 |
| *** nati_ueno has joined #openstack-neutron | 13:55 | |
| *** VenuMurthy has joined #openstack-neutron | 13:55 | |
| zzelle | feleouet: hi | 13:56 |
| *** s3wong has joined #openstack-neutron | 13:56 | |
| enikanorov_ | ihrachyshka: do you think much of lock waits are caused by notifications? | 13:57 |
| enikanorov_ | i doubt this | 13:57 |
| feleouet | zzelle: hi | 13:57 |
| *** kopparam has quit IRC | 13:57 | |
| *** vimal7370 has joined #openstack-neutron | 13:57 | |
| *** kopparam has joined #openstack-neutron | 13:57 | |
| *** busterswt has quit IRC | 13:58 | |
| ihrachyshka | enikanorov_: well, I've only started to read ml2 driver, fixed one of those locks, and discovered another one by chance. For what I see in the code, there is no discipline in nested calls of other methods, so these are all potential locks. | 13:58 |
| zzelle | feleouet: i updated the code on flow apply defer: https://review.openstack.org/77578 | 13:59 |
| zzelle | feleouet: could you have a look ? i am not really fluent with l2 pop and have difficulties to check the good behavior | 13:59 |
| *** armax has joined #openstack-neutron | 14:00 | |
| *** busterswt has joined #openstack-neutron | 14:00 | |
| *** devvesa has joined #openstack-neutron | 14:00 | |
| enikanorov_ | ihrachyshka: i agree that there are convoluted cases of notifications under transaction | 14:01 |
| enikanorov_ | ihrachyshka: i think we need to sharp-shoot them one by one | 14:01 |
| feleouet | zzelle: ok, I'll have a look, and give a try | 14:02 |
| ihrachyshka | enikanorov_: how? have you checked the patch I've sent for one of them, sharp-shooting one? this made API more vague and exposed internals to caller | 14:02 |
| *** hoangdo has joined #openstack-neutron | 14:02 | |
| ihrachyshka | enikanorov_: I guess if we go along that line, we'll end up with maccaroni code | 14:02 |
| enikanorov_ | ihrachyshka: what patch do you mean? | 14:03 |
| ihrachyshka | enikanorov_: https://review.openstack.org/100934 | 14:03 |
| hoangdo | hello guys, do you know how to use "neutron router-gateway-set" with my defined port, without creating a new one. | 14:04 |
| enikanorov_ | ihrachyshka: yep, that is sharp-shooting and i'm fine with that | 14:05 |
| enikanorov_ | we really should have a discipline in using transactions rather than create complex helpers to avoid shooting ourselves in the foot | 14:06 |
| ihrachyshka | enikanorov_: I disagree. shard-shooting like that means passing objects of vague intent thru the stack back to some root caller, polluting arguments with do_notify like arguments etc. We have at least 3 levels of nesting in e.g. delete_floatingip() calling to delete_port() calling to disassoaciate_floatingips(). Do you want us to pass arguments thru the large stack instead of isolating notification request to the code that | 14:09 |
| ihrachyshka | really results in it? | 14:09 |
| enikanorov_ | ihrachyshka: that's the question of current logic. it's the problem that it is convoluted and leads to such need. avoiding it with a hack is not a good way to go imo | 14:10 |
| ihrachyshka | do you call scheduled actions mechanism a hack? | 14:11 |
| mestery | ihrachyshka: Yes, give me a few minutes, on with someone else now. | 14:11 |
| * mestery lets ihrachyshka and enikanorov_ continue on and will read the backscroll. | 14:11 | |
| ihrachyshka | mestery: roger | 14:11 |
| *** otherwiseguy has joined #openstack-neutron | 14:12 | |
| ihrachyshka | I would say, returning routers' set back to caller and requiring it handle notifications on our behalf is a hack | 14:12 |
| *** nati_ueno_i has joined #openstack-neutron | 14:13 | |
| *** rolledback has joined #openstack-neutron | 14:13 | |
| *** nati_ueno_i has quit IRC | 14:13 | |
| enikanorov_ | ihrachyshka: that's a question of how we reuse code and why we do so under transaction | 14:13 |
| *** nati_ueno_i has joined #openstack-neutron | 14:14 | |
| enikanorov_ | may be something needs to be split into more methods | 14:14 |
| ihrachyshka | enikanorov_: can you give an example of how you would handle that in case of the patch I've posted? | 14:15 |
| *** ramishra has joined #openstack-neutron | 14:15 | |
| *** nati_ueno has quit IRC | 14:16 | |
| zzelle | hoangdo: it is not possible to perform neutron router_gateway-set with a port | 14:16 |
| enikanorov_ | ihrachyshka: yes | 14:16 |
| enikanorov_ | ihrachyshka: simple refactoring: disassociate_floatingips has db part and notification part | 14:16 |
| *** rolledback has quit IRC | 14:17 | |
| salv-orlando | hello esco-franc | 14:17 |
| enikanorov_ | ihrachyshka: extract db part, make that method to run db stuff without transaction (with a comment that it should be called under transaction), then make another method that will do db stuff (extracted method) and notify | 14:17 |
| ihrachyshka | enikanorov_: you mean, removing do_notify argument and always require a caller to issue notifications? notifications depend on result of db part though, so we still need to pass the set thru the stach | 14:17 |
| ihrachyshka | *stack | 14:17 |
| hoangdo | zzelle: thanks zzelle, is there no way to hack it. I need to use my own MAC somehow. | 14:18 |
| *** nati_ueno_i is now known as nati_ueno | 14:18 | |
| *** rolledback has joined #openstack-neutron | 14:18 | |
| ihrachyshka | enikanorov_: actually, that was the initial version of the patch. See: https://review.openstack.org/#/c/100934/2/neutron/db/l3_db.py | 14:19 |
| enikanorov_ | do_notify is a parameter that defines how method is used, which means that method could actually be split on versions for each use case | 14:19 |
| ihrachyshka | enikanorov_: this solves the problem of unstable state of the function (no argument controlling its internals), but still exposure of internals thru return result | 14:19 |
| *** yano has quit IRC | 14:20 | |
| enikanorov_ | not sure what internals are you talking about | 14:20 |
| *** nlahouti has joined #openstack-neutron | 14:20 | |
| ihrachyshka | enikanorov_: set of routers returned, passed to notify part later | 14:20 |
| *** xuhanp has joined #openstack-neutron | 14:21 | |
| *** annegent_ has quit IRC | 14:21 | |
| *** annegent_ has joined #openstack-neutron | 14:21 | |
| ihrachyshka | enikanorov_: caller is not interested in those internals, it just wants to disassociate floating ip. but instead, it's required to issue some functions that are not related to its primary goal | 14:22 |
| openstackgerrit | A change was merged to openstack/neutron-specs: Dell ml2 Mechanism Driver https://review.openstack.org/92495 | 14:22 |
| *** roeyc has quit IRC | 14:22 | |
| *** carl_baldwin has joined #openstack-neutron | 14:22 | |
| *** yano has joined #openstack-neutron | 14:22 | |
| enikanorov_ | ihrachyshka: that's fine to pass those 'internals' as notify_routers_updated should be protected method | 14:23 |
| *** nlahouti has quit IRC | 14:24 | |
| *** nplanel_ has quit IRC | 14:26 | |
| *** mlavalle has joined #openstack-neutron | 14:28 | |
| ihrachyshka | enikanorov_: what do you mean protected? not avail for RPC? | 14:29 |
| *** Demitar has joined #openstack-neutron | 14:30 | |
| ihrachyshka | enikanorov_: I don't think that's fine. As I said above, we have at least three-folded nesting, with different types of notifications issued by relevant code (security groups, routers...). We would need to pass multiple sets back in stack to fix all those. And this will become messier with each change we introduce | 14:31 |
| openstackgerrit | Chengli Xu proposed a change to openstack/neutron: l2-pop notifies agent of active ports only https://review.openstack.org/102543 | 14:31 |
| *** VenuMurthy has quit IRC | 14:32 | |
| enikanorov_ | ihrachyshka: protected in OOP-sense | 14:32 |
| ihrachyshka | again, we don't always know whether we've the method that opened transaction, so we'll need to introduce do_notify= like arguments, or check context.session.is_active to decide whether that's a good time to issue notifications. | 14:32 |
| *** nlahouti has joined #openstack-neutron | 14:32 | |
| enikanorov_ | (in python that would just mean it's name would be _notify_routers_updated which is merely a style) | 14:32 |
| ihrachyshka | enikanorov_: roger | 14:33 |
| ihrachyshka | s/we've/we're | 14:33 |
| *** vivek-ebay has joined #openstack-neutron | 14:33 | |
| enikanorov_ | ihrachyshka: well, my opinion that three-folded nesting is just a problem of code composition | 14:33 |
| enikanorov_ | and decomposition requires passing data around - that is totally fine | 14:34 |
| ihrachyshka | enikanorov_: if we can hide those details, that's even better | 14:34 |
| enikanorov_ | hide from whom? | 14:35 |
| ihrachyshka | enikanorov_: I agree that this can be a problem of composition, though I suspect such rework would be huge and wouldn't land at least in Juno. we are also looking into getting those issues fixed in Icehouse. | 14:36 |
| ihrachyshka | enikanorov_: from caller | 14:36 |
| openstackgerrit | A change was merged to openstack/neutron-specs: DB migration refactoring https://review.openstack.org/95738 | 14:36 |
| *** nlahouti has quit IRC | 14:37 | |
| enikanorov_ | ihrachyshka: well i suspect that notifications within a transaction is smaller portion of the problem | 14:38 |
| *** radez_g0n3 is now known as radez | 14:39 | |
| enikanorov_ | we have lots of lock wait timeout bugs not related to notifications | 14:39 |
| ihrachyshka | enikanorov_: can you describe some of the scenarios? | 14:39 |
| *** kraterekometa has joined #openstack-neutron | 14:39 | |
| kraterekometa | hi | 14:40 |
| enikanorov_ | ihrachyshka: can't do this right now. I just can say that i've briefly checked a few cases of timeouts where no notifications were issued | 14:40 |
| *** spandhe has joined #openstack-neutron | 14:41 | |
| ihrachyshka | enikanorov_: ok, in the meantime, I'll look thru the list of timeout bugs you've posted on meeting wiki page | 14:41 |
| *** WackoRobie has quit IRC | 14:41 | |
| *** ijw has joined #openstack-neutron | 14:41 | |
| *** WackoRobie has joined #openstack-neutron | 14:41 | |
| enikanorov_ | ihrachyshka: i suggest you search bugs for 'Lock wait' string as there are more than what's on meeting wiki page | 14:41 |
| *** esco-franc has quit IRC | 14:42 | |
| ihrachyshka | enikanorov_: well, that's a good start at least. thanks for suggestion. | 14:42 |
| *** kraterekometa is now known as esco-franc | 14:42 | |
| *** morganfainberg_Z is now known as morganfainberg | 14:42 | |
| enikanorov_ | should be > 10 bugs with that | 14:42 |
| *** vivek-ebay has quit IRC | 14:42 | |
| *** spandhe_ has joined #openstack-neutron | 14:43 | |
| *** zhiyan is now known as zhiyan_ | 14:43 | |
| *** kopparam_ has joined #openstack-neutron | 14:45 | |
| *** spandhe has quit IRC | 14:45 | |
| *** spandhe_ is now known as spandhe | 14:45 | |
| *** WackoRobie has quit IRC | 14:46 | |
| openstackgerrit | Avishay Balderman proposed a change to openstack/neutron-specs: LBaaS L7 Switching - design document https://review.openstack.org/99709 | 14:46 |
| *** ijw has quit IRC | 14:47 | |
| *** scott-millward has joined #openstack-neutron | 14:47 | |
| *** kopparam has quit IRC | 14:48 | |
| *** markmcclain has joined #openstack-neutron | 14:48 | |
| *** sweston_ has joined #openstack-neutron | 14:48 | |
| *** rwsu has joined #openstack-neutron | 14:49 | |
| *** zhipeng has quit IRC | 14:50 | |
| *** sweston_ has quit IRC | 14:50 | |
| *** vivek-eb_ has joined #openstack-neutron | 14:50 | |
| *** zhipeng has joined #openstack-neutron | 14:50 | |
| enikanorov_ | salv-orlando: how do we commit e-r queries? any bugs needed for that, or just patches? | 14:53 |
| *** sbfox has joined #openstack-neutron | 14:53 | |
| *** shakamunyi has quit IRC | 14:54 | |
| ihrachyshka | enikanorov_: I've taken https://bugs.launchpad.net/neutron/+bug/1311970 for example. as far as I see, the scenario could go as follows: one of the thread is executing delete_port(). It opens transaction, then calls to disassociate_floatingips(). The latter method issues notification. We're yielded while being locked on the port. We yield to another thread that is updating port status to DOWN. It locks, we get the failure. | 14:54 |
| *** sbfox has quit IRC | 14:54 | |
| enikanorov_ | ihrachyshka: that could be | 14:56 |
| *** shakamunyi has joined #openstack-neutron | 14:56 | |
| *** dfarrell07 has joined #openstack-neutron | 14:56 | |
| ihrachyshka | enikanorov_: the problem with those bugs is that we can't say for sure, so 'could be' is the most we may get :) | 14:57 |
| *** shakamunyi is now known as shakamunyi_otp | 14:57 | |
| *** shakamunyi_otp is now known as shakamunyi | 14:57 | |
| *** shakamunyi is now known as shakamunyi_otp | 14:57 | |
| enikanorov_ | ihrachyshka: yeah... | 14:57 |
| ihrachyshka | enikanorov_: this one is actually can be fixed by my patch | 14:57 |
| ihrachyshka | [if the reasoning is correct] | 14:57 |
| *** dfarrell07 has quit IRC | 14:57 | |
| *** dfarrell07 has joined #openstack-neutron | 14:58 | |
| *** mwagner_lap has quit IRC | 15:00 | |
| *** banix has joined #openstack-neutron | 15:01 | |
| *** swat30 has quit IRC | 15:03 | |
| mestery | HenryG: In your latest devstack patch, why do the files/[apt,rpms,rpms-suse]/opendaylight files show up as having been deleted and re-added? | 15:05 |
| *** Guest73428 has joined #openstack-neutron | 15:06 | |
| HenryG | mestery: They are renamed from 'opendaylight' to 'openvswitch'. I don't know why it showed up that way. | 15:06 |
| *** TrevorV_ has joined #openstack-neutron | 15:06 | |
| *** dfarrell07 has quit IRC | 15:07 | |
| *** zhhuabj has quit IRC | 15:07 | |
| * mestery needs glasses | 15:07 | |
| mestery | HenryG: Either that, or I need to stop doing 5 things at the same time. | 15:07 |
| Guest73428 | hi, I just installed neutron of openstack havana in centos 6.5,and I run neutron-db-check: ERROR:unable to find version information for the neutron database. | 15:07 |
| *** dfarrell07 has joined #openstack-neutron | 15:08 | |
| *** kopparam_ has quit IRC | 15:08 | |
| *** kopparam has joined #openstack-neutron | 15:09 | |
| Guest73428 | anybody know the reasons? | 15:09 |
| ihrachyshka | enikanorov_: ok, looking into another case: https://bugs.launchpad.net/neutron/+bug/1313851 there, sync_routers() is called. it opens txn, then calls to _ensure_host_set_on_ports(), which lower in the stack calls to plugin.update_port(), which notifies and yields. then we switch to a thread that updates port state -> boom | 15:10 |
| *** nlahouti has joined #openstack-neutron | 15:10 | |
| *** evgenyf has quit IRC | 15:10 | |
| *** swat30 has joined #openstack-neutron | 15:10 | |
| *** rolledback has quit IRC | 15:10 | |
| *** alagalah has joined #openstack-neutron | 15:11 | |
| ihrachyshka | that one is *not* fixed by my patch, that's another instance | 15:11 |
| *** Guest73428 has left #openstack-neutron | 15:13 | |
| *** kopparam has quit IRC | 15:13 | |
| *** zz_blogan is now known as blogan | 15:14 | |
| ihrachyshka | enikanorov_: mestery: please check above. I need to leave now. I'll return back later to gather your comments, if any :) | 15:15 |
| mestery | ihrachyshka: Will do, thanks! | 15:15 |
| *** mwagner_lap has joined #openstack-neutron | 15:16 | |
| *** trad511 has joined #openstack-neutron | 15:18 | |
| *** shivharis has joined #openstack-neutron | 15:20 | |
| *** shakamunyi_otp has quit IRC | 15:20 | |
| *** shakamunyi has joined #openstack-neutron | 15:20 | |
| *** shakamunyi has quit IRC | 15:20 | |
| shivharis | mestery: ping | 15:20 |
| *** rolledback has joined #openstack-neutron | 15:21 | |
| *** vivek-eb_ has quit IRC | 15:21 | |
| mestery | shivharis: whats up? | 15:22 |
| shivharis | mestery: would you like to delegate this bug: | 15:22 |
| shivharis | https://bugs.launchpad.net/neutron/+bug/1204456 | 15:22 |
| *** rolledback has quit IRC | 15:23 | |
| *** ihrachyshka has quit IRC | 15:23 | |
| mestery | shivharis: Yes, do you want it? | 15:23 |
| shivharis | i can bring this up in the ML2 meeting, i have garnered some volunteers | 15:23 |
| mestery | shivharis: I'll remove myself and let you delegate it. thanks! | 15:23 |
| shivharis | k, anytime | 15:24 |
| *** WackoRobie has joined #openstack-neutron | 15:28 | |
| *** rolledback has joined #openstack-neutron | 15:34 | |
| *** esco-franc has quit IRC | 15:36 | |
| *** ijw has joined #openstack-neutron | 15:38 | |
| *** ajo|mac has joined #openstack-neutron | 15:39 | |
| *** annegent_ has quit IRC | 15:40 | |
| *** WackoRob_ has joined #openstack-neutron | 15:40 | |
| *** WackoRobie has quit IRC | 15:40 | |
| *** manishg has joined #openstack-neutron | 15:41 | |
| *** manishg has quit IRC | 15:41 | |
| *** ijw_ has joined #openstack-neutron | 15:41 | |
| *** Sukhdev has joined #openstack-neutron | 15:42 | |
| *** annegent_ has joined #openstack-neutron | 15:43 | |
| *** ijw has quit IRC | 15:43 | |
| Sukhdev | mestery:Ping | 15:43 |
| *** nati_ueno has quit IRC | 15:43 | |
| *** annegent_ has quit IRC | 15:43 | |
| *** geekinutah has joined #openstack-neutron | 15:44 | |
| mestery | Sukhdev: Whats up? | 15:44 |
| Sukhdev | mestery: We have couple of patches that are ready to go and need core approvals - what is the best way to get approvals? | 15:45 |
| *** ajo|mac has quit IRC | 15:46 | |
| *** ijw_ has quit IRC | 15:46 | |
| *** manishg has joined #openstack-neutron | 15:49 | |
| *** ranger81 has joined #openstack-neutron | 15:50 | |
| manishg | carl_baldwin - I do not find any way to validate string length based on what is in schema. Is there _any_ place in neutron that does it? This is re physnet name validation. | 15:53 |
| Sukhdev | mestery: still there? | 15:54 |
| *** ranger81 has quit IRC | 15:54 | |
| *** xuhanp has quit IRC | 15:54 | |
| HenryG | sc68cal: ping | 15:56 |
| *** slogan_away has joined #openstack-neutron | 15:56 | |
| openstackgerrit | Paul Michali proposed a change to openstack/neutron: VPNaaS: separate out validation logic for ref impl https://review.openstack.org/102351 | 15:57 |
| *** rcurran has joined #openstack-neutron | 15:57 | |
| *** padkrish has joined #openstack-neutron | 15:58 | |
| *** yamahata has quit IRC | 15:58 | |
| *** slogan_away is now known as slogan | 15:59 | |
| carl_baldwin | manishg: Could you ping me a bit later? I’m in a couple of meetings at the moment. :) | 15:59 |
| manishg | carl_baldwin - np. will do . | 16:00 |
| *** spandhe has quit IRC | 16:03 | |
| mestery | Sukhdev: Patches or BPs? If patches, shoot them my way and I' | 16:03 |
| mestery | Sukhdev: have a look | 16:03 |
| *** ijw has joined #openstack-neutron | 16:03 | |
| *** manishg has quit IRC | 16:05 | |
| *** manishg has joined #openstack-neutron | 16:05 | |
| *** sbfox has joined #openstack-neutron | 16:05 | |
| openstackgerrit | Vishal Agarwal proposed a change to openstack/neutron: Fix re-creation of the pool directory https://review.openstack.org/102572 | 16:06 |
| *** puck has quit IRC | 16:15 | |
| *** vjardin has joined #openstack-neutron | 16:15 | |
| *** _cjones_ has joined #openstack-neutron | 16:18 | |
| *** _cjones_ has quit IRC | 16:19 | |
| *** _cjones_ has joined #openstack-neutron | 16:20 | |
| *** puck has joined #openstack-neutron | 16:22 | |
| *** packet has joined #openstack-neutron | 16:24 | |
| *** yfauser1 has joined #openstack-neutron | 16:24 | |
| *** yfauser1 has left #openstack-neutron | 16:24 | |
| *** amotoki has joined #openstack-neutron | 16:26 | |
| *** afazekas_ has quit IRC | 16:26 | |
| *** a_le has joined #openstack-neutron | 16:27 | |
| *** crc32 has joined #openstack-neutron | 16:27 | |
| *** jecarey has quit IRC | 16:28 | |
| *** sweston_ has joined #openstack-neutron | 16:29 | |
| *** yfauser has joined #openstack-neutron | 16:29 | |
| *** yfauser1 has joined #openstack-neutron | 16:31 | |
| *** yfauser1 has left #openstack-neutron | 16:32 | |
| *** yfauser has quit IRC | 16:33 | |
| *** jorgem has joined #openstack-neutron | 16:34 | |
| *** jobewan has joined #openstack-neutron | 16:34 | |
| *** s3wong_ has joined #openstack-neutron | 16:36 | |
| *** shakamunyi has joined #openstack-neutron | 16:37 | |
| *** dave_tucker is now known as dave_tucker_zzz | 16:39 | |
| *** crc32 has quit IRC | 16:40 | |
| *** s3wong has quit IRC | 16:40 | |
| openstackgerrit | Andreas Jaeger proposed a change to openstack/python-neutronclient: Improve help strings https://review.openstack.org/89274 | 16:41 |
| *** devvesa has quit IRC | 16:42 | |
| *** max_lobur has joined #openstack-neutron | 16:42 | |
| max_lobur | Hi folks! I'm interesting whether IPv6 sub-team have a separate channel or all the guys here | 16:43 |
| *** jorgem1 has joined #openstack-neutron | 16:44 | |
| *** jorgem has quit IRC | 16:44 | |
| *** Zemeio has joined #openstack-neutron | 16:47 | |
| *** crc32 has joined #openstack-neutron | 16:48 | |
| *** suresh12 has joined #openstack-neutron | 16:50 | |
| *** doude has quit IRC | 16:52 | |
| *** amaretskiy has quit IRC | 16:54 | |
| *** vivek-ebay has joined #openstack-neutron | 16:54 | |
| *** mwagner_lap has quit IRC | 16:54 | |
| *** s3wong_ is now known as s3wong | 16:56 | |
| *** safchain has quit IRC | 16:56 | |
| *** ranger81 has joined #openstack-neutron | 16:56 | |
| *** amarao has quit IRC | 16:59 | |
| *** jpich has quit IRC | 16:59 | |
| *** achanda has joined #openstack-neutron | 17:00 | |
| banix | Sukhdev: shivharis says he wont be available now; need to find another time | 17:02 |
| Sukhdev | banix: sure | 17:02 |
| *** slogan is now known as slogan_away | 17:02 | |
| Sukhdev | banix: afternoons work best for me - | 17:02 |
| Sukhdev | banix: but, it is late for you :-) | 17:03 |
| Sukhdev | shivharis: are you here? | 17:03 |
| *** rcurran has quit IRC | 17:03 | |
| banix | Sukhdev: afternoon your time is ok for me. say upto 4pm PST. | 17:03 |
| *** baoli has joined #openstack-neutron | 17:03 | |
| *** otherwiseguy has quit IRC | 17:04 | |
| shivharis | Sukhdev: I am here but in another meeting with folks here... | 17:04 |
| shivharis | Sukhdev: ttyl | 17:04 |
| Sukhdev | banix: ah ha - that is not afternoon :0 | 17:04 |
| Sukhdev | banix: do you have my email? | 17:04 |
| *** alagalah_ has joined #openstack-neutron | 17:04 | |
| banix | Sukhdev: can find it | 17:04 |
| *** doude has joined #openstack-neutron | 17:05 | |
| Sukhdev | banix: sukhdev@arista.com | 17:05 |
| *** nlahouti has quit IRC | 17:05 | |
| *** padkrish has quit IRC | 17:05 | |
| Sukhdev | banix: have shivharis email as well - lets discuss on the email and come up with a time slot and take it from there | 17:05 |
| banix | Sukhdev: sounds good | 17:06 |
| Sukhdev | banix: /have/you have/ | 17:06 |
| Sukhdev | banix: OK - cool so, will look for your email | 17:06 |
| banix | Sukhdev: bye now | 17:06 |
| Sukhdev | banix: bye | 17:06 |
| chuckC | Sukhdev: banix: can you inlclude me also, would like to listen in | 17:06 |
| chuckC | ccarlino@hp.com | 17:06 |
| *** alagalah has quit IRC | 17:07 | |
| banix | chuckC: sure thing | 17:07 |
| chuckC | banix: thanks | 17:07 |
| Sukhdev | chuckC: sure | 17:07 |
| openstackgerrit | A change was merged to openstack/python-neutronclient: Ensure .status_code is defined for all NeutronClientExceptions https://review.openstack.org/98318 | 17:08 |
| Sukhdev | mestery: Ping | 17:08 |
| *** yamamoto has quit IRC | 17:08 | |
| *** spandhe has joined #openstack-neutron | 17:09 | |
| *** padkrish has joined #openstack-neutron | 17:09 | |
| *** nlahouti has joined #openstack-neutron | 17:10 | |
| *** leenheer has joined #openstack-neutron | 17:13 | |
| *** harlowja_away is now known as harlowja | 17:13 | |
| *** spandhe has quit IRC | 17:14 | |
| *** yamamoto has joined #openstack-neutron | 17:14 | |
| openstackgerrit | Jakub Libosvar proposed a change to openstack/neutron: Exit rpc_loop when SIGTERM is recieved in ovs-agent https://review.openstack.org/102519 | 17:15 |
| jlibosva | armax: hi, thanks for review on the nova stuff. Could you please take a look on Brent's comment on L907 https://review.openstack.org/#/c/98386/5/nova/network/neutronv2/api.py ? That was the reason why we wanted you to take a look :) | 17:16 |
| *** jaypipes has joined #openstack-neutron | 17:17 | |
| *** rolledback has quit IRC | 17:17 | |
| *** spandhe has joined #openstack-neutron | 17:17 | |
| *** dave_tucker_zzz is now known as dave_tucker | 17:19 | |
| *** banix_ has joined #openstack-neutron | 17:19 | |
| *** shivharis has quit IRC | 17:20 | |
| openstackgerrit | A change was merged to openstack/neutron: Imported Translations from Transifex https://review.openstack.org/102113 | 17:20 |
| openstackgerrit | A change was merged to openstack/python-neutronclient: Set firewall_rules only after appending all rules https://review.openstack.org/99175 | 17:21 |
| *** jlibosva has quit IRC | 17:21 | |
| *** VijayB has joined #openstack-neutron | 17:21 | |
| mestery | Sukhdev: Pong, sorry, juggling many things today | 17:23 |
| *** VijayB has quit IRC | 17:24 | |
| Sukhdev | mestery: no worries - same here - could not complete earlier conversation - had to run to chair ML2 meeting | 17:24 |
| mestery | Sukhdev: Cool! Great working doign that BTW. | 17:24 |
| Sukhdev | mestery: We have three specs that ready for core approvals - | 17:24 |
| *** VijayB_ has joined #openstack-neutron | 17:24 | |
| mestery | Sukhdev: OK, point them our way | 17:25 |
| Sukhdev | mestery: shall I send you the email? | 17:25 |
| mestery | Sukhdev: Please, that would work best. | 17:25 |
| Sukhdev | mestery: cool - I will be sending you the email within next few minutes | 17:25 |
| openstackgerrit | A change was merged to openstack/neutron: NSX: return 400 if dscp set for trusted queue https://review.openstack.org/101201 | 17:25 |
| *** banix_ has quit IRC | 17:26 | |
| *** baojg has quit IRC | 17:26 | |
| openstackgerrit | A change was merged to openstack/neutron: NSX sync cache: add a flag to skip item deletion https://review.openstack.org/101191 | 17:26 |
| *** SumitNaiksatam has joined #openstack-neutron | 17:27 | |
| *** yamamoto has quit IRC | 17:28 | |
| *** yamamoto has joined #openstack-neutron | 17:29 | |
| *** yamamoto_ has joined #openstack-neutron | 17:29 | |
| *** yamamoto has quit IRC | 17:31 | |
| *** yamamoto_ is now known as yamamoto | 17:31 | |
| *** yamamoto_ has joined #openstack-neutron | 17:31 | |
| *** otherwiseguy has joined #openstack-neutron | 17:31 | |
| *** Zemeio has quit IRC | 17:31 | |
| *** mandeep has joined #openstack-neutron | 17:33 | |
| *** sbfox has quit IRC | 17:36 | |
| *** saripurigopi has joined #openstack-neutron | 17:37 | |
| *** yamamoto has quit IRC | 17:37 | |
| openstackgerrit | Bradley Klein proposed a change to openstack/python-neutronclient: Add keystone v3 auth support https://review.openstack.org/92390 | 17:38 |
| *** yamamoto has joined #openstack-neutron | 17:38 | |
| *** yamamoto has quit IRC | 17:40 | |
| *** sbfox has joined #openstack-neutron | 17:40 | |
| *** Sukhdev has quit IRC | 17:42 | |
| *** dave_tucker is now known as dave_tucker_zzz | 17:42 | |
| *** Shrews has joined #openstack-neutron | 17:43 | |
| *** Shrews has left #openstack-neutron | 17:43 | |
| *** rolledback has joined #openstack-neutron | 17:43 | |
| *** jaypipes has quit IRC | 17:44 | |
| *** jaypipes has joined #openstack-neutron | 17:45 | |
| *** zzelle has quit IRC | 17:46 | |
| *** kopparam has joined #openstack-neutron | 17:47 | |
| *** Zemeio has joined #openstack-neutron | 17:49 | |
| manishg | carl_baldwin , rkukura , mestery - is there any way to validate input in neutron to make sure the maxlen of the param is within bounds of the column size (specified in sqlalchemy model) ? particular example is physical_network name -- specified as sa.string(64) in model. How to validate it? Is it required or do we depend on db to throw error for these cases? any examples in the code? | 17:53 |
| *** Zemeio has quit IRC | 17:54 | |
| *** kopparam has quit IRC | 17:54 | |
| carl_baldwin | manishg: Hi, thanks for your patience. Things are beginning to settle down for me here so I can chat now. | 17:54 |
| openstackgerrit | Rossella Sblendido proposed a change to openstack/neutron: Introduce bulk calls for get device details https://review.openstack.org/66899 | 17:54 |
| *** dave_tucker_zzz is now known as dave_tucker | 17:55 | |
| *** a_le has quit IRC | 17:55 | |
| *** yamamoto_ has quit IRC | 17:55 | |
| *** jecarey has joined #openstack-neutron | 17:56 | |
| carl_baldwin | manishg: I had found a page yesterday that looked like it was relevant to this. But, I’m having trouble finding it. | 17:56 |
| * carl_baldwin goes to look through browser history. | 17:56 | |
| carl_baldwin | manishg: Does this help? http://stackoverflow.com/questions/2317081/sqlalchemy-maximum-column-length | 17:57 |
| openstackgerrit | Rossella Sblendido proposed a change to openstack/neutron: Introduce bulk calls for get device details https://review.openstack.org/66899 | 17:58 |
| *** vjardin has quit IRC | 18:00 | |
| *** nati_ueno has joined #openstack-neutron | 18:00 | |
| *** sbfox has quit IRC | 18:01 | |
| *** sbfox has joined #openstack-neutron | 18:02 | |
| *** Sukhdev has joined #openstack-neutron | 18:02 | |
| *** rm_work|away is now known as rm_work | 18:02 | |
| Sukhdev | mestery: I sent the email with three specs requiring core approval | 18:02 |
| *** padkrish has quit IRC | 18:03 | |
| *** nlahouti has quit IRC | 18:03 | |
| mestery | Sukhdev: Got it, thanks! | 18:03 |
| Sukhdev | mestery: BTW, can you also re-look at this one as well https://review.openstack.org/#/c/97125/ | 18:04 |
| *** alagalah_ has quit IRC | 18:05 | |
| sbfox | Hey all, Im having a huge problem with neutron syncdb mirgating from havana 2.2 to icehouse | 18:05 |
| *** nlahouti has joined #openstack-neutron | 18:05 | |
| sbfox | Getting key integrity errors, are there any known problems/fixed you are aware of? | 18:05 |
| manishg | carl_baldwin - let me take a look. what is interesting is nowhere else in neutron code it is validated. so question is why do we want to impose it on this? | 18:06 |
| *** padkrish has joined #openstack-neutron | 18:07 | |
| manishg | carl_baldwin - let me try that out. | 18:07 |
| manishg | another question - how do you internationalize a new exception message? | 18:07 |
| sbfox | - sqlalchemy.exc.OperationalError: (OperationalError) (1061, "Duplicate key name 'uniq_member0pool_id0address0port'") 'ALTER TABLE members ADD CONSTRAINT uniq_member0pool_id0address0port UNIQUE (pool_id, address, protocol_port)' () | 18:08 |
| *** coolsvap is now known as coolsvap|afk | 18:11 | |
| carl_baldwin | manishg: You may have a good point there. I was just focusing on the how. Have you asked rkukura why he included the comment? | 18:11 |
| carl_baldwin | manishg: Where do the values come from that are being validated? Are they user input? What are the chances that a name might exceed the column width? | 18:12 |
| manishg | carl_baldwin - I think his comment was generic - meaning 'do validation' and not much more than that. I haven't been able to get in touch with rkukura. pinged a few times. | 18:13 |
| *** afazekas_ has joined #openstack-neutron | 18:13 | |
| manishg | yes, it's user input and there is a possibility of going wide. same as network name, subnet name, etc. etc. | 18:13 |
| carl_baldwin | manishg: based on this response then I would say it is important to validate it. | 18:14 |
| manishg | none of the others seem to be validated. so prbly relying on db to throw it out. carl_baldwin - yep, agree it should be validated but none of the other params (which ar user inputs) seem to be validated. | 18:14 |
| *** geekinutah has quit IRC | 18:15 | |
| manishg | so I'm kind of curious if there is another mechanism that we rely on. if so, perhaps the same framework will take care of this (not a good idea to do something one off as that will confuse folks and might even start a new [but bad] pattern that people will start top copy/ paste) | 18:15 |
| carl_baldwin | manishg: I don’t think that is a good enough reason alone. Can you give an example of another fixed width field that is not validated? | 18:15 |
| manishg | network name. | 18:15 |
| carl_baldwin | manishg: That is a question worth answering. | 18:15 |
| manishg | subnet name. | 18:15 |
| manishg | and we can take any of the other strings in network/ subnet/ port and any of the extensions :) | 18:16 |
| *** vimal7370 has quit IRC | 18:16 | |
| carl_baldwin | Both network name and subnet name should be validated somehow. Have you tried to supply a name to one of these that exceeds the limit? | 18:16 |
| manishg | I didn't find any validation on maxlen. maybe I missed something. | 18:16 |
| manishg | yeah, I'm going to try it now… ! | 18:17 |
| *** mandeep has quit IRC | 18:17 | |
| carl_baldwin | manishg: Another question that I failed to ask before is “what is the penantly if no validation is done?" | 18:17 |
| carl_baldwin | Does the operation fail or is the name somehow truncated? If the latter, what is the consequence? | 18:18 |
| manishg | carl_baldwin : I'll have to check it out. it depends on how the sql will handle it. currently looks like goes thru! | 18:19 |
| manishg | name = sa.Column(sa.String(255)) -- ths is for Network name. | 18:19 |
| carl_baldwin | What happens to characters 256 - the end? | 18:20 |
| manishg | and I added a 400 character long name. it seems to have gone thru without truncation. so I'm curious …. need to check it out! | 18:20 |
| carl_baldwin | What db engine? | 18:20 |
| manishg | net-list shows me full name. mysql. | 18:20 |
| openstackgerrit | Murali Birru proposed a change to openstack/neutron: L2 enhancement for L3 DVR https://review.openstack.org/87730 | 18:22 |
| openstackgerrit | Murali Birru proposed a change to openstack/neutron: Add L3 Scheduler Changes for Distributed Routers https://review.openstack.org/89694 | 18:22 |
| openstackgerrit | Murali Birru proposed a change to openstack/neutron: Add L3 Extension for Distributed Routers https://review.openstack.org/84223 | 18:22 |
| marun | manishg, carl_baldwin: The wsgi validators support length validation (via max_len parameter) but I can't see any attributes that actually specify it. | 18:23 |
| marun | see neutron/api/v2/attributes.py | 18:23 |
| *** jecarey has quit IRC | 18:23 | |
| manishg | hi marun, I'm trying to see a way to pull the max-len from sqalchemy model so the literal is not specified in two places. | 18:24 |
| marun | manishg: ideally the models would be annotated with enough metadata to drive validation directly | 18:25 |
| manishg | marun, are you talking about the attribute map? I see you specify the validator there and for string perhaps there is a maxlen as an option somewhere. | 18:25 |
| marun | manishg: anything else is going to be hacky | 18:25 |
| marun | manishg: the validation functions take a max_len parameter but I'm not sure how it can be specified | 18:26 |
| *** yamamoto has joined #openstack-neutron | 18:26 | |
| manishg | marun, thanks. yeah, so I'm simply putting in a simple patch that validates a user-specified attribute "physical_network" -- other than making sure that it's non-zero length (i.e. avoid empty string) is there any other validation needed? | 18:26 |
| *** mwagner_lap has joined #openstack-neutron | 18:27 | |
| manishg | the discussion was around the fact that we should validate maxlen. My point is that the db perhaps takes care of it and nowhere else in neutron code it is done. We should not expect the patch to be doing it. | 18:28 |
| manishg | I can put a literal in there based on value in model but that is hacky. Now if model changes , the literal in validation code could be out-of-sync with model. | 18:28 |
| carl_baldwin | manishg: Agreed that is not the right approach ^ | 18:29 |
| manishg | so my point is to not validate maxlen for physical_network name in the patch. | 18:29 |
| marun | manishg: I think there need to be limits on the size somewhere or ddos is possible before the db is ever reached, but as you say if it's not done everywhere it probably doesn't belong in the patch in question. | 18:29 |
| marun | manishg: if maxlen is not validated elsewhere, maybe file a bug to that effect so that it doesn't get lost. | 18:29 |
| manishg | nova uses some validators for such stuff but in neutron we don't have it. and it didn't make sense to pull in a pkg for a patch. it is something separate. yes, makes sense to me marun. I'll file a bug and suggest some of the ways of doing it. | 18:30 |
| carl_baldwin | marun: That sounds reasonable. | 18:31 |
| *** morganfainberg_L has quit IRC | 18:31 | |
| manishg | also , if a string is added how to i18n it ? | 18:31 |
| carl_baldwin | manishg: Looking deeper in to the patch I see this is user input coming through the ml2 conf. | 18:31 |
| *** yamamoto has quit IRC | 18:31 | |
| carl_baldwin | So, ddos attack wouldn’t be possible in this case. | 18:31 |
| manishg | carl_baldwin, yes in that particular patch it's from conf. | 18:31 |
| manishg | but not always. | 18:32 |
| manishg | you could add "*" and then I think anything will be allowed (I'll have to check that). | 18:32 |
| carl_baldwin | manishg: The string should be like this to be translated: _(“Translate me”) | 18:32 |
| *** leenheer has quit IRC | 18:32 | |
| manishg | ah, thanks carl_baldwin. | 18:32 |
| carl_baldwin | manishg: ^ That is all you need to do for the string. | 18:32 |
| carl_baldwin | manishg: Looks like your patch is only one ‘_’ short. :) | 18:33 |
| *** esco-franc has joined #openstack-neutron | 18:33 | |
| esco-franc | hi | 18:33 |
| *** slogan_away has quit IRC | 18:33 | |
| manishg | yeah. will fix it :) thanks. | 18:33 |
| manishg | carl_baldwin - how about "if not all(blah)" vs "for elem in blah: if not elem: " ? should I be changing that? | 18:34 |
| manishg | all we want at this point is empty length validation. | 18:35 |
| *** joshh has quit IRC | 18:36 | |
| carl_baldwin | manishg: If all you’re checking is that the strings are non-zero length then personally I would accept either. But I don’t speak for everyone. | 18:36 |
| manishg | got it. thanks. | 18:36 |
| carl_baldwin | manishg: Since this is configuration file input I’m not very concerned about length validation either. | 18:37 |
| *** leenheer has joined #openstack-neutron | 18:37 | |
| *** geekinutah has joined #openstack-neutron | 18:37 | |
| carl_baldwin | that’s just me | 18:37 |
| manishg | carl_baldwin - thanks for your input . appreciate it. | 18:37 |
| *** igordcard has quit IRC | 18:38 | |
| carl_baldwin | manishg: Of course, the same could be said about validating for non-zero length. ;) | 18:38 |
| carl_baldwin | which is probably why rkukura left it in the first place. | 18:39 |
| manishg | thing with zero-len is that it will make it into db. but beyond max-len is probably caught somewhere (I'm hoping) - and if not there will be some general framework that will catch 'em all. but empty string could be valid config in some cases (optional field). e.g. local type. | 18:40 |
| *** afazekas_ has quit IRC | 18:41 | |
| carl_baldwin | Fair enough. | 18:41 |
| manishg | so I was thinking zero-len is probly worth checking and raise during startup. | 18:41 |
| *** ramishra_ has joined #openstack-neutron | 18:41 | |
| *** ramishra has quit IRC | 18:42 | |
| openstackgerrit | Sayaji Patil proposed a change to openstack/neutron: Making http_timeout a configurable parameter https://review.openstack.org/102604 | 18:42 |
| *** jmsoares has quit IRC | 18:42 | |
| *** jp_at_hp has quit IRC | 18:44 | |
| *** yamamoto has joined #openstack-neutron | 18:46 | |
| *** bandarji has joined #openstack-neutron | 18:47 | |
| *** a_le has joined #openstack-neutron | 18:47 | |
| openstackgerrit | Manish Godara proposed a change to openstack/neutron: validate flat networks physical name https://review.openstack.org/102295 | 18:51 |
| *** yamamoto has quit IRC | 18:51 | |
| *** kopparam has joined #openstack-neutron | 18:51 | |
| openstackgerrit | guang-yee proposed a change to openstack/python-neutronclient: Add keystone v3 auth support https://review.openstack.org/92390 | 18:51 |
| *** harlowja has quit IRC | 18:52 | |
| *** german__ has joined #openstack-neutron | 18:54 | |
| *** leenheer has quit IRC | 18:55 | |
| *** chuckC has quit IRC | 18:56 | |
| *** kopparam has quit IRC | 18:56 | |
| *** padkrish has quit IRC | 19:02 | |
| *** mwagner_lap has quit IRC | 19:02 | |
| *** afazekas_ has joined #openstack-neutron | 19:03 | |
| *** leenheer has joined #openstack-neutron | 19:04 | |
| *** a_le has quit IRC | 19:04 | |
| *** openstackgerrit has quit IRC | 19:04 | |
| *** Longgeek has quit IRC | 19:04 | |
| *** a_le has joined #openstack-neutron | 19:05 | |
| *** leenheer has quit IRC | 19:05 | |
| *** openstackgerrit has joined #openstack-neutron | 19:06 | |
| *** lori is now known as lori|away | 19:07 | |
| *** Longgeek has joined #openstack-neutron | 19:08 | |
| *** achanda_ has joined #openstack-neutron | 19:14 | |
| *** achanda has quit IRC | 19:17 | |
| *** Sukhdev has quit IRC | 19:18 | |
| *** liusheng has quit IRC | 19:20 | |
| *** liusheng has joined #openstack-neutron | 19:20 | |
| *** german__ has quit IRC | 19:23 | |
| *** chuckC has joined #openstack-neutron | 19:23 | |
| *** TrevorV__ has joined #openstack-neutron | 19:23 | |
| *** german__ has joined #openstack-neutron | 19:24 | |
| *** jecarey has joined #openstack-neutron | 19:24 | |
| *** esco-franc has quit IRC | 19:26 | |
| *** stackarr has joined #openstack-neutron | 19:29 | |
| *** Sukhdev has joined #openstack-neutron | 19:31 | |
| *** alexpilotti has quit IRC | 19:32 | |
| *** eezhova has quit IRC | 19:34 | |
| *** TrevorV__ has quit IRC | 19:34 | |
| *** esco-franc has joined #openstack-neutron | 19:40 | |
| *** padkrish has joined #openstack-neutron | 19:40 | |
| *** lykinsbd has joined #openstack-neutron | 19:40 | |
| *** ihrachyshka has joined #openstack-neutron | 19:41 | |
| *** rolledback has quit IRC | 19:42 | |
| *** ajo|mac has joined #openstack-neutron | 19:44 | |
| *** _cjones_ has quit IRC | 19:44 | |
| *** ajo|mac has quit IRC | 19:44 | |
| *** _cjones_ has joined #openstack-neutron | 19:45 | |
| *** crc32 has quit IRC | 19:45 | |
| SpamapS | Hi! should I need to set a DSN for ml2 on compute nodes? | 19:46 |
| *** yamamoto has joined #openstack-neutron | 19:46 | |
| *** a_le has quit IRC | 19:46 | |
| *** ajo|mac has joined #openstack-neutron | 19:47 | |
| *** a_le has joined #openstack-neutron | 19:47 | |
| SpamapS | I'd hope that neutron-openvswitch-agent would only need to contact RabbitMQ. | 19:47 |
| *** crc32 has joined #openstack-neutron | 19:49 | |
| *** ramishra_ has quit IRC | 19:50 | |
| *** ramishra has joined #openstack-neutron | 19:51 | |
| *** ihrachyshka has quit IRC | 19:51 | |
| *** yamamoto has quit IRC | 19:51 | |
| *** a_le has quit IRC | 19:51 | |
| *** ihrachyshka has joined #openstack-neutron | 19:51 | |
| openstackgerrit | Terry Wilson proposed a change to openstack/neutron: Add sanity check for nova notification support https://review.openstack.org/102619 | 19:51 |
| *** shakamunyi has quit IRC | 19:52 | |
| *** leenheer has joined #openstack-neutron | 19:52 | |
| *** erecio has quit IRC | 19:52 | |
| *** shakamunyi has joined #openstack-neutron | 19:53 | |
| ihrachyshka | mestery: hey, I'm back. were there any additional thoughts on db lock timeouts while I was off? | 19:54 |
| *** ramishra has quit IRC | 19:55 | |
| *** padkrish has quit IRC | 19:55 | |
| *** dave_tucker is now known as dave_tucker_zzz | 19:55 | |
| *** erecio has joined #openstack-neutron | 19:57 | |
| *** padkrish has joined #openstack-neutron | 19:58 | |
| *** otherwiseguy has quit IRC | 19:59 | |
| *** mdorman has joined #openstack-neutron | 20:02 | |
| *** nlahouti has quit IRC | 20:03 | |
| *** alagalah has joined #openstack-neutron | 20:04 | |
| *** openstackgerrit has quit IRC | 20:05 | |
| *** Longgeek has quit IRC | 20:05 | |
| *** openstackgerrit has joined #openstack-neutron | 20:06 | |
| mestery | ihrachyshka: Unfourtanetly, no, I've been consumed by other things. :) | 20:07 |
| *** max_lobur has quit IRC | 20:07 | |
| ihrachyshka | mestery: np, at least I didn't miss anything :) | 20:08 |
| mestery | ihrachyshka: :P | 20:08 |
| openstackgerrit | Carl Baldwin proposed a change to openstack/neutron: Add L3 Extension for Distributed Routers https://review.openstack.org/84223 | 20:11 |
| *** sbfox1 has joined #openstack-neutron | 20:12 | |
| *** blogan is now known as zz_blogan | 20:13 | |
| *** sbfox1 has quit IRC | 20:13 | |
| *** sbfox1 has joined #openstack-neutron | 20:13 | |
| *** rolledback has joined #openstack-neutron | 20:14 | |
| *** sbfox has quit IRC | 20:14 | |
| *** alagalah has quit IRC | 20:14 | |
| openstackgerrit | Yi Sun proposed a change to openstack/neutron: This code change adds service group and service object extension https://review.openstack.org/67784 | 20:14 |
| *** alagalah has joined #openstack-neutron | 20:14 | |
| *** a_le has joined #openstack-neutron | 20:15 | |
| *** rolledback has quit IRC | 20:15 | |
| *** sbfox has joined #openstack-neutron | 20:17 | |
| *** sbfox1 has quit IRC | 20:17 | |
| *** amarao has joined #openstack-neutron | 20:17 | |
| *** mwagner_lap has joined #openstack-neutron | 20:18 | |
| *** achanda has joined #openstack-neutron | 20:18 | |
| *** rolledback has joined #openstack-neutron | 20:19 | |
| openstackgerrit | Rossella Sblendido proposed a change to openstack/neutron: Introduce bulk calls for get device details https://review.openstack.org/66899 | 20:19 |
| *** padkrish has quit IRC | 20:20 | |
| *** achanda_ has quit IRC | 20:20 | |
| *** padkrish has joined #openstack-neutron | 20:20 | |
| *** afazekas_ has quit IRC | 20:21 | |
| *** JoeHazzers has quit IRC | 20:21 | |
| *** dsneddon has joined #openstack-neutron | 20:21 | |
| *** rolledback has quit IRC | 20:23 | |
| *** JoeHazzers has joined #openstack-neutron | 20:23 | |
| *** dsneddon has quit IRC | 20:23 | |
| *** arosen has joined #openstack-neutron | 20:23 | |
| *** dsneddon has joined #openstack-neutron | 20:24 | |
| arosen | markmcclain: salv-orlando armax ping https://review.openstack.org/#/c/99461/ | 20:24 |
| *** dave_tucker_zzz is now known as dave_tucker | 20:24 | |
| ihrachyshka | rossella_s: hey. can I ask you not to upload patchsets too often? it's hard to track what's changed in each of them. | 20:24 |
| ihrachyshka | rossella_s: also a brief description of changes in comments after each upload would be useful for reviewers | 20:24 |
| arosen | ihrachyshka: gerrit has a way to diff between patchset ranges that's pretty helpful in those cases. | 20:25 |
| *** dsneddon has quit IRC | 20:25 | |
| rossella_s | arosen: thanks! | 20:25 |
| *** dsneddon has joined #openstack-neutron | 20:25 | |
| *** a_le has quit IRC | 20:25 | |
| *** padkrish has quit IRC | 20:25 | |
| openstackgerrit | Kevin Benton proposed a change to openstack/neutron: Remove __init__ method from TunnelCallback mixin https://review.openstack.org/101121 | 20:25 |
| rossella_s | ihrachyshka: I try not to upload too many patch sets...but it happens that I miss something and I have to push again | 20:25 |
| ihrachyshka | arosen: yes, sure. the problem is that I have comments unhandled while new patchsets are still coming. I think it's worth applying all comments before uploading a new set. | 20:25 |
| *** a_le has joined #openstack-neutron | 20:26 | |
| openstackgerrit | Carl Baldwin proposed a change to openstack/neutron: L2 Agent-side additions to support DVR https://review.openstack.org/87730 | 20:26 |
| arosen | ihrachyshka: yea i understand :/ | 20:26 |
| *** cuddyt has joined #openstack-neutron | 20:27 | |
| salv-orlando | arosen: for your patch I would move the call to update port out of the transaction. I misread the code. I thought it was already this way. The other question however is what this would imply? | 20:27 |
| ihrachyshka | rossella_s: there's nothing bad if your latest patchset has some bugs, take your time to fix those issues and apply other changes requested by reviewers, then upload. no one is in hurry. :) | 20:28 |
| rossella_s | ihrachyshka: I haven't addressed only the comments that I haven't seen. I think I answered to all your first round comments | 20:28 |
| arosen | salv-orlando: in my 1st patch set that's what i did: https://review.openstack.org/#/c/99461/2/neutron/db/l3_db.py | 20:28 |
| *** networkstatic has joined #openstack-neutron | 20:29 | |
| arosen | salv-orlando: right i think moving it out we'll need to have to do rollback manually (and hope that works) | 20:29 |
| ihrachyshka | rossella_s: no, I think classes still reimplement the same method with copy-pasted code; unneeded resync variable is still there. maybe other changes needed, I haven't checked too long | 20:29 |
| arosen | but that's what we do in the nsx plugin today. | 20:29 |
| arosen | I'm fine doing either route | 20:29 |
| arosen | salv-orlando: support tickets just keep pilling up for this issue :P | 20:30 |
| *** a_le has quit IRC | 20:30 | |
| *** rolledback has joined #openstack-neutron | 20:30 | |
| rossella_s | ihrachyshka: I see your point ...anyway that patch was approved already | 20:32 |
| *** padkrish has joined #openstack-neutron | 20:32 | |
| ihrachyshka | salv-orlando: what you would achieve by moving it out of that transaction? | 20:32 |
| salv-orlando | arosen: so I’m not super-dumb. I just reviewed your patch looking at the wrong patchset | 20:32 |
| salv-orlando | arosen: I am just mildly dumb | 20:32 |
| ihrachyshka | rossella_s: does it mean my comments are not to be addressed? | 20:32 |
| rossella_s | ihrachyshka: I mean was approved before your comments. Your comments are mostly nits | 20:33 |
| rossella_s | ihrachyshka: if you have a look at the story, that patch was approved like twice at least | 20:33 |
| ihrachyshka | rossella_s: no, sorry, they are not. copy-paste is not good, leaving unused code is not good. | 20:33 |
| salv-orlando | ihrachyshka: we are trying to avoid creating chances for lock wait timeout because of eventlet yields. an update port within a transaction potentially increases the transaction lenght by a lot - and that in turn will increase the risk of a lock wait timeout | 20:33 |
| arosen | salv-orlando: my vote would be to merge this patch and just hope it doesn't break anything. If it does we'll come back and fix it. But i'm fine to take a different approach if there is one | 20:33 |
| *** nlahouti has joined #openstack-neutron | 20:33 | |
| salv-orlando | arosen: I’m sure if somebody from openstack-qa and openstack-infra are reading what you wrote, they would be taking out a contract on you | 20:34 |
| ihrachyshka | salv-orlando: ok, I see. so you want to reduce area of failure. still the area is there, just a bit less. :) | 20:34 |
| *** otherwiseguy has joined #openstack-neutron | 20:34 | |
| salv-orlando | ihrachyshka: in theory as long as eventlet can yield within a transaction this can happen. | 20:35 |
| *** moha_hunt has joined #openstack-neutron | 20:35 | |
| openstackgerrit | Carl Baldwin proposed a change to openstack/neutron: L2 Agent-side additions to support DVR https://review.openstack.org/87730 | 20:35 |
| salv-orlando | if withing a transaction you do i/o calls such a few rest calls to a 3rd party backend, then you’re like a turkey voting for christmas ;) | 20:35 |
| openstackgerrit | Carl Baldwin proposed a change to openstack/neutron: RPC additions to support DVR https://review.openstack.org/102332 | 20:35 |
| openstackgerrit | Carl Baldwin proposed a change to openstack/neutron: L2 Model additions to support DVR https://review.openstack.org/102101 | 20:35 |
| ihrachyshka | salv-orlando: exactly. so as a way to mitigate existing problem, that's fine. I just wanted to make sure you don't consider it as a fix for the issue, rather a half-fix | 20:36 |
| rossella_s | ihrachyshka: I agree that copy-paste is not good but I don't think a mixin is a good solution...I will leave the decision to some core member. I am just a bit frustrated, that patch has been there for months and was already approved. | 20:36 |
| ihrachyshka | salv-orlando: well, not REST only. any network interaction, like RPC call (we do it a lot) | 20:37 |
| ihrachyshka | rossella_s: I've provided you another alternative in the comments (importing a function into class). please read them. | 20:38 |
| salv-orlando | ihrachyshka: I was suggesting REST because they’re often synchronous, slower, and repeated | 20:38 |
| *** nlahouti has quit IRC | 20:38 | |
| arosen | salv-orlando: :) | 20:38 |
| salv-orlando | but yes, RPC is the same stuff | 20:38 |
| ihrachyshka | rossella_s: I feel your pain | 20:38 |
| ihrachyshka | rossella_s: I'm sorry I haven't reached the patch earlier, and I know latest oslo.messaging port required quite some rebase on your side. | 20:38 |
| salv-orlando | arosen: the fact is the different approach is also tedious… manually handling rollback in case of failures in update_port | 20:39 |
| *** jorgem1 is now known as jorgem | 20:39 | |
| salv-orlando | unless we say that in that case we just port the port in error state? | 20:39 |
| *** achanda has quit IRC | 20:39 | |
| ihrachyshka | salv-orlando: have you seen my draft blueprint to handle some of those lock issues by postponing notifications to after the time of db session close? | 20:40 |
| *** jordandh has quit IRC | 20:41 | |
| boris-42 | mestery Hi! | 20:41 |
| boris-42 | markmcclain hi | 20:41 |
| mestery | boris-42: Howdy! | 20:41 |
| boris-42 | mestery infra patch for neutron rally jobs was merged | 20:41 |
| boris-42 | mestery we need to get this patch https://review.openstack.org/#/c/97803/ now in neutron | 20:42 |
| mestery | boris-42: Excellent! | 20:42 |
| * mestery looks. | 20:42 | |
| *** a_le has joined #openstack-neutron | 20:43 | |
| *** nlahouti has joined #openstack-neutron | 20:43 | |
| mestery | boris-42: I'll review this in detail today yet, but my initial reaction is this could be a bug rather than a BP, I don't think it needs a full BP. | 20:43 |
| openstackgerrit | AARON ZHANG proposed a change to openstack/neutron: Assign Cisco nw profile to multi-tenants in single request https://review.openstack.org/102640 | 20:44 |
| *** rolledback has quit IRC | 20:44 | |
| boris-42 | mestery so jfyi all rally jobs in neutron will fail unit this is merged | 20:44 |
| boris-42 | mestery they are non voting so no worries | 20:44 |
| *** arosen has quit IRC | 20:44 | |
| *** arosen has joined #openstack-neutron | 20:44 | |
| mestery | boris-42: Can you file a bug and re-submit this? That's my initial feedback, and would satisfy yamamoto as well | 20:45 |
| boris-42 | mestery will do | 20:46 |
| *** yamamoto has joined #openstack-neutron | 20:46 | |
| *** karimb has joined #openstack-neutron | 20:46 | |
| *** a_le has quit IRC | 20:47 | |
| ihrachyshka | rossella_s: I don't want your rightful frustration to evolve, so again, let me restate: I'm open to help you, I'm avail for your questions regarding the review, and I'll give your patch a priority in my review queue. hopefully, we'll be able to settle down all remaining issues with it this week. is it ok for you? | 20:49 |
| *** a_le has joined #openstack-neutron | 20:49 | |
| *** yfauser has joined #openstack-neutron | 20:49 | |
| *** gabriel-bezerra has quit IRC | 20:49 | |
| carl_baldwin | rossella_s: ihrachyshka: The question of code duplication in this patch came up long ago during the Montreal code sprint. The problem is bigger than just this patch and we settled on allowing it for consistency. I think I recall the conversation correctly. | 20:49 |
| ihrachyshka | carl_baldwin: allowing what for consistency? | 20:50 |
| *** yfauser has left #openstack-neutron | 20:50 | |
| carl_baldwin | The duplication. The new methods aren’t the only ones. | 20:50 |
| *** rolledback has joined #openstack-neutron | 20:50 | |
| rossella_s | carl_baldwin: yes I think you recall correctly, at least that's the same I remember :) | 20:50 |
| *** yamamoto has quit IRC | 20:51 | |
| carl_baldwin | ihrachyshka: In my opinion, the patch could be accepted as is and someone (maybe you) could address the overall code duplication problem in a refactor. | 20:51 |
| ihrachyshka | carl_baldwin: hm, I don't think this is a good approach. we can do better in this particular case, it doesn't require too much effort to make 'the right way'. it's not about rework month long | 20:51 |
| *** gabriel-bezerra has joined #openstack-neutron | 20:52 | |
| ihrachyshka | carl_baldwin: plus, there are other comments not addressed. passing topic argument to self.call when unneeded, leaving redundant code after changes applied | 20:52 |
| carl_baldwin | ihrachyshka: I disagree. Since the problem is more systematic I’d rather follow the pattern with this patch and fix it with a refactor. | 20:52 |
| *** jordandh has joined #openstack-neutron | 20:52 | |
| *** networkstatic has quit IRC | 20:53 | |
| ihrachyshka | carl_baldwin: it's like saying that since we have plenty of translatable debug messages in the code, we're fine to introduce new | 20:53 |
| *** ajo|mac has quit IRC | 20:53 | |
| carl_baldwin | ihrachyshka: That is a different case. That presents new work to the translators. | 20:54 |
| *** ajo|mac has joined #openstack-neutron | 20:54 | |
| ihrachyshka | and that presents new work for other developers, no? | 20:54 |
| *** yfauser has joined #openstack-neutron | 20:54 | |
| *** yfauser has left #openstack-neutron | 20:54 | |
| ihrachyshka | now if there is a bug in duplicated code, someone will need to go thru all the methods and fix it everywhere, or handle refactoring before fixing a bug | 20:55 |
| *** dguitarbite has quit IRC | 20:56 | |
| *** erecio has quit IRC | 20:56 | |
| *** radez is now known as radez_g0n3 | 20:56 | |
| *** rolledback has quit IRC | 20:57 | |
| ihrachyshka | now, I would be ok if rossella_s stepped in to solve those issues in separate patches of the same authorship. but I haven't heard that. instead, you propose *me* to do the work. :) | 20:57 |
| carl_baldwin | ihrachyshka: I haven’t had a chance to review all of your other comments. Since the patch has progressed a bit, would you mind highlighting the unaddressed comments that are code problems? | 20:58 |
| *** rolledback has joined #openstack-neutron | 20:58 | |
| rossella_s | ihrachyshka: I think I volunteered to take care of this in a comment :) | 20:58 |
| ihrachyshka | carl_baldwin: sure, I was just unsure whether the latest upload is the settled one | 20:58 |
| carl_baldwin | ihrachyshka: It was merely a suggestion. | 20:59 |
| ihrachyshka | rossella_s: well, sorry then for missing that one. let me update the latest patch with all my concerns, and you'll update there how we're going to handle them. ok? | 20:59 |
| rossella_s | ihrachyshka carl_baldwin: I think the comments not addressed are 1) remove topic 2) avoid duplicated code 3) avoid using the resync variable | 20:59 |
| HenryG | ihrachyshka: any idea why my devstack neutron is failing with "oslo.config.cfg.DuplicateOptError: duplicate option: rpc_backend"? | 21:00 |
| rossella_s | ihrachyshka: thanks! next time I will be more consistent with replies to comments and patch submission | 21:00 |
| ihrachyshka | HenryG: probably because old .pyc files are left from old RPC layer. remove neutron/openstack/common/[notifier|rpc] | 21:00 |
| rossella_s | carl_baldwin: thanks a lot :) | 21:00 |
| HenryG | ihrachyshka: Thanks! let me try that | 21:01 |
| *** ajo|mac has quit IRC | 21:02 | |
| *** achanda has joined #openstack-neutron | 21:05 | |
| openstackgerrit | Baodong (Robert) Li proposed a change to openstack/neutron: Support Router Advertisement Daemon (radvd) for IPv6 https://review.openstack.org/102648 | 21:07 |
| *** VijayB_ has quit IRC | 21:08 | |
| *** geekinutah has quit IRC | 21:09 | |
| kevinbenton | salv-orlando, mestery: ping re https://review.openstack.org/#/c/101447/ | 21:09 |
| mestery | kevinbenton: Yo, | 21:09 |
| *** WackoRob_ has quit IRC | 21:09 | |
| *** WackoRobie has joined #openstack-neutron | 21:10 | |
| kevinbenton | mestery: that patch seems to be passing tripleo now consistently | 21:10 |
| *** VijayB_ has joined #openstack-neutron | 21:10 | |
| mestery | kevinbenton: Nice! Now we just need it to pass Jenkins ;) | 21:10 |
| * mestery looks at the jenkins errors | 21:10 | |
| *** VijayB_ has quit IRC | 21:10 | |
| kevinbenton | mestery: yes, this is a new trick the unit tests have shown | 21:10 |
| salv-orlando | kevinbenton: I would love to +A but then jenkins will be angry | 21:10 |
| salv-orlando | and nobody want to see an angry jenkins | 21:11 |
| mestery | salv-orlando: angry jenkins is really mean | 21:11 |
| kevinbenton | salv-orlando, mestery: patch set 3 passed jenkins once | 21:11 |
| ihrachyshka | rossella_s: carl_baldwin: I've walked thru the code and posted my comments. please take a look. | 21:11 |
| kevinbenton | and then started this junk | 21:11 |
| *** geekinutah has joined #openstack-neutron | 21:11 | |
| kevinbenton | and i can’t reproduce locally | 21:12 |
| carl_baldwin | ihrachyshka: Thanks. | 21:12 |
| * carl_baldwin goes to look | 21:12 | |
| * rossella_s goes to look | 21:13 | |
| openstackgerrit | A change was merged to openstack/python-neutronclient: Improve help strings https://review.openstack.org/89274 | 21:14 |
| kevinbenton | salv-orlando, mestery: oh well. merge conflict anyway so i have to manually rebase | 21:14 |
| *** WackoRobie has quit IRC | 21:14 | |
| *** sballe has joined #openstack-neutron | 21:16 | |
| salv-orlando | kevinbenton: passed jenkins once but the last run left a -1 becayuse of unit tests failures… I don’t think those are intermittent | 21:16 |
| rossella_s | ihrachyshka, carl_baldwin so my summary was almost correct... 1) remove topic 2) avoid duplicated code 3) avoid using the resync variable | 21:16 |
| *** amarao has quit IRC | 21:16 | |
| rossella_s | there's another issue and it's about adding a comment | 21:16 |
| *** julim has quit IRC | 21:16 | |
| kevinbenton | salv-orlando: yes, this failure is new behavior that i can’t reproduce locally | 21:16 |
| rossella_s | ihrachyshka, carl_baldwin for 1 and 2 I'd do that in another patch, since those are issues that are affecting other parts of the code | 21:17 |
| openstackgerrit | Terry Wilson proposed a change to openstack/neutron: Add sanity check for nova notification support https://review.openstack.org/102619 | 21:17 |
| ihrachyshka | rossella_s: also a debug message translatable | 21:17 |
| salv-orlando | kevinbenton: even if you rebase your patch to master HEAD? | 21:18 |
| rossella_s | ihrachyshka, carl_baldwin probably it's better to comment on the review | 21:18 |
| kevinbenton | salv-orlando: does jenkins do that now? | 21:18 |
| carl_baldwin | rossella_s: ihrachyshka: I’ll accept that for 1 and 2. | 21:18 |
| kevinbenton | salv-orlando: this is the check queue. i thought it tested as-is | 21:18 |
| salv-orlando | ah right. well that’s strange. | 21:18 |
| salv-orlando | I hope we don’t have intermittent failures in unit tests | 21:19 |
| *** vivek-ebay has quit IRC | 21:19 | |
| rossella_s | ihrachyshka: is that ok for you? regarding 1 and 2? | 21:19 |
| kevinbenton | salv-orlando: although the merge conflict with master is almost exactly this problem... | 21:19 |
| carl_baldwin | kevinbenton: salv-orlando: The check queeu does a merge to HEAD I believe. | 21:19 |
| kevinbenton | carl_baldwin: argh | 21:19 |
| rossella_s | ihrachyshka, carl_baldwin: the rest of the comments, I will address them now | 21:19 |
| salv-orlando | carl_baldwin: I’m to tired to remember how that works | 21:19 |
| kevinbenton | carl_baldwin: that’s new | 21:19 |
| salv-orlando | but if conflicts when you recheck | 21:19 |
| salv-orlando | then if must do the rebase | 21:19 |
| kevinbenton | salv-orlando: yes, rebasing | 21:20 |
| carl_baldwin | kevinbenton: It has been there for a few months. It is how I know I need to rebase. | 21:20 |
| *** julim has joined #openstack-neutron | 21:20 | |
| kevinbenton | carl_baldwin: i don’t like it :-) | 21:20 |
| kevinbenton | carl_baldwin: because it doesn’t tell me that it’s doing that | 21:21 |
| carl_baldwin | kevinbenton: salv-orlando: They may have added the merge step when Jenkins started rechecking patches automatically with new comments on older patches. | 21:21 |
| ihrachyshka | rossella_s: ok, I'm slightly fine with postponing 1 and 2 if they will be addressed in your consequent patches. | 21:21 |
| *** rolledback has quit IRC | 21:21 | |
| rossella_s | ihrachyshka: deal! And thanks for your time and effort to improve my code | 21:22 |
| *** SridharG has quit IRC | 21:22 | |
| *** TrevorV_ has quit IRC | 21:22 | |
| ihrachyshka | nah, I feel like being a pain in the ass :) | 21:23 |
| *** HenryG has quit IRC | 21:23 | |
| *** julim has quit IRC | 21:24 | |
| *** SridharG has joined #openstack-neutron | 21:24 | |
| carl_baldwin | ihrachyshka: thanks for your feedback. I read back through some of the patch sets. | 21:24 |
| openstackgerrit | Kevin Benton proposed a change to openstack/neutron: Revert "Revert "ovs-agent: Ensure integration bridge is created"" https://review.openstack.org/101463 | 21:26 |
| kevinbenton | mestery, salv-orlando: ^^ step 1 | 21:26 |
| *** harlowja has joined #openstack-neutron | 21:28 | |
| kevinbenton | mestery, salv-orlando: looks like the patch that caused the issue was the agent name change | 21:29 |
| kevinbenton | mestery, salv-orlando: removed the call to get_local_port_mac() | 21:29 |
| salv-orlando | kevinbenton: it’s too late for me to remember git history... | 21:29 |
| salv-orlando | which patch? | 21:29 |
| kevinbenton | salv-orlando: https://github.com/openstack/neutron/commit/2ad79a3f9ac8f690cc960128549317c0c3811e19 | 21:30 |
| *** Sukhdev has quit IRC | 21:31 | |
| ihrachyshka | carl_baldwin: rossella_s: ok, I need to have some sleep. if you're keen to push it quick, you may assume that once my other comments are addressed, I'm putting +1 for the patch and stop blocking it. | 21:31 |
| *** a_le has quit IRC | 21:31 | |
| *** a_le has joined #openstack-neutron | 21:32 | |
| rossella_s | ihrachyshka: thanks a lot again!! Have a good rest | 21:32 |
| openstackgerrit | Kevin Benton proposed a change to openstack/neutron: OVS agent: Correct bridge setup ordering https://review.openstack.org/101447 | 21:32 |
| kevinbenton | mestery, salv-orlando: ^^ step 2 | 21:33 |
| openstackgerrit | Rossella Sblendido proposed a change to openstack/neutron: Introduce bulk calls for get device details https://review.openstack.org/66899 | 21:33 |
| salv-orlando | kevinbenton: step0 is to let jenkins finish check queue on https://review.openstack.org/#/c/101447/ | 21:34 |
| salv-orlando | in this way we’ll send both patche together through the gate | 21:34 |
| salv-orlando | because 101463 would cause the triple0 breakage | 21:34 |
| kevinbenton | salv-orlando: indeed | 21:34 |
| *** jaypipes has quit IRC | 21:34 | |
| salv-orlando | and in this way we should minimize the broken window to a few minutes | 21:34 |
| ihrachyshka | rossella_s: ah, you was quick, so I've already put +1. now going to bed, for real :) | 21:35 |
| *** ihrachyshka has quit IRC | 21:35 | |
| *** SridharG has quit IRC | 21:37 | |
| *** SridharG has joined #openstack-neutron | 21:38 | |
| *** SridharG has quit IRC | 21:39 | |
| *** esco-franc has quit IRC | 21:39 | |
| openstackgerrit | Fawad Khaliq proposed a change to openstack/neutron: Fixed dhcp & gateway ip conflict in PLUMgrid plugin https://review.openstack.org/102654 | 21:43 |
| *** yamamoto has joined #openstack-neutron | 21:46 | |
| *** karimb has quit IRC | 21:46 | |
| *** geekinutah has quit IRC | 21:47 | |
| *** puck has quit IRC | 21:47 | |
| *** chuckC has quit IRC | 21:49 | |
| *** dsneddon has quit IRC | 21:51 | |
| *** yamamoto has quit IRC | 21:51 | |
| *** geekinutah has joined #openstack-neutron | 21:55 | |
| *** vivek-ebay has joined #openstack-neutron | 21:56 | |
| openstackgerrit | Carl Baldwin proposed a change to openstack/neutron: Add L3 Scheduler Changes for Distributed Routers https://review.openstack.org/89694 | 21:56 |
| *** VijayB_ has joined #openstack-neutron | 21:56 | |
| mestery | salv-orlando kevinbenton: I'm on board with steps [0,1,2] | 21:58 |
| mestery | Had to step away for a meeting, apologies. | 21:58 |
| *** ajo|mac has joined #openstack-neutron | 21:59 | |
| *** yfauser has joined #openstack-neutron | 22:00 | |
| *** geekinutah has quit IRC | 22:01 | |
| *** dims__ has quit IRC | 22:01 | |
| *** puck has joined #openstack-neutron | 22:01 | |
| *** padkrish has quit IRC | 22:01 | |
| *** padkrish has joined #openstack-neutron | 22:02 | |
| *** yfauser has left #openstack-neutron | 22:02 | |
| *** nlahouti has quit IRC | 22:02 | |
| *** padkrish_ has joined #openstack-neutron | 22:03 | |
| openstackgerrit | fumihiko kakuma proposed a change to openstack/neutron: OFAgent: Share codes of l2-population in OVS agent https://review.openstack.org/87440 | 22:03 |
| kevinbenton | mestery: no prob. i’ll ping you when the tests finish | 22:03 |
| *** nlahouti has joined #openstack-neutron | 22:04 | |
| *** padkrish has quit IRC | 22:06 | |
| *** carl_baldwin has quit IRC | 22:06 | |
| *** markmcclain has quit IRC | 22:08 | |
| *** puck has quit IRC | 22:10 | |
| *** cuddyt has quit IRC | 22:10 | |
| *** dsneddon has joined #openstack-neutron | 22:10 | |
| *** chuckC has joined #openstack-neutron | 22:12 | |
| *** achanda has quit IRC | 22:18 | |
| *** achanda_ has joined #openstack-neutron | 22:20 | |
| *** dims__ has joined #openstack-neutron | 22:25 | |
| *** jobewan has quit IRC | 22:27 | |
| *** slogan has joined #openstack-neutron | 22:30 | |
| *** shakamunyi has quit IRC | 22:31 | |
| *** VijayB_ has quit IRC | 22:32 | |
| *** a_le has quit IRC | 22:32 | |
| *** a_le has joined #openstack-neutron | 22:32 | |
| *** slogan has quit IRC | 22:34 | |
| *** doude has quit IRC | 22:35 | |
| *** mdorman has quit IRC | 22:35 | |
| *** openstackgerrit has quit IRC | 22:35 | |
| *** gduan has joined #openstack-neutron | 22:35 | |
| *** mdorman has joined #openstack-neutron | 22:36 | |
| *** openstackgerrit has joined #openstack-neutron | 22:37 | |
| *** garyduan has quit IRC | 22:38 | |
| *** a_le has quit IRC | 22:38 | |
| *** a_le has joined #openstack-neutron | 22:39 | |
| *** geekinutah has joined #openstack-neutron | 22:39 | |
| *** ijw has quit IRC | 22:40 | |
| *** suresh12 has quit IRC | 22:41 | |
| *** markmcclain has joined #openstack-neutron | 22:45 | |
| *** kopparam has joined #openstack-neutron | 22:45 | |
| *** yamamoto has joined #openstack-neutron | 22:46 | |
| *** suresh12 has joined #openstack-neutron | 22:46 | |
| openstackgerrit | guang-yee proposed a change to openstack/python-neutronclient: Add keystone v3 auth support https://review.openstack.org/92390 | 22:46 |
| *** shakamunyi has joined #openstack-neutron | 22:46 | |
| *** shakamunyi has quit IRC | 22:47 | |
| openstackgerrit | guang-yee proposed a change to openstack/python-neutronclient: Add keystone v3 auth support https://review.openstack.org/92390 | 22:47 |
| *** yamamoto_ has joined #openstack-neutron | 22:48 | |
| *** markmcclain has quit IRC | 22:49 | |
| *** a_le has quit IRC | 22:49 | |
| *** kopparam has quit IRC | 22:50 | |
| *** a_le has joined #openstack-neutron | 22:50 | |
| *** yamamoto has quit IRC | 22:51 | |
| *** shakamunyi has joined #openstack-neutron | 22:52 | |
| *** yamamoto_ has quit IRC | 22:52 | |
| *** jecarey has quit IRC | 22:53 | |
| *** puck has joined #openstack-neutron | 22:55 | |
| *** garyduan has joined #openstack-neutron | 22:55 | |
| *** gduan has quit IRC | 22:56 | |
| *** rms_13 has joined #openstack-neutron | 22:58 | |
| *** jorgem has quit IRC | 22:59 | |
| *** suresh12 has quit IRC | 23:01 | |
| *** markmcclain has joined #openstack-neutron | 23:01 | |
| *** stackarr has quit IRC | 23:01 | |
| *** vivek-ebay has quit IRC | 23:01 | |
| openstackgerrit | Paul Michali proposed a change to openstack/neutron: VPNaaS: separate out validation logic for ref impl https://review.openstack.org/102351 | 23:03 |
| *** vivek-ebay has joined #openstack-neutron | 23:03 | |
| *** vivek-ebay has quit IRC | 23:03 | |
| *** seizadi has joined #openstack-neutron | 23:04 | |
| *** vivek-ebay has joined #openstack-neutron | 23:04 | |
| *** sbfox has quit IRC | 23:05 | |
| *** suresh12 has joined #openstack-neutron | 23:07 | |
| *** mdorman has quit IRC | 23:07 | |
| *** a_le has quit IRC | 23:07 | |
| *** ijw has joined #openstack-neutron | 23:08 | |
| *** ijw has quit IRC | 23:09 | |
| *** ijw has joined #openstack-neutron | 23:09 | |
| *** diegows has quit IRC | 23:10 | |
| kevinbenton | mestery, salv-orlando: tests passed on https://review.openstack.org/#/c/101447/ | 23:11 |
| mestery | kevinbenton: OK, thanks! Going +2 on the dependent patch now | 23:11 |
| salv-orlando | done | 23:12 |
| *** seizadi has quit IRC | 23:13 | |
| *** leenheer has quit IRC | 23:14 | |
| *** otherwiseguy has quit IRC | 23:14 | |
| *** Sukhdev has joined #openstack-neutron | 23:14 | |
| *** seizadi has joined #openstack-neutron | 23:15 | |
| *** moha_hunt has quit IRC | 23:16 | |
| *** padkrish_ has quit IRC | 23:19 | |
| *** sbfox has joined #openstack-neutron | 23:20 | |
| *** nlahouti has quit IRC | 23:20 | |
| openstackgerrit | AARON ZHANG proposed a change to openstack/python-neutronclient: Repeat add-tenant and remove-tenant option in cli https://review.openstack.org/102671 | 23:21 |
| *** padkrish has joined #openstack-neutron | 23:23 | |
| *** kbringard has joined #openstack-neutron | 23:25 | |
| *** mdorman has joined #openstack-neutron | 23:26 | |
| *** shakamunyi has quit IRC | 23:27 | |
| *** gildub has joined #openstack-neutron | 23:27 | |
| *** shakamunyi has joined #openstack-neutron | 23:28 | |
| *** mdorman has quit IRC | 23:29 | |
| *** banix has quit IRC | 23:30 | |
| *** padkrish has quit IRC | 23:30 | |
| jamespd | I am trying to understand why the ovs agent br-ex creates the phy-br-ex <> int-br-ex link. shouldn't there never need to be direct connection between these two bridges because all traffic should be flowing through the L3 agent? | 23:30 |
| *** s3wong has quit IRC | 23:31 | |
| *** nlahouti has joined #openstack-neutron | 23:31 | |
| kevinbenton | jamespd: that sounds like a configuration issue. integration bridges setting i think | 23:32 |
| kevinbenton | jamespd: something associating a physnet to br-ex | 23:32 |
| *** VijayB has joined #openstack-neutron | 23:32 | |
| jamespd | yeah, I have br-ex in bridge_mappings. This is an error, yes? | 23:34 |
| *** nlahouti has quit IRC | 23:36 | |
| kevinbenton | jamespd: yeah, you shouldn’t need that if the l3 agents are configured to use br-ex | 23:36 |
| *** padkrish has joined #openstack-neutron | 23:36 | |
| kevinbenton | jamespd: for external networks | 23:36 |
| kevinbenton | jamespd: which is this setting https://github.com/openstack/neutron/blob/master/etc/l3_agent.ini#L46 | 23:37 |
| *** sballe has quit IRC | 23:38 | |
| jamespd | My problem is that every time I try to create an external network, it gets added to to the 'default' physnet which I am using for the integration bridge. | 23:39 |
| *** SumitNaiksatam has quit IRC | 23:39 | |
| jamespd | so I use 'provider:physical_network public' when creating the network | 23:39 |
| jamespd | which seems to require that 'public:br-ex' be in bridge_mappings | 23:40 |
| *** nlahouti has joined #openstack-neutron | 23:40 | |
| jamespd | I'm not sure how else to make sure that the public network isn't associated with my integration bridge physnet. | 23:40 |
| jamespd | if that makes any sense :/ | 23:40 |
| *** kbringard has quit IRC | 23:41 | |
| kevinbenton | jamespd: well if you have that setting for l3_agent.ini as br-ex, the l3 agent will just plug into that directly so it doesn’t matter if it goes into the default physnet | 23:42 |
| kevinbenton | jamespd: however, if you go the route you went, which is probably more ‘correct’ it will create the link between br-int and br-ex | 23:43 |
| kevinbenton | jamespd: but that link isn’t going to pass traffic so you shouldn’t have to worry | 23:43 |
| kevinbenton | jamespd: if you use ovs-ofctl to dump the flows you should see a default deny | 23:43 |
| jamespd | right. so I have that link, and it does work just fine. | 23:43 |
| jamespd | but it worried me that I had an extra link... I worry that it suggests some config error. | 23:44 |
| kevinbenton | jamespd: oh, nope | 23:44 |
| kevinbenton | jamespd: completely expected if you have a physnet for your external network | 23:44 |
| kevinbenton | jamespd: so the l3 agent external interface will plugin into br-int like every other interface | 23:45 |
| * jamespd nods | 23:45 | |
| kevinbenton | jamespd: and an openflow rule will be installed to rewrite the vlan to the external vlan as it goes to br-ex | 23:45 |
| kevinbenton | jamespd: and another one to rewrite the external vlan back to the br-int local one for returning traffic | 23:46 |
| *** yamamoto has joined #openstack-neutron | 23:46 | |
| *** dave_tucker is now known as dave_tucker_zzz | 23:47 | |
| *** r-daneel has joined #openstack-neutron | 23:48 | |
| *** vivek-ebay has quit IRC | 23:49 | |
| openstackgerrit | Itsuro Oda proposed a change to openstack/neutron: Do not schedule reserved DHCP port to agents https://review.openstack.org/98278 | 23:51 |
| *** yamamoto has quit IRC | 23:51 | |
| *** ajo|mac has quit IRC | 23:52 | |
| *** r-daneel has quit IRC | 23:52 | |
| jamespd | kevinbenton: thanks for the explanation. I think I am following. | 23:56 |
| *** padkrish has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!