| yamamoto | nlahouti: have you asked mark to revisit https://review.openstack.org/#/c/89211/ ? | 00:00 |
|---|---|---|
| *** zarric has quit IRC | 00:00 | |
| nlahouti | yamamoto: Yes. | 00:00 |
| nlahouti | yamamoto: I had brief discussion with him. | 00:00 |
| nlahouti | yamamoto: will continue talking to him tomorrow. | 00:01 |
| *** vivek-ebay has quit IRC | 00:01 | |
| yamamoto | nlahouti: does he have more reasons to keep -2 than bp approval? | 00:02 |
| *** nati_ueno has quit IRC | 00:02 | |
| *** ianw has quit IRC | 00:02 | |
| *** ianw has joined #openstack-neutron | 00:03 | |
| nlahouti | yamamoto: his concern was to improve spec and addess how the API is not going to be broken ... | 00:04 |
| nlahouti | yamamoto: I need to talk to him more to get more details and understand what his concern is | 00:04 |
| nlahouti | yamamoto: he didn't have time today, I have to talk to him tomorrow. | 00:05 |
| yamamoto | nlahouti: ok. thank you | 00:06 |
| nlahouti | yamamoto: np. thx always for your great review | 00:06 |
| *** ijw_ has joined #openstack-neutron | 00:07 | |
| *** sbfox has quit IRC | 00:08 | |
| *** otherwiseguy has joined #openstack-neutron | 00:11 | |
| *** ijw_ has quit IRC | 00:11 | |
| *** flwang_ has quit IRC | 00:13 | |
| openstackgerrit | Armando Migliaccio proposed a change to openstack/neutron: NSX: unify the two distributed routing extensions https://review.openstack.org/108177 | 00:15 |
| *** _cjones_ has quit IRC | 00:15 | |
| *** baoli has joined #openstack-neutron | 00:16 | |
| *** _cjones_ has joined #openstack-neutron | 00:16 | |
| *** dfarrell07 has quit IRC | 00:16 | |
| *** tomoe_ has quit IRC | 00:17 | |
| *** tomoe_ has joined #openstack-neutron | 00:17 | |
| *** nati_ueno has joined #openstack-neutron | 00:18 | |
| *** hemanthravi has quit IRC | 00:20 | |
| *** _cjones_ has quit IRC | 00:20 | |
| *** xgerman has quit IRC | 00:22 | |
| openstackgerrit | Angus Lees proposed a change to openstack/neutron: Unset HTTP_PROXY before attempting to connect to test WSGI server https://review.openstack.org/110853 | 00:22 |
| *** shashankhegde has quit IRC | 00:23 | |
| openstackgerrit | Angus Lees proposed a change to openstack/neutron: Ignore http_proxy while connecting to test WSGI server https://review.openstack.org/110853 | 00:24 |
| openstackgerrit | Brandon Logan proposed a change to openstack/neutron: Implement managers for synchronous haproxy driver https://review.openstack.org/108174 | 00:28 |
| openstackgerrit | Brandon Logan proposed a change to openstack/neutron: Plugin/DB additions for version 2 of LBaaS API https://review.openstack.org/105609 | 00:28 |
| openstackgerrit | Brandon Logan proposed a change to openstack/neutron: Implement synchronous haproxy driver methods https://review.openstack.org/108173 | 00:28 |
| openstackgerrit | Brandon Logan proposed a change to openstack/neutron: Tests for extension, db and plugin for LBaaS V2 https://review.openstack.org/105610 | 00:28 |
| openstackgerrit | Brandon Logan proposed a change to openstack/neutron: New extension for version 2 of LBaaS API https://review.openstack.org/105331 | 00:28 |
| openstackgerrit | Brandon Logan proposed a change to openstack/neutron: Implement Jinja templates for haproxy config https://review.openstack.org/106867 | 00:28 |
| *** zhhuabj has joined #openstack-neutron | 00:28 | |
| openstackgerrit | YAMAMOTO Takashi proposed a change to openstack/neutron: ofagent: Implement physical_interface_mappings https://review.openstack.org/113166 | 00:29 |
| *** dfarrell07 has joined #openstack-neutron | 00:30 | |
| *** armax has left #openstack-neutron | 00:32 | |
| *** a_le has quit IRC | 00:32 | |
| openstackgerrit | Itsuro Oda proposed a change to openstack/neutron: Add endpoint_type parameter to MetaInterfaceDriver https://review.openstack.org/113411 | 00:36 |
| *** ZZelle has quit IRC | 00:38 | |
| openstackgerrit | Swaminathan Vasudevan proposed a change to openstack/neutron: Implement migration of legacy routers to distributed https://review.openstack.org/105855 | 00:41 |
| *** netuser has quit IRC | 00:41 | |
| *** nati_ueno has quit IRC | 00:42 | |
| *** SumitNaiksatam has quit IRC | 00:43 | |
| *** popw has joined #openstack-neutron | 00:44 | |
| *** banix has joined #openstack-neutron | 00:45 | |
| *** nlahouti has quit IRC | 00:46 | |
| *** marun has quit IRC | 00:46 | |
| *** marun has joined #openstack-neutron | 00:47 | |
| *** a_le has joined #openstack-neutron | 00:47 | |
| *** a_le has quit IRC | 00:56 | |
| openstackgerrit | Rajeev Grover proposed a change to openstack/neutron: Fix interface add for dvr with gateway https://review.openstack.org/113412 | 00:57 |
| *** VijayB_ has quit IRC | 01:00 | |
| *** a_le has joined #openstack-neutron | 01:02 | |
| *** SumitNaiksatam has joined #openstack-neutron | 01:02 | |
| *** amcrn has quit IRC | 01:02 | |
| *** packet has joined #openstack-neutron | 01:03 | |
| *** ijw has quit IRC | 01:09 | |
| *** moha_hunt has quit IRC | 01:09 | |
| *** packet has quit IRC | 01:10 | |
| *** mlavalle has quit IRC | 01:11 | |
| openstackgerrit | Rajeev Grover proposed a change to openstack/neutron: Fix interface add for dvr with gateway https://review.openstack.org/113412 | 01:12 |
| *** rudrarugge has quit IRC | 01:15 | |
| *** baoli has quit IRC | 01:15 | |
| *** a_le has quit IRC | 01:17 | |
| *** padkrish has quit IRC | 01:17 | |
| *** a_le has joined #openstack-neutron | 01:18 | |
| *** baoli has joined #openstack-neutron | 01:18 | |
| *** xianghuihui has quit IRC | 01:21 | |
| openstackgerrit | Sayaji Patil proposed a change to openstack/neutron: Implements sync mechanism between Neutron and Nuage VSD https://review.openstack.org/113418 | 01:23 |
| *** xianghui has joined #openstack-neutron | 01:26 | |
| *** SumitNaiksatam has quit IRC | 01:28 | |
| *** SumitNaiksatam has joined #openstack-neutron | 01:29 | |
| *** magesh has quit IRC | 01:31 | |
| *** rudrarugge has joined #openstack-neutron | 01:33 | |
| *** xianghui has quit IRC | 01:34 | |
| *** xianghuihui has joined #openstack-neutron | 01:34 | |
| *** sbfox has joined #openstack-neutron | 01:34 | |
| *** aranjan_ has joined #openstack-neutron | 01:35 | |
| *** carl_baldwin has quit IRC | 01:35 | |
| *** SumitNaiksatam has quit IRC | 01:36 | |
| *** yamahata has joined #openstack-neutron | 01:36 | |
| *** AbsinthMind has quit IRC | 01:38 | |
| *** SumitNaiksatam has joined #openstack-neutron | 01:38 | |
| *** aranjan has quit IRC | 01:38 | |
| *** marun has quit IRC | 01:38 | |
| *** aranjan_ has quit IRC | 01:39 | |
| *** marun has joined #openstack-neutron | 01:40 | |
| *** SumitNaiksatam has quit IRC | 01:40 | |
| *** banix has quit IRC | 01:42 | |
| *** nati_ueno has joined #openstack-neutron | 01:42 | |
| *** a_le has quit IRC | 01:45 | |
| openstackgerrit | Angus Lees proposed a change to openstack/neutron: Purge use of "PRED and A or B" poor-mans-ternary https://review.openstack.org/110849 | 01:48 |
| *** dfarrell07 has quit IRC | 01:50 | |
| *** rook has joined #openstack-neutron | 01:50 | |
| openstackgerrit | Itsuro Oda proposed a change to openstack/neutron: Add endpoint_type parameter to MetaInterfaceDriver https://review.openstack.org/113411 | 01:55 |
| *** SridharR_ has joined #openstack-neutron | 02:00 | |
| *** trinaths has joined #openstack-neutron | 02:01 | |
| openstackgerrit | Franck Yelles proposed a change to openstack/neutron: Implements: blueprint dhcp-host-routes-and-dns-support-for-nuage-plugin https://review.openstack.org/113421 | 02:01 |
| *** SridharRamaswamy has quit IRC | 02:04 | |
| *** yano has quit IRC | 02:05 | |
| *** SridharR_ has quit IRC | 02:05 | |
| *** leenheer has joined #openstack-neutron | 02:05 | |
| *** sbfox has quit IRC | 02:05 | |
| *** ijw has joined #openstack-neutron | 02:08 | |
| *** magesh_ has joined #openstack-neutron | 02:09 | |
| *** leenheer has quit IRC | 02:09 | |
| *** yano has joined #openstack-neutron | 02:11 | |
| openstackgerrit | Michael Smith proposed a change to openstack/neutron: Fixes an issue with FIP re-association https://review.openstack.org/112427 | 02:12 |
| *** AbsinthMind1 has joined #openstack-neutron | 02:12 | |
| *** ijw has quit IRC | 02:13 | |
| *** flwang_ has joined #openstack-neutron | 02:14 | |
| *** flwang_ has quit IRC | 02:18 | |
| *** [1]trinaths has joined #openstack-neutron | 02:20 | |
| *** banix has joined #openstack-neutron | 02:20 | |
| *** AbsinthMind1 has quit IRC | 02:21 | |
| *** nlahouti has joined #openstack-neutron | 02:21 | |
| *** trinaths has quit IRC | 02:22 | |
| *** [1]trinaths is now known as trinaths | 02:22 | |
| openstackgerrit | Isaku Yamahata proposed a change to openstack/neutron: add auth token to context https://review.openstack.org/111756 | 02:22 |
| openstackgerrit | yangxurong proposed a change to openstack/neutron: Use call to report state when ovs_agent starts up https://review.openstack.org/108693 | 02:22 |
| *** zarric has joined #openstack-neutron | 02:23 | |
| *** _cjones_ has joined #openstack-neutron | 02:23 | |
| *** _cjones_ has quit IRC | 02:23 | |
| *** sbalukoff has quit IRC | 02:23 | |
| *** _cjones_ has joined #openstack-neutron | 02:24 | |
| *** VijayB_ has joined #openstack-neutron | 02:25 | |
| *** _cjones_ has quit IRC | 02:25 | |
| *** magesh_ has quit IRC | 02:25 | |
| *** trinaths has quit IRC | 02:25 | |
| *** _cjones_ has joined #openstack-neutron | 02:25 | |
| *** nlahouti_ has joined #openstack-neutron | 02:27 | |
| *** nlahouti has quit IRC | 02:28 | |
| *** vivek-ebay has joined #openstack-neutron | 02:28 | |
| *** nlahouti has joined #openstack-neutron | 02:29 | |
| *** nlahouti_ has quit IRC | 02:29 | |
| *** _cjones_ has quit IRC | 02:30 | |
| *** vivek-eb_ has joined #openstack-neutron | 02:35 | |
| *** vivek-ebay has quit IRC | 02:35 | |
| *** nlahouti has quit IRC | 02:36 | |
| *** magesh_ has joined #openstack-neutron | 02:36 | |
| *** nlahouti has joined #openstack-neutron | 02:43 | |
| *** nlahouti has quit IRC | 02:43 | |
| *** nlahouti has joined #openstack-neutron | 02:44 | |
| *** diegows has quit IRC | 02:49 | |
| *** WackoRobie has quit IRC | 02:53 | |
| *** WackoRobie has joined #openstack-neutron | 02:53 | |
| *** magesh_ has quit IRC | 03:01 | |
| *** magesh_ has joined #openstack-neutron | 03:01 | |
| *** baoli has quit IRC | 03:03 | |
| *** jgrimm has joined #openstack-neutron | 03:05 | |
| *** VijayB_ has quit IRC | 03:06 | |
| *** aranjan has joined #openstack-neutron | 03:09 | |
| openstackgerrit | Angus Lees proposed a change to openstack/neutron: Remove unused arg to config.setup_logging() https://review.openstack.org/111180 | 03:10 |
| *** leenheer has joined #openstack-neutron | 03:12 | |
| *** aranjan has quit IRC | 03:12 | |
| *** aranjan has joined #openstack-neutron | 03:13 | |
| *** leenheer has quit IRC | 03:13 | |
| *** tomoe_ has quit IRC | 03:14 | |
| *** tomoe_ has joined #openstack-neutron | 03:14 | |
| openstackgerrit | A change was merged to openstack/neutron: Implement Midonet Juno Network Api calls https://review.openstack.org/103820 | 03:20 |
| *** tomoe_ has quit IRC | 03:25 | |
| *** tomoe_ has joined #openstack-neutron | 03:25 | |
| *** yamamot__ has joined #openstack-neutron | 03:26 | |
| *** AlexF_ has joined #openstack-neutron | 03:27 | |
| *** yamamoto has quit IRC | 03:29 | |
| *** nati_ueno has quit IRC | 03:31 | |
| *** leenheer has joined #openstack-neutron | 03:31 | |
| *** AbsinthMind has joined #openstack-neutron | 03:33 | |
| *** rwsu has quit IRC | 03:38 | |
| *** SumitNaiksatam has joined #openstack-neutron | 03:38 | |
| *** crc32 has quit IRC | 03:41 | |
| *** sbfox has joined #openstack-neutron | 03:41 | |
| *** Xurong has quit IRC | 03:49 | |
| *** banix has quit IRC | 03:50 | |
| *** WackoRobie has quit IRC | 03:54 | |
| *** WackoRobie has joined #openstack-neutron | 03:54 | |
| *** chandankumar has joined #openstack-neutron | 03:55 | |
| *** zarric has quit IRC | 03:56 | |
| *** WackoRobie has quit IRC | 03:58 | |
| *** zarric has joined #openstack-neutron | 03:59 | |
| *** magesh_ has quit IRC | 03:59 | |
| *** AbsinthMind has quit IRC | 04:00 | |
| *** leenheer has quit IRC | 04:02 | |
| openstackgerrit | A change was merged to openstack/python-neutronclient: Provide support for nested objects https://review.openstack.org/111474 | 04:03 |
| openstackgerrit | A change was merged to openstack/neutron: MLNX Agent: Process port_update notifications in the main agent loop https://review.openstack.org/94882 | 04:04 |
| *** rook has quit IRC | 04:04 | |
| *** zarric has quit IRC | 04:04 | |
| *** popw has quit IRC | 04:05 | |
| *** popw has joined #openstack-neutron | 04:05 | |
| *** a_le has joined #openstack-neutron | 04:06 | |
| *** ijw has joined #openstack-neutron | 04:10 | |
| *** leenheer has joined #openstack-neutron | 04:11 | |
| *** HenryG is now known as HenryG_afk | 04:14 | |
| *** sgordon has quit IRC | 04:14 | |
| *** ijw has quit IRC | 04:15 | |
| *** flwang_ has joined #openstack-neutron | 04:15 | |
| *** a_le has quit IRC | 04:15 | |
| *** a_le has joined #openstack-neutron | 04:15 | |
| *** carl_baldwin has joined #openstack-neutron | 04:17 | |
| *** marun has quit IRC | 04:17 | |
| openstackgerrit | Angus Lees proposed a change to openstack/python-neutronclient: Avoid modifying default function arguments https://review.openstack.org/113435 | 04:18 |
| *** magesh_ has joined #openstack-neutron | 04:18 | |
| *** flwang_ has quit IRC | 04:19 | |
| openstackgerrit | Armando Migliaccio proposed a change to openstack/neutron: Fix PortNotFound exception during sync_routers https://review.openstack.org/113400 | 04:20 |
| *** chandankumar has quit IRC | 04:20 | |
| openstackgerrit | A change was merged to openstack/python-neutronclient: Fix unit tests to succeed on any PYTHONHASHSEED https://review.openstack.org/110991 | 04:21 |
| *** leenheer has quit IRC | 04:22 | |
| *** WackoRobie has joined #openstack-neutron | 04:25 | |
| *** leenheer has joined #openstack-neutron | 04:25 | |
| *** WackoRobie has quit IRC | 04:26 | |
| *** WackoRobie has joined #openstack-neutron | 04:27 | |
| *** networkstatic has quit IRC | 04:27 | |
| *** nati_ueno has joined #openstack-neutron | 04:27 | |
| *** yfried has quit IRC | 04:29 | |
| *** AlexF_ has quit IRC | 04:30 | |
| *** sbalukoff has joined #openstack-neutron | 04:30 | |
| *** leenheer has quit IRC | 04:32 | |
| *** nati_ueno has quit IRC | 04:32 | |
| *** leenheer has joined #openstack-neutron | 04:35 | |
| *** WackoRobie has quit IRC | 04:37 | |
| *** leenheer has quit IRC | 04:38 | |
| *** trinaths has joined #openstack-neutron | 04:43 | |
| *** leenheer has joined #openstack-neutron | 04:43 | |
| *** magesh_ has quit IRC | 04:46 | |
| *** dfarrell07 has joined #openstack-neutron | 04:48 | |
| *** hajay_ has joined #openstack-neutron | 04:55 | |
| *** gildub has quit IRC | 04:56 | |
| *** gildub has joined #openstack-neutron | 04:58 | |
| *** kopparam has joined #openstack-neutron | 05:02 | |
| *** WackoRobie has joined #openstack-neutron | 05:02 | |
| *** rharwood has quit IRC | 05:07 | |
| *** Longgeek has joined #openstack-neutron | 05:07 | |
| *** ianw has quit IRC | 05:08 | |
| *** ianw has joined #openstack-neutron | 05:08 | |
| *** rharwood has joined #openstack-neutron | 05:08 | |
| *** shashankhegde has joined #openstack-neutron | 05:09 | |
| *** WackoRobie has quit IRC | 05:09 | |
| openstackgerrit | Armando Migliaccio proposed a change to openstack/neutron: Make plugin and l3plugin available as mixin's properties https://review.openstack.org/113400 | 05:09 |
| *** enikanorov has quit IRC | 05:10 | |
| *** enikanorov has joined #openstack-neutron | 05:11 | |
| *** garyk has joined #openstack-neutron | 05:12 | |
| *** carl_baldwin has quit IRC | 05:13 | |
| *** jaypipes has joined #openstack-neutron | 05:14 | |
| *** garyk1 has joined #openstack-neutron | 05:14 | |
| *** garyk has quit IRC | 05:14 | |
| openstackgerrit | Armando Migliaccio proposed a change to openstack/neutron: Make plugin and l3plugin available as mixin's properties https://review.openstack.org/113400 | 05:15 |
| openstackgerrit | Rudra Rugge proposed a change to openstack/neutron: Opencontrail plug-in implementation for core resources https://review.openstack.org/96630 | 05:15 |
| *** Zemeio has quit IRC | 05:15 | |
| *** ramishra has joined #openstack-neutron | 05:15 | |
| *** wanghong has quit IRC | 05:15 | |
| *** chandankumar has joined #openstack-neutron | 05:16 | |
| *** wanghong has joined #openstack-neutron | 05:16 | |
| *** chandankumar has quit IRC | 05:17 | |
| *** chandankumar has joined #openstack-neutron | 05:17 | |
| *** ramishra has quit IRC | 05:17 | |
| *** ramishra has joined #openstack-neutron | 05:17 | |
| *** aleksandr_null has quit IRC | 05:20 | |
| *** aleksandr_null has joined #openstack-neutron | 05:21 | |
| *** s3wong has joined #openstack-neutron | 05:22 | |
| openstackgerrit | Armando Migliaccio proposed a change to openstack/neutron: Make plugin and l3plugin available as mixin's properties https://review.openstack.org/113400 | 05:22 |
| *** seizadi has joined #openstack-neutron | 05:22 | |
| *** otherwiseguy has quit IRC | 05:22 | |
| *** otherwiseguy has joined #openstack-neutron | 05:23 | |
| *** WackoRobie has joined #openstack-neutron | 05:24 | |
| *** AlexF_ has joined #openstack-neutron | 05:25 | |
| openstackgerrit | Armando Migliaccio proposed a change to openstack/neutron: Make plugin and l3plugin available as mixin's properties https://review.openstack.org/113400 | 05:27 |
| *** nati_ueno has joined #openstack-neutron | 05:28 | |
| *** WackoRobie has quit IRC | 05:29 | |
| *** ramishra has quit IRC | 05:29 | |
| *** rotbeard has joined #openstack-neutron | 05:29 | |
| *** ramishra has joined #openstack-neutron | 05:30 | |
| *** vivek-ebay has joined #openstack-neutron | 05:31 | |
| *** nati_ueno has quit IRC | 05:33 | |
| *** ramishra has quit IRC | 05:34 | |
| *** vivek-eb_ has quit IRC | 05:34 | |
| *** ramishra has joined #openstack-neutron | 05:35 | |
| *** vivek-ebay has quit IRC | 05:35 | |
| *** k4n0 has joined #openstack-neutron | 05:37 | |
| *** WackoRobie has joined #openstack-neutron | 05:38 | |
| *** gangil has joined #openstack-neutron | 05:39 | |
| *** WackoRobie has quit IRC | 05:40 | |
| *** ianw has quit IRC | 05:42 | |
| *** ianw has joined #openstack-neutron | 05:42 | |
| *** WackoRobie has joined #openstack-neutron | 05:43 | |
| *** ianw has quit IRC | 05:43 | |
| openstackgerrit | Angus Lees proposed a change to openstack/python-neutronclient: Remove incorrect super() call https://review.openstack.org/113439 | 05:43 |
| *** ianw has joined #openstack-neutron | 05:43 | |
| *** rharwood has quit IRC | 05:43 | |
| *** ianw has quit IRC | 05:43 | |
| *** ianw has joined #openstack-neutron | 05:44 | |
| *** ianw has quit IRC | 05:44 | |
| *** yfried has joined #openstack-neutron | 05:44 | |
| *** ianw has joined #openstack-neutron | 05:44 | |
| *** ianw has quit IRC | 05:45 | |
| *** ianw has joined #openstack-neutron | 05:45 | |
| *** ianw has quit IRC | 05:46 | |
| *** ianw has joined #openstack-neutron | 05:46 | |
| *** rharwood has joined #openstack-neutron | 05:46 | |
| openstackgerrit | Armando Migliaccio proposed a change to openstack/neutron: Fix PortNotFound exception during sync_routers https://review.openstack.org/113441 | 05:48 |
| *** Xurong has joined #openstack-neutron | 05:48 | |
| *** leenheer has quit IRC | 05:48 | |
| *** ianw has quit IRC | 05:50 | |
| *** ianw has joined #openstack-neutron | 05:51 | |
| *** leenheer has joined #openstack-neutron | 05:53 | |
| *** otherwiseguy has quit IRC | 05:53 | |
| *** ajo_ has quit IRC | 05:53 | |
| *** ajo__ has joined #openstack-neutron | 05:53 | |
| openstackgerrit | A change was merged to openstack/neutron: Fixes an issue with FIP re-association https://review.openstack.org/112427 | 05:54 |
| *** leenheer has quit IRC | 06:00 | |
| *** s3wong has quit IRC | 06:00 | |
| *** ianw has quit IRC | 06:01 | |
| *** tomoe_ has quit IRC | 06:01 | |
| openstackgerrit | Preeti Mirji proposed a change to openstack/neutron: Adm policy can't use other tenants unshared rules https://review.openstack.org/108952 | 06:01 |
| *** reed has quit IRC | 06:01 | |
| *** tomoe_ has joined #openstack-neutron | 06:01 | |
| *** sgordon`` has quit IRC | 06:01 | |
| *** rharwood has quit IRC | 06:01 | |
| openstackgerrit | Armando Migliaccio proposed a change to openstack/neutron: Improve structure of external_gateway_added https://review.openstack.org/112385 | 06:01 |
| *** reed has joined #openstack-neutron | 06:02 | |
| openstackgerrit | OpenStack Proposal Bot proposed a change to openstack/neutron: Imported Translations from Transifex https://review.openstack.org/112761 | 06:04 |
| *** sgordon has joined #openstack-neutron | 06:06 | |
| *** rharwood has joined #openstack-neutron | 06:07 | |
| *** SumitNaiksatam has quit IRC | 06:07 | |
| *** evgenyf has joined #openstack-neutron | 06:07 | |
| *** shashankhegde has quit IRC | 06:08 | |
| *** SumitNaiksatam has joined #openstack-neutron | 06:08 | |
| *** yamamoto has joined #openstack-neutron | 06:09 | |
| *** yamamo___ has joined #openstack-neutron | 06:10 | |
| *** ijw has joined #openstack-neutron | 06:12 | |
| *** oda-g has quit IRC | 06:12 | |
| *** yamamot__ has quit IRC | 06:12 | |
| *** yamamoto has quit IRC | 06:14 | |
| *** AlexF_ has quit IRC | 06:14 | |
| trinaths | Hi Reviewers, May I have your time for review of my new patchset at https://review.openstack.org/#/c/109659/ | 06:15 |
| openstackgerrit | YAMAMOTO Takashi proposed a change to openstack/neutron: ofagent: Implement physical_interface_mappings https://review.openstack.org/113166 | 06:15 |
| openstackgerrit | YAMAMOTO Takashi proposed a change to openstack/neutron: l2pop: Allow network types overridable https://review.openstack.org/112947 | 06:15 |
| openstackgerrit | YAMAMOTO Takashi proposed a change to openstack/neutron: ofagent: Local arp responder for VLAN https://review.openstack.org/112949 | 06:15 |
| *** AlexF_ has joined #openstack-neutron | 06:15 | |
| *** flwang_ has joined #openstack-neutron | 06:16 | |
| *** ijw has quit IRC | 06:16 | |
| *** nati_ueno has joined #openstack-neutron | 06:17 | |
| *** Longgeek has quit IRC | 06:19 | |
| *** irenab has joined #openstack-neutron | 06:19 | |
| *** dfarrell07 has quit IRC | 06:20 | |
| *** flwang_ has quit IRC | 06:20 | |
| *** Longgeek has joined #openstack-neutron | 06:21 | |
| *** pradipta_away is now known as pradipta | 06:23 | |
| *** ollivier has joined #openstack-neutron | 06:26 | |
| *** yamamo___ has quit IRC | 06:28 | |
| *** a_le has quit IRC | 06:31 | |
| *** akash has joined #openstack-neutron | 06:31 | |
| *** gangil has left #openstack-neutron | 06:31 | |
| *** devvesa has joined #openstack-neutron | 06:33 | |
| *** atmark has joined #openstack-neutron | 06:33 | |
| atmark | Hello, I'm having problem starting neutron-server. I'm getting this error in the log file | 06:35 |
| atmark | Reconnecting to AMQP server on 10.107.1.1:5672 | 06:35 |
| trinaths | atmark: check your neutron.conf whether amqp is configured correctly | 06:35 |
| atmark | rabbitmq is running in the background | 06:35 |
| atmark | [root@controller-node ~(keystone_admin)]# lsof -i :5672 | 06:35 |
| atmark | COMMAND PID USER FD TYPE DEVICE SIZE/OFF NODE NAME | 06:36 |
| atmark | beam.smp 37027 rabbitmq 15u IPv6 232688 0t0 TCP *:amqp (LISTEN) | 06:36 |
| trinaths | atmark: use paste.openstack.org to showcase logs | 06:36 |
| *** akash has quit IRC | 06:37 | |
| *** lordd_ has joined #openstack-neutron | 06:39 | |
| atmark | trinaths: http://sprunge.us/ibYJ | 06:39 |
| *** hajay_ has quit IRC | 06:42 | |
| *** nlahouti has quit IRC | 06:42 | |
| atmark | trinaths: am I missing something? | 06:43 |
| openstackgerrit | Robin Wang proposed a change to openstack/neutron: Neutron-ovs-agent crashes when OVS is restarted/dead https://review.openstack.org/113450 | 06:43 |
| *** gangil has joined #openstack-neutron | 06:44 | |
| *** bvandenh has joined #openstack-neutron | 06:44 | |
| *** gangil has quit IRC | 06:44 | |
| *** gangil has joined #openstack-neutron | 06:44 | |
| trinaths | atmark: neutron.conf is okay | 06:45 |
| *** gangil has quit IRC | 06:45 | |
| trinaths | atmark: can you restart and check the server log. | 06:45 |
| *** gangil has joined #openstack-neutron | 06:45 | |
| atmark | AMQP server on 10.107.1.1:5672 is unreachable: Socket closed. Trying again in 3 seconds. | 06:48 |
| *** AlexF_ has quit IRC | 06:51 | |
| *** lordd_ has quit IRC | 06:52 | |
| trinaths | atmark: where is this amqp actually running? Is on Openstack controller or some other server | 06:53 |
| *** WackoRobie has quit IRC | 06:56 | |
| atmark | it's on the controller | 06:57 |
| *** pasquier-s has joined #openstack-neutron | 06:57 | |
| *** yamamoto has joined #openstack-neutron | 06:58 | |
| openstackgerrit | Evgeny Fedoruk proposed a change to openstack/neutron: New common util module for Barbican TLS containers https://review.openstack.org/109849 | 06:59 |
| *** ivar-lazzaro has quit IRC | 07:00 | |
| *** ivar-lazzaro has joined #openstack-neutron | 07:00 | |
| *** lordd_ has joined #openstack-neutron | 07:01 | |
| trinaths | atmark: okay. can make it localhost for rabbit_host and try it | 07:02 |
| *** ivar-lazzaro has quit IRC | 07:02 | |
| *** ivar-lazzaro1 has joined #openstack-neutron | 07:02 | |
| *** a_le has joined #openstack-neutron | 07:02 | |
| *** ramishra has quit IRC | 07:02 | |
| *** ramishra has joined #openstack-neutron | 07:02 | |
| atmark | trinaths: jus jut a sec | 07:05 |
| atmark | just a sec* | 07:06 |
| *** ramishra has quit IRC | 07:07 | |
| atmark | trinaths: still the same | 07:08 |
| atmark | AMQP server on 127.0.0.1:5672 is unreachable: Socket closed. Trying again in 15 seconds. | 07:08 |
| *** matrohon has joined #openstack-neutron | 07:08 | |
| *** a_le has quit IRC | 07:08 | |
| *** yamamoto has quit IRC | 07:10 | |
| *** dbite has joined #openstack-neutron | 07:10 | |
| *** SridharG has joined #openstack-neutron | 07:12 | |
| *** ijw has joined #openstack-neutron | 07:13 | |
| *** nlahouti has joined #openstack-neutron | 07:13 | |
| *** nlahouti has quit IRC | 07:17 | |
| *** ijw has quit IRC | 07:17 | |
| openstackgerrit | Angus Lees proposed a change to openstack/python-neutronclient: Remove incorrect super() call https://review.openstack.org/113439 | 07:18 |
| trinaths | atmark: can you restart amqp | 07:18 |
| *** jgallard has joined #openstack-neutron | 07:20 | |
| openstackgerrit | Rudra Rugge proposed a change to openstack/neutron: Opencontrail plug-in implementation for core resources https://review.openstack.org/96630 | 07:21 |
| *** jgallard has quit IRC | 07:23 | |
| *** jgallard has joined #openstack-neutron | 07:23 | |
| *** nagyz has joined #openstack-neutron | 07:24 | |
| *** ramishra has joined #openstack-neutron | 07:27 | |
| *** sballe has joined #openstack-neutron | 07:27 | |
| *** seizadi has quit IRC | 07:27 | |
| *** salv-orlando has quit IRC | 07:27 | |
| *** jschwarz has joined #openstack-neutron | 07:28 | |
| *** yamamoto has joined #openstack-neutron | 07:28 | |
| *** afazekas has joined #openstack-neutron | 07:31 | |
| *** yamamoto has quit IRC | 07:31 | |
| *** jistr has joined #openstack-neutron | 07:36 | |
| *** trinaths has quit IRC | 07:38 | |
| *** rudrarugge has quit IRC | 07:40 | |
| *** trinaths has joined #openstack-neutron | 07:42 | |
| *** AlexF_ has joined #openstack-neutron | 07:49 | |
| openstackgerrit | Angus Lees proposed a change to openstack/neutron: Raise correct NotImplementedError exception https://review.openstack.org/113457 | 07:50 |
| *** flwang_ has joined #openstack-neutron | 07:51 | |
| *** gangil has quit IRC | 07:53 | |
| *** jpich has joined #openstack-neutron | 07:53 | |
| *** gangil has joined #openstack-neutron | 07:53 | |
| *** Krast has joined #openstack-neutron | 07:55 | |
| *** gangil has quit IRC | 07:58 | |
| *** gangil has joined #openstack-neutron | 07:58 | |
| *** gangil has quit IRC | 07:58 | |
| *** gangil has joined #openstack-neutron | 07:58 | |
| *** networkstatic has joined #openstack-neutron | 07:58 | |
| *** ramishra has quit IRC | 07:58 | |
| *** networkstatic has quit IRC | 07:58 | |
| *** ramishra has joined #openstack-neutron | 07:59 | |
| *** fifieldt__ is now known as fifieldt | 07:59 | |
| *** Krast has quit IRC | 08:02 | |
| *** yamamoto has joined #openstack-neutron | 08:02 | |
| openstackgerrit | Robin Wang proposed a change to openstack/neutron: Neutron-ovs-agent crashes when OVS is restarted/dead https://review.openstack.org/113450 | 08:02 |
| *** sbfox has quit IRC | 08:03 | |
| *** ramishra has quit IRC | 08:03 | |
| *** AlexF_ has quit IRC | 08:04 | |
| *** dosaboy_ has joined #openstack-neutron | 08:05 | |
| openstackgerrit | Chirag Shahani proposed a change to openstack/neutron: Add template attr. for subnet, router create in Nuage plugin https://review.openstack.org/112749 | 08:05 |
| *** yamamoto has quit IRC | 08:06 | |
| *** Krast has joined #openstack-neutron | 08:06 | |
| *** Dafna has joined #openstack-neutron | 08:07 | |
| openstackgerrit | John Schwarz proposed a change to openstack/neutron: Allow filtering routers by network-ids https://review.openstack.org/111580 | 08:09 |
| openstackgerrit | Robin Wang proposed a change to openstack/neutron: Neutron-ovs-agent crashes when OVS is restarted/dead(Alternative Patch I) https://review.openstack.org/110538 | 08:13 |
| *** jlibosva has joined #openstack-neutron | 08:14 | |
| *** ijw has joined #openstack-neutron | 08:14 | |
| *** nlahouti has joined #openstack-neutron | 08:15 | |
| *** ihrachyshka has joined #openstack-neutron | 08:16 | |
| *** nlahouti has quit IRC | 08:19 | |
| *** ijw has quit IRC | 08:19 | |
| *** ihrachyshka has quit IRC | 08:20 | |
| *** ihrachyshka has joined #openstack-neutron | 08:21 | |
| openstackgerrit | Robin Wang proposed a change to openstack/neutron: Neutron-ovs-agent crashes when OVS is restarted/dead(Alternative Patch II) https://review.openstack.org/113450 | 08:22 |
| openstackgerrit | Kevin Benton proposed a change to openstack/neutron: External Attachment Point Extension Mixin https://review.openstack.org/97173 | 08:23 |
| *** zhhuabj has quit IRC | 08:24 | |
| *** zhhuabj has joined #openstack-neutron | 08:25 | |
| openstackgerrit | shihanzhang proposed a change to openstack/neutron: Refactor security group rpc call https://review.openstack.org/111876 | 08:25 |
| *** amuller has joined #openstack-neutron | 08:27 | |
| *** yfried is now known as yfried_afk | 08:27 | |
| *** AlexF_ has joined #openstack-neutron | 08:27 | |
| *** yamamoto has joined #openstack-neutron | 08:28 | |
| *** ramishra has joined #openstack-neutron | 08:29 | |
| trinaths | Hi Reviewers, May I have your time for review of my new patchset at https://review.openstack.org/#/c/109659/ | 08:31 |
| *** yamamoto has quit IRC | 08:33 | |
| *** nati_ueno has quit IRC | 08:33 | |
| *** ramishra has quit IRC | 08:34 | |
| *** nati_ueno has joined #openstack-neutron | 08:34 | |
| *** dosaboy_ has quit IRC | 08:36 | |
| *** drankis has quit IRC | 08:37 | |
| *** tomoe_ has quit IRC | 08:38 | |
| *** tomoe_ has joined #openstack-neutron | 08:38 | |
| *** Krast has quit IRC | 08:40 | |
| *** Krast has joined #openstack-neutron | 08:40 | |
| *** AlexF_ has quit IRC | 08:41 | |
| *** nlahouti has joined #openstack-neutron | 08:46 | |
| *** dosaboy_ has joined #openstack-neutron | 08:46 | |
| *** dosaboy_ is now known as dosaboy-alt | 08:46 | |
| *** nagyz has quit IRC | 08:46 | |
| *** dosaboy-alt has quit IRC | 08:47 | |
| *** nlahouti has quit IRC | 08:50 | |
| *** pradipta is now known as pradipta_away | 08:51 | |
| openstackgerrit | Franck Yelles proposed a change to openstack/neutron: Adding support of DNS nameserver and Host routes for the Nuage Plugin https://review.openstack.org/113421 | 08:53 |
| *** AlexF_ has joined #openstack-neutron | 08:55 | |
| *** zhhuabj has quit IRC | 08:56 | |
| *** zhhuabj has joined #openstack-neutron | 08:56 | |
| *** gangil has quit IRC | 09:03 | |
| *** a_le has joined #openstack-neutron | 09:03 | |
| *** Hayim has quit IRC | 09:06 | |
| *** AlexF_ has quit IRC | 09:07 | |
| openstackgerrit | Robin Wang proposed a change to openstack/neutron: Neutron-ovs-agent crashes when OVS is restarted/dead(Alternative Patch I) https://review.openstack.org/110538 | 09:07 |
| *** Hayim has joined #openstack-neutron | 09:07 | |
| *** a_le has quit IRC | 09:08 | |
| *** AlexF_ has joined #openstack-neutron | 09:08 | |
| *** nati_ueno has quit IRC | 09:12 | |
| *** nlahouti has joined #openstack-neutron | 09:16 | |
| *** claudiub has joined #openstack-neutron | 09:18 | |
| *** dosaboy has quit IRC | 09:20 | |
| openstackgerrit | Miroslav Los proposed a change to openstack/neutron: Fix performance of get_bridge_for_tap_interface and its use https://review.openstack.org/113467 | 09:20 |
| *** nlahouti has quit IRC | 09:21 | |
| *** dosaboy has joined #openstack-neutron | 09:23 | |
| openstackgerrit | Miroslav Los proposed a change to openstack/neutron: Fix performance of get_bridge_for_tap_interface https://review.openstack.org/113467 | 09:24 |
| *** aranjan has quit IRC | 09:24 | |
| *** evgenyf has quit IRC | 09:27 | |
| *** ramishra has joined #openstack-neutron | 09:28 | |
| *** yamamoto has joined #openstack-neutron | 09:29 | |
| *** ajo_ has joined #openstack-neutron | 09:29 | |
| openstackgerrit | Miroslav Los proposed a change to openstack/neutron: Fix performance of get_bridge_for_tap_interface https://review.openstack.org/113467 | 09:29 |
| *** geekinut1h has joined #openstack-neutron | 09:30 | |
| *** AlexF_ has quit IRC | 09:32 | |
| *** ajo__ has quit IRC | 09:32 | |
| *** evgenyf has joined #openstack-neutron | 09:32 | |
| *** geekinutah has quit IRC | 09:33 | |
| *** yamamoto has quit IRC | 09:34 | |
| *** AlexF_ has joined #openstack-neutron | 09:36 | |
| *** oreillyd has joined #openstack-neutron | 09:37 | |
| openstackgerrit | Ilya Shakhat proposed a change to openstack/python-neutronclient: Replace httpretty with requests_mock https://review.openstack.org/113265 | 09:38 |
| trinaths | Hi Reviewers, May I have your time for review of my new patchset at https://review.openstack.org/#/c/109659/ | 09:46 |
| openstackgerrit | Koteswara Rao Kelam proposed a change to openstack/python-neutronclient: firewall policy update for a rule is not working https://review.openstack.org/104132 | 09:47 |
| *** zhiyan has quit IRC | 09:47 | |
| *** jdag has quit IRC | 09:48 | |
| *** ctracey has quit IRC | 09:49 | |
| *** skath has quit IRC | 09:50 | |
| *** jdag_ has joined #openstack-neutron | 09:50 | |
| *** sweston has quit IRC | 09:50 | |
| *** nimrodsun has quit IRC | 09:52 | |
| *** zhiyan has joined #openstack-neutron | 09:52 | |
| openstackgerrit | Robin Wang proposed a change to openstack/neutron: OVS-Agent crashes when OVS restarted/dead(Alternative Patch II) https://review.openstack.org/113450 | 09:53 |
| *** ctracey has joined #openstack-neutron | 09:53 | |
| *** sweston has joined #openstack-neutron | 09:54 | |
| *** skath has joined #openstack-neutron | 09:54 | |
| *** dgollub has joined #openstack-neutron | 09:54 | |
| *** jdag_ has quit IRC | 09:55 | |
| *** flaviof_zzz is now known as flaviof | 09:55 | |
| *** zhiyan has quit IRC | 09:57 | |
| openstackgerrit | John Davidge proposed a change to openstack/neutron: Do not assume order of new_peers list elements https://review.openstack.org/112034 | 09:57 |
| openstackgerrit | Robin Wang proposed a change to openstack/neutron: OVS-Agent crashes when OVS dead(Alternative I) https://review.openstack.org/110538 | 09:57 |
| *** nimrodsun has joined #openstack-neutron | 09:57 | |
| openstackgerrit | Robin Wang proposed a change to openstack/neutron: OVS-Agent crashes when OVS dead(Alternative II) https://review.openstack.org/113450 | 09:58 |
| *** AlexF_ has quit IRC | 09:59 | |
| *** zhiyan has joined #openstack-neutron | 09:59 | |
| *** jdag_ has joined #openstack-neutron | 10:00 | |
| *** dave_tucker_zzz is now known as dave_tucker | 10:00 | |
| *** raorn_ is now known as raorn | 10:03 | |
| *** nati_ueno has joined #openstack-neutron | 10:09 | |
| *** aranjan has joined #openstack-neutron | 10:11 | |
| *** alexpilotti has joined #openstack-neutron | 10:12 | |
| *** jgallard has quit IRC | 10:12 | |
| *** ianw has joined #openstack-neutron | 10:14 | |
| *** nati_ueno has quit IRC | 10:14 | |
| *** nlahouti has joined #openstack-neutron | 10:18 | |
| *** pasquier-s has quit IRC | 10:20 | |
| *** AlexF_ has joined #openstack-neutron | 10:21 | |
| *** nlahouti has quit IRC | 10:22 | |
| *** aranjan has quit IRC | 10:24 | |
| *** ajo__ has joined #openstack-neutron | 10:26 | |
| *** sgordon_ has joined #openstack-neutron | 10:28 | |
| *** ajo_ has quit IRC | 10:28 | |
| *** yamamoto has joined #openstack-neutron | 10:29 | |
| *** ajo_ has joined #openstack-neutron | 10:30 | |
| *** ajo__ has quit IRC | 10:30 | |
| *** AlexF_ has quit IRC | 10:31 | |
| *** yamamoto has quit IRC | 10:33 | |
| *** AlexF_ has joined #openstack-neutron | 10:34 | |
| *** yfried_afk is now known as yfried | 10:35 | |
| *** dgollub has quit IRC | 10:43 | |
| *** sballe has quit IRC | 10:46 | |
| *** Krast has quit IRC | 10:50 | |
| *** RockKuo_Home has joined #openstack-neutron | 10:51 | |
| *** aranjan has joined #openstack-neutron | 10:51 | |
| *** ZZelle has joined #openstack-neutron | 10:52 | |
| jschwarz | ilyashakhat_, ping | 10:52 |
| *** aranjan has quit IRC | 10:53 | |
| *** kopparam has quit IRC | 10:53 | |
| ZZelle | yfried, hi | 10:53 |
| ilyashakhat_ | jschwarz: hi! | 10:53 |
| yfried | sbadia: ZZelle? | 10:53 |
| yfried | ZZelle: sorry. hi | 10:53 |
| *** aranjan has joined #openstack-neutron | 10:53 | |
| jschwarz | ilyashakhat_, can you look at https://bugs.launchpad.net/python-neutronclient/+bug/1353536 please? | 10:54 |
| *** aranjan has quit IRC | 10:54 | |
| ilyashakhat_ | jschwarz: ok, I'll take it | 10:55 |
| ZZelle | yfried, i am trying to understand how i can address your 2nd comment in https://review.openstack.org/#/c/103498/9/tempest/api/network/admin/test_provider_extension.py | 10:55 |
| jschwarz | It would appear the bug is at https://github.com/openstack/python-neutronclient/blob/master/neutronclient/neutron/v2_0/lb/healthmonitor.py L77 - the parameter 'timeout' is already defined elsewhere, causing a collision and making the parameter to always not appear | 10:55 |
| *** aranjan has joined #openstack-neutron | 10:55 | |
| ZZelle | yfried, i am not really tempest fluent | 10:55 |
| jschwarz | ilyashakhat_, actually I don't mind taking it, just wanted to make sure I found the right cause for it :) | 10:55 |
| *** ramishra has quit IRC | 10:56 | |
| *** ramishra has joined #openstack-neutron | 10:57 | |
| jschwarz | It is already defined https://github.com/openstack/python-neutronclient/blob/master/neutronclient/shell.py L597 | 10:57 |
| ilyashakhat_ | jschwarz: it was added in https://review.openstack.org/#/c/105386/ | 10:57 |
| *** sgordon_ has quit IRC | 10:58 | |
| *** reed has quit IRC | 10:58 | |
| yfried | ZZelle: did you read the bp I linked to? | 10:58 |
| jschwarz | I agree | 10:59 |
| *** claudiub has quit IRC | 11:00 | |
| ZZelle | yfried, i understand the BP but not where to put things | 11:00 |
| jschwarz | ilyashakhat_, you mind if I take the fix? I'll send you a link to the patch when it's up | 11:00 |
| *** aranjan has quit IRC | 11:00 | |
| ilyashakhat_ | jschwarz: yes, please do it | 11:01 |
| ilyashakhat_ | btw, what will be the fix? rollback of global option or rename of LB-specific? | 11:01 |
| yfried | ZZelle: let's move this to #openstack-qa | 11:01 |
| jschwarz | rename of LB-specific to something like '--monitor-timeout' and assigning it to the original variable at L93 | 11:02 |
| jschwarz | unless you have a better idea | 11:03 |
| ilyashakhat_ | the concern is about backward compatibility, because we change the cli | 11:03 |
| *** Hayim has quit IRC | 11:04 | |
| jschwarz | oh, I see.. | 11:04 |
| jschwarz | I imagine a similar solution can be implemented for the global option, though I'd like to talk to Kevin about it first | 11:05 |
| *** jschwarz is now known as jschwarz|away | 11:07 | |
| ilyashakhat_ | agree, renaming of global option looks better | 11:08 |
| *** yamamoto has joined #openstack-neutron | 11:10 | |
| *** Longgeek has quit IRC | 11:10 | |
| *** Longgeek has joined #openstack-neutron | 11:10 | |
| *** nati_ueno has joined #openstack-neutron | 11:11 | |
| *** yamahata has quit IRC | 11:11 | |
| *** GonZo2000 has joined #openstack-neutron | 11:14 | |
| *** GonZo2000 has quit IRC | 11:14 | |
| *** GonZo2000 has joined #openstack-neutron | 11:14 | |
| *** Longgeek has quit IRC | 11:14 | |
| *** nati_ueno has quit IRC | 11:15 | |
| *** vishnoianil has joined #openstack-neutron | 11:16 | |
| vishnoianil | Hi All | 11:16 |
| vishnoianil | I am trying to test FWaaS service and facing issue related to iptables rule | 11:17 |
| vishnoianil | anybody up for taking a shot at it ? | 11:17 |
| vishnoianil | I am using icehouse | 11:17 |
| *** vishnoianil is now known as avishnoi | 11:17 | |
| *** nlahouti has joined #openstack-neutron | 11:20 | |
| trinaths | avishnoi,: what is the issue you are facing | 11:21 |
| openstackgerrit | Evgeny Fedoruk proposed a change to openstack/neutron: Removing sorted() function from assertEqual() https://review.openstack.org/113494 | 11:23 |
| *** flaviof is now known as flaviof_zzz | 11:23 | |
| *** kopparam has joined #openstack-neutron | 11:24 | |
| *** nlahouti has quit IRC | 11:24 | |
| *** drankis has joined #openstack-neutron | 11:24 | |
| *** kopparam has quit IRC | 11:28 | |
| *** ramishra has quit IRC | 11:29 | |
| *** ramishra has joined #openstack-neutron | 11:30 | |
| *** Longgeek has joined #openstack-neutron | 11:34 | |
| *** ramishra has quit IRC | 11:34 | |
| *** gildub has quit IRC | 11:34 | |
| *** sgordon_ has joined #openstack-neutron | 11:35 | |
| *** sambetts has joined #openstack-neutron | 11:35 | |
| *** jab is now known as bradjones | 11:36 | |
| *** baoli has joined #openstack-neutron | 11:38 | |
| *** gildub has joined #openstack-neutron | 11:38 | |
| *** pasquier-s has joined #openstack-neutron | 11:38 | |
| *** dgollub has joined #openstack-neutron | 11:39 | |
| *** WackoRobie has joined #openstack-neutron | 11:40 | |
| *** AlexF_ has quit IRC | 11:40 | |
| openstackgerrit | Jakub Libosvar proposed a change to openstack/neutron: Import all neutron modules in autogenerate https://review.openstack.org/112928 | 11:40 |
| openstackgerrit | Jakub Libosvar proposed a change to openstack/neutron: Change autogenerate to be unconditional https://review.openstack.org/108700 | 11:40 |
| *** lordd_ has quit IRC | 11:42 | |
| openstackgerrit | YAMAMOTO Takashi proposed a change to openstack/neutron: ofagent: Implement physical_interface_mappings https://review.openstack.org/113166 | 11:45 |
| openstackgerrit | YAMAMOTO Takashi proposed a change to openstack/neutron: ofagent: Enable local arp responder for local VMs https://review.openstack.org/113160 | 11:45 |
| openstackgerrit | YAMAMOTO Takashi proposed a change to openstack/neutron: l2pop: Allow network types overridable https://review.openstack.org/112947 | 11:45 |
| openstackgerrit | YAMAMOTO Takashi proposed a change to openstack/neutron: l2pop: get_agent_ports: Don't yield (None, {}) https://review.openstack.org/113159 | 11:45 |
| openstackgerrit | YAMAMOTO Takashi proposed a change to openstack/neutron: ofagent: merge br-tun into br-int https://review.openstack.org/112488 | 11:45 |
| openstackgerrit | YAMAMOTO Takashi proposed a change to openstack/neutron: ofagent: Local arp responder for VLAN https://review.openstack.org/112949 | 11:45 |
| openstackgerrit | YAMAMOTO Takashi proposed a change to openstack/neutron: ML2: Make get_device_details report mac address as well https://review.openstack.org/112948 | 11:45 |
| *** ramishra has joined #openstack-neutron | 11:48 | |
| *** jschwarz|away is now known as jschwarz | 11:48 | |
| jschwarz | ilyashakhat_, sorry I went afk all of a sudden | 11:48 |
| jschwarz | will contact Kevin Benton and see what he says about it | 11:49 |
| ilyashakhat_ | ok | 11:49 |
| *** nlahouti has joined #openstack-neutron | 11:51 | |
| *** WackoRobie has quit IRC | 11:51 | |
| *** WackoRobie has joined #openstack-neutron | 11:52 | |
| jschwarz | kevinbenton, ping | 11:52 |
| avishnoi | trinaths: i created two private network for a tenant | 11:53 |
| avishnoi | and i am using vxlan based network | 11:53 |
| *** aranjan has joined #openstack-neutron | 11:53 | |
| avishnoi | trinaths: i created router to connect both the private network | 11:53 |
| avishnoi | trinaths: i created firewall with rule, which blocks ICMP traffic between the two VM connected to these two networks | 11:54 |
| avishnoi | trinaths: but its not blocking the ICMP traffic | 11:55 |
| *** nlahouti has quit IRC | 11:55 | |
| avishnoi | trinaths: when i dump iptables rules on the router namespace, i don't see any new rules that deny the traffic | 11:55 |
| avishnoi | trinaths: but i observer strange behaviour, when i restart l3-agent, it installs those icmp related rules in the iptables | 11:56 |
| *** diegows has joined #openstack-neutron | 11:57 | |
| *** aranjan has quit IRC | 11:57 | |
| avishnoi | trinaths: i further dig into iptables_manage.py, and it seems like we call defer_apply_on in process_router and call defer_apply_off in this function when we see if there is any external gw interface is attached to the router | 11:58 |
| *** flwang_ has quit IRC | 11:58 | |
| avishnoi | trinaths: which is not the case in my setup, so i created external network and attached it to the router and now its working absolutely fine | 11:58 |
| matrohon | yamamoto : hi | 11:59 |
| avishnoi | trinaths: so just wanted to know, if its a requirement that router need to be connected to the external network? | 11:59 |
| openstackgerrit | YAMAMOTO Takashi proposed a change to openstack/neutron: ofagent: Implement physical_interface_mappings https://review.openstack.org/113166 | 12:02 |
| openstackgerrit | YAMAMOTO Takashi proposed a change to openstack/neutron: ofagent: Local arp responder for VLAN https://review.openstack.org/112949 | 12:02 |
| *** AlexF_ has joined #openstack-neutron | 12:04 | |
| trinaths | avishnoi: Not required. but you need a router configured. | 12:05 |
| *** jgallard has joined #openstack-neutron | 12:06 | |
| *** prad has joined #openstack-neutron | 12:07 | |
| jschwarz | pcm_, ping | 12:09 |
| *** trinaths has quit IRC | 12:10 | |
| *** salv-orlando has joined #openstack-neutron | 12:10 | |
| *** banix has joined #openstack-neutron | 12:17 | |
| *** AlexF_ has quit IRC | 12:17 | |
| openstackgerrit | Evgeny Fedoruk proposed a change to openstack/neutron: Removing sorted() function from assertEqual() https://review.openstack.org/113494 | 12:18 |
| *** lordd_ has joined #openstack-neutron | 12:19 | |
| *** nlahouti has joined #openstack-neutron | 12:22 | |
| *** dbite has quit IRC | 12:24 | |
| *** kopparam has joined #openstack-neutron | 12:24 | |
| *** dbite has joined #openstack-neutron | 12:25 | |
| *** ihrachyshka_ has joined #openstack-neutron | 12:25 | |
| *** SridharG has quit IRC | 12:25 | |
| openstackgerrit | Robin Wang proposed a change to openstack/neutron: OVS-Agent crashes when OVS dead(Alternative I) https://review.openstack.org/110538 | 12:25 |
| *** nlahouti has quit IRC | 12:26 | |
| *** baoli has quit IRC | 12:26 | |
| *** ihrachyshka has quit IRC | 12:28 | |
| *** yamamoto has quit IRC | 12:31 | |
| *** swat30 has quit IRC | 12:32 | |
| *** saju_m has joined #openstack-neutron | 12:33 | |
| *** swat30 has joined #openstack-neutron | 12:33 | |
| *** ramishra has quit IRC | 12:35 | |
| *** ramishra_ has joined #openstack-neutron | 12:36 | |
| *** kopparam has quit IRC | 12:36 | |
| *** saju_m has quit IRC | 12:38 | |
| *** lordd_ has quit IRC | 12:39 | |
| *** saju_m has joined #openstack-neutron | 12:39 | |
| *** zhipeng has quit IRC | 12:43 | |
| *** Dafna has quit IRC | 12:44 | |
| pcm_ | jschwarz: pong | 12:44 |
| jschwarz | was gonna ask you if you saw the answer I wrote on your comment on my patch, but you already beat me to it :) | 12:45 |
| *** gildub has quit IRC | 12:46 | |
| *** AlexF_ has joined #openstack-neutron | 12:48 | |
| openstackgerrit | Britt Houser proposed a change to openstack/neutron: Change nexus_dict values to be array https://review.openstack.org/113009 | 12:52 |
| *** nlahouti has joined #openstack-neutron | 12:53 | |
| *** k4n0 has quit IRC | 12:53 | |
| *** aranjan has joined #openstack-neutron | 12:53 | |
| *** banix has quit IRC | 12:54 | |
| *** jistr has quit IRC | 12:56 | |
| *** rotbeard has quit IRC | 12:56 | |
| *** nlahouti has quit IRC | 12:57 | |
| *** jistr has joined #openstack-neutron | 12:57 | |
| *** heyongli has joined #openstack-neutron | 12:57 | |
| *** aranjan has quit IRC | 12:58 | |
| *** nati_ueno has joined #openstack-neutron | 12:58 | |
| *** AlexF_ has quit IRC | 13:00 | |
| *** thomasem has joined #openstack-neutron | 13:00 | |
| ZZelle | Hi, everyone | 13:00 |
| mestery | ZZelle: Hi my friend! | 13:01 |
| mestery | ZZelle: How are you doing? | 13:01 |
| *** absubram_ has joined #openstack-neutron | 13:01 | |
| ZZelle | Better than ever, summer under the rain :s | 13:01 |
| mestery | ZZelle: Ha! :) | 13:02 |
| *** HenryG_afk is now known as HenryG | 13:02 | |
| *** kbringard has joined #openstack-neutron | 13:02 | |
| ZZelle | mestery, one (culture) question, do you if there is any intentions to deprecate mlnx and bigswitch plugins? | 13:04 |
| ZZelle | mestery, it seems their ml2 mechanisms are providing the same features | 13:04 |
| *** AbsinthMind has joined #openstack-neutron | 13:04 | |
| openstackgerrit | Vinod Kumar proposed a change to openstack/neutron: Fix dvr router getting scheduled on router update https://review.openstack.org/112576 | 13:04 |
| *** julim has joined #openstack-neutron | 13:04 | |
| *** julim has quit IRC | 13:06 | |
| mestery | ZZelle: I've not heard anything yet, but perhaps | 13:07 |
| *** rpothier has joined #openstack-neutron | 13:07 | |
| ZZelle | mestery, when you look at mlnx README, it seems we should use mlnx ml2 mechanism | 13:08 |
| mestery | ZZelle: Interesting | 13:08 |
| ZZelle | mestery, let me find the link | 13:08 |
| *** WackoRobie has quit IRC | 13:08 | |
| *** WackoRobie has joined #openstack-neutron | 13:09 | |
| ZZelle | mestery, https://wiki.openstack.org/wiki/Mellanox-Neutron | 13:09 |
| *** AlexF_ has joined #openstack-neutron | 13:10 | |
| *** SridharG has joined #openstack-neutron | 13:10 | |
| ZZelle | irenab, hi | 13:11 |
| mestery | ZZelle: cool | 13:13 |
| *** WackoRobie has quit IRC | 13:13 | |
| *** shashankhegde has joined #openstack-neutron | 13:13 | |
| ZZelle | mestery, when i look at bigswitch plugin and ml2 mechanism it seems the first one can be deprecated in favor of the mechanism | 13:14 |
| mestery | ZZelle: +100 | 13:14 |
| mestery | ZZelle: Also, the Cisco standalone plugin will be deprecated this cycle as well in favor of the Nexus ML2 driver | 13:14 |
| *** lordd_ has joined #openstack-neutron | 13:14 | |
| ZZelle | mestery, sure? it seems (based on module/package names) the plugin supports more cisco appliances? | 13:15 |
| *** kbringard has quit IRC | 13:15 | |
| HenryG | ZZelle: You are correct | 13:15 |
| *** BillTheKat has joined #openstack-neutron | 13:15 | |
| *** kbringard has joined #openstack-neutron | 13:15 | |
| HenryG | ZZelle: Only the Nexus sub-plugin will be deprecated. The n1kv sub-plugin will remain. | 13:16 |
| ZZelle | HenryG, ok | 13:16 |
| ZZelle | kevinbenton, hi | 13:17 |
| *** jgallard has quit IRC | 13:17 | |
| *** phil_h has joined #openstack-neutron | 13:18 | |
| openstackgerrit | Avishay Balderman proposed a change to openstack/neutron: Plugin/DB additions for version 2 of LBaaS API https://review.openstack.org/105609 | 13:19 |
| openstackgerrit | Avishay Balderman proposed a change to openstack/neutron: Tests for extension, db and plugin for LBaaS V2 https://review.openstack.org/105610 | 13:19 |
| openstackgerrit | Avishay Balderman proposed a change to openstack/neutron: New extension for version 2 of LBaaS API https://review.openstack.org/105331 | 13:19 |
| openstackgerrit | Avishay Balderman proposed a change to openstack/neutron: Radware LBaaS v2 driver https://review.openstack.org/113516 | 13:19 |
| *** ivar-lazzaro1 has quit IRC | 13:19 | |
| *** mrsnivvel has quit IRC | 13:21 | |
| *** changbl has quit IRC | 13:21 | |
| *** AlexF_ has quit IRC | 13:21 | |
| ZZelle | mestery, only core plugins "owners" can choose to deprecate them? | 13:22 |
| mestery | ZZelle: Yes | 13:22 |
| *** BillTheKat has quit IRC | 13:22 | |
| *** nati_ueno has quit IRC | 13:22 | |
| ZZelle | mestery, so let hope we have more owners like yamamoto! | 13:23 |
| *** nlahouti has joined #openstack-neutron | 13:23 | |
| mestery | ZZelle: ++ | 13:24 |
| *** julim has joined #openstack-neutron | 13:24 | |
| *** blogan has quit IRC | 13:27 | |
| *** ptoohill has quit IRC | 13:28 | |
| *** TrevorV has quit IRC | 13:28 | |
| *** nlahouti has quit IRC | 13:28 | |
| *** evgenyf has quit IRC | 13:29 | |
| *** jgallard has joined #openstack-neutron | 13:30 | |
| *** garywils has joined #openstack-neutron | 13:30 | |
| *** AlexF_ has joined #openstack-neutron | 13:31 | |
| irenab | ZZelle: hi | 13:31 |
| irenab | ZZelle: hi | 13:31 |
| irenab | hi | 13:32 |
| *** yamamoto has joined #openstack-neutron | 13:32 | |
| ZZelle | irenab, one question about mlnx plugin and ml2 mechanism | 13:32 |
| irenab | hi | 13:32 |
| *** baoli has joined #openstack-neutron | 13:32 | |
| *** irenab has left #openstack-neutron | 13:32 | |
| *** irenab has joined #openstack-neutron | 13:32 | |
| irenab | ZZelle:hi | 13:32 |
| *** aveiga has joined #openstack-neutron | 13:32 | |
| *** ptoohill has joined #openstack-neutron | 13:33 | |
| ZZelle | irenab, when i look at the wiki page on mlnx, it seems ml2 mechanism is the prefered way to interact with mlnx | 13:33 |
| *** blogan has joined #openstack-neutron | 13:33 | |
| ZZelle | irenab, am i right? | 13:33 |
| ZZelle | irenab, https://wiki.openstack.org/wiki/Mellanox-Neutron | 13:34 |
| *** TrevorV has joined #openstack-neutron | 13:34 | |
| irenab | ZZelle: yes. Initally in Havana there was monolitic plugin, and in Icehouse ML2 MD was added to provide same functionality | 13:34 |
| *** zhipeng has joined #openstack-neutron | 13:34 | |
| irenab | ZZelle: any specific issue that concerns you? | 13:35 |
| ZZelle | irenab, any intention to deprecate the mlnx plugin if the ml2 mechanism provides the same feature? | 13:35 |
| irenab | ZZelle: yes. Actually right now, looking into mail thread regarding the way to deprecate plugin :-) | 13:36 |
| *** yamamoto has quit IRC | 13:36 | |
| ZZelle | ok, great | 13:36 |
| ZZelle | mestery, ^ | 13:36 |
| irenab | ZZelle: had to reconnect and do not see previous content. Any concern that mestery has? | 13:37 |
| *** claudiub has joined #openstack-neutron | 13:37 | |
| *** netuser has joined #openstack-neutron | 13:39 | |
| *** TrevorV has quit IRC | 13:39 | |
| *** garywils is now known as gary | 13:39 | |
| *** gary is now known as gaz | 13:39 | |
| *** ptoohill has quit IRC | 13:39 | |
| *** gaz is now known as gazza | 13:40 | |
| *** blogan has quit IRC | 13:40 | |
| netuser | QUESTION: When I run stack.sh, in neutron only 1 network is created, no public network is created ?????? | 13:40 |
| *** reaper has joined #openstack-neutron | 13:40 | |
| *** reaper has quit IRC | 13:41 | |
| pcm_ | netuser: What's your localrc have? | 13:41 |
| netuser | disable_service n-net, enable_service q-svc, enable_service q-agt, enable_service q-dhcp, enable_service q-13, enable_service q-meta, enable_service neutron | 13:43 |
| *** rook has joined #openstack-neutron | 13:43 | |
| netuser | also | 13:43 |
| netuser | RECLONE=yes | 13:43 |
| *** ivar-lazzaro has joined #openstack-neutron | 13:43 | |
| netuser | I only see private network after running stack.sh, even though I'm supposed to have 2: private and public | 13:44 |
| *** phil_h has quit IRC | 13:45 | |
| *** crose has joined #openstack-neutron | 13:45 | |
| *** ptoohill has joined #openstack-neutron | 13:45 | |
| pcm_ | netuser: Check lib/neutron script to see what conditions are needed to create the second network. | 13:45 |
| *** blogan has joined #openstack-neutron | 13:46 | |
| pcm_ | netuser: YOu can also look at the stack.sh.log (if set) and see what nets it is creating. | 13:46 |
| *** AbsinthMind has quit IRC | 13:46 | |
| *** TrevorV has joined #openstack-neutron | 13:46 | |
| openstackgerrit | Avishay Balderman proposed a change to openstack/neutron: Plugin/DB additions for version 2 of LBaaS API https://review.openstack.org/105609 | 13:46 |
| openstackgerrit | Avishay Balderman proposed a change to openstack/neutron: Tests for extension, db and plugin for LBaaS V2 https://review.openstack.org/105610 | 13:46 |
| openstackgerrit | Avishay Balderman proposed a change to openstack/neutron: New extension for version 2 of LBaaS API https://review.openstack.org/105331 | 13:46 |
| openstackgerrit | Avishay Balderman proposed a change to openstack/neutron: Radware LBaaS v2 driver https://review.openstack.org/113525 | 13:47 |
| pcm_ | netuser: There is a create_neutron_initial_network function that creates the network/subnet | 13:48 |
| openstackgerrit | Ihar Hrachyshka proposed a change to openstack/neutron: Use jsonutils instead of stdlib json https://review.openstack.org/112708 | 13:48 |
| *** AbsinthMind has joined #openstack-neutron | 13:49 | |
| *** avishnoi has left #openstack-neutron | 13:49 | |
| *** jistr has quit IRC | 13:49 | |
| netuser | pcm_:Yeah, I'm looking at it right now | 13:49 |
| pcm_ | netuser: Should have l3 service enabled to get public net. | 13:49 |
| pcm_ | netuser: seems like you do. | 13:49 |
| *** jistr has joined #openstack-neutron | 13:50 | |
| *** irenab has quit IRC | 13:50 | |
| *** AbsinthMind has quit IRC | 13:50 | |
| *** jprovazn has joined #openstack-neutron | 13:51 | |
| netuser | pcm_: It also doesn't create br-ex bridge | 13:51 |
| *** WackoRobie has joined #openstack-neutron | 13:51 | |
| *** marun has joined #openstack-neutron | 13:51 | |
| netuser | which is for public network | 13:51 |
| pcm_ | netuser: May need 'OVS_PHYSICAL_BRIDGE=br-ex' in localrc. | 13:52 |
| *** flaviof_zzz is now known as flaviof | 13:53 | |
| *** aranjan has joined #openstack-neutron | 13:53 | |
| *** evgenyf has joined #openstack-neutron | 13:53 | |
| *** ramishra_ has quit IRC | 13:54 | |
| *** ramishra has joined #openstack-neutron | 13:54 | |
| netuser | pcm_: I ran another installation on another vm without setting OVS_PHYSICAL_BRIDGE and stack.sh created br-ex automatically | 13:54 |
| *** nlahouti has joined #openstack-neutron | 13:54 | |
| *** FJB has joined #openstack-neutron | 13:55 | |
| *** aranjan has quit IRC | 13:57 | |
| *** julim has quit IRC | 13:58 | |
| *** heyongli has quit IRC | 13:59 | |
| *** flwang_ has joined #openstack-neutron | 13:59 | |
| *** xuhanp has joined #openstack-neutron | 13:59 | |
| *** nlahouti has quit IRC | 13:59 | |
| pcm_ | netuser: Not sure if that should be the default used. You'd have to check the scripts. However, I had one system w/o br-ex and had to set that variable. | 13:59 |
| *** julim has joined #openstack-neutron | 14:01 | |
| *** ramishra has quit IRC | 14:02 | |
| *** flwang_ has quit IRC | 14:03 | |
| *** WackoRobie has quit IRC | 14:04 | |
| *** WackoRobie has joined #openstack-neutron | 14:04 | |
| openstackgerrit | Sergey Shnaidman proposed a change to openstack/neutron: Add None as one of valid ipv6 attributes https://review.openstack.org/113145 | 14:05 |
| *** afazekas has quit IRC | 14:05 | |
| pcm_ | netuser: You can post your whole localrc using paste.openstack.org so folks here can look at your config and comment. | 14:06 |
| *** banix has joined #openstack-neutron | 14:06 | |
| openstackgerrit | Sergey Shnaidman proposed a change to openstack/neutron: Add None as one of valid ipv6 attributes https://review.openstack.org/113145 | 14:06 |
| *** gazza has quit IRC | 14:08 | |
| *** armax has joined #openstack-neutron | 14:10 | |
| *** phil_h has joined #openstack-neutron | 14:13 | |
| *** hanzhou has joined #openstack-neutron | 14:14 | |
| *** jistr has quit IRC | 14:14 | |
| *** otherwiseguy has joined #openstack-neutron | 14:15 | |
| *** FJB has quit IRC | 14:15 | |
| *** jistr has joined #openstack-neutron | 14:15 | |
| *** FJB has joined #openstack-neutron | 14:19 | |
| *** FJB has quit IRC | 14:19 | |
| pcm_ | netuser: Key will be to see what the log shows and then you can do failed commands manually and see what happens. | 14:19 |
| *** FJB has joined #openstack-neutron | 14:19 | |
| *** afazekas has joined #openstack-neutron | 14:20 | |
| *** jistr has quit IRC | 14:21 | |
| *** ijw has joined #openstack-neutron | 14:21 | |
| *** jistr has joined #openstack-neutron | 14:21 | |
| *** gazza has joined #openstack-neutron | 14:22 | |
| *** jgallard has quit IRC | 14:25 | |
| *** nlahouti has joined #openstack-neutron | 14:25 | |
| *** ijw has quit IRC | 14:25 | |
| *** dfarrell07 has joined #openstack-neutron | 14:27 | |
| *** nlahouti has quit IRC | 14:29 | |
| *** tongli has joined #openstack-neutron | 14:30 | |
| *** seizadi has joined #openstack-neutron | 14:31 | |
| *** yamamoto has joined #openstack-neutron | 14:32 | |
| *** salv-orlando has quit IRC | 14:33 | |
| *** ivar-lazzaro has quit IRC | 14:35 | |
| *** Dafna has joined #openstack-neutron | 14:35 | |
| *** nlahouti has joined #openstack-neutron | 14:36 | |
| netuser | pcm:_ OK, I fixed it. | 14:36 |
| netuser | pcm:_ I have another question | 14:36 |
| pcm_ | netuser: What was the fix? | 14:37 |
| *** yamamoto has quit IRC | 14:37 | |
| netuser | pcm:_ I just ran stack.sh | 14:37 |
| netuser | and it worked for some reason | 14:38 |
| *** jprovazn has quit IRC | 14:38 | |
| netuser | pcm_: that was weird | 14:38 |
| pcm_ | netuser: yes | 14:38 |
| netuser | pcm_: I have a port group/bridge in vmware vcenter which I use for vm deployment | 14:38 |
| netuser | pcm_: how can I use it in neutron | 14:39 |
| *** mrmoje has joined #openstack-neutron | 14:39 | |
| pcm_ | netuser: Afraid I don't know about VMWare at all. | 14:39 |
| netuser | pcm_: I ran: sudo ovs-vsctl add-br br112 , sudo ovs-vsctl add-port br112 eth3 | 14:39 |
| netuser | pcm_: how can I tell neutron to use that bridge | 14:40 |
| *** networkstatic has joined #openstack-neutron | 14:40 | |
| *** nlahouti has quit IRC | 14:40 | |
| *** gazza has quit IRC | 14:40 | |
| *** SridharG has quit IRC | 14:41 | |
| pcm_ | netuser: Not really sure. Maybe specifying eth3 as PUBLIC_INTERFACE (if that is what you want to use it for)? Not really sure. I usually just do a plain setup for development. | 14:43 |
| pcm_ | netuser: You may want to ask on openstack IRC for usage Openstack questions as there may be people there better able to answer. | 14:43 |
| pcm_ | this IRC is typically for development related questions in Neutron. | 14:44 |
| *** sbfox has joined #openstack-neutron | 14:44 | |
| *** markmcclain has joined #openstack-neutron | 14:44 | |
| netuser | pcm_: I'm actually doing dev as well, I just need to setup neutron | 14:45 |
| larsks | netuser: You probably want to ask "how do i configure" questions on #openstack. | 14:45 |
| netuser | pcm_: and isn't setup a part of the devepment ? | 14:45 |
| *** vivek-ebay has joined #openstack-neutron | 14:46 | |
| larsks | Whereas this is a good place for "I am writing a neutron plugin and I need help understanding the internal api". | 14:46 |
| pcm_ | netuser: #openstack is for usage questions. #openstack-neutron is for Neutron coding questions. | 14:47 |
| *** dfarrell07 has quit IRC | 14:47 | |
| *** ollivier has quit IRC | 14:48 | |
| pcm_ | netuser: Not that people here can't answer your questions, but it's not the focus of the list. Just trying to guide you to the best place to get help. | 14:50 |
| netuser | pcm_: ok, go tit | 14:51 |
| pcm_ | netuser: I'm not very versed in the different config setups. Usually just have a basic config and then tweak it as needed once and a while. | 14:52 |
| *** AlexF_ has quit IRC | 14:52 | |
| openstackgerrit | Wei Wang proposed a change to openstack/neutron: "if dict.get(key):" over "if key in dict and dict[key]:" https://review.openstack.org/112511 | 14:53 |
| *** aranjan has joined #openstack-neutron | 14:53 | |
| *** ihrachyshka_ has quit IRC | 14:55 | |
| *** jdag_ is now known as jdag | 14:55 | |
| *** AlexF_ has joined #openstack-neutron | 14:57 | |
| *** carl_baldwin has joined #openstack-neutron | 14:58 | |
| *** nati_ueno has joined #openstack-neutron | 14:58 | |
| *** yeungp has joined #openstack-neutron | 14:58 | |
| *** sbfox has quit IRC | 14:59 | |
| *** markmcclain1 has joined #openstack-neutron | 15:00 | |
| *** mrmoje has quit IRC | 15:00 | |
| *** markmcclain has quit IRC | 15:01 | |
| *** jlibosva has quit IRC | 15:01 | |
| *** devvesa has quit IRC | 15:02 | |
| *** nati_ueno has quit IRC | 15:03 | |
| *** dfarrell07 has joined #openstack-neutron | 15:04 | |
| *** radez_g0n3 is now known as radez | 15:04 | |
| openstackgerrit | A change was merged to openstack/neutron: Add delete operations for the ODL MechanismDriver https://review.openstack.org/96459 | 15:06 |
| *** WackoRobie has quit IRC | 15:06 | |
| *** WackoRobie has joined #openstack-neutron | 15:07 | |
| *** aranjan has quit IRC | 15:07 | |
| openstackgerrit | Wei Wang proposed a change to openstack/neutron: DuplicateSecurityGroupRuleInPost should point out rule id https://review.openstack.org/113551 | 15:07 |
| *** gazza has joined #openstack-neutron | 15:08 | |
| *** afazekas has quit IRC | 15:08 | |
| *** nlahouti has joined #openstack-neutron | 15:09 | |
| *** WackoRobie has quit IRC | 15:09 | |
| *** WackoRobie has joined #openstack-neutron | 15:09 | |
| *** dbite has quit IRC | 15:12 | |
| openstackgerrit | Armando Migliaccio proposed a change to openstack/neutron: Fix AttributeError when setting external gateway on DVR router https://review.openstack.org/112146 | 15:12 |
| *** banix has quit IRC | 15:12 | |
| *** amuller_ has joined #openstack-neutron | 15:13 | |
| *** WackoRobie has quit IRC | 15:14 | |
| *** amuller has quit IRC | 15:15 | |
| *** ramishra has joined #openstack-neutron | 15:15 | |
| *** absubram_ has quit IRC | 15:15 | |
| *** xgerman has joined #openstack-neutron | 15:16 | |
| *** jorgem has joined #openstack-neutron | 15:20 | |
| *** mlavalle has joined #openstack-neutron | 15:20 | |
| *** mlavalle has quit IRC | 15:20 | |
| openstackgerrit | Miguel Angel Ajo proposed a change to openstack/neutron: Watch child processes and take action when any of them die https://review.openstack.org/112798 | 15:20 |
| *** ramishra has quit IRC | 15:21 | |
| *** ramishra has joined #openstack-neutron | 15:21 | |
| *** afazekas has joined #openstack-neutron | 15:21 | |
| *** lordd_ has quit IRC | 15:21 | |
| *** ijw has joined #openstack-neutron | 15:22 | |
| *** jckasper has quit IRC | 15:22 | |
| *** lordd_ has joined #openstack-neutron | 15:23 | |
| *** vivek-eb_ has joined #openstack-neutron | 15:24 | |
| *** changbl has joined #openstack-neutron | 15:24 | |
| *** vivek-ebay has quit IRC | 15:26 | |
| *** skolekonov has quit IRC | 15:26 | |
| *** ijw has quit IRC | 15:26 | |
| armax | enikanorov: ping | 15:27 |
| enikanorov_ | armax: pong | 15:27 |
| *** amotoki has joined #openstack-neutron | 15:27 | |
| armax | enikanorov_: re patch 113144' | 15:27 |
| armax | enikanorov_: are you still working on it? | 15:28 |
| *** xuhanp has quit IRC | 15:28 | |
| *** dbite has joined #openstack-neutron | 15:28 | |
| enikanorov_ | armax: I am. what are your suggestions? | 15:28 |
| armax | enikanorov_: did you look into Rossella’s comment further? | 15:28 |
| enikanorov_ | armax: yes, i've replied to it | 15:28 |
| *** regXboi has joined #openstack-neutron | 15:28 | |
| *** saju_m has quit IRC | 15:29 | |
| armax | enikanorov_: ok, in relation to the log noise, have you had a chance to look at the NotFound errors for the firewall? | 15:29 |
| armax | enikanorov_: I’ve been meaning to look at those for a while | 15:29 |
| enikanorov_ | no, I didn't do it for firewall. we did fix similar issue for lbaas, which ended with lowering log level for such cases | 15:30 |
| *** AbsinthMind has joined #openstack-neutron | 15:30 | |
| enikanorov_ | because that's the nature of API tests that delete objects before they get reflected on the backend | 15:31 |
| enikanorov_ | so agents try to access already-deleted objects, i guess it's the same for firewall | 15:31 |
| armax | enikanorov_: well, this is the same type of failure that bug #1355409 is about | 15:31 |
| *** jschwarz has quit IRC | 15:32 | |
| enikanorov_ | understood. it's just strange that it started to appear a not so long ago | 15:32 |
| armax | enikanorov_: you mean the firewall errors? | 15:32 |
| *** FJB has quit IRC | 15:32 | |
| *** yamamoto has joined #openstack-neutron | 15:32 | |
| enikanorov_ | Im mean PortNotFound | 15:32 |
| armax | enikanorov_: true, I tried to root cause it | 15:33 |
| armax | enikanorov_: but I couldn’t find anything obvious…that code has been like that for a while | 15:33 |
| armax | enikanorov_: perhaps it was a time bomb waiting to explode :) | 15:33 |
| enikanorov_ | maybe it's something on tempest side, i don't know | 15:34 |
| openstackgerrit | Tri Hoang Vo proposed a change to openstack/neutron: Detect if iproute2 allows delete namespace https://review.openstack.org/106365 | 15:34 |
| enikanorov_ | then it might make sense to keep those traces in the logs | 15:34 |
| enikanorov_ | armax: just to not mask the issue | 15:34 |
| *** mlavalle has joined #openstack-neutron | 15:35 | |
| *** afazekas has quit IRC | 15:35 | |
| *** rwsu has joined #openstack-neutron | 15:35 | |
| armax | enikanorov_: we should really cleaning up the logs from all the errors and start voting -1 on traces | 15:35 |
| armax | enikanorov_: *should be cleaning | 15:36 |
| *** magesh_ has joined #openstack-neutron | 15:36 | |
| enikanorov_ | armax: agree on that also | 15:37 |
| armax | enikanorov_: only then it becomes more clear if it’s a regression, or an hidden timing issue | 15:37 |
| *** yamamoto has quit IRC | 15:37 | |
| armax | enikanorov_: do you know if there’s anyone working on that? | 15:37 |
| *** magesh_ has quit IRC | 15:38 | |
| enikanorov_ | armax: no, i don't know anybody looking into that except us | 15:39 |
| armax | enikanorov_: ok, I guess that would be us then :) | 15:40 |
| *** hanzhou has quit IRC | 15:41 | |
| *** ramishra_ has joined #openstack-neutron | 15:48 | |
| *** ramishra_ has quit IRC | 15:48 | |
| *** shashankhegde has quit IRC | 15:48 | |
| *** ramishra_ has joined #openstack-neutron | 15:48 | |
| *** FJB has joined #openstack-neutron | 15:49 | |
| *** ramishra has quit IRC | 15:51 | |
| *** jgallard has joined #openstack-neutron | 15:53 | |
| *** aranjan has joined #openstack-neutron | 15:53 | |
| *** ivar-lazzaro has joined #openstack-neutron | 15:54 | |
| *** reed has joined #openstack-neutron | 15:54 | |
| *** emagana has joined #openstack-neutron | 15:55 | |
| *** ihrachyshka has joined #openstack-neutron | 15:55 | |
| *** dgollub has quit IRC | 15:55 | |
| *** annegent_ has joined #openstack-neutron | 15:55 | |
| *** aranjan has quit IRC | 15:58 | |
| *** matrohon has quit IRC | 15:59 | |
| *** banix has joined #openstack-neutron | 15:59 | |
| *** nati_ueno has joined #openstack-neutron | 15:59 | |
| *** flwang_ has joined #openstack-neutron | 16:00 | |
| mestery | amuller_: Does this one (https://blueprints.launchpad.net/neutron/+spec/tenant-delete) still have a chance for Juno? Will you push code by FPF next Thursday? | 16:01 |
| amuller_ | mestery: Good question... I've been working on the VRRP spec very intensely in the past few weeks | 16:02 |
| openstackgerrit | Elena Ezhova proposed a change to openstack/neutron: Send HTTP exceptions in the format expected by neutronclient https://review.openstack.org/113570 | 16:03 |
| *** jgallard has quit IRC | 16:03 | |
| *** a_le has joined #openstack-neutron | 16:03 | |
| *** nati_ueno has quit IRC | 16:03 | |
| amuller_ | mestery: I have code up, one patch is missing. I'm going to try for Juno. | 16:04 |
| mestery | amuller_: Awesome! | 16:04 |
| *** flwang_ has quit IRC | 16:04 | |
| *** jckasper has joined #openstack-neutron | 16:04 | |
| *** ihrachyshka has quit IRC | 16:06 | |
| *** tongli has quit IRC | 16:09 | |
| *** vivek-ebay has joined #openstack-neutron | 16:10 | |
| *** seizadi has quit IRC | 16:10 | |
| *** _cjones_ has joined #openstack-neutron | 16:11 | |
| *** ajo__ has joined #openstack-neutron | 16:12 | |
| *** yfried_ has joined #openstack-neutron | 16:13 | |
| *** sbfox has joined #openstack-neutron | 16:13 | |
| *** vivek-eb_ has quit IRC | 16:14 | |
| *** yfried has quit IRC | 16:14 | |
| *** bvandenh has quit IRC | 16:14 | |
| *** vivek-ebay has quit IRC | 16:14 | |
| *** evgenyf has quit IRC | 16:15 | |
| *** amuller__ has joined #openstack-neutron | 16:15 | |
| *** yfried__ has joined #openstack-neutron | 16:16 | |
| *** ajo_ has quit IRC | 16:16 | |
| *** FJB has quit IRC | 16:16 | |
| *** SridharRamaswamy has joined #openstack-neutron | 16:16 | |
| *** WackoRobie has joined #openstack-neutron | 16:17 | |
| *** pasquier-s has quit IRC | 16:17 | |
| *** ajo_ has joined #openstack-neutron | 16:18 | |
| *** amuller_ has quit IRC | 16:19 | |
| openstackgerrit | Wei Wang proposed a change to openstack/neutron: SecurityGroupRuleExists should point out rule id inseand of group id https://review.openstack.org/113551 | 16:19 |
| *** yfried_ has quit IRC | 16:20 | |
| *** ajo__ has quit IRC | 16:20 | |
| *** yfried__ has quit IRC | 16:22 | |
| *** shashankhegde has joined #openstack-neutron | 16:22 | |
| *** hajay has joined #openstack-neutron | 16:23 | |
| *** bjornar_ has joined #openstack-neutron | 16:23 | |
| *** sbalukoff has quit IRC | 16:25 | |
| openstackgerrit | stephen-ma proposed a change to openstack/neutron: Delete DVR namespaces on node after removing last VM https://review.openstack.org/112465 | 16:25 |
| *** hajay has left #openstack-neutron | 16:25 | |
| *** jistr has quit IRC | 16:25 | |
| openstackgerrit | Armando Migliaccio proposed a change to openstack/neutron: Fix KeyError during sync_routers https://review.openstack.org/113580 | 16:28 |
| *** pballand has joined #openstack-neutron | 16:28 | |
| *** yamamoto has joined #openstack-neutron | 16:32 | |
| *** shashankhegde has quit IRC | 16:35 | |
| *** chandankumar has quit IRC | 16:36 | |
| *** AlexF_ has quit IRC | 16:37 | |
| *** yamamoto has quit IRC | 16:37 | |
| *** vivek-ebay has joined #openstack-neutron | 16:40 | |
| diegows | Any good presentation about neutron to recommend? We are going to talk about it in OpenStack Latinamerica hangout today and I don't have time to reinvent the wheel. Something short just to start the discussion, english, spanish, doesn't matter | 16:41 |
| diegows | I mean, any good slide :) | 16:41 |
| diegows | I'll keep the credits to the origin author of course | 16:41 |
| *** ramishra_ has quit IRC | 16:41 | |
| *** markmcclain1 has quit IRC | 16:42 | |
| EmilienM | diegows: this one is good: http://www.slideshare.net/enovance/mc-clainopenstack-in-action | 16:43 |
| *** sbfox has quit IRC | 16:45 | |
| *** thomasem_ has joined #openstack-neutron | 16:46 | |
| diegows | EmilienM, looks good | 16:46 |
| diegows | thanks | 16:46 |
| *** thomasem has quit IRC | 16:47 | |
| *** banix has quit IRC | 16:47 | |
| *** banix has joined #openstack-neutron | 16:48 | |
| *** aranjan has joined #openstack-neutron | 16:48 | |
| openstackgerrit | Carl Baldwin proposed a change to openstack/neutron: Preserve link local IP allocations for DVR fip ns across restart https://review.openstack.org/109486 | 16:48 |
| *** ramishra has joined #openstack-neutron | 16:48 | |
| *** dsneddon has joined #openstack-neutron | 16:49 | |
| *** seizadi has joined #openstack-neutron | 16:49 | |
| anteaya | so actually I think mestery should weigh in on this thread first emagana: http://lists.openstack.org/pipermail/openstack-dev/2014-August/042746.html | 16:50 |
| *** seizadi has quit IRC | 16:50 | |
| * mestery looks | 16:50 | |
| *** thomasem_ has quit IRC | 16:50 | |
| *** ramishra_ has joined #openstack-neutron | 16:50 | |
| emagana | anteaya: Thanks for following up | 16:50 |
| *** ramishra_ has quit IRC | 16:51 | |
| emagana | mestery: I believe we are missing the way we "audit" plugins/drivers CI | 16:51 |
| *** oreillyd has quit IRC | 16:51 | |
| *** ramishra_ has joined #openstack-neutron | 16:51 | |
| anteaya | emagana: good point | 16:51 |
| mestery | emagana: I'm looking for someone to lead that role, interested? :) | 16:51 |
| mestery | It's a full time job. | 16:51 |
| mestery | Ideally we want it automated | 16:52 |
| emagana | anteaya: Is there any process in place in Infra for validation of CIs? | 16:52 |
| anteaya | no | 16:52 |
| anteaya | actually it is a known pain point | 16:52 |
| emagana | mestery: If you want me to not have any more friends in the community, sure! | 16:52 |
| anteaya | and something we need some resources on | 16:52 |
| anteaya | while automating is a direction | 16:52 |
| mestery | emagana: hahahahahahahaha | 16:52 |
| emagana | mestery: ha ha ha ha | 16:52 |
| anteaya | we need to agree on what we are automating | 16:52 |
| emagana | mestery: Let's start for "manual" validation, once we have a process in place we can propose an automation | 16:53 |
| anteaya | so it will be largely manual for a long while until we even have an agreed upon set of data that we can collect and evaluate to demonstrate "validation" | 16:53 |
| anteaya | emagana: jinx | 16:53 |
| emagana | anteaya: same thought! | 16:53 |
| mestery | Works for me | 16:53 |
| anteaya | :D | 16:53 |
| anteaya | so are you willing to take this on? | 16:53 |
| anteaya | do you have time with documentation? | 16:54 |
| anteaya | I want to work with you but documentation can't suffer as a result | 16:54 |
| *** ramishra has quit IRC | 16:54 | |
| emagana | anteaya: Documentation is my priority right now, specially for this cycle because Neutron will sufeer a lot of changes | 16:54 |
| emagana | However, I am not the writer most of the time, the great docs team help me to do that | 16:55 |
| *** AbsinthMind has quit IRC | 16:55 | |
| emagana | I mainly help the Docs team to identidy what needs to be changed what is the best way to add the new content | 16:56 |
| *** SumitNaiksatam has quit IRC | 16:56 | |
| emagana | anteaya & mestery: I can start reviewing what plugins/drivers we have in the tree and create a matrix (if it does not exist already) that should be reviewed weekly | 16:57 |
| mestery | emagana: We have that matrix, I can point you at it. | 16:57 |
| emagana | I am ware of the Drivelog | 16:57 |
| mestery | emagana: Before you duplicate any work, lets syncup on this. :) | 16:57 |
| mestery | emagana: DriverLog is anteaya' | 16:57 |
| mestery | favorites thing | 16:57 |
| mestery | ^^^ | 16:57 |
| mestery | :) | 16:57 |
| emagana | I thought DriveLog was a Mirantis activity :-( | 16:58 |
| emagana | My bad! | 16:58 |
| *** ihrachyshka has joined #openstack-neutron | 16:58 | |
| *** a_le has quit IRC | 16:58 | |
| *** amuller__ has quit IRC | 16:58 | |
| *** SridharG has joined #openstack-neutron | 16:59 | |
| mestery | I think the issues with DriverLog are around it's usage of certified, etc. | 16:59 |
| *** amuller__ has joined #openstack-neutron | 16:59 | |
| mestery | There's a thread on that, not sure how it ended up though. | 16:59 |
| *** hajay__ has joined #openstack-neutron | 16:59 | |
| *** a_le has joined #openstack-neutron | 16:59 | |
| *** hajay__ has quit IRC | 17:00 | |
| *** arosen has joined #openstack-neutron | 17:00 | |
| *** nati_ueno has joined #openstack-neutron | 17:00 | |
| emagana | mestery: I can weekly run a validation of the plugins/drivers | 17:00 |
| *** hajay__ has joined #openstack-neutron | 17:00 | |
| *** jpich has quit IRC | 17:00 | |
| emagana | I can report what is missing and failing or something weird.. | 17:01 |
| mestery | emagana: That's what the weekly third party meeting does already, it would be good to coordinate there. | 17:01 |
| mestery | Make sense? | 17:01 |
| mestery | That would be helpful though! | 17:01 |
| mestery | I did that for a while, but my plate is constantly overfilled and things fall off all the time | 17:01 |
| emagana | mestery: Understood! | 17:02 |
| *** mestery has quit IRC | 17:02 | |
| *** WackoRobie has quit IRC | 17:02 | |
| *** magesh_ has joined #openstack-neutron | 17:02 | |
| *** hajay__ has left #openstack-neutron | 17:02 | |
| *** mestery has joined #openstack-neutron | 17:03 | |
| emagana | So far, it is just you coordinating that meeting, sorry I have not attend it lately | 17:03 |
| *** WackoRobie has joined #openstack-neutron | 17:03 | |
| *** thomasem has joined #openstack-neutron | 17:03 | |
| emagana | happy to not support any plugin/driver anymore! | 17:03 |
| emagana | :-) | 17:03 |
| *** shashankhegde has joined #openstack-neutron | 17:04 | |
| *** ijw has joined #openstack-neutron | 17:04 | |
| *** rudrajit has joined #openstack-neutron | 17:05 | |
| *** nati_ueno has quit IRC | 17:05 | |
| *** seizadi has joined #openstack-neutron | 17:06 | |
| *** a_le has quit IRC | 17:06 | |
| *** WackoRobie has quit IRC | 17:07 | |
| *** a_le has joined #openstack-neutron | 17:07 | |
| *** jschwarz has joined #openstack-neutron | 17:07 | |
| *** ianw has quit IRC | 17:08 | |
| emagana | mestery: This is the meeting right: https://wiki.openstack.org/wiki/Meetings#Third_Party_Meeting | 17:08 |
| emagana | mestery: Is there one specific for Neutron? I remember you started one | 17:08 |
| *** SridharG has quit IRC | 17:09 | |
| mestery | emagana: Nope, that's the one which is active | 17:12 |
| *** SumitNaiksatam has joined #openstack-neutron | 17:12 | |
| emagana | mestery: good! | 17:13 |
| *** bjornar_ has quit IRC | 17:13 | |
| anteaya | great | 17:14 |
| anteaya | so yes a weekly update on plugin/driver data would be great | 17:14 |
| *** claudiub has quit IRC | 17:14 | |
| anteaya | if you just start by gathering any data and displaying it | 17:14 |
| *** kaiserpathos has joined #openstack-neutron | 17:14 | |
| anteaya | we can iterate on how to present it for consumption | 17:14 |
| anteaya | and then we can figure out what the data means | 17:15 |
| anteaya | and when a data point is considered actionable | 17:15 |
| anteaya | emagana: is that fair? | 17:15 |
| *** krtaylor has quit IRC | 17:15 | |
| emagana | anteaya: it sounds good to me! | 17:15 |
| emagana | I will have the report ready for the meeting | 17:15 |
| anteaya | emagana: anything you start with will be great | 17:16 |
| anteaya | just gather something and we can go from there | 17:16 |
| anteaya | and thank you | 17:16 |
| emagana | anteaya: perfect! | 17:16 |
| anteaya | thank you! | 17:17 |
| * anteaya does a happy dance | 17:17 | |
| emagana | anteaya: On a very different topic, do you know if the multinode spec has been delayed? | 17:17 |
| emagana | anteaya: I dont see any reviews on it besides Carl and Rosella's ones | 17:17 |
| anteaya | ah | 17:17 |
| anteaya | okay so some info you probably don't have | 17:17 |
| anteaya | you did a great job on the spec in MN and I thank you | 17:18 |
| anteaya | I was wrong | 17:18 |
| anteaya | since I thought it was a seed spec for work on mulinode in germany | 17:18 |
| anteaya | and it wasn't | 17:18 |
| anteaya | so my fault there | 17:18 |
| anteaya | sdague and attila worked on it in germany | 17:18 |
| anteaya | and as far as I know multinode capability is available | 17:19 |
| anteaya | so today is tuesday | 17:19 |
| anteaya | so show up at the infra meeting today | 17:19 |
| anteaya | and you can add a discussion item to today's agenda | 17:19 |
| anteaya | or hope for some open discussion time | 17:19 |
| anteaya | also I suggest attending the qa meeting on thursday | 17:20 |
| anteaya | emagana: sorry you didn't have this info prior to now | 17:20 |
| emagana | anteaya: n.p. | 17:21 |
| *** igordcard has joined #openstack-neutron | 17:21 | |
| anteaya | so does that help? | 17:21 |
| emagana | So, this spec is not required anymore? Is that correct? | 17:21 |
| anteaya | actually come to the infra meeting today | 17:21 |
| anteaya | let's get some clarity on that | 17:21 |
| emagana | ok.. let me chack the time.. | 17:21 |
| emagana | check* | 17:21 |
| anteaya | clarkb has some thoughts and I would like you to hear from him | 17:21 |
| emagana | anteaya: Its a 1900 UTC right? | 17:22 |
| * anteaya checks too | 17:22 | |
| anteaya | 1900 yes: https://wiki.openstack.org/wiki/Meetings/InfraTeamMeeting | 17:22 |
| *** ijw_ has joined #openstack-neutron | 17:24 | |
| mtreinish | armax: hey on https://review.openstack.org/#/c/113414/ if you update the commit msg I'll give it a +2 right now | 17:24 |
| armax | mtreinish: yes, sir, right on it | 17:25 |
| openstackgerrit | Rossella Sblendido proposed a change to openstack/neutron: Remove SELECT FOR UPDATE from db_base_plugin https://review.openstack.org/100963 | 17:25 |
| mtreinish | armax: cool, thanks | 17:25 |
| armax | mtreinish: my bad for not being more thorough | 17:26 |
| *** bjornar_ has joined #openstack-neutron | 17:26 | |
| *** SridharRamaswamy has quit IRC | 17:26 | |
| *** aranjan_ has joined #openstack-neutron | 17:26 | |
| *** SridharRamaswamy has joined #openstack-neutron | 17:27 | |
| *** AbsinthMind has joined #openstack-neutron | 17:28 | |
| *** ijw_ has quit IRC | 17:28 | |
| mtreinish | armax: no worries, I'm equally guilty of nondescript commit msgs | 17:28 |
| mtreinish | armax: on that second patch it needs a rebase | 17:28 |
| *** SridharR_ has joined #openstack-neutron | 17:28 | |
| armax | mtreinish: it merges fine though | 17:28 |
| mtreinish | it's based on an old rev of the todo removal patch, you just need to push the rebase button | 17:29 |
| armax | mtreinish: I’ll rebase it one more time once I am done with the commit message…making sure I don’t make grammar mistakes or typos | 17:29 |
| armax | :) | 17:29 |
| *** aranjan has quit IRC | 17:30 | |
| *** aranjan_ has quit IRC | 17:31 | |
| *** SridharRamaswamy has quit IRC | 17:31 | |
| *** yamamoto has joined #openstack-neutron | 17:32 | |
| *** dsneddon has quit IRC | 17:32 | |
| armax | mtreinish: done | 17:33 |
| *** ramishra has joined #openstack-neutron | 17:33 | |
| armax | mtreinish: thanks again | 17:33 |
| *** yfried__ has joined #openstack-neutron | 17:36 | |
| *** absubram_ has joined #openstack-neutron | 17:36 | |
| *** ramishra_ has quit IRC | 17:36 | |
| *** emagana has quit IRC | 17:36 | |
| *** yamamoto has quit IRC | 17:37 | |
| *** emagana has joined #openstack-neutron | 17:37 | |
| mtreinish | armax: sure np. I'm glad to remove that bit. IIRC I was opposed to adding it in the first place :) | 17:38 |
| armax | mtreinish: I’ll keep an eye if anything regresses | 17:38 |
| armax | mtreinish: but as far as I can tell if the cleanup is not there, code would go astray, it’d be good if mlavalle could chime in too | 17:39 |
| *** salv-orlando has joined #openstack-neutron | 17:40 | |
| *** ramishra_ has joined #openstack-neutron | 17:40 | |
| mtreinish | thanks, but I doubt it will, the bug which caused us to add that was fixed a long time ago. I think someone else would have noticed if it were still a problem | 17:40 |
| mtreinish | armax: yeah getting his input would be good | 17:40 |
| *** SayajiPatil has joined #openstack-neutron | 17:42 | |
| *** ramishra has quit IRC | 17:42 | |
| *** sbalukoff has joined #openstack-neutron | 17:47 | |
| *** flwang_ has joined #openstack-neutron | 17:48 | |
| *** ramishra_ has quit IRC | 17:49 | |
| *** vishwanathj has joined #openstack-neutron | 17:49 | |
| *** spandhe_ has joined #openstack-neutron | 17:50 | |
| *** flwang_ has quit IRC | 17:52 | |
| *** kaiserpathos has quit IRC | 17:53 | |
| *** padkrish has joined #openstack-neutron | 17:53 | |
| *** dbite has quit IRC | 17:56 | |
| openstackgerrit | Armando Migliaccio proposed a change to openstack/neutron: Fix PortNotFound exception during sync_routers https://review.openstack.org/113441 | 17:58 |
| *** mestery has quit IRC | 18:00 | |
| *** nati_ueno has joined #openstack-neutron | 18:01 | |
| *** VijayB has joined #openstack-neutron | 18:01 | |
| *** aveiga has quit IRC | 18:02 | |
| *** amuller__ has quit IRC | 18:03 | |
| *** jschwarz has quit IRC | 18:03 | |
| *** thomasem has quit IRC | 18:04 | |
| *** jckasper has quit IRC | 18:04 | |
| *** jschwarz has joined #openstack-neutron | 18:04 | |
| *** banix has quit IRC | 18:05 | |
| *** garyduan has joined #openstack-neutron | 18:05 | |
| *** nati_ueno has quit IRC | 18:05 | |
| *** jckasper has joined #openstack-neutron | 18:06 | |
| *** gduan has quit IRC | 18:06 | |
| *** jschwarz has quit IRC | 18:07 | |
| *** garyk1 has quit IRC | 18:08 | |
| *** otherwiseguy has quit IRC | 18:08 | |
| *** thomasem has joined #openstack-neutron | 18:09 | |
| *** AbsinthMind has quit IRC | 18:10 | |
| *** BillTheKat has joined #openstack-neutron | 18:10 | |
| *** AbsinthMind has joined #openstack-neutron | 18:11 | |
| *** k_stev has joined #openstack-neutron | 18:11 | |
| *** sbfox has joined #openstack-neutron | 18:12 | |
| *** AbsinthMind1 has joined #openstack-neutron | 18:13 | |
| *** gazza has quit IRC | 18:15 | |
| *** sbfox1 has joined #openstack-neutron | 18:15 | |
| *** AbsinthMind has quit IRC | 18:15 | |
| *** sbfox has quit IRC | 18:17 | |
| *** ajo__ has joined #openstack-neutron | 18:17 | |
| *** _cjones_ has quit IRC | 18:18 | |
| *** _cjones_ has joined #openstack-neutron | 18:19 | |
| *** padkrish has quit IRC | 18:19 | |
| *** krtaylor has joined #openstack-neutron | 18:19 | |
| *** padkrish has joined #openstack-neutron | 18:20 | |
| *** ajo_ has quit IRC | 18:20 | |
| openstackgerrit | Rudra Rugge proposed a change to openstack/neutron: Opencontrail plug-in implementation for core resources https://review.openstack.org/96630 | 18:20 |
| *** padkrish_ has joined #openstack-neutron | 18:20 | |
| *** marun has quit IRC | 18:21 | |
| *** marun has joined #openstack-neutron | 18:21 | |
| *** markmcclain has joined #openstack-neutron | 18:21 | |
| *** a_le has quit IRC | 18:23 | |
| *** a_le has joined #openstack-neutron | 18:23 | |
| *** nlahouti has quit IRC | 18:24 | |
| *** padkrish has quit IRC | 18:24 | |
| *** AbsinthMind1 has quit IRC | 18:27 | |
| *** a_le has quit IRC | 18:28 | |
| *** nlahouti has joined #openstack-neutron | 18:28 | |
| *** a_le has joined #openstack-neutron | 18:29 | |
| *** ajo_ has joined #openstack-neutron | 18:30 | |
| *** spandhe_ has quit IRC | 18:31 | |
| *** a_le has quit IRC | 18:31 | |
| *** ajo__ has quit IRC | 18:31 | |
| *** a_le has joined #openstack-neutron | 18:32 | |
| *** padkrish_ has quit IRC | 18:32 | |
| *** Longgeek has quit IRC | 18:32 | |
| *** yamamoto has joined #openstack-neutron | 18:32 | |
| *** crc32 has joined #openstack-neutron | 18:32 | |
| *** padkrish has joined #openstack-neutron | 18:32 | |
| *** crc32 has quit IRC | 18:33 | |
| *** crc32 has joined #openstack-neutron | 18:34 | |
| *** dbite has joined #openstack-neutron | 18:34 | |
| *** vivek-eb_ has joined #openstack-neutron | 18:34 | |
| *** dsneddon has joined #openstack-neutron | 18:35 | |
| *** vivek-eb_ has quit IRC | 18:35 | |
| *** dconde has joined #openstack-neutron | 18:35 | |
| *** spandhe_ has joined #openstack-neutron | 18:36 | |
| *** vivek-eb_ has joined #openstack-neutron | 18:36 | |
| *** rudrarugge has joined #openstack-neutron | 18:36 | |
| *** yamamoto has quit IRC | 18:36 | |
| openstackgerrit | Swaminathan Vasudevan proposed a change to openstack/neutron: Fix for floatingip-delete not removing fip_gw port https://review.openstack.org/111421 | 18:36 |
| *** padkrish_ has joined #openstack-neutron | 18:37 | |
| *** padkrish has quit IRC | 18:37 | |
| *** vivek-ebay has quit IRC | 18:37 | |
| *** ivar-lazzaro has quit IRC | 18:37 | |
| *** alexpilotti has quit IRC | 18:39 | |
| mlavalle | armax: did you oing me? | 18:40 |
| mlavalle | ping me? | 18:40 |
| armax | mlavalle: yeah, was looking for you to comment on: https://review.openstack.org/#/c/113414/ | 18:40 |
| openstackgerrit | Bob Melander proposed a change to openstack/neutron: Adds router service plugin for CSR1kv https://review.openstack.org/101002 | 18:41 |
| mlavalle | armax: ok, I'll take a look later today. is that ok? | 18:41 |
| armax | mlavalle: yup no worries, thanks | 18:41 |
| mlavalle | armax: :-) | 18:41 |
| *** banix has joined #openstack-neutron | 18:41 | |
| *** banix has quit IRC | 18:43 | |
| *** banix has joined #openstack-neutron | 18:44 | |
| *** networkstatic has quit IRC | 18:45 | |
| *** catohornet has joined #openstack-neutron | 18:45 | |
| *** sdmonov has joined #openstack-neutron | 18:50 | |
| *** ivar-lazzaro has joined #openstack-neutron | 18:50 | |
| *** ijw_ has joined #openstack-neutron | 18:51 | |
| *** WackoRobie has joined #openstack-neutron | 18:52 | |
| *** seizadi has quit IRC | 18:53 | |
| *** ijw has quit IRC | 18:54 | |
| *** banix has quit IRC | 18:55 | |
| *** ianw has joined #openstack-neutron | 18:58 | |
| *** vivek-eb_ has quit IRC | 18:58 | |
| *** spandhe_ has quit IRC | 18:58 | |
| *** BillTheKat has quit IRC | 18:59 | |
| *** spandhe_ has joined #openstack-neutron | 19:01 | |
| *** sambetts has quit IRC | 19:03 | |
| openstackgerrit | Marga Millet proposed a change to openstack/neutron: Raise exception for network delete with subnets presents https://review.openstack.org/111349 | 19:03 |
| *** reed has quit IRC | 19:03 | |
| netuser | How can I tell configure neutron to use a bridge ? | 19:05 |
| *** rpothier has quit IRC | 19:06 | |
| *** igordcard has quit IRC | 19:06 | |
| *** annegent_ has quit IRC | 19:10 | |
| openstackgerrit | Sam Betts proposed a change to openstack/neutron: Syncronize disable_dhcp_helper to avoid race condition https://review.openstack.org/113617 | 19:11 |
| *** jckasper_ has joined #openstack-neutron | 19:11 | |
| *** nlahouti has quit IRC | 19:12 | |
| *** netuser has quit IRC | 19:13 | |
| *** jckasper has quit IRC | 19:14 | |
| openstackgerrit | Sam Betts proposed a change to openstack/neutron: Syncronize disable_dhcp_helper to avoid race condition https://review.openstack.org/113617 | 19:15 |
| *** otherwiseguy has joined #openstack-neutron | 19:16 | |
| *** phil_h has quit IRC | 19:17 | |
| *** ihrachyshka has quit IRC | 19:21 | |
| openstackgerrit | Armando Migliaccio proposed a change to openstack/neutron: Fix KeyError during sync_routers https://review.openstack.org/113580 | 19:22 |
| openstackgerrit | Carl Baldwin proposed a change to openstack/neutron: Refactor l2_pop code to pass mac/ip info more readably https://review.openstack.org/112178 | 19:24 |
| *** ijw has joined #openstack-neutron | 19:25 | |
| *** magesh_ has quit IRC | 19:26 | |
| openstackgerrit | Armando Migliaccio proposed a change to openstack/neutron: Fix 404 error fetching metadata when using DVR https://review.openstack.org/112376 | 19:26 |
| *** mestery has joined #openstack-neutron | 19:27 | |
| openstackgerrit | Armando Migliaccio proposed a change to openstack/neutron: Fix KeyError during sync_routers https://review.openstack.org/113580 | 19:27 |
| *** ijw has quit IRC | 19:29 | |
| *** annegent_ has joined #openstack-neutron | 19:32 | |
| *** yamamoto has joined #openstack-neutron | 19:32 | |
| *** AbsinthMind has joined #openstack-neutron | 19:33 | |
| *** WackoRobie has quit IRC | 19:37 | |
| *** yamamoto has quit IRC | 19:37 | |
| *** shashankhegde has quit IRC | 19:41 | |
| *** WackoRobie has joined #openstack-neutron | 19:43 | |
| *** annegent_ has quit IRC | 19:43 | |
| *** ajo_ has quit IRC | 19:44 | |
| *** rook is now known as rook-out | 19:47 | |
| *** flwang_ has joined #openstack-neutron | 19:49 | |
| *** marun has quit IRC | 19:50 | |
| *** marun has joined #openstack-neutron | 19:50 | |
| *** dconde has quit IRC | 19:53 | |
| *** flwang_ has quit IRC | 19:53 | |
| *** SumitNaiksatam has quit IRC | 19:55 | |
| *** VijayB has quit IRC | 19:56 | |
| *** AlexF has joined #openstack-neutron | 19:57 | |
| *** shashankhegde has joined #openstack-neutron | 19:57 | |
| *** seizadi has joined #openstack-neutron | 19:58 | |
| armax | haleyb: ping | 20:00 |
| *** jckasper__ has joined #openstack-neutron | 20:01 | |
| *** gangil has joined #openstack-neutron | 20:02 | |
| *** gangil has quit IRC | 20:02 | |
| *** gangil has joined #openstack-neutron | 20:02 | |
| *** amotoki has quit IRC | 20:02 | |
| *** thomasem has quit IRC | 20:02 | |
| *** jorgem has quit IRC | 20:03 | |
| *** jckasper_ has quit IRC | 20:04 | |
| openstackgerrit | badveli_vishnuus proposed a change to openstack/neutron: Add service group as a firewall customized service https://review.openstack.org/106274 | 20:05 |
| *** sbfox1 has quit IRC | 20:05 | |
| *** dave_tucker is now known as dave_tucker_zzz | 20:06 | |
| *** saju_m has joined #openstack-neutron | 20:08 | |
| *** gildub has joined #openstack-neutron | 20:08 | |
| *** baoli has quit IRC | 20:11 | |
| *** saju_m has quit IRC | 20:13 | |
| *** rms_13_ has joined #openstack-neutron | 20:15 | |
| *** nlahouti has joined #openstack-neutron | 20:15 | |
| *** bjornar_ has quit IRC | 20:15 | |
| *** dconde has joined #openstack-neutron | 20:16 | |
| *** sbfox has joined #openstack-neutron | 20:17 | |
| *** mlavalle has quit IRC | 20:18 | |
| *** nlahouti has quit IRC | 20:19 | |
| openstackgerrit | Aaron Rosen proposed a change to openstack/neutron: NSX: update_lswitch method can add duplicate version tags https://review.openstack.org/113366 | 20:22 |
| jogo | arosen: ping re nova neutron | 20:25 |
| arosen | jogo: pong | 20:26 |
| arosen | what's up? | 20:26 |
| *** ijw has joined #openstack-neutron | 20:26 | |
| jogo | mordred had a good idea in openstack-infra | 20:26 |
| jogo | nova should have 'nova boot --gimme-floating-ip' | 20:26 |
| *** thomasem has joined #openstack-neutron | 20:26 | |
| jogo | as that stuff is tricky to get right as a user | 20:27 |
| jogo | arosen: thoughts? | 20:27 |
| *** thomasem has quit IRC | 20:27 | |
| *** magesh_ has joined #openstack-neutron | 20:27 | |
| *** thomasem has joined #openstack-neutron | 20:27 | |
| *** nlahouti has joined #openstack-neutron | 20:28 | |
| arosen | jogo: Hrm, personally I'm not a big fan of moving more orchestration into nova :/ | 20:28 |
| arosen | I think horizon should have a checkbox allocate floating and then horizon can call out to neutron to create the floatingip | 20:28 |
| *** amcrn has joined #openstack-neutron | 20:28 | |
| jogo | arosen: sure, but we awnt to support CLI users for this as well | 20:29 |
| arosen | jogo: eventually i hope that nova doesn't have to do any write operations to neutron only reads. | 20:29 |
| *** mordred has joined #openstack-neutron | 20:29 | |
| jogo | arosen: we can move things into nova client instead maybe | 20:29 |
| mordred | jogo: if it's just in teh client, then you'll still have the race condition | 20:29 |
| arosen | jogo: one sec i actually think there is a way the novaclient can kinda do this today. | 20:30 |
| arosen | mordred: i'd prefer the race condition on the client side ;) | 20:30 |
| mordred | jogo: which is that if one API call creates a floating ip, and another program says "I want to see if there are any unclaimed floating ips" | 20:30 |
| mordred | arosen: the problem we have in nodepool is that $sometimes, floating ips don't get cleaned up when we delete servers | 20:31 |
| *** emagana has quit IRC | 20:32 | |
| mordred | that means we either have to have a program that cleans that and perhaps deletes an ip that was in the middle of the create/attach logic | 20:32 |
| mordred | or we have to have create look for unused ips and then race to attach it | 20:32 |
| arosen | mordred: how are you deleting the floatingips? | 20:32 |
| mordred | arosen: we delete them in our delete server code | 20:32 |
| mordred | explicitly | 20:32 |
| *** yamamoto has joined #openstack-neutron | 20:32 | |
| *** emagana has joined #openstack-neutron | 20:32 | |
| arosen | mordred: does that call out to neutron to delete them? | 20:32 |
| mordred | arosen: https://github.com/openstack-infra/nodepool/blob/master/nodepool/provider_manager.py#L483 | 20:33 |
| mordred | right now we just use the novaclient | 20:33 |
| jogo | so mordred: a bit of backround arosen doe a lot o fwork on the neutron integration in nova | 20:34 |
| * arosen let me look at what nova does with this api call. | 20:34 | |
| mordred | jogo: awesome | 20:34 |
| arosen | mordred: does it raise an error or just not delete the floatingip? | 20:34 |
| mordred | arosen: the reason we occasionally leak floating ips is unknown | 20:35 |
| mordred | but at times, we'll wind up with a few hundred unattached floating ips | 20:35 |
| mordred | which we should never have | 20:35 |
| mordred | but, you know, clouds :) | 20:36 |
| mordred | the thing is - the _simple_ case is so hard to do and get right | 20:36 |
| mordred | which is "I want a usable server" | 20:36 |
| mordred | it takes a ton of steps | 20:36 |
| *** julim has quit IRC | 20:36 | |
| *** mlavalle has joined #openstack-neutron | 20:36 | |
| mordred | we mostly get it right - I mean, we spin up between 10k and 20k a day ... so it's obviously possible | 20:36 |
| *** emagana has quit IRC | 20:36 | |
| *** yamamoto has quit IRC | 20:37 | |
| openstackgerrit | Carl Baldwin proposed a change to openstack/neutron: Preserve link local IP allocations for DVR fip ns across restart https://review.openstack.org/109486 | 20:37 |
| *** networkstatic has joined #openstack-neutron | 20:37 | |
| jogo | salv-orlando: ^ | 20:37 |
| *** crc32 has quit IRC | 20:38 | |
| salv-orlando | jogo: are you pointing me to that patch or what mordred said? | 20:38 |
| salv-orlando | arosen, mordred, jogo: I might be a bit more informed since I’ve had previous discussion on this with potential contributors which then found me too annoying to deal with ;) | 20:38 |
| jogo | salv-orlando: the arosen mordred backlog | 20:38 |
| arosen | mordred: looking at the TaskManager in nodepool/task_manager.py does it not log the return of the commands it runs on the client side? | 20:39 |
| salv-orlando | so about 1 year ago there was somebody (Steven_weston I think) that looked at how nova implements this feature | 20:39 |
| arosen | mordred: my guess is neutron raises and error and we never successfully delete the floatingip in neutron :( | 20:39 |
| arosen | mordred: I'm not quite sure though without seeing the neutron-server logs. | 20:40 |
| mordred | arosen: nod. I can try to capture more information | 20:40 |
| mordred | oh - well, I mean, I'll never have access to those | 20:40 |
| mordred | but if it's something that would return an error through to the novaclient call, I can find and log that for sure | 20:40 |
| arosen | mordred: how does nova boot --gimme-floating-ip help with this? | 20:40 |
| *** saju_m has joined #openstack-neutron | 20:40 | |
| salv-orlando | arosen: basically leaving to either neutron or nova/neutron integration ensure that there are no leaks | 20:41 |
| *** seizadi has quit IRC | 20:41 | |
| arosen | mordred: there definitely should be some kind of error raised on the nova-client i imagine. Does nodepool retry on 500 errors? | 20:41 |
| *** seizadi has joined #openstack-neutron | 20:41 | |
| salv-orlando | yet what I’m skeptical about that is the fact that a floating IP once allocated, in neutron logic, can be easily reused if it’s not associated | 20:41 |
| mordred | arosen: it's purely going to be either moving on or erroring in that case based on exceptinos thrown by novaclient | 20:41 |
| mordred | salv-orlando: yah - that's the race condition that makes it hard to close the loop on my side | 20:42 |
| mordred | salv-orlando: without keeping a database of allocation associations or something | 20:42 |
| mordred | don't get me wrong - for the times when I want the flexibility, floating-ip concept is great ... | 20:43 |
| mordred | but many times I just wanta server with an ip, and I want the ip to go away wehn the server goes away | 20:43 |
| nlahouti | markmcclain: Hi Mark, do you have time to continue our discussion? | 20:44 |
| salv-orlando | mordred: understood. We’ve seen that requirement a lot of times. The fact is that if I get it right the floating IP which are “leaked” are not associated to any internal port. Is that correct? | 20:44 |
| mordred | yeah | 20:44 |
| mordred | salv-orlando: and then, for a human it's easy | 20:45 |
| *** gduan has joined #openstack-neutron | 20:45 | |
| mordred | but to deal with it programmatically either to reap un-associated ips, or to re-use unassociated ips on subsequent create both have race conditions | 20:45 |
| *** seizadi has quit IRC | 20:45 | |
| *** seizadi has joined #openstack-neutron | 20:45 | |
| mordred | of course, I could probably just put the re-use logic into a tight loop | 20:45 |
| *** garyduan has quit IRC | 20:46 | |
| mordred | so that create always first tries to use an existing one, and attach always handles failure by going back to the start of looking for an exisitng ip and maybe creating one if needed | 20:46 |
| mordred | but eek, you know? | 20:46 |
| salv-orlando | mordred: that would work I think. But “try and use an existing one” might not be super obvious with neutron API. Mostly because when listing floating IPs *I think* we don’t have a way to return only those not associated | 20:48 |
| arosen | mordred: yea i understand this definitely sucks. We definitely don't want the client to have to do so much work. | 20:48 |
| salv-orlando | not that this is something hard to fix | 20:48 |
| *** clarkb has joined #openstack-neutron | 20:48 | |
| arosen | salv-orlando: it's easy enough to filter on the client side if there isn't a port_id | 20:49 |
| mordred | salv-orlando: currently when I do this by hand, I run neutron floating-ip list on the commadn line and then grep for ' \- ' | 20:49 |
| salv-orlando | arosen: so we fixed the thing about not being able to filter on none attributes there? | 20:49 |
| jogo | marun: ^ backlog | 20:49 |
| *** saju_m has quit IRC | 20:49 | |
| clarkb | mordred: that doesn't work with neutron client, it works with nova client | 20:50 |
| clarkb | ebcause the two clients use two different output formats | 20:50 |
| arosen | salv-orlando: i'm just saying if we don't support filtering for listing floatingips that aren't allocated we can just have the client list all the floating-ips and they can determine on their end. | 20:50 |
| marun | Why isn't it possible to allocate a port and associate it with a floating ip in one step? | 20:50 |
| salv-orlando | arosen: sure. I was just trying to remove sources for burden on the client side. | 20:50 |
| marun | It's a common enough use-case. | 20:50 |
| arosen | salv-orlando: gotcha | 20:50 |
| *** saju_m has joined #openstack-neutron | 20:50 | |
| marun | (i mean server-side, in neutron) | 20:51 |
| arosen | marun: the problem is actually on the delete side. | 20:51 |
| marun | arosen: why? | 20:51 |
| arosen | marun: delete a port and it's floatingip | 20:51 |
| marun | arosen: and again, that isn't reasonable to do in one step? | 20:51 |
| *** ezobn__ has joined #openstack-neutron | 20:52 | |
| arosen | marun: only because our api doesn't do any kind of cascading deletes between though resources | 20:52 |
| arosen | we treat them independently; | 20:52 |
| marun | arosen: so, let's fix that. | 20:52 |
| salv-orlando | arosen: I think marun is talking about binding the life cycle of floating IP to that of the port where it’s associated | 20:52 |
| marun | salv-orlando: +1 | 20:52 |
| arosen | marun: it's possible someone doesn't want to give up the floatingip when they delete the port though. | 20:52 |
| salv-orlando | so when you delete a port you just don’t disassociate the floating IPs, you delete them too. | 20:52 |
| marun | arosen: so make it optional | 20:52 |
| marun | arosen: but still support it | 20:52 |
| arosen | marun: sure we could do that but that would be an additional extension to the api right? | 20:53 |
| salv-orlando | marun: technically yes. And an annoying one given how crappy is the ext framework in this case. | 20:53 |
| *** kbringard has quit IRC | 20:54 | |
| marun | For now an extension, sure. | 20:54 |
| salv-orlando | I think we should make that simpler by not providing the automation when one creates a port (ie: saying something like —auto-allocate-floatingips) | 20:54 |
| openstackgerrit | Magesh GV proposed a change to openstack/neutron: Group Policy API-2: Classifiers, Actions, and Rules https://review.openstack.org/101816 | 20:54 |
| marun | But so long as the cli and orchestration frameworks can take advantage, I'm not sure it matters that it's not in the core API (yet). | 20:54 |
| openstackgerrit | Magesh GV proposed a change to openstack/neutron: Group Policy Mapping Plugin-1: EP, EPG, L2 Policy, L3 Policy https://review.openstack.org/101957 | 20:54 |
| openstackgerrit | Magesh GV proposed a change to openstack/neutron: Group Policy: Resource Mapping Driver https://review.openstack.org/105272 | 20:54 |
| openstackgerrit | Magesh GV proposed a change to openstack/neutron: Group Policy Plugin-1: EP, EPG, L2 Policy, L3 Policy https://review.openstack.org/96393 | 20:54 |
| openstackgerrit | Magesh GV proposed a change to openstack/neutron: Group Policy: Implicit Policy Driver https://review.openstack.org/104727 | 20:54 |
| openstackgerrit | Magesh GV proposed a change to openstack/neutron: Group Policy DB-3: Contracts https://review.openstack.org/103755 | 20:54 |
| openstackgerrit | Magesh GV proposed a change to openstack/neutron: Group Policy API-3: Contracts https://review.openstack.org/103486 | 20:55 |
| openstackgerrit | Magesh GV proposed a change to openstack/neutron: Group Policy API-1: EP, EPG, L2 Policy, L3 Policy https://review.openstack.org/95900 | 20:55 |
| openstackgerrit | Magesh GV proposed a change to openstack/neutron: Group Policy Mapping API-1: EP, EPG, L2 Policy, L3 Policy https://review.openstack.org/100716 | 20:55 |
| openstackgerrit | Magesh GV proposed a change to openstack/neutron: Group Policy DB-2: Classifiers, Actions, and Rules https://review.openstack.org/103456 | 20:55 |
| salv-orlando | we should do that when we associate a floating ip. With an attribute called “bind” | 20:55 |
| openstackgerrit | Magesh GV proposed a change to openstack/neutron: Group Policy Plugin-3: Contracts https://review.openstack.org/103798 | 20:55 |
| marun | Huh | 20:55 |
| openstackgerrit | Magesh GV proposed a change to openstack/neutron: Group Policy DB-1: EP, EPG, L2 Policy, L3 Policy https://review.openstack.org/96050 | 20:55 |
| openstackgerrit | Magesh GV proposed a change to openstack/neutron: Group Policy Mapping DB-1: EP, EPG, L2 Policy, L3 Policy https://review.openstack.org/101795 | 20:55 |
| openstackgerrit | Magesh GV proposed a change to openstack/neutron: Group Policy Plugin-2: Classifiers, Actions, and Rules https://review.openstack.org/103465 | 20:55 |
| openstackgerrit | Magesh GV proposed a change to openstack/neutron: Group Policy: One Convergence NVSD Group Policy Driver https://review.openstack.org/113649 | 20:55 |
| marun | I think he's trying to tell us something | 20:55 |
| salv-orlando | and then you tell me I should like group policy | 20:55 |
| marun | But I'm not sure it's what he's intending. | 20:55 |
| marun | hah | 20:55 |
| salv-orlando | possibly that all these problems would be solved if we move to group policy? | 20:55 |
| marun | So I hear. | 20:55 |
| salv-orlando | where were we? | 20:56 |
| *** ijw has quit IRC | 20:56 | |
| marun | So, implement in an extension? | 20:56 |
| marun | salv-orlando: I'm not sure I understood your last point | 20:56 |
| salv-orlando | yeah but not on port | 20:56 |
| marun | re: auto-allocate-floatingips | 20:56 |
| *** ijw has joined #openstack-neutron | 20:56 | |
| salv-orlando | on floating ip association. | 20:56 |
| arosen | It's funny we have two different work flows in the api actually | 20:57 |
| arosen | One can do this: | 20:57 |
| salv-orlando | there we would pass a flag that binds the floating IP lifecycle to the port where it’s being associated | 20:57 |
| arosen | neutron port-create net1; neutron floatingip-create public; neutron floatingip-associate blah balh; | 20:57 |
| arosen | or: neutron port-create net1; neutron floatingip-create --port-id=1f12a922-e8c9-44af-960c-f0ab958cbef9 Colo-InterConnect | 20:57 |
| marun | salv-orlando: ah, gotcha. So the associate could optionally become stickier | 20:57 |
| marun | arosen: so associate on create vs associate after create | 20:58 |
| arosen | yup | 20:58 |
| arosen | my gut though tells me that the client side should be dealing with clean up here if they get an error to an API call. | 20:59 |
| *** emagana has joined #openstack-neutron | 20:59 | |
| marun | arosen: I think that might be a reasonable stop-gap | 20:59 |
| arosen | marun: what's the stop-gap? Doing it on the client side? | 21:00 |
| arosen | retrying the delete/ | 21:00 |
| marun | arosen: But I think having the option of tighter binding between ports and floating ips on the server-side should be the goal | 21:00 |
| marun | arosen: yes | 21:00 |
| marun | arosen: client-side can paper over server deficiencies, and maybe that's a good short-term goal | 21:00 |
| marun | arosen: but long-term I think fixing the server-side makes more sense | 21:01 |
| *** ijw has quit IRC | 21:01 | |
| salv-orlando | marun, arosen: yes, as long as we do not add a lot of orchestration among neutron resources - that would be error prone. | 21:01 |
| *** saju_m has quit IRC | 21:01 | |
| arosen | marun: right but i don't know if fixing the server by providing an API call to cascade the delete is the right fix. I'm guessing we'll still have the same underlying issue there somewhere for why the delete is failing. | 21:01 |
| salv-orlando | orchestration like creating a floating ip as a side effect of a port-create call | 21:01 |
| *** VijayB has joined #openstack-neutron | 21:02 | |
| salv-orlando | arosen: you are probably correct as well. Why the cascaded delete might be desirable, it makes sense to go first to the bottom of the problem is manifesting in nodepool | 21:02 |
| salv-orlando | arosen: did mordred already tell you where to find the nodepools logs? I don’t know if they’re accessible through logstash. | 21:03 |
| clarkb | they are not | 21:03 |
| mordred | they are not | 21:03 |
| clarkb | we don't put our system logs in logstash because they erally shouldn't all be public | 21:03 |
| clarkb | test logs are 100% public | 21:03 |
| arosen | i'm digging into the python-novaclient bindings looking to see if there might be another way. | 21:03 |
| clarkb | but we are happy to provide people with logs | 21:03 |
| clarkb | one thing (I don't know if mordred covered this before), but nova/neutron don't do a good job of keeping track of floating ip state aiui. So if you have a shared tenant/project/whateveritiscallednow you have to do all sorts of manual tracking yourself | 21:04 |
| clarkb | otherwise you could delete allocated floating ips because you raced the associating to a node | 21:05 |
| *** sdmonov has quit IRC | 21:05 | |
| *** jckasper_ has joined #openstack-neutron | 21:05 | |
| marun | clarkb: that sounds accurate | 21:05 |
| salv-orlando | clarkb: do you mean tracking allocation/association state through nova compute floatingips extension? | 21:06 |
| *** VijayB has quit IRC | 21:06 | |
| arosen | clarkb: i know this isn't a great solution but why not ensure the floatingip is deleted before we issue the call to delete the server. This way we would know the mapping. | 21:06 |
| clarkb | salv-orlando: I mean that floating ip delete errored a month ago. neutron and nova don't expose that info | 21:06 |
| arosen | salv-orlando: it looks like there is no way to tell which floatingips to garbage collect as they could be new ones that were just allocated | 21:07 |
| salv-orlando | clarkb: I see. that’s what arosen is trying to understand then, and why we’d like to look at the logs | 21:07 |
| clarkb | arosen: correct | 21:07 |
| *** VijayB has joined #openstack-neutron | 21:07 | |
| openstackgerrit | Kevin Benton proposed a change to openstack/neutron: Big Switch: Separate L3 functions into L3 service https://review.openstack.org/113653 | 21:08 |
| arosen | clarkb: alternatively if this api call was made we could also avoid that: neutron floatingip-create --port-id=port...id network ; | 21:08 |
| arosen | that creates the floatingip and port binding in one call to neutron | 21:08 |
| salv-orlando | arosen: putting for one second aside the fact that floating ip delete errors out and that probably goes unnoticed in nova, would you say that all floating IPs without an associate port can be garbage collected? | 21:08 |
| clarkb | arosen: does novaclient expose that? | 21:08 |
| salv-orlando | clarkb: not. | 21:09 |
| *** jckasper__ has quit IRC | 21:09 | |
| arosen | salv-orlando: no i don't think they can be garbaged collected because i believe the workflow from the nova side is create floatingip (floating ip will have no port binding) then associate | 21:09 |
| salv-orlando | becuase the nova’s floating IP extension does not do that, and the —auto-allocate-floating-ips feature is not available when using neutorn. | 21:09 |
| *** bradjones has quit IRC | 21:10 | |
| clarkb | arosen: we do delete floating ips first https://git.openstack.org/cgit/openstack-infra/nodepool/tree/nodepool/provider_manager.py#n503 | 21:10 |
| salv-orlando | arosen: ok let’s make the assumption we don’t want to change the current workflow then ;) | 21:10 |
| arosen | clarkb: i wonder if this line is raising then: https://git.openstack.org/cgit/openstack-infra/nodepool/tree/nodepool/provider_manager.py#n509 | 21:11 |
| salv-orlando | clarkb: and you do that after destroyng the server, right? | 21:11 |
| clarkb | salv-orlando: no server deletion is the end of that method | 21:11 |
| clarkb | arosen: I think the actual issue is during server creation | 21:11 |
| salv-orlando | clarkb: right sorry about that. I should have scrolled a bit more! | 21:12 |
| arosen | clarkb: interesting, so do you know if these leaked floatingips are all from failed jobs? | 21:12 |
| clarkb | arosen: I don't. last time I looked at this it apepared related to creating servers that failed. the errored server would cleanup but the floating ip wouldn't always | 21:13 |
| arosen | clarkb: The workflow for creation is: launch-server; create-floatingip; associate-with-instance? | 21:13 |
| salv-orlando | clarkb: here? https://git.openstack.org/cgit/openstack-infra/nodepool/tree/nodepool/provider_manager.py#n333 | 21:13 |
| *** vishwanathj has quit IRC | 21:13 | |
| arosen | clarkb: oh so we don't wait for the server to go active first either i guess? | 21:13 |
| arosen | before we associate it | 21:13 |
| *** vishwanathj has joined #openstack-neutron | 21:13 | |
| clarkb | salv-orlando: up a call level iirc, let me find it | 21:13 |
| clarkb | arosen: its been a while. I am refreshing my memory | 21:14 |
| arosen | clarkb: i see you do wait for it to go active here: | 21:15 |
| clarkb | arosen: https://git.openstack.org/cgit/openstack-infra/nodepool/tree/nodepool/nodepool.py#n383 | 21:15 |
| arosen | https://git.openstack.org/cgit/openstack-infra/nodepool/tree/nodepool/nodepool.py#n392 | 21:15 |
| clarkb | ya it is synchronous so maybe that wasn't it | 21:15 |
| arosen | clarkb: okay that's good. If it wasn't waiting then there could definitely be a race condition were the port wasn't created in neutron | 21:16 |
| *** dconde has quit IRC | 21:16 | |
| *** radez is now known as radez_g0n3 | 21:17 | |
| *** mlavalle has quit IRC | 21:17 | |
| *** yamamoto has joined #openstack-neutron | 21:17 | |
| salv-orlando | clarkb: let’s go back for a sec to https://git.openstack.org/cgit/openstack-infra/nodepool/tree/nodepool/provider_manager.py#n508 | 21:18 |
| *** vishwanathj has quit IRC | 21:18 | |
| salv-orlando | this spawns a task which remove (ie: disassociates) the floating ip - I think submit task does not wait for a response from the API server | 21:18 |
| *** seizadi has quit IRC | 21:19 | |
| *** seizadi has joined #openstack-neutron | 21:19 | |
| salv-orlando | then the next line tries to remove that, ans does again submittask | 21:19 |
| *** emagana has quit IRC | 21:20 | |
| openstackgerrit | A change was merged to openstack/neutron: MLNX Agent: ensure removed ports get treated on resyncs https://review.openstack.org/94888 | 21:20 |
| *** rudrarugge has quit IRC | 21:21 | |
| openstackgerrit | Chirag Shahani proposed a change to openstack/neutron: Add template attr. for subnet, router create in Nuage plugin https://review.openstack.org/112749 | 21:21 |
| salv-orlando | if the task started at line 508 did not complete yet, the one at line 509 will error out… I’m not sure however if the exception raised by the task will appear in delete_server. If it does not, then the server will be deleted, the floating IP disassociated there (if it’s still available) | 21:21 |
| salv-orlando | but the floating Ip, as a db resource will not be deleted. | 21:21 |
| *** emagana has joined #openstack-neutron | 21:21 | |
| salv-orlando | clarkb: if my reasoning if correct, that might explain the leakage - maybe | 21:21 |
| *** sbfox has quit IRC | 21:22 | |
| *** yamamoto has quit IRC | 21:24 | |
| *** yfried__ has quit IRC | 21:25 | |
| *** WackoRobie has quit IRC | 21:27 | |
| *** WackoRobie has joined #openstack-neutron | 21:28 | |
| *** padkrish_ has quit IRC | 21:29 | |
| nlahouti | markmcclain: ping | 21:30 |
| *** padkrish_ has joined #openstack-neutron | 21:31 | |
| *** WackoRobie has quit IRC | 21:32 | |
| *** marun has quit IRC | 21:33 | |
| *** ivar-lazzaro has quit IRC | 21:33 | |
| *** ivar-lazzaro has joined #openstack-neutron | 21:34 | |
| salv-orlando | nlahouti: sorry I’m having a delay reviewing your patch as I had to deal with a few higher priority tasks. Will try and sort this review out tomorrow. | 21:34 |
| nlahouti | salv-orlando: really appreciate it | 21:34 |
| *** crc32 has joined #openstack-neutron | 21:36 | |
| clarkb | salv-orlando: sorry grabbed lunch, reading now | 21:42 |
| salv-orlando | clarkb: stil here | 21:42 |
| arosen | clarkb: salv-orlando https://review.openstack.org/113659 | 21:43 |
| clarkb | arosen: I am not sure that is true via novaclient? at least maybe not how we use it? | 21:43 |
| clarkb | arosen: oh wait its just the association that is removed | 21:43 |
| clarkb | I see | 21:43 |
| salv-orlando | arosen: you’re the usual commit-grabber. | 21:44 |
| clarkb | arosen: salv-orlando 113659 can still fail like the situation salv-orlando described? | 21:44 |
| arosen | clarkb: i tested it locally http://paste.openstack.org/show/94049/ | 21:44 |
| salv-orlando | clarkb, arosen: I just need to check one thing in niova’s floating ip extension | 21:45 |
| salv-orlando | hold on | 21:45 |
| ZZelle | kevinbenton, hi | 21:45 |
| clarkb | salv-orlando: ya ok, tasks are submitted and processed in order, but if one failed asynchronously the other would charge ahead and fail too | 21:45 |
| salv-orlando | arosen, clarkb: cool. Delete includes disassociate | 21:45 |
| kevinbenton | ZZelle: hey, what’s up? | 21:45 |
| salv-orlando | arosen’s patch should fix the potential race | 21:46 |
| arosen | salv-orlando: yup my pasted shows that :) | 21:46 |
| ZZelle | kevinbenton, one question about bigswitch plugin/ml2 mechanism | 21:46 |
| salv-orlando | arosen: I did not even bother looking at that… you could have fabricated it ;) | 21:46 |
| arosen | :) | 21:46 |
| kevinbenton | ZZelle: shoot | 21:46 |
| ZZelle | kevinbenton, am i wrong or ML2 + bigswitch mechanism provides the same feature as bigswitch plugin | 21:46 |
| ZZelle | ? | 21:46 |
| clarkb | salv-orlando: arosen: can yo udelete the floating ip before the server is deleted? | 21:47 |
| salv-orlando | clarkb: yes you can. | 21:47 |
| clarkb | salv-orlando: arosen: it isn't an error to delete and associated floating ip? | 21:47 |
| arosen | still though 509 could raise in error i guess https://git.openstack.org/cgit/openstack-infra/nodepool/tree/nodepool/provider_manager.py#n509 it looks like retrying on that could be helpful. | 21:47 |
| kevinbenton | ZZelle: not quite. right now the Big Switch plugin assumes the controller owns the vswitch | 21:47 |
| kevinbenton | ZZelle: the ML2 driver assumes the agent configures the vswitch | 21:47 |
| salv-orlando | clarkb: you are right but if you novaclient the delete call will also try and disassociate | 21:47 |
| salv-orlando | * if you use novaclient | 21:48 |
| kevinbenton | ZZelle: we are working on converging them, this L3 separation work that I just pushed is a first step | 21:48 |
| clarkb | salv-orlando: right but the order of that code is delete floating ip, delete node | 21:48 |
| arosen | clarkb: it looks like nova lets you delete a floatingip even if it's associated to a server: see: http://paste.openstack.org/show/94049/ (but this could be fabricated ;) ) | 21:48 |
| salv-orlando | you can point them to the code ;) | 21:48 |
| clarkb | I don't think this will fix the problem | 21:48 |
| clarkb | I swear neutron has errored on me trying to delete associated items | 21:49 |
| arosen | clarkb: i agree i don't think this will really fix anything. Let me ask you this. If line 508 raises an exception does 509 still get executed? | 21:49 |
| salv-orlando | clarkb, arosen: https://git.openstack.org/cgit/openstack/nova/tree/nova/api/openstack/compute/contrib/floating_ips.py#n193 | 21:49 |
| clarkb | arosen: no | 21:49 |
| ZZelle | kevinbenton, ok just wondering if they behaves the same and if the plugin could be deprecated ==> so no | 21:49 |
| clarkb | salv-orlando: yes, but nodepool is deleting the floating ip before deleting the node | 21:50 |
| clarkb | salv-orlando: so delting floating ip will fail then the node won't be deleted | 21:50 |
| kevinbenton | ZZelle: not quite yet. possibly in Kilo though | 21:50 |
| *** flwang_ has joined #openstack-neutron | 21:50 | |
| arosen | clarkb: actually so i think this might help. Previously you were saying you would see a lot of floatingips that were not mapped to server. In this case the only way this would happen now is if there was a failure because we're deleting the floatingip directly | 21:50 |
| clarkb | really it appears that nodepool needs a floatingip table to track that state and manage associations itself then do deletions on floating ips like ti delets nodes | 21:50 |
| *** sdmonov has joined #openstack-neutron | 21:50 | |
| clarkb | I think this is the bug that mordred is trying to point out when it comes to clients needing to track all this state themselves | 21:51 |
| salv-orlando | clarkb: ok so that’s important - an error during any asynchronous task will raise an exception in delete_server preventing deletion of the node | 21:51 |
| ZZelle | kevinbenton, great | 21:51 |
| clarkb | salv-orlando: sort of, we will retry deletions so it is ok if we eventually succeed | 21:51 |
| clarkb | salv-orlando: in this case we will never succeed | 21:51 |
| clarkb | (we have to retry for other raisisn) | 21:51 |
| *** dconde has joined #openstack-neutron | 21:51 | |
| salv-orlando | clarkb, arosen: this might sound stupid at this point, but do we have a bug report where I could get more details? I don’t want to sound procedural, but I’m not understanding if we are dealing with leaked floating ips, leaked node, or both | 21:52 |
| arosen | clarkb: I think with this change you will actually now be able to garbage collect safely. | 21:53 |
| clarkb | arosen: no, I think it will prevent any garbage collection because the delete floating ip will always fail | 21:53 |
| clarkb | salv-orlando: nodepool is leaking floating ips | 21:53 |
| *** armax has quit IRC | 21:53 | |
| clarkb | salv-orlando: this is happening because nodepool does not treat floating ips as discrete resources. instead it relies on their association to nodes whcih may be deleted to find them | 21:53 |
| arosen | clarkb: you said if the delete_floatingip fails you don't continue on to delete the server? | 21:54 |
| clarkb | salv-orlando: the "bug" here is that nova and neutron should handle this for us | 21:54 |
| mordred | salv-orlando: mainly ... we're trying to be good citizens since we're such heavy openstack users | 21:54 |
| *** flwang_ has quit IRC | 21:54 | |
| clarkb | arosen: https://git.openstack.org/cgit/openstack-infra/nodepool/tree/nodepool/provider_manager.py#n509 is where we delete, I need to check if that is a synchronous or async task | 21:54 |
| mordred | salv-orlando: in this case, I'd go so far as to say that the fact that I, as an end user, have to know ANYTHING about floating ips to accomplish the simple task of "give me a server with an IP" is a leakage of an implementation detail into an interface | 21:55 |
| mordred | I understand it may be hard/impossible to fix | 21:55 |
| *** sgordon_ has quit IRC | 21:55 | |
| clarkb | arosen: but #509 will fail the first tiem it runs. best case no floating ip is deleted and nova disassociates the node from the ip | 21:55 |
| clarkb | arosen: at that point we have leaked the floating ip | 21:55 |
| clarkb | worst case that method short circuits and we never delete the node | 21:55 |
| mordred | and I don't even know if there _is_ a fix or who would own it - but I'd happily file a bug if that's helpful | 21:55 |
| *** oda-g has joined #openstack-neutron | 21:56 | |
| mordred | I similarly, as a user, almost never want to know anything about ports or routers. in fact, I do not actually want to know anything about ports or routers. | 21:57 |
| salv-orlando | clarkb: that my analysis as well, under the assumptions that the task is aunc | 21:57 |
| mordred | my neutron got messed up one time and I had to learn about them ... and it was neat that they existed! | 21:58 |
| *** jckasper_ has quit IRC | 21:58 | |
| mordred | but they don't have any bearing on me getting my workload issues taken care of day in and day out | 21:58 |
| clarkb | arosen: it looks like it is synchronous via https://git.openstack.org/cgit/openstack-infra/nodepool/tree/nodepool/task_manager.py#n43 | 21:58 |
| clarkb | so we submit the task then that thread waits for it to happen. if an exception is thrown that is reraised and nodedelete won't happen | 21:58 |
| *** ijw has joined #openstack-neutron | 21:59 | |
| salv-orlando | mordred: I *might* try and sidestep the conversation saying that then the group policy API will relieve you of that and give you the ability of handling everything with concepts such as endpoints, contracts, and policies. But I won’t dare.. | 21:59 |
| *** yamamoto has joined #openstack-neutron | 21:59 | |
| arosen | clarkb: Then do you guys manually clean up the server? | 21:59 |
| salv-orlando | mordred: your usability concerns are correct. | 21:59 |
| clarkb | arosen: no, it seems to be working now because we disassociate first | 22:00 |
| openstackgerrit | YAMAMOTO Takashi proposed a change to openstack/neutron: ofagent: Implement physical_interface_mappings https://review.openstack.org/113166 | 22:00 |
| openstackgerrit | YAMAMOTO Takashi proposed a change to openstack/neutron: ofagent: Enable local arp responder for local VMs https://review.openstack.org/113160 | 22:00 |
| openstackgerrit | YAMAMOTO Takashi proposed a change to openstack/neutron: l2pop: Allow network types overridable https://review.openstack.org/112947 | 22:00 |
| openstackgerrit | YAMAMOTO Takashi proposed a change to openstack/neutron: ofagent: merge br-tun into br-int https://review.openstack.org/112488 | 22:00 |
| openstackgerrit | YAMAMOTO Takashi proposed a change to openstack/neutron: ofagent: Local arp responder for VLAN https://review.openstack.org/112949 | 22:00 |
| clarkb | arosen: that initial disassociate may fail, but then we will try again | 22:00 |
| mordred | salv-orlando: hehe | 22:00 |
| clarkb | arosen: I think the issue with your patch is we will never try to disassociate first so we can't get beyond #509 | 22:00 |
| mordred | salv-orlando: yeah - if you told me I'd need to thin about endpoints, contracts, and policies, I would quite diretly just start using amazon | 22:00 |
| mordred | :) | 22:00 |
| *** sdmonov has quit IRC | 22:01 | |
| salv-orlando | mordred: unless you’re a commercial lawyer. Then you should be comfortable with contracts. | 22:01 |
| mordred | salv-orlando: if I was a commercial lawyer, I'd own many more private jets than I currently do | 22:01 |
| *** SridharRamaswamy has joined #openstack-neutron | 22:02 | |
| salv-orlando | anyway - how to address these concerns. I think there are 2 classes of usability issues - 1) things we do wrong in the API and should be fixed 2) things that are fine for an api that exposes “primitives” but that 90% of users won’t never be bothered with | 22:02 |
| *** amotoki has joined #openstack-neutron | 22:03 | |
| salv-orlando | for the first we’ve been looking at API changes for addressing the network/subnet dicotomy for instance, but the discussion never came to a concrete code proposal | 22:03 |
| *** Sridhar__ has joined #openstack-neutron | 22:04 | |
| salv-orlando | for the second actually, my opinion is that we should do that outside of neutron, but I’m still battling with this as on other hand I don’t want to add another API endpoint which does orchestration. To this aim, Heat would be an option, but what about situations where it’s not dpeloyed? | 22:04 |
| salv-orlando | anyway back to our problem. | 22:04 |
| emagana | mestery: It seems that you approve this one: https://blueprints.launchpad.net/neutron/+spec/bsn-l3-service-plugin | 22:04 |
| emagana | mestery: I couldn't find the spec related to this BP. | 22:05 |
| emagana | I should also ask to kevinbenton | 22:05 |
| salv-orlando | clarkb: let’s say that we have grounds to suspect there is a race there, and that we have an hypothesis that however cannot be confirmed until we are sure the tasks are async | 22:05 |
| salv-orlando | where there means delete_server in nodepool | 22:06 |
| *** SridharR_ has quit IRC | 22:06 | |
| arosen | clarkb: where do you try again? I'm not sure I see the benefit of putting tasks in the task queue if they are executed synchronously ? | 22:06 |
| *** SridharRamaswamy has quit IRC | 22:06 | |
| *** banix has joined #openstack-neutron | 22:07 | |
| arosen | (02:54:08 PM) clarkb: arosen: I think the issue with your patch is we will never try to disassociate first so we can't get beyond #509 | 22:07 |
| arosen | ^why do you say this? | 22:07 |
| clarkb | arosen: we retry with an internal "cron". we put them in a queue to do rate limiting | 22:07 |
| emagana | mestery or kevinbenton: are you around? | 22:07 |
| clarkb | arosen: you cannot sanely do async api calls to openstack | 22:07 |
| clarkb | you get ratelimited and blacklisted | 22:07 |
| *** amotoki has quit IRC | 22:08 | |
| kevinbenton | emagana: https://review.openstack.org/#/c/106269/ | 22:08 |
| emagana | kevinbenton: Thanks.. I could not find it in the BP in launchpad | 22:09 |
| arosen | clarkb: sorry one more question. So when this occurs i guess these api calls to nova eventually return 200's because nodepool is able to delete the instance? | 22:09 |
| emagana | reviewing your patch right now ;) | 22:09 |
| clarkb | arosen: yes, I believe that is true | 22:09 |
| clarkb | arosen: otherwise we would retry the failing method repeatedly | 22:10 |
| mordred | salv-orlando: yah - also, needing heat for "gimme a server that works kthxbai" is a bit of overkill | 22:10 |
| kevinbenton | emagana: ok. i updated the whiteboard on launchpad with a link | 22:10 |
| emagana | kevinbenton: great, other reviewers will need it as well. | 22:10 |
| kevinbenton | emagana: i didn’t think reviewers read the specs :-) | 22:11 |
| *** yeungp has quit IRC | 22:11 | |
| emagana | kevinbenton: I do.. I like to amke sure that the BP is covering everything in the spec | 22:12 |
| emagana | make* | 22:12 |
| *** lordd_ has quit IRC | 22:12 | |
| *** seizadi1 has joined #openstack-neutron | 22:13 | |
| *** seizadi has quit IRC | 22:13 | |
| arosen | clarkb: hrm it seems like everything should be cleaned up fine then :( | 22:13 |
| emagana | kevinbenton: I know.. I am a weird guy! | 22:14 |
| kevinbenton | emagana: that’s the dream of the spec! | 22:15 |
| nlahouti | markmcclain: ping | 22:16 |
| emagana | kevinbenton: +2 | 22:16 |
| *** lordd_ has joined #openstack-neutron | 22:17 | |
| *** emagana has quit IRC | 22:17 | |
| *** emagana has joined #openstack-neutron | 22:18 | |
| *** regXboi has quit IRC | 22:19 | |
| ZZelle | hi, some cores can have a look to https://review.openstack.org/105472 ? | 22:20 |
| *** changbl has quit IRC | 22:21 | |
| *** emagana has quit IRC | 22:22 | |
| *** armax has joined #openstack-neutron | 22:22 | |
| rms_13_ | cores, can someone take a look at https://review.openstack.org/#/c/112171/ | 22:26 |
| clarkb | arosen: salv-orlando: mordred: ok new theory, the disassociate succeeds, the delete floating ip fails, the node delete succeeds | 22:28 |
| clarkb | arosen: salv-orlando: mordred: in that situations we would never retry the floating ip delete | 22:28 |
| clarkb | arosen: so I think your suggestion to retry the floating ip delete may be the thing to do | 22:29 |
| *** Sridhar__ has quit IRC | 22:29 | |
| salv-orlando | clarkb: which is my initial one - but in order to be credible we need to be sure that a failure in delete floating ip might go unnocited | 22:29 |
| clarkb | salv-orlando: ya so let me go pull up logs | 22:29 |
| salv-orlando | and therefore the tasks started at lines 508 and 509 should be async | 22:29 |
| *** SridharRamaswamy has joined #openstack-neutron | 22:29 | |
| *** ijw has quit IRC | 22:30 | |
| arosen | clarkb: why do you say you guys aren't retrying on delete_floatingip? | 22:30 |
| *** SridharR_ has joined #openstack-neutron | 22:32 | |
| clarkb | arosen: because once the node is deleted we will never run that function for that node again | 22:32 |
| *** emagana has joined #openstack-neutron | 22:33 | |
| *** emagana has quit IRC | 22:33 | |
| *** ivar-lazzaro has quit IRC | 22:33 | |
| *** ivar-lazzaro has joined #openstack-neutron | 22:34 | |
| *** emagana has joined #openstack-neutron | 22:34 | |
| arosen | I've actually reproduced that here but i'm confused why that happends. I put a raise TypeError here: https://git.openstack.org/cgit/openstack-infra/nodepool/tree/nodepool/provider_manager.py#n164 and saw nodepool retry and retry | 22:34 |
| *** SridharRamaswamy has quit IRC | 22:34 | |
| arosen | I moved the raise TypeError to herer: https://git.openstack.org/cgit/openstack-infra/nodepool/tree/nodepool/provider_manager.py#n169 and it looks like it only tried trice. | 22:35 |
| *** sbfox has joined #openstack-neutron | 22:35 | |
| clarkb | huh | 22:35 |
| *** jaypipes has quit IRC | 22:36 | |
| arosen | clarkb: AH I completely see why this is occuring | 22:37 |
| arosen | It's getting the floatingip binding on the server from list-servers | 22:37 |
| arosen | once we disassociate it's no longer there. | 22:37 |
| clarkb | yes | 22:38 |
| arosen | so it no longer finds the floatingip it had | 22:38 |
| arosen | doh :) | 22:38 |
| clarkb | our resources are nodes not ips | 22:38 |
| clarkb | and this is the impedance mismatch we are trying to explain | 22:38 |
| clarkb | we don't care about managing floating ips as discrete things | 22:38 |
| arosen | clarkb: in that case i think my patch should completely solve this issue for you. We'll just keep retrying to delete the floatingip | 22:39 |
| lifeless | some people do sometimes | 22:39 |
| clarkb | lifeless: yes, I think having the ability to is good | 22:39 |
| lifeless | but most folk don't most of the time | 22:39 |
| clarkb | lifeless: but making the default thing the hard thing that most people don't want is a bug | 22:39 |
| lifeless | clarkb: I totally agree :) | 22:39 |
| clarkb | arosen: oh I see, except we will need two round trips to do it | 22:39 |
| lifeless | this is why I spat the dummy when nova proposed making port management for vms entirely manual | 22:39 |
| *** AlexF has quit IRC | 22:40 | |
| clarkb | arosen: actually no | 22:40 |
| clarkb | because deleteip will fail, we will then delete the node, then nothing will call deleteip again because the node is gone | 22:40 |
| arosen | clarkb: two round trips to getServerFromList? | 22:40 |
| arosen | clarkb: one sec let me test. | 22:40 |
| clarkb | arosen: your change removes line 508 so that we rely on nova to disassociate. that must happen before we delete ip | 22:41 |
| clarkb | so deleteip will throw and exception. if we ever manage to dlete teh server we will not come back and delete the ip | 22:41 |
| clarkb | I think the "proper" fix here is the thing we are tring to avoid and that is treating floating ips as a new resource that nodepool tracks | 22:41 |
| *** shashankhegde has quit IRC | 22:42 | |
| arosen | clarkb: That patch cause us to keep trying delete_floatingip | 22:43 |
| arosen | we'll never delete the server | 22:43 |
| *** banix has quit IRC | 22:44 | |
| arosen | clarkb: http://paste.openstack.org/show/94062/ | 22:44 |
| clarkb | oh right, order is important. we could put a retry loop at line 509 | 22:44 |
| *** prad has quit IRC | 22:44 | |
| arosen | clarkb: if you scroll through though logs you'll see we keep raising typerror | 22:44 |
| arosen | Man this was a tricky bug :) | 22:44 |
| clarkb | ya, and it never drops through to delete server | 22:44 |
| clarkb | so it fails again and again :) | 22:45 |
| arosen | right which is what we want so hopefully we'll eventually clean it up. | 22:45 |
| *** hemanthravi has joined #openstack-neutron | 22:46 | |
| arosen | clarkb: I'll update the commit message now that we know what's going on :) | 22:47 |
| clarkb | arosen: maybe we should go back to -infra and fill jeblair in? he may have other ideas on fixing this | 22:47 |
| clarkb | arosen: wait that commit doesn't work as is I thought? | 22:47 |
| *** carl_baldwin has quit IRC | 22:48 | |
| arosen | clarkb: no it does. I mean just to update the commit message to saw why it fixes the issue. Initially i though it just removed an unneeded call | 22:48 |
| clarkb | arosen: I don't think it does | 22:48 |
| arosen | why? | 22:48 |
| clarkb | arosen: deleteip will not succeed until after we disassociate the ip | 22:48 |
| clarkb | we will never disassociate the ip because we do not delete the node | 22:48 |
| clarkb | because the function short circuits on deleteip failing | 22:48 |
| clarkb | it is the same situation as raising your TypeError | 22:49 |
| arosen | clarkb: we'll eventually delete the node once we are able to delete the floatingip | 22:49 |
| *** stack__ has joined #openstack-neutron | 22:49 | |
| *** banix has joined #openstack-neutron | 22:49 | |
| clarkb | arosen: I don't think we can delete the floatingip until we disassocaite the node from the ip | 22:49 |
| arosen | clarkb: https://git.openstack.org/cgit/openstack/nova/tree/nova/api/openstack/compute/contrib/floating_ips.py#n193 | 22:50 |
| arosen | clarkb: i showed in my pastebin that you can ? | 22:50 |
| *** thomasem_ has joined #openstack-neutron | 22:50 | |
| arosen | clarkb: that said I think there is another bad thing that can happen here. If you look at that link i sent you it's possible for line 196 to succeed which means we'll be in the same situation :( | 22:50 |
| clarkb | oh it is the delete of the ip that disassociates | 22:51 |
| clarkb | I thought it was the delete of the node | 22:51 |
| clarkb | ok I am on board now | 22:51 |
| arosen | clarkb: the delete of the node also disassociates the floatingip but the problem is then we won't be able to find the ip anymore | 22:51 |
| clarkb | arosen: yup | 22:52 |
| clarkb | ok, that is where I was stuck. I didn't realize delete of floating ip did it too | 22:52 |
| *** shashankhegde has joined #openstack-neutron | 22:52 | |
| clarkb | in this case if it fails it will make all the deletion happen again | 22:52 |
| clarkb | and again and again so we should be good | 22:52 |
| clarkb | oh wait | 22:52 |
| clarkb | that won't fix it | 22:53 |
| clarkb | it has the same bug we have today | 22:53 |
| *** thomasem has quit IRC | 22:53 | |
| clarkb | arosen: if https://git.openstack.org/cgit/openstack/nova/tree/nova/api/openstack/compute/contrib/floating_ips.py#n196 succeeds but https://git.openstack.org/cgit/openstack/nova/tree/nova/api/openstack/compute/contrib/floating_ips.py#n203 fails we will not get the ip from the floating ip list ? | 22:53 |
| arosen | clarkb: right that was what i was saying :) | 22:53 |
| *** emagana has quit IRC | 22:53 | |
| arosen | clarkb: but interestingly enough neutron allows us to delete the floatingip even if it associated. | 22:54 |
| arosen | clarkb: so we could actually change this code to say if using neutron don't call disassociate_floating_ip | 22:54 |
| *** thomasem_ has quit IRC | 22:54 | |
| arosen | clarkb: I got to run to a 4:00 :(. I'll try and reconnect there though. | 22:54 |
| clarkb | and delete it directly? and only then delete the node? | 22:55 |
| kevinbenton | rkukura: do you have that document handy that shows what information is available during the various states of port creation/binding? | 22:55 |
| *** julim has joined #openstack-neutron | 22:56 | |
| *** jeblair has joined #openstack-neutron | 22:57 | |
| salv-orlando | clarkb: I can take it from arosen | 22:57 |
| salv-orlando | as he’s now afk | 22:57 |
| mordred | so, currently, I don't think we have any nova-network+floating-ip providers | 22:58 |
| mordred | so "if neutron: delete floating ip first" is probably fine, yeah? | 22:59 |
| *** shashankhegde has quit IRC | 22:59 | |
| clarkb | mordred: but only via the neutron client | 22:59 |
| clarkb | mordred: if you do it via nova client it will do the wrong leaky thing | 22:59 |
| clarkb | mordred: so we would need to maintain another task manager for neutron... | 22:59 |
| mordred | clarkb: well ... hrm | 23:00 |
| salv-orlando | uhm. I sense like you guys don’t want to use the neutroclient though... | 23:00 |
| mordred | salv-orlando: it's just more complexity on our side ... | 23:00 |
| clarkb | salv-orlando: well on principle no :) | 23:00 |
| mordred | what I really want is openstacksdk | 23:00 |
| clarkb | we can do it | 23:00 |
| salv-orlando | clarkb, mordred: understood | 23:00 |
| *** julim has quit IRC | 23:00 | |
| mordred | and to have that use the right tools as it sees fit | 23:00 |
| salv-orlando | I think we can fix https://git.openstack.org/cgit/openstack/nova/tree/nova/api/openstack/compute/contrib/floating_ips.py#n178 to do the right thing with neutron | 23:01 |
| *** ijw_ has quit IRC | 23:01 | |
| salv-orlando | and you can still use the novaclient | 23:01 |
| salv-orlando | and everybody will be happy | 23:01 |
| * mordred hands salv-orlando a cheesecake | 23:01 | |
| clarkb | ya so we could go kick novaclient too | 23:01 |
| salv-orlando | well not mordred as he wants the “gimme a machine with a public ip kthxbie” thing | 23:01 |
| mordred | salv-orlando: one of these days I'll get my wish there!!!! | 23:02 |
| clarkb | salv-orlando: that said, is it "correct" to delete a floating ip without disassociating it? | 23:02 |
| clarkb | to me one implies the other | 23:02 |
| clarkb | because how can you be associated with a deleted thing? | 23:02 |
| salv-orlando | clarkb: uhm… so when you delete a floating ip neutron implicitly removes any association with active ports (that’s VM for you) | 23:03 |
| clarkb | salv-orlando: doesn't noav need to know about that? | 23:04 |
| *** crose has quit IRC | 23:04 | |
| clarkb | otherwise the ip listing will be wrong | 23:04 |
| salv-orlando | clarkb: you might be right. I forgot that other thing about ip listing. | 23:04 |
| salv-orlando | that however might be a different bug... | 23:05 |
| *** networkstatic has quit IRC | 23:07 | |
| *** _cjones_ has quit IRC | 23:07 | |
| *** drjones has joined #openstack-neutron | 23:07 | |
| *** ZZelle has quit IRC | 23:08 | |
| *** banix has quit IRC | 23:10 | |
| *** dbite has quit IRC | 23:10 | |
| clarkb | salv-orlando: mordred: if we use neutron to list associations we will get teh correct mapping right? | 23:10 |
| clarkb | salv-orlando: mordred: so nodepool may just need to be careful to use neutronclient for those other things | 23:10 |
| salv-orlando | yes but in that case you will see association by ports, not vms | 23:10 |
| clarkb | oh | 23:11 |
| salv-orlando | clarkb: let’s just go a step back. If line 196 succeeds and line 203 fails then the floating IP is never removed from the nova list - causing it to appear as stale. | 23:13 |
| clarkb | salv-orlando: the association is removed so we won't find it | 23:14 |
| clarkb | salv-orlando: we are looking for IPs associated to the node that is being deleted | 23:14 |
| salv-orlando | clarkb: right however if we change the code there, as also arosen said, to do things differently with neutron, by not disassociating first, then the siingle delete operation would do both disassociate and delete - and that will come for “free”. In this way, if that fails the IP should not appear as disassociated and you should be able to find it. | 23:15 |
| *** emagana has joined #openstack-neutron | 23:15 | |
| salv-orlando | however, when then you delete the VM… this will trigger a process in neutron which disassociates the floating IP but does not delete it, so you might back at square 1 :/ | 23:15 |
| *** emagana has quit IRC | 23:20 | |
| *** AbsinthMind has quit IRC | 23:23 | |
| *** SridharRamaswamy has joined #openstack-neutron | 23:24 | |
| nlahouti | markmcclain: ping | 23:25 |
| *** sbfox has quit IRC | 23:25 | |
| *** dconde has quit IRC | 23:25 | |
| *** Sridhar__ has joined #openstack-neutron | 23:26 | |
| *** diegows has quit IRC | 23:26 | |
| *** SridharR_ has quit IRC | 23:27 | |
| *** emagana has joined #openstack-neutron | 23:28 | |
| *** SridharRamaswamy has quit IRC | 23:28 | |
| *** lordd_ has quit IRC | 23:32 | |
| *** lordd_ has joined #openstack-neutron | 23:33 | |
| *** reed has joined #openstack-neutron | 23:34 | |
| *** padkrish_ has quit IRC | 23:34 | |
| *** lordd_ has quit IRC | 23:35 | |
| *** padkrish has joined #openstack-neutron | 23:35 | |
| *** lordd_ has joined #openstack-neutron | 23:35 | |
| *** cgoncalves has quit IRC | 23:36 | |
| *** cgoncalves has joined #openstack-neutron | 23:37 | |
| *** emagana has quit IRC | 23:39 | |
| *** emagana has joined #openstack-neutron | 23:39 | |
| *** padkrish has quit IRC | 23:39 | |
| *** emagana has quit IRC | 23:42 | |
| *** emagana has joined #openstack-neutron | 23:43 | |
| openstackgerrit | Doug Wiegley proposed a change to openstack/neutron: A10 Networks LBaaS v1 Driver https://review.openstack.org/113702 | 23:44 |
| *** nati_ueno has joined #openstack-neutron | 23:45 | |
| *** Sridhar__ has quit IRC | 23:46 | |
| *** SridharRamaswamy has joined #openstack-neutron | 23:46 | |
| *** emagana has quit IRC | 23:47 | |
| *** flwang_ has joined #openstack-neutron | 23:51 | |
| *** SridharRamaswamy has quit IRC | 23:51 | |
| *** otherwiseguy has quit IRC | 23:54 | |
| *** flwang_ has quit IRC | 23:56 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!