| *** tflynn_ has quit IRC | 00:01 | |
| *** tflynn has joined #openstack-neutron | 00:01 | |
| *** merooney has quit IRC | 00:02 | |
| *** tflynn has quit IRC | 00:03 | |
| *** tflynn has joined #openstack-neutron | 00:04 | |
| *** lucas__ has joined #openstack-neutron | 00:04 | |
| *** ivar-lazzaro has joined #openstack-neutron | 00:06 | |
| *** nlahouti1 has joined #openstack-neutron | 00:07 | |
| *** vivek-ebay has quit IRC | 00:07 | |
| *** minwang2 has quit IRC | 00:07 | |
| *** vivek-ebay has joined #openstack-neutron | 00:08 | |
| *** minwang2 has joined #openstack-neutron | 00:08 | |
| *** minwang2 has quit IRC | 00:08 | |
| *** lucas__ has quit IRC | 00:08 | |
| *** abregman has quit IRC | 00:09 | |
| *** nlahouti has quit IRC | 00:10 | |
| *** merooney has joined #openstack-neutron | 00:11 | |
| *** jasondotstar has quit IRC | 00:11 | |
| *** merooney has quit IRC | 00:11 | |
| *** thurloat is now known as thurloat_isgone | 00:11 | |
| *** ijw has quit IRC | 00:12 | |
| *** merooney has joined #openstack-neutron | 00:12 | |
| *** _cjones_ has quit IRC | 00:14 | |
| *** sridhar_ram has joined #openstack-neutron | 00:14 | |
| *** merooeny_ has joined #openstack-neutron | 00:14 | |
| *** achanda has quit IRC | 00:15 | |
| *** merooney has quit IRC | 00:16 | |
| *** jasondotstar has joined #openstack-neutron | 00:17 | |
| *** merooeny_ has quit IRC | 00:18 | |
| *** xek_ has joined #openstack-neutron | 00:19 | |
| *** xek has quit IRC | 00:20 | |
| *** isq_ has joined #openstack-neutron | 00:20 | |
| *** isq has quit IRC | 00:20 | |
| *** Piet has quit IRC | 00:21 | |
| *** david-lyle has quit IRC | 00:24 | |
| *** ujjain has quit IRC | 00:24 | |
| *** madhu_ak has quit IRC | 00:25 | |
| *** ujjain has joined #openstack-neutron | 00:26 | |
| *** ujjain has quit IRC | 00:26 | |
| *** ujjain has joined #openstack-neutron | 00:26 | |
| openstackgerrit | John Nielsen proposed openstack/neutron: Extend vxlan_group option to allow a range of group addresses https://review.openstack.org/204725 | 00:28 |
|---|---|---|
| *** veena has quit IRC | 00:29 | |
| *** shwetaap1 has quit IRC | 00:30 | |
| *** changbl has joined #openstack-neutron | 00:34 | |
| *** vivek-ebay has quit IRC | 00:34 | |
| *** vivek-ebay has joined #openstack-neutron | 00:35 | |
| *** mwagner_lap has joined #openstack-neutron | 00:40 | |
| *** changzhi_ has joined #openstack-neutron | 00:43 | |
| *** sridhar_ram has quit IRC | 00:46 | |
| *** eckelcu has quit IRC | 00:46 | |
| *** xek_ has quit IRC | 00:46 | |
| *** xek_ has joined #openstack-neutron | 00:47 | |
| *** txomon|fon has joined #openstack-neutron | 00:50 | |
| *** MentalRay_ has joined #openstack-neutron | 00:50 | |
| *** changzhi_ has quit IRC | 00:54 | |
| *** zhenguo has joined #openstack-neutron | 00:57 | |
| *** ivar-laz_ has joined #openstack-neutron | 00:58 | |
| *** ritesh has joined #openstack-neutron | 01:00 | |
| *** ivar-lazzaro has quit IRC | 01:01 | |
| *** ivar-laz_ has quit IRC | 01:02 | |
| *** eckelcu has joined #openstack-neutron | 01:02 | |
| *** erkules_ has joined #openstack-neutron | 01:03 | |
| *** thurloat_isgone has quit IRC | 01:03 | |
| *** thorst has joined #openstack-neutron | 01:03 | |
| *** erkules has quit IRC | 01:05 | |
| *** lucas____ has joined #openstack-neutron | 01:05 | |
| *** sthillma has quit IRC | 01:06 | |
| *** thurloat_isgone has joined #openstack-neutron | 01:06 | |
| *** thurloat_isgone is now known as thurloat | 01:06 | |
| *** fzdarsky has quit IRC | 01:10 | |
| *** lucas____ has quit IRC | 01:10 | |
| *** tflynn_ has joined #openstack-neutron | 01:10 | |
| *** jasondotstar has quit IRC | 01:12 | |
| *** tflynn has quit IRC | 01:13 | |
| *** thorst has quit IRC | 01:14 | |
| *** tflynn_ has quit IRC | 01:15 | |
| *** armax has joined #openstack-neutron | 01:15 | |
| *** shivrao has quit IRC | 01:16 | |
| *** regXboi has joined #openstack-neutron | 01:16 | |
| regXboi | gus: ping - how goes it? | 01:16 |
| *** jasondotstar has joined #openstack-neutron | 01:17 | |
| gus | regXboi: hey. Good I guess. I find it impossible to predict the trajectory of specs - I still don't see where the two +2s are going to come from and any new reviewer can take it off in a wildly different direction. | 01:18 |
| *** r-daneel has quit IRC | 01:19 | |
| regXboi | I hear you - the only thing I can think of is to hunt down the PTL and some of the cores and ask for their candid opinions on whether this has a valid chance of landing in liberty or not | 01:19 |
| regXboi | if it doesn't then I'd consider going with the whole thing in neutron with a planned refactoring in mitaka | 01:20 |
| regXboi | it's ugly, but ..... (holds up hands and shrugs) | 01:20 |
| *** salv-orlando has joined #openstack-neutron | 01:20 | |
| gus | regXboi: do you think it would get into neutron without a similar open ended discussion process? | 01:21 |
| regXboi | gus: that's an honest question and I don't know ... | 01:21 |
| regXboi | gus: I think mestery is at least partially on board with the idea of getting rid of shell in favor of pyroute2 | 01:21 |
| openstackgerrit | Miguel Lavalle proposed openstack/neutron: Add dns_label processing for Ports https://review.openstack.org/200952 | 01:21 |
| regXboi | but that doesn't close off the open ended discussions | 01:22 |
| *** jasondotstar has quit IRC | 01:22 | |
| *** breitz has quit IRC | 01:22 | |
| gus | regXboi: yep, we certainly feel pressure from rootwrap performance to do something different in neutron (hence rootwrap-daemon). | 01:23 |
| regXboi | gus: that doesn't really help the problem completely - using shell in front of netlink hurts in too many cases | 01:24 |
| * gus is generally pessimistic of the OpenStack decision process so would be reluctant to suggest any sort of timelines without having 2 +2's lined up beforehand. | 01:24 | |
| * regXboi feels that pain | 01:24 | |
| *** salv-orlando has quit IRC | 01:25 | |
| *** patrickeast has quit IRC | 01:25 | |
| openstackgerrit | Miguel Lavalle proposed openstack/neutron: Add dns_label processing for Ports https://review.openstack.org/200952 | 01:25 |
| *** thurloat is now known as thurloat_isgone | 01:27 | |
| *** moha_hunt has joined #openstack-neutron | 01:27 | |
| gus | regXboi: a day to write the actual feature, 3 days to make it work with eventlet, 2 weeks to make it work in our testing environment, and I'd guess wildly perhaps a month? to get the spec approved. | 01:27 |
| gus | :/ | 01:27 |
| clarkb | whats the difficulty with making it work in the test environment? can you just switch neutron to it instead of the rootwrapped shell calls? If neutron is doing it in python that should be trivial ya? | 01:29 |
| gus | oh, I made it work so that was past tense. The difficulty was with all the mocked out logging - with privsep we have non-trivial code on the privileged side, so we need to capture logs/stacktraces from there too during unittests. | 01:30 |
| *** nlahouti1 has quit IRC | 01:30 | |
| clarkb | gotcha | 01:31 |
| gus | Most of the fixes involved replacing stringio-based in-process logging with honest-to-goodness logging to files again, so that we can slurp it back regardless of which process generated it. | 01:31 |
| *** s3wong has quit IRC | 01:31 | |
| *** jbell8 has joined #openstack-neutron | 01:32 | |
| *** eckelcu has quit IRC | 01:32 | |
| regXboi | sigh - same as it ever was | 01:32 |
| *** lucas____ has joined #openstack-neutron | 01:33 | |
| gus | .. and everytime I peek behind the unittest curtain, I get lost in a maze of unittest classes that just delegate to each other so I don't claim that those 2 weeks were representative of what someone else might have spent :P | 01:33 |
| *** minwang2 has joined #openstack-neutron | 01:34 | |
| *** amuller has quit IRC | 01:36 | |
| *** breitz has joined #openstack-neutron | 01:37 | |
| *** lucas____ has quit IRC | 01:37 | |
| *** stanzgy has joined #openstack-neutron | 01:38 | |
| *** cdl1 has quit IRC | 01:46 | |
| *** miyagishi_t has joined #openstack-neutron | 01:49 | |
| openstackgerrit | Adolfo Duarte proposed openstack/neutron: HA for DVR - Neutron Server side code changes https://review.openstack.org/143169 | 01:51 |
| *** yamamoto has joined #openstack-neutron | 01:51 | |
| *** banix has joined #openstack-neutron | 01:56 | |
| *** daneyon has quit IRC | 01:57 | |
| *** changzhi_ has joined #openstack-neutron | 01:57 | |
| *** breitz has quit IRC | 01:57 | |
| *** erkules has joined #openstack-neutron | 02:00 | |
| *** pal has joined #openstack-neutron | 02:02 | |
| openstackgerrit | shihanzhang proposed openstack/neutron: Install arp spoofing protection flow after setting port tag https://review.openstack.org/199425 | 02:02 |
| *** erkules_ has quit IRC | 02:03 | |
| *** abhiraut has quit IRC | 02:04 | |
| *** minwang2 has quit IRC | 02:07 | |
| lifeless | gus: :/ | 02:07 |
| *** minwang2 has joined #openstack-neutron | 02:08 | |
| *** shwetaap has joined #openstack-neutron | 02:15 | |
| *** crc32 has joined #openstack-neutron | 02:16 | |
| *** crc32 has quit IRC | 02:16 | |
| *** crc32 has joined #openstack-neutron | 02:16 | |
| *** shwetaap has quit IRC | 02:17 | |
| *** dboik has joined #openstack-neutron | 02:18 | |
| *** minwang2 has quit IRC | 02:18 | |
| *** david-lyle has joined #openstack-neutron | 02:19 | |
| *** johnbelamaric has quit IRC | 02:20 | |
| *** hitalia has quit IRC | 02:21 | |
| *** minwang2 has joined #openstack-neutron | 02:22 | |
| *** david-lyle has quit IRC | 02:23 | |
| *** shwetaap has joined #openstack-neutron | 02:26 | |
| *** Piet has joined #openstack-neutron | 02:26 | |
| *** dboik has quit IRC | 02:27 | |
| *** dboik has joined #openstack-neutron | 02:27 | |
| *** ajmiller_ has joined #openstack-neutron | 02:32 | |
| *** minwang2 has quit IRC | 02:32 | |
| *** spandhe has quit IRC | 02:33 | |
| *** wei-wang_ has joined #openstack-neutron | 02:34 | |
| *** lucas__ has joined #openstack-neutron | 02:34 | |
| *** baoli has joined #openstack-neutron | 02:35 | |
| *** pal has quit IRC | 02:35 | |
| *** minwang2 has joined #openstack-neutron | 02:36 | |
| *** achanda has joined #openstack-neutron | 02:37 | |
| *** lucas__ has quit IRC | 02:39 | |
| *** boris-42 has quit IRC | 02:40 | |
| *** armax has quit IRC | 02:41 | |
| regXboi | is there anybody only who can talk to the experimental tempest DVR job? | 02:42 |
| regXboi | er *online* | 02:42 |
| *** cdl has joined #openstack-neutron | 02:43 | |
| clarkb | I know somewhat how it is set up if thats what you want to know | 02:44 |
| clarkb | (the multinide dvr job) | 02:44 |
| regXboi | not the multinode dvr job, but the one that turns dvr on by default | 02:44 |
| regXboi | I'm looking to find the code that executes it so that I can understand the delta to the base tempest job | 02:45 |
| *** minwang2 has quit IRC | 02:46 | |
| *** saksham has quit IRC | 02:46 | |
| *** minwang2 has joined #openstack-neutron | 02:47 | |
| clarkb | https://git.openstack.org/cgit/openstack-infra/project-config/tree/jenkins/jobs/neutron.yaml#n54 is the job definition | 02:47 |
| clarkb | the jey bit being the DVR toggle for devstack-gate which is likely where you find the interesting bits (there and devstack) | 02:48 |
| clarkb | *key bit | 02:48 |
| *** banix has quit IRC | 02:48 | |
| regXboi | thx - moe cloning | 02:49 |
| regXboi | er *more | 02:49 |
| *** oomichi has joined #openstack-neutron | 02:51 | |
| *** minwang2 has quit IRC | 02:51 | |
| *** hakimo has joined #openstack-neutron | 02:51 | |
| *** hakimo_ has quit IRC | 02:53 | |
| *** vivek-ebay has quit IRC | 03:00 | |
| *** MentalRay_ has quit IRC | 03:03 | |
| regXboi | well now that begs an interesting question | 03:03 |
| *** spandhe has joined #openstack-neutron | 03:03 | |
| regXboi | why would setting DVR to 1 involve q-fwaas | 03:03 |
| clarkb | no clue | 03:04 |
| *** jbell8 has quit IRC | 03:06 | |
| *** jbell8 has joined #openstack-neutron | 03:07 | |
| regXboi | ok, but I managed to answer my question - so many thanks! | 03:07 |
| *** dboik has quit IRC | 03:08 | |
| *** dboik has joined #openstack-neutron | 03:09 | |
| *** spandhe_ has joined #openstack-neutron | 03:09 | |
| *** spandhe has quit IRC | 03:09 | |
| *** spandhe_ is now known as spandhe | 03:09 | |
| *** mragupat has joined #openstack-neutron | 03:12 | |
| *** mragupat_ has joined #openstack-neutron | 03:13 | |
| *** dboik has quit IRC | 03:14 | |
| *** mragupat has quit IRC | 03:16 | |
| *** baoli has quit IRC | 03:20 | |
| *** yamahata has joined #openstack-neutron | 03:21 | |
| openstackgerrit | Zhi Chang proposed openstack/neutron: Keep dns nameserver order consistency https://review.openstack.org/199588 | 03:22 |
| *** coolsvap|away is now known as coolsvap | 03:26 | |
| *** zhhuabj_ has quit IRC | 03:33 | |
| *** david-lyle has joined #openstack-neutron | 03:34 | |
| *** zhhuabj has joined #openstack-neutron | 03:34 | |
| *** armax has joined #openstack-neutron | 03:35 | |
| *** ajmiller_ has quit IRC | 03:37 | |
| *** regXboi has quit IRC | 03:37 | |
| *** pritesh has joined #openstack-neutron | 03:38 | |
| *** pal has joined #openstack-neutron | 03:40 | |
| *** fawadkhaliq has joined #openstack-neutron | 03:42 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/python-neutronclient: Updated from global requirements https://review.openstack.org/204422 | 03:43 |
| *** fedexo has joined #openstack-neutron | 03:46 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/neutron: Updated from global requirements https://review.openstack.org/206825 | 03:47 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/neutron-vpnaas: Updated from global requirements https://review.openstack.org/206826 | 03:48 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/python-neutronclient: Updated from global requirements https://review.openstack.org/204422 | 03:50 |
| *** mkolesni_ has joined #openstack-neutron | 03:52 | |
| *** achanda has quit IRC | 03:55 | |
| *** amotoki has joined #openstack-neutron | 03:56 | |
| *** tictactech_ has quit IRC | 03:59 | |
| *** vivek-eb_ has joined #openstack-neutron | 04:03 | |
| *** padkrish has joined #openstack-neutron | 04:06 | |
| *** padkrish has quit IRC | 04:06 | |
| *** padkrish has joined #openstack-neutron | 04:07 | |
| *** dboik has joined #openstack-neutron | 04:09 | |
| *** ramanjaneya_ has joined #openstack-neutron | 04:09 | |
| *** mragupat_ has quit IRC | 04:09 | |
| *** vishwanathj has joined #openstack-neutron | 04:10 | |
| *** dboik has quit IRC | 04:14 | |
| *** fedexo has quit IRC | 04:21 | |
| *** lucas__ has joined #openstack-neutron | 04:38 | |
| *** armax has quit IRC | 04:42 | |
| *** lucas__ has quit IRC | 04:42 | |
| *** padkrish has quit IRC | 04:45 | |
| *** yfried|afk is now known as yfried | 04:45 | |
| *** fedexo has joined #openstack-neutron | 04:48 | |
| *** azbiswas has joined #openstack-neutron | 04:52 | |
| *** ub has joined #openstack-neutron | 04:53 | |
| *** ekuris has joined #openstack-neutron | 04:54 | |
| *** ub2 has quit IRC | 04:54 | |
| *** shaohe_robot has quit IRC | 04:54 | |
| *** garyk has joined #openstack-neutron | 04:54 | |
| *** shaohe_robot has joined #openstack-neutron | 04:55 | |
| *** yfried is now known as yfried|afk | 04:57 | |
| *** padkrish has joined #openstack-neutron | 04:58 | |
| *** yfried|afk is now known as yfried | 05:03 | |
| *** moha_hunt has quit IRC | 05:03 | |
| *** pritesh has quit IRC | 05:03 | |
| *** pritesh has joined #openstack-neutron | 05:04 | |
| *** daneyon has joined #openstack-neutron | 05:05 | |
| *** daneyon_ has joined #openstack-neutron | 05:07 | |
| *** ub has quit IRC | 05:08 | |
| *** Bhavya has joined #openstack-neutron | 05:09 | |
| *** yfried has quit IRC | 05:09 | |
| *** daneyon has quit IRC | 05:10 | |
| *** lajos-katona has joined #openstack-neutron | 05:14 | |
| *** rotbeard has joined #openstack-neutron | 05:14 | |
| *** shwetaap has quit IRC | 05:19 | |
| *** changzhi has quit IRC | 05:24 | |
| *** achanda has joined #openstack-neutron | 05:26 | |
| *** numan has joined #openstack-neutron | 05:32 | |
| *** gampel has joined #openstack-neutron | 05:33 | |
| *** nlahouti has joined #openstack-neutron | 05:34 | |
| *** azbiswas has quit IRC | 05:36 | |
| *** shz has quit IRC | 05:38 | |
| *** azbiswas has joined #openstack-neutron | 05:38 | |
| *** crc32 has quit IRC | 05:39 | |
| *** lucas__ has joined #openstack-neutron | 05:40 | |
| *** trinaths has joined #openstack-neutron | 05:42 | |
| *** lucas__ has quit IRC | 05:44 | |
| *** fedexo has quit IRC | 05:45 | |
| *** fedexo has joined #openstack-neutron | 05:46 | |
| *** itzikb has joined #openstack-neutron | 05:50 | |
| *** yamamoto has quit IRC | 05:50 | |
| *** nlahouti has quit IRC | 05:51 | |
| openstackgerrit | Hirofumi Ichihara proposed openstack/neutron: Enable to update external network subnet's gateway-ip https://review.openstack.org/194939 | 05:51 |
| *** nlahouti has joined #openstack-neutron | 05:51 | |
| openstackgerrit | Ramanjaneya Reddy Palleti proposed openstack/python-neutronclient: Support QoS neutron-client (1/2). https://review.openstack.org/189655 | 05:52 |
| *** abregman has joined #openstack-neutron | 05:56 | |
| *** fedexo has quit IRC | 05:57 | |
| *** Murali has joined #openstack-neutron | 05:58 | |
| openstackgerrit | Ramanjaneya Reddy Palleti proposed openstack/python-neutronclient: Support CLI changes for QoS (2/2). https://review.openstack.org/198277 | 05:58 |
| *** yapeng has joined #openstack-neutron | 05:59 | |
| *** nlahouti has quit IRC | 05:59 | |
| *** scheuran has joined #openstack-neutron | 05:59 | |
| *** yalie has quit IRC | 06:00 | |
| *** yalie1 has joined #openstack-neutron | 06:01 | |
| *** yalie1 has left #openstack-neutron | 06:01 | |
| *** park_heijlong_ has joined #openstack-neutron | 06:02 | |
| *** kiran-r has joined #openstack-neutron | 06:05 | |
| *** ddepaoli has joined #openstack-neutron | 06:06 | |
| openstackgerrit | Zhi Chang proposed openstack/neutron: Keep dns nameserver order consistency https://review.openstack.org/199588 | 06:06 |
| *** afazekas has joined #openstack-neutron | 06:09 | |
| *** yapeng has quit IRC | 06:10 | |
| openstackgerrit | shihanzhang proposed openstack/neutron: Install arp spoofing protection flow after setting port tag https://review.openstack.org/199425 | 06:11 |
| *** hichihara has quit IRC | 06:13 | |
| *** vivek-eb_ has quit IRC | 06:15 | |
| *** sthillma has joined #openstack-neutron | 06:19 | |
| *** lukasa has joined #openstack-neutron | 06:19 | |
| *** sthillma_ has joined #openstack-neutron | 06:21 | |
| *** josecastroleon has joined #openstack-neutron | 06:22 | |
| *** yamamoto has joined #openstack-neutron | 06:22 | |
| *** sthillma has quit IRC | 06:24 | |
| *** sthillma_ is now known as sthillma | 06:24 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/neutron: Imported Translations from Transifex https://review.openstack.org/206885 | 06:24 |
| *** hitalia has joined #openstack-neutron | 06:25 | |
| *** pritesh has quit IRC | 06:26 | |
| *** jbell8 has quit IRC | 06:27 | |
| *** shz has joined #openstack-neutron | 06:28 | |
| *** moshele has joined #openstack-neutron | 06:29 | |
| *** arajagopal has joined #openstack-neutron | 06:29 | |
| *** ub has joined #openstack-neutron | 06:32 | |
| *** zhenguo has quit IRC | 06:32 | |
| *** armax has joined #openstack-neutron | 06:33 | |
| *** ekuris has quit IRC | 06:34 | |
| *** yfried has joined #openstack-neutron | 06:35 | |
| *** Marga_ has quit IRC | 06:36 | |
| *** ekuris has joined #openstack-neutron | 06:37 | |
| *** SridharG has joined #openstack-neutron | 06:38 | |
| *** yfried has quit IRC | 06:39 | |
| *** yfried has joined #openstack-neutron | 06:39 | |
| *** zhenguo has joined #openstack-neutron | 06:40 | |
| *** Marga_ has joined #openstack-neutron | 06:40 | |
| *** trinaths has left #openstack-neutron | 06:41 | |
| *** yfried is now known as yfried|afk | 06:41 | |
| *** lucas__ has joined #openstack-neutron | 06:42 | |
| *** abehl has joined #openstack-neutron | 06:44 | |
| *** azbiswas_ has joined #openstack-neutron | 06:45 | |
| *** lucas__ has quit IRC | 06:46 | |
| *** azbiswas has quit IRC | 06:48 | |
| oomichi | gmann: oops, nice catch. will do it soon :( | 06:49 |
| oomichi | sorry, I am on wrong channel.. | 06:49 |
| *** tfukushima has joined #openstack-neutron | 06:51 | |
| *** ramanjaneya_ has quit IRC | 06:51 | |
| *** achanda has quit IRC | 06:54 | |
| *** alex_klimov has joined #openstack-neutron | 06:56 | |
| *** azbiswas_ has quit IRC | 06:57 | |
| *** achanda has joined #openstack-neutron | 06:57 | |
| *** jlibosva has joined #openstack-neutron | 06:59 | |
| *** tongli_ has quit IRC | 07:00 | |
| *** azbiswas has joined #openstack-neutron | 07:00 | |
| *** abehl has quit IRC | 07:03 | |
| *** txomon|fon has quit IRC | 07:04 | |
| *** yamahata has quit IRC | 07:05 | |
| *** otherwiseguy has quit IRC | 07:05 | |
| openstackgerrit | Sridhar Gaddam proposed openstack/neutron: Update dhcp agent cache for network:dhcp ports https://review.openstack.org/205888 | 07:09 |
| *** arajagopal has quit IRC | 07:10 | |
| *** otherwiseguy has joined #openstack-neutron | 07:10 | |
| *** salv-orlando has joined #openstack-neutron | 07:12 | |
| *** azbiswas_ has joined #openstack-neutron | 07:12 | |
| *** nmagnezi has joined #openstack-neutron | 07:12 | |
| *** kawa2014 has joined #openstack-neutron | 07:13 | |
| *** mickeys has joined #openstack-neutron | 07:14 | |
| *** adam_g has quit IRC | 07:14 | |
| *** yfried|afk is now known as yfried | 07:14 | |
| *** azbiswas has quit IRC | 07:15 | |
| *** sthillma has quit IRC | 07:16 | |
| *** abehl has joined #openstack-neutron | 07:16 | |
| *** nyechiel has joined #openstack-neutron | 07:18 | |
| *** ronis has joined #openstack-neutron | 07:19 | |
| *** bharath has joined #openstack-neutron | 07:22 | |
| *** gampel1 has joined #openstack-neutron | 07:22 | |
| *** txomon|fon has joined #openstack-neutron | 07:25 | |
| *** txomon|fon has joined #openstack-neutron | 07:25 | |
| *** armax has quit IRC | 07:25 | |
| *** gampel has quit IRC | 07:26 | |
| *** achanda has quit IRC | 07:27 | |
| *** tmorin has joined #openstack-neutron | 07:31 | |
| openstackgerrit | Yushiro FURUKAWA proposed openstack/neutron-specs: Packet logging API for Neutron https://review.openstack.org/203509 | 07:35 |
| *** sthillma has joined #openstack-neutron | 07:36 | |
| *** ihrachyshka has joined #openstack-neutron | 07:38 | |
| *** jlanoux has joined #openstack-neutron | 07:38 | |
| *** tmorin has quit IRC | 07:39 | |
| *** jschwarz has joined #openstack-neutron | 07:39 | |
| openstackgerrit | Zhi Chang proposed openstack/neutron: Keep dns nameserver order consistency https://review.openstack.org/199588 | 07:40 |
| *** y_sawai has joined #openstack-neutron | 07:42 | |
| *** mickeys has quit IRC | 07:43 | |
| *** lucas__ has joined #openstack-neutron | 07:43 | |
| miyagishi_t | kevinbenton: ping | 07:44 |
| openstackgerrit | shihanzhang proposed openstack/neutron: Use fanout RPC message to nofity the security group's change https://review.openstack.org/189171 | 07:46 |
| *** gampel1 has left #openstack-neutron | 07:46 | |
| *** yog__ has joined #openstack-neutron | 07:47 | |
| *** abehl has quit IRC | 07:48 | |
| *** lucas__ has quit IRC | 07:48 | |
| *** y_sawai has quit IRC | 07:49 | |
| *** crc32 has joined #openstack-neutron | 07:52 | |
| *** tmorin has joined #openstack-neutron | 07:53 | |
| *** y_sawai has joined #openstack-neutron | 07:58 | |
| *** pal has quit IRC | 07:59 | |
| *** yog__ has quit IRC | 07:59 | |
| *** padkrish has quit IRC | 07:59 | |
| openstackgerrit | IWAMOTO Toshihiro proposed openstack/neutron: OVS-agent: Introduce Ryu based OpenFlow implementation https://review.openstack.org/153946 | 08:00 |
| *** abehl has joined #openstack-neutron | 08:01 | |
| *** y_sawai has quit IRC | 08:02 | |
| *** matrohon has joined #openstack-neutron | 08:03 | |
| openstackgerrit | Ihar Hrachyshka proposed openstack/neutron: [neutron-db-manage] remove old HEAD file when updating for branches https://review.openstack.org/206924 | 08:03 |
| jschwarz | ihrachyshka, so I've encountered a way to DoS the qos plugin | 08:05 |
| jschwarz | :) | 08:05 |
| ihrachyshka | jschwarz, woohoo | 08:05 |
| ihrachyshka | jschwarz, tell me | 08:05 |
| jschwarz | ihrachyshka, create a policy | 08:06 |
| jschwarz | ihrachyshka, create a rule | 08:06 |
| jschwarz | ihrachyshka, delete the rule | 08:06 |
| jschwarz | ihrachyshka, http://pastebin.com/e5E1MdGi | 08:06 |
| jschwarz | ihrachyshka, ??? | 08:06 |
| jschwarz | ihrachyshka, profit | 08:06 |
| jschwarz | the error was on a simple "list rules" request | 08:07 |
| *** daneyon_ has quit IRC | 08:07 | |
| ihrachyshka | jschwarz, what's in db in terms of qos_rules/qos_bw_limit_rules? | 08:07 |
| *** safchain has joined #openstack-neutron | 08:08 | |
| jschwarz | qos_bandwidth_limit_rules is empty | 08:08 |
| *** lucas__ has joined #openstack-neutron | 08:08 | |
| *** jistr has joined #openstack-neutron | 08:08 | |
| ihrachyshka | I suspect qos_rules is there? | 08:08 |
| jschwarz | qos_rules has a line | 08:08 |
| ihrachyshka | yeap | 08:08 |
| ihrachyshka | jschwarz, may you recheck with https://review.openstack.org/205762 ? | 08:08 |
| jschwarz | the id equals the rule I deleted and the qos_policy_id matches the policy id | 08:09 |
| ihrachyshka | (you would need to flush the db) | 08:09 |
| jschwarz | flush and run db-manage, right? | 08:09 |
| ihrachyshka | that model split causes more headache than there benefits (especially since there are none) | 08:09 |
| ihrachyshka | yeap | 08:09 |
| jschwarz | can I just delete the neutron database and run db-manage instead? :P | 08:09 |
| ihrachyshka | sure, that's the idea | 08:10 |
| ihrachyshka | you meant devstack? no need ot | 08:10 |
| ihrachyshka | *to | 08:10 |
| jschwarz | no | 08:10 |
| jschwarz | just 'delete database neutron;' or something | 08:10 |
| openstackgerrit | IWAMOTO Toshihiro proposed openstack/neutron: OVS-agent: Switch the default to "native" of_interface https://review.openstack.org/206329 | 08:10 |
| *** azbiswas_ has quit IRC | 08:12 | |
| *** sthillma has quit IRC | 08:12 | |
| *** azbiswas has joined #openstack-neutron | 08:12 | |
| *** lucas__ has quit IRC | 08:12 | |
| ihrachyshka | jschwarz, that should be enough; drop database; db-manage upgrade | 08:12 |
| jschwarz | done | 08:12 |
| jschwarz | let me check | 08:12 |
| bharath | Hi, I am learning neutron +ovs flow, I have one doubt. why does neutron creating a bridge every time when we launch an instance? I am referring to this diagram http://docs.openstack.org/admin-guide-cloud/content/figures/14/a/a/common/figures/under-the-hood-scenario-1-ovs-compute.png | 08:13 |
| jschwarz | ihrachyshka, looks good | 08:13 |
| ihrachyshka | jschwarz, awesome | 08:13 |
| ihrachyshka | jschwarz, I should work on pushing it asap with your tests | 08:14 |
| *** ygbo has joined #openstack-neutron | 08:14 | |
| jschwarz | ihrachyshka, would not object. saw that your patch got in yesterday | 08:14 |
| bharath | can't we apply iptables on tap device ? | 08:14 |
| jschwarz | ihrachyshka, so lets get a +2+A stamp on my first 2 and then we do whatever we want :) | 08:14 |
| ihrachyshka | garyk, are you with us? | 08:15 |
| ihrachyshka | garyk, qos has some desperate need for reviews | 08:15 |
| garyk | ihrachyshka: please point me to the reviews | 08:16 |
| ihrachyshka | garyk, https://review.openstack.org/#/c/205796/ | 08:16 |
| ihrachyshka | garyk, https://review.openstack.org/#/c/205815/ | 08:16 |
| ihrachyshka | garyk, https://review.openstack.org/#/c/205762/ | 08:16 |
| *** yog__ has joined #openstack-neutron | 08:16 | |
| ihrachyshka | garyk, also https://review.openstack.org/206525 would be good. | 08:17 |
| *** dguitarbite has joined #openstack-neutron | 08:18 | |
| openstackgerrit | Cedric Brandily proposed openstack/neutron: Python 3: set webob.Response.text instead of webob.Response.body https://review.openstack.org/204791 | 08:18 |
| openstackgerrit | Cedric Brandily proposed openstack/neutron: Python 3: use a hash to sort dictionaries https://review.openstack.org/204792 | 08:18 |
| *** tflynn has joined #openstack-neutron | 08:24 | |
| jschwarz | garyk, ihrachyshka, thanks | 08:24 |
| *** jasondotstar has joined #openstack-neutron | 08:25 | |
| *** tflynn_ has joined #openstack-neutron | 08:26 | |
| *** neiljerr` has joined #openstack-neutron | 08:26 | |
| *** tfukushima has quit IRC | 08:26 | |
| *** tfukushima has joined #openstack-neutron | 08:28 | |
| *** tflynn has quit IRC | 08:29 | |
| *** azbiswas has quit IRC | 08:33 | |
| openstackgerrit | Kevin Benton proposed openstack/neutron: Initialize port_info dict as blank in OVS agent https://review.openstack.org/206764 | 08:34 |
| *** john-davidge has joined #openstack-neutron | 08:36 | |
| *** erkules has quit IRC | 08:38 | |
| *** erkules has joined #openstack-neutron | 08:38 | |
| *** john-davidge_ has joined #openstack-neutron | 08:39 | |
| ihrachyshka | kevinbenton, ^^ heh, I just reported https://bugs.launchpad.net/neutron/+bug/1479265 | 08:39 |
| openstack | Launchpad bug 1479265 in neutron "ovs agent crashes with UnboundLocalError in case self.scan_ancillary_ports fails" [Undecided,New] | 08:39 |
| kevinbenton | ihrachyshka: assaf beat you to it! | 08:40 |
| ihrachyshka | I cry every time I see that spaghetti | 08:40 |
| ihrachyshka | kevinbenton, I think it's actually different | 08:40 |
| kevinbenton | ihrachyshka: gross | 08:40 |
| *** john-davidge has quit IRC | 08:40 | |
| *** john-davidge_ is now known as john-davidge | 08:40 | |
| kevinbenton | ihrachyshka: i'll let you fix that one ;) | 08:40 |
| ihrachyshka | heh | 08:40 |
| kevinbenton | ihrachyshka: or i can do it in the morning | 08:40 |
| ihrachyshka | kevinbenton, I am not blocked on it, just reported to feel good when rechecking ;) | 08:41 |
| openstackgerrit | ZongKai LI proposed openstack/neutron: [WIP]support neutron-ovs-dvr-multigateway https://review.openstack.org/191671 | 08:41 |
| *** l6unchpad has joined #openstack-neutron | 08:46 | |
| *** tflynn_ has quit IRC | 08:50 | |
| miyagishi_t | kevinbenton: I have a question about implementation plan of Distributed SNAT. | 08:51 |
| miyagishi_t | kevinbenton: I thought implementation plan based on etherpad: https://etherpad.openstack.org/p/decentralized-snat | 08:51 |
| miyagishi_t | kevinbenton: I understood that "True Distributed SNAT" and "Carrier Grade NAT" solves Public IP consumption problem. | 08:51 |
| miyagishi_t | kevinbenton: However, in my understanding, these plans have to manage outside of OpenStack. | 08:52 |
| miyagishi_t | kevinbenton: If so, I think it is difficult to become standard feature. | 08:52 |
| miyagishi_t | kevinbenton: If my understanding is wrong, I want to understand correctly, would you tell me detail of these plans | 08:52 |
| miyagishi_t | kevinbenton: or who should I ask to about this? | 08:52 |
| *** ZZelle has quit IRC | 08:53 | |
| *** ZZelle has joined #openstack-neutron | 08:53 | |
| kevinbenton | miyagishi_t: if you have a proposal to do distributed SNAT that would work with the current implementation, that could go into neutron | 08:54 |
| *** ip0353 has joined #openstack-neutron | 08:55 | |
| openstackgerrit | Ihar Hrachyshka proposed openstack/neutron: Initialize ancillary_port_info dict as blank in OVS agent https://review.openstack.org/206941 | 08:55 |
| ihrachyshka | kevinbenton, pretty dumb to have another patch for that one but ^^ | 08:55 |
| kevinbenton | miyagishi_t: however, getting the upstream router to send to the correct neutron node requires configuration on the upstream router | 08:55 |
| kevinbenton | ihrachyshka: that's special :) | 08:56 |
| *** yushiro has joined #openstack-neutron | 08:56 | |
| *** crc32 has quit IRC | 08:56 | |
| kevinbenton | ihrachyshka: LGTM. that whole main loop thing needs to be broken down even more | 08:57 |
| openstackgerrit | Ihar Hrachyshka proposed openstack/neutron: Initialize ancillary_port_info dict as blank in OVS agent https://review.openstack.org/206941 | 08:57 |
| kevinbenton | ihrachyshka: but i think that's what rossella_s is working on anyway | 08:57 |
| *** gongysh_ has quit IRC | 08:57 | |
| ihrachyshka | kevinbenton, yeah. she is a brave one to get into that | 08:57 |
| *** pal has joined #openstack-neutron | 08:57 | |
| rossella_s | ihrachyshka, kevinbenton yeah my plan is to break it down, it's too long | 08:58 |
| *** spandhe has quit IRC | 08:59 | |
| rossella_s | I should update my patches | 08:59 |
| *** doude has joined #openstack-neutron | 09:00 | |
| ihrachyshka | garyk, are you on https://review.openstack.org/#/c/205762/ ? | 09:00 |
| *** nijaba has quit IRC | 09:02 | |
| garyk | ihrachyshka: can you ping me when it gets the +1 from jenikns | 09:03 |
| *** coolsvap is now known as coolsvap|away | 09:04 | |
| *** amotoki has quit IRC | 09:05 | |
| *** slaweq_ has joined #openstack-neutron | 09:05 | |
| ihrachyshka | garyk, ah ok, it had one before I updated commit message | 09:06 |
| *** nijaba has joined #openstack-neutron | 09:07 | |
| garyk | ihrachyshka: i had one comment | 09:07 |
| *** nyechiel has quit IRC | 09:07 | |
| jschwarz | ihrachyshka, https://review.openstack.org/#/c/205796/ didn't get -1 from jenkins and the patch isn't on the zuul list... | 09:08 |
| jschwarz | is there a problem with the gate again? :< | 09:08 |
| *** nyechiel has joined #openstack-neutron | 09:10 | |
| ihrachyshka | jschwarz, meh. I hope not. I will check with infra in an hour, or you can do it | 09:10 |
| jschwarz | will do it myself | 09:10 |
| ihrachyshka | infra is bumpy these days. | 09:10 |
| ihrachyshka | jschwarz, thanks | 09:11 |
| * ihrachyshka is away for an hour | 09:11 | |
| *** l6unchpad has quit IRC | 09:12 | |
| *** coolsvap|away is now known as coolsvap | 09:12 | |
| openstackgerrit | Zhi Chang proposed openstack/neutron: Keep dns nameserver order consistency https://review.openstack.org/199588 | 09:14 |
| *** ihrachyshka has quit IRC | 09:15 | |
| *** fzdarsky has joined #openstack-neutron | 09:19 | |
| *** alexpilotti has joined #openstack-neutron | 09:21 | |
| *** hitalia has quit IRC | 09:22 | |
| *** l6unchpad has joined #openstack-neutron | 09:24 | |
| openstackgerrit | Rossella Sblendido proposed openstack/neutron: Introduce get_ports_attributes in OVSBridge https://review.openstack.org/200279 | 09:25 |
| *** yamamoto has quit IRC | 09:26 | |
| *** abregman is now known as abregman|brb | 09:28 | |
| *** yfried is now known as yfried|afk | 09:29 | |
| -openstackstatus- NOTICE: Currently our CI system is broken, jobs are not getting processed at all. | 09:29 | |
| *** ChanServ changes topic to "Currently our CI system is broken, jobs are not getting processed at all." | 09:29 | |
| *** tfukushima has quit IRC | 09:32 | |
| *** azbiswas has joined #openstack-neutron | 09:34 | |
| *** yamamoto has joined #openstack-neutron | 09:35 | |
| *** azbiswas has quit IRC | 09:42 | |
| *** anilvenkata has joined #openstack-neutron | 09:45 | |
| *** alex_xu has quit IRC | 09:48 | |
| openstackgerrit | Kevin Benton proposed openstack/neutron: Initialize port_info dict as blank in OVS agent https://review.openstack.org/206764 | 09:48 |
| *** yfried|afk is now known as yfried | 09:49 | |
| *** alex_xu has joined #openstack-neutron | 09:49 | |
| *** coolsvap is now known as coolsvap|away | 09:50 | |
| *** ihrachyshka has joined #openstack-neutron | 09:53 | |
| ihrachyshka | jschwarz, so gate stuck and we have no ETA to get it back? | 09:55 |
| *** doude has quit IRC | 09:55 | |
| *** jasondotstar has quit IRC | 09:57 | |
| *** coolsvap|away is now known as coolsvap | 09:58 | |
| *** yfried is now known as yfried|afk | 09:59 | |
| miyagishi_t | kevinbenton: Thanks for your reply. | 09:59 |
| miyagishi_t | kevinbenton: IMHO if SNAT would be distributed with current implementation, it will become high IP consumption. So I think it needs some changes of neutron. | 09:59 |
| miyagishi_t | kevinbenton: few days ago, I proposed my implementation plan in openstack-dev ML: http://lists.openstack.org/pipermail/openstack-dev/2015-July/070478.html | 10:00 |
| miyagishi_t | kevinbenton: In an ideal situation, this plan can be reduction of IP consumption than simply decentralized plan - "One IP Address per Router per Host". | 10:00 |
| miyagishi_t | kevinbenton: However in a worst situation, this plan would use too many public IPs, so this doesn't become a solution to the root of the problem. | 10:00 |
| miyagishi_t | kevinbenton: Therefore, I want to understand more correctly about "True Distributed SNAT" and "Carrier Grade NAT". Would you tell me detail of these plans? | 10:00 |
| *** ihrachyshka has quit IRC | 10:00 | |
| *** fzdarsky has quit IRC | 10:00 | |
| *** ygbo has quit IRC | 10:00 | |
| *** abehl has quit IRC | 10:00 | |
| *** alex_klimov has quit IRC | 10:00 | |
| *** ub has quit IRC | 10:00 | |
| *** garyk has quit IRC | 10:00 | |
| *** sbelous has quit IRC | 10:00 | |
| *** Kennan has quit IRC | 10:00 | |
| *** mitz has quit IRC | 10:00 | |
| *** yuanying has quit IRC | 10:00 | |
| *** dguerri` has quit IRC | 10:00 | |
| *** sc68cal has quit IRC | 10:00 | |
| *** mitz has joined #openstack-neutron | 10:00 | |
| *** park_heijlong_ has quit IRC | 10:00 | |
| *** l6unchpa_ has joined #openstack-neutron | 10:00 | |
| *** garyk has joined #openstack-neutron | 10:00 | |
| *** tfukushima has joined #openstack-neutron | 10:00 | |
| *** ihrachyshka has joined #openstack-neutron | 10:01 | |
| *** alex_klimov has joined #openstack-neutron | 10:01 | |
| *** yuanying has joined #openstack-neutron | 10:01 | |
| *** Kennan has joined #openstack-neutron | 10:01 | |
| *** dguerri` has joined #openstack-neutron | 10:01 | |
| *** ub has joined #openstack-neutron | 10:01 | |
| *** sbelous has joined #openstack-neutron | 10:01 | |
| *** dguerri` is now known as dguerri | 10:01 | |
| *** dguerri has joined #openstack-neutron | 10:01 | |
| *** sc68cal has joined #openstack-neutron | 10:02 | |
| *** kiran-r has quit IRC | 10:03 | |
| *** abehl has joined #openstack-neutron | 10:03 | |
| *** ygbo has joined #openstack-neutron | 10:03 | |
| *** fzdarsky has joined #openstack-neutron | 10:03 | |
| *** clduser_ has joined #openstack-neutron | 10:05 | |
| *** lajos-katona1 has joined #openstack-neutron | 10:05 | |
| *** txomon|fon has quit IRC | 10:05 | |
| *** l6unchpad has quit IRC | 10:05 | |
| *** jschwarz has quit IRC | 10:05 | |
| *** clduser has quit IRC | 10:05 | |
| *** ildikov has quit IRC | 10:05 | |
| *** Piet has quit IRC | 10:05 | |
| *** ildikov has joined #openstack-neutron | 10:05 | |
| *** txomon|fon has joined #openstack-neutron | 10:06 | |
| *** lajos-katona has quit IRC | 10:06 | |
| *** jschwarz has joined #openstack-neutron | 10:06 | |
| *** nyechiel_ has joined #openstack-neutron | 10:07 | |
| openstackgerrit | Thomas Morin proposed openstack/neutron-specs: Neutron Extension for BGPVPN https://review.openstack.org/177740 | 10:08 |
| *** rdo has quit IRC | 10:08 | |
| *** yushiro has quit IRC | 10:09 | |
| *** jasondotstar has joined #openstack-neutron | 10:09 | |
| *** nyechiel has quit IRC | 10:09 | |
| *** yushiro has joined #openstack-neutron | 10:09 | |
| *** rdo has joined #openstack-neutron | 10:10 | |
| *** kiran-r has joined #openstack-neutron | 10:12 | |
| *** slaweq_ has quit IRC | 10:13 | |
| *** zhenguo has quit IRC | 10:13 | |
| *** josecastroleon has quit IRC | 10:15 | |
| openstackgerrit | shihanzhang proposed openstack/neutron: Support distributed dhcp agent https://review.openstack.org/184423 | 10:17 |
| *** pc_m has joined #openstack-neutron | 10:18 | |
| *** doude has joined #openstack-neutron | 10:18 | |
| openstackgerrit | Sridhar Gaddam proposed openstack/neutron: Update dhcp agent cache for network:dhcp ports https://review.openstack.org/205888 | 10:19 |
| moshele | jlibosva: hi | 10:24 |
| jlibosva | moshele: hey | 10:24 |
| moshele | jlibosva: this cause a problem in the unitest https://github.com/openstack/neutron/blob/feature/qos/neutron/plugins/ml2/drivers/openvswitch/agent/common/config.py#L103-105 we adding another qos driver | 10:24 |
| moshele | jlibosva: error with duplicate option: agent_driver | 10:25 |
| moshele | jlibosva: I think we should rename the section to ovs_qos or something like that | 10:25 |
| moshele | jlibosva: or just call the parameter ovs_qos_agent_driver | 10:26 |
| jlibosva | moshele: I don't get it. What another qos driver are you adding? Why does it have to do anything with ovs agent? | 10:26 |
| jlibosva | moshele: can you show me the patch? | 10:27 |
| jlibosva | moshele: ajo's idea was that every agent will have its own default | 10:27 |
| moshele | jlibosva: it is not upstream but I am adding sriov support so I need to add it to sriov config | 10:27 |
| *** yushiro has quit IRC | 10:27 | |
| jlibosva | moshele: you should add to sriov agent a new config opt | 10:28 |
| moshele | jlibosva: I add qos/agent_driver for sriov, but the problem is when you run the unit test | 10:29 |
| ihrachyshka | moshele, you should not register both ovs and sriov options in tests. | 10:29 |
| jlibosva | moshele: then I think the problem is in test. unittest is a testing of a small piece in separation. sriov agent shouldn't know anything about configs for ovs agent, right? | 10:30 |
| moshele | jlibosva: but it happens when you run them all, right? | 10:30 |
| jlibosva | moshele: aaah, I think I see the problem | 10:31 |
| jlibosva | moshele: you mean that all unittests share the same conf obj? | 10:31 |
| moshele | jlibosva: yes | 10:31 |
| jlibosva | I think we have had this problem before ... | 10:31 |
| moshele | jlibosva: what was the solution? | 10:32 |
| jlibosva | checking | 10:32 |
| jlibosva | moshele: so your code won't be public? | 10:32 |
| moshele | jlibosva: it will be but is not ready yet | 10:34 |
| jlibosva | moshele: can you try to add self.addCleanup(cfg.CONF.reset) | 10:35 |
| jlibosva | moshele: to where the option is used? | 10:36 |
| moshele | jlibosva: sure | 10:36 |
| *** fzdarsky has quit IRC | 10:37 | |
| *** abregman|brb is now known as abregman | 10:37 | |
| jschwarz | ihrachyshka, yes | 10:37 |
| *** rossella_s has quit IRC | 10:37 | |
| *** rossella_s has joined #openstack-neutron | 10:38 | |
| *** aix has quit IRC | 10:38 | |
| ihrachyshka | jlibosva, should we use some config fixture for that? | 10:39 |
| ihrachyshka | jlibosva, moshele http://docs.openstack.org/developer/oslo.config/fixture.html | 10:39 |
| jlibosva | ihrachyshka: didn't know about that. I don't know how will it be done cause the opt is register on module import | 10:41 |
| jlibosva | or isn't it? | 10:41 |
| jlibosva | it s | 10:41 |
| jlibosva | anyway, I don't think it's necessary to make a config section for agent_drivers just because it makes mess in unittests, I'd rather spend time on investigation how to fix unittests | 10:42 |
| ihrachyshka | jlibosva, well, we may need to rework registration not to do it. it's just wrong. | 10:42 |
| *** Ramanjaneya has quit IRC | 10:42 | |
| ihrachyshka | instead, there should be a function that receives conf object and registers options on it | 10:43 |
| jlibosva | I'm not sure, ovs agent has its own dedicated config module where it does the registering. Noone but ovs agent should import this module. | 10:44 |
| *** yfried|afk is now known as yfried | 10:48 | |
| *** stanzgy has quit IRC | 10:48 | |
| *** changzhi_ has quit IRC | 10:49 | |
| *** lennyb_ has quit IRC | 10:49 | |
| moshele | jlibosva: the self.addCleanup(cfg.CONF.reset) didn't work or I used it wrong | 10:50 |
| moshele | jlibosva: the problem is in the when doing the import so we can not reset | 10:50 |
| moshele | jlibosva, ihrachyshka | 10:51 |
| moshele | : should I use oslo.config/fixture? | 10:51 |
| jlibosva | moshele: can you put your WIP patch somewhere so I can have a look? | 10:52 |
| ihrachyshka | moshele, can we avoid the import? yeah, a patch would help. | 10:52 |
| jlibosva | moshele: even by format-patch | 10:52 |
| moshele | jlibosva: give me a sec I will send you the path in mail | 10:53 |
| jlibosva | cool | 10:53 |
| *** coolsvap is now known as coolsvap|away | 10:55 | |
| *** yamamoto has quit IRC | 10:57 | |
| moshele | jlibosva: sent to you by mail | 10:57 |
| moshele | ihrachyshka: did you see my comment in https://review.openstack.org/#/c/206038/ ? | 10:58 |
| ihrachyshka | moshele, sorry, will check now | 10:59 |
| *** yamamoto has joined #openstack-neutron | 10:59 | |
| openstackgerrit | Ihar Hrachyshka proposed openstack/neutron-fwaas: Killed existing downgrade rules in migration scripts https://review.openstack.org/206984 | 10:59 |
| *** shaohe_robot has quit IRC | 11:00 | |
| openstackgerrit | Ihar Hrachyshka proposed openstack/neutron-fwaas: Killed existing downgrade rules in migration scripts https://review.openstack.org/206984 | 11:00 |
| *** shaohe_robot has joined #openstack-neutron | 11:01 | |
| openstackgerrit | Cedric Brandily proposed openstack/neutron: Python 3: set webob.Response.text instead of webob.Response.body https://review.openstack.org/204791 | 11:01 |
| openstackgerrit | Cedric Brandily proposed openstack/neutron: Python 3: use a hash to sort dictionaries https://review.openstack.org/204792 | 11:01 |
| *** jsheeren has joined #openstack-neutron | 11:04 | |
| jlibosva | moshele: I think the best would be what ihrachyshka suggested - make a function that does the registering and triggering it in agen'st "main" | 11:05 |
| ihrachyshka | ideally, imports should not have side effects like that | 11:05 |
| *** zhhuabj has quit IRC | 11:06 | |
| moshele | jlibosva, ihrachyshka | 11:06 |
| moshele | : ok I will do it | 11:06 |
| jlibosva | moshele: if you want, I can do that for ovs and you can base your patch on mine | 11:06 |
| moshele | jlibosva: ok sure | 11:07 |
| *** miyagishi_t has quit IRC | 11:10 | |
| *** sbelous has quit IRC | 11:10 | |
| *** zhhuabj has joined #openstack-neutron | 11:10 | |
| *** jprovazn has joined #openstack-neutron | 11:11 | |
| *** sbelous has joined #openstack-neutron | 11:11 | |
| openstackgerrit | Ihar Hrachyshka proposed openstack/neutron-vpnaas: script.py.mako: added license header and missing branch_labels https://review.openstack.org/206989 | 11:13 |
| openstackgerrit | Ihar Hrachyshka proposed openstack/neutron-vpnaas: Killed existing downgrade rules in migration scripts https://review.openstack.org/206990 | 11:13 |
| openstackgerrit | Cedric Brandily proposed openstack/neutron: Python 3: compare response.body to bytes in namespace_proxy test https://review.openstack.org/206991 | 11:18 |
| openstackgerrit | Cedric Brandily proposed openstack/neutron: Python 3: hmac requires bytes key/msg https://review.openstack.org/206993 | 11:21 |
| jlibosva | moshele: I did this: https://review.openstack.org/#/c/206994/ | 11:23 |
| *** jasondot_ has joined #openstack-neutron | 11:23 | |
| jlibosva | moshele: I haven't tested it yet though :) | 11:23 |
| *** baoli has joined #openstack-neutron | 11:24 | |
| *** phil_h has joined #openstack-neutron | 11:24 | |
| jschwarz | ihrachyshka, lets discuss the API server's URI :p | 11:25 |
| ihrachyshka | moshele, I've replied on sriov | 11:25 |
| *** jasondotstar has quit IRC | 11:25 | |
| ihrachyshka | jschwarz, wanna bj or here? | 11:25 |
| jschwarz | here is fine | 11:26 |
| ihrachyshka | I'm good | 11:26 |
| jschwarz | ihrachyshka, the URI for getting (showing) a bw limit rule is: '/qos/policies/%(policy)s/bandwidth_limit_rule/%(rule)s' | 11:26 |
| jschwarz | why do we need the '/policies/%(policy)s/' for? the policy id isn't saved in the qos_bandwidth_limit_rule sql table so it's not needed | 11:27 |
| *** gampel has joined #openstack-neutron | 11:29 | |
| jschwarz | ihrachyshka, looking at neutron/services/qos/qos_plugin.py, the get_policy_bandwidth_limit_rule function is getting a policy_id but completely ignores it | 11:29 |
| jschwarz | so... it's not really needed | 11:29 |
| jschwarz | removing it will simplify quite a bit of stuff on the neutronclient department I think | 11:29 |
| ihrachyshka | jschwarz, will it mean we'll post rules as /qos/<rule_type>s/%(rule)s? | 11:30 |
| jschwarz | ihrachyshka, erm | 11:31 |
| *** Bhavya has quit IRC | 11:31 | |
| jschwarz | ihrachyshka, probably, yes. which is not a good thinjg | 11:31 |
| jschwarz | s/thinjg/thing/ | 11:31 |
| *** baoli has quit IRC | 11:31 | |
| ihrachyshka | jschwarz, just curious: how is it not a good thing, specifically? | 11:32 |
| jschwarz | ihrachyshka, well, we need some way to associate a rule to a policy | 11:32 |
| ihrachyshka | qos_policy_id? | 11:32 |
| jschwarz | right now it's done with the /qos/policies/<policy_id>/<type>/<rule_id> and that also associates the rule to the policy | 11:32 |
| *** Guest100_ has joined #openstack-neutron | 11:32 | |
| jschwarz | if we remove the prefix we'll have to add an API for associating a rule to a policy | 11:33 |
| ihrachyshka | well, passing policy_id would do the same. | 11:33 |
| jschwarz | hmmmmmm | 11:33 |
| jschwarz | yes. | 11:33 |
| ihrachyshka | I really think we complicated a lot of things going with that multi level thing | 11:33 |
| jschwarz | I agree | 11:33 |
| ihrachyshka | as we did for models and other stuf | 11:33 |
| ihrachyshka | it should not be that complex. | 11:33 |
| jschwarz | so perhaps it's time we simplify things? | 11:34 |
| ihrachyshka | it's like a net_id needed to create a subnet | 11:34 |
| ihrachyshka | we don't embedd subnets though | 11:34 |
| ihrachyshka | jschwarz, it's always time, assuming api tests are in place. | 11:34 |
| jschwarz | creating the subnet_id neccesitate the net_id as a parameter, right? | 11:34 |
| ihrachyshka | moshele, ^^ what do you think? | 11:34 |
| ihrachyshka | jschwarz, yes, it does | 11:34 |
| jschwarz | ihrachyshka, luckily the API tests are merging :) | 11:34 |
| ihrachyshka | jschwarz, you can base on it | 11:35 |
| *** mwagner_lap has quit IRC | 11:35 | |
| *** oomichi has quit IRC | 11:36 | |
| *** pal has quit IRC | 11:37 | |
| moshele | ihrachyshka: can you download iproute2 and got to commit f89a2a05ffa94ac5bec9f50751f761215356092b | 11:38 |
| moshele | ihrachyshka: the commit introduce the new parameters but didn't update the usage | 11:39 |
| jschwarz | ihrachyshka, so I'm gonna start looking at changing the API server.. | 11:39 |
| jschwarz | ihrachyshka, obviously I won't get to the functional tests anytime soon ^_^ | 11:39 |
| moshele | ihrachyshka: so I am not sure how help usage is better, I am relaying on the snapshot of this commit and above | 11:40 |
| ihrachyshka | moshele, do we have a later snapshot with help text? | 11:40 |
| ihrachyshka | moshele, another option would be to call the latest one and fall back to the old one on failure. | 11:42 |
| ihrachyshka | you could also cache the 'too old' state somewhere to avoid calls to the command that is not there yet. | 11:43 |
| *** sbelous has quit IRC | 11:43 | |
| ihrachyshka | moshele, finally, we may just require the new version. | 11:43 |
| moshele | ihrachyshka: they both should work if that is the problem I can just use rate | 11:43 |
| *** itzikb has quit IRC | 11:43 | |
| *** sbelous has joined #openstack-neutron | 11:44 | |
| ihrachyshka | jschwarz, ack | 11:44 |
| sc68cal | morning all | 11:45 |
| *** itzikb has joined #openstack-neutron | 11:45 | |
| *** thorst has joined #openstack-neutron | 11:47 | |
| *** Ramanjaneya has joined #openstack-neutron | 11:48 | |
| *** l6unchpa_ has quit IRC | 11:48 | |
| *** fawadkhaliq has quit IRC | 11:48 | |
| *** jsheeren has quit IRC | 11:49 | |
| moshele | ihrachyshka: let keep it simple currently all the destro support "rate" option as it is not deprecate in iproute2 and I will remove the support for max_tx_rate , what do you think? | 11:50 |
| jlibosva | ihrachyshka: moshele I see we use a lot of qos_ prefix for modules. I think it makes more sense to drop qos_ prefix for modules that belong to qos package. e.g. neutron/services/qos/qos_consts -> neutron/services/qos/consts etc. | 11:51 |
| jlibosva | yeah, I think this is super important to have :) | 11:51 |
| *** yfried has quit IRC | 11:51 | |
| *** veena has joined #openstack-neutron | 11:51 | |
| *** coolsvap|away is now known as coolsvap | 11:51 | |
| moshele | jlibosva: I agree | 11:52 |
| ihrachyshka | sc68cal, hey man :) | 11:54 |
| *** shaohe_robot1 has joined #openstack-neutron | 11:54 | |
| ihrachyshka | moshele, I agree to kill redundant stuff | 11:55 |
| *** josecastroleon has joined #openstack-neutron | 11:55 | |
| ihrachyshka | jlibosva, agreed. after most stuff is done, we'll clean the tree | 11:55 |
| ihrachyshka | for now, let's avoid moving pieces | 11:55 |
| *** fawadkhaliq has joined #openstack-neutron | 11:55 | |
| *** johnbelamaric has joined #openstack-neutron | 11:55 | |
| pc_m | ihrachyshka: hi | 11:57 |
| ihrachyshka | pc_m, hi | 11:57 |
| pc_m | ihrachyshka: Had some questions on the new migration (doing a new one for VPN). | 11:57 |
| *** shaohe_robot has quit IRC | 11:57 | |
| ihrachyshka | pc_m, sure | 11:57 |
| pc_m | ihrachyshka: In the expand files, what is the branch_label? | 11:57 |
| ihrachyshka | pc_m, it should be only in the very first script in a new branch | 11:58 |
| ihrachyshka | pc_m, it marks the whole branch of scripts with the name | 11:58 |
| ihrachyshka | and you can use the name to refer to the branch (and => it's latest head) | 11:58 |
| ihrachyshka | so you can db-manage upgrade liberty_expand@head | 11:58 |
| *** phil_h has quit IRC | 11:58 | |
| ihrachyshka | and get all scripts in the branch applied | 11:59 |
| pc_m | for the second expand script then, it just has a depends on of the previous release (e.g. kilo) and no branch label? | 11:59 |
| pc_m | and how does migration know to apply both scripts for the branch? | 12:01 |
| pc_m | or does the second script need to depend on the first? | 12:01 |
| *** tfukushima has quit IRC | 12:01 | |
| ihrachyshka | pc_m, second one only has down_revision to the first one | 12:02 |
| ihrachyshka | no kilo needed | 12:02 |
| ihrachyshka | kilo will be enforced by the first one | 12:02 |
| *** yamamoto has quit IRC | 12:02 | |
| ihrachyshka | (it should have it in depends_on) | 12:03 |
| *** pal has joined #openstack-neutron | 12:03 | |
| ihrachyshka | pc_m, so other than the first script, consequent are pretty the same as before | 12:03 |
| pc_m | ihrachyshka: So first has down_revision = None, depends_on = kilo, and branch label. Second one has depends_on = <first> | 12:04 |
| *** amuller has joined #openstack-neutron | 12:04 | |
| pc_m | Is that correct? | 12:04 |
| ihrachyshka | pc_m, no, second one has down_revision = <first> | 12:04 |
| ihrachyshka | otherwise correct | 12:04 |
| ihrachyshka | depends_on is used when you want to logically untange a series from another branch while making sure the order is still correct when applied | 12:05 |
| pc_m | ihrachyshka: gotcha. And no depends_on for second, right? | 12:05 |
| ihrachyshka | pc_m, yes, depends_on is used 1) in first patch, to refer to the last one in the previous cycle 2) in case you need to make an expand script depending on a contraction one, or vice versa | 12:06 |
| ihrachyshka | using depends_on there instead of down_revision allows us to avoid mixing branch labels and keeping them for proper scripts only | 12:06 |
| *** aix has joined #openstack-neutron | 12:07 | |
| pc_m | ihrachyshka: and I take it that the contract scripts are symmetrical? | 12:07 |
| ihrachyshka | otherwise, if you would put down_revision for a script from another branch, you would automatically apply the labels from that other branch to all scripts in the one you want to update | 12:07 |
| ihrachyshka | pc_m, symmetrical?.. | 12:07 |
| *** phil_h has joined #openstack-neutron | 12:07 | |
| *** l6unchpad has joined #openstack-neutron | 12:07 | |
| *** pal has quit IRC | 12:07 | |
| ihrachyshka | ah, you mean, same rules? sure | 12:08 |
| pc_m | ihrachyshka: Same thing... first commit has down_revision=none, depends_on,... | 12:08 |
| ihrachyshka | they are basically two independent series, just sometimes you may need to depends_on between them to keep order | 12:08 |
| ihrachyshka | yes, same thing | 12:08 |
| *** mkolesni__ has joined #openstack-neutron | 12:08 | |
| ihrachyshka | gsagie, hi man. any news on non-native impl for ovs qos driver? | 12:09 |
| *** mkolesni_ has quit IRC | 12:09 | |
| *** yfried has joined #openstack-neutron | 12:09 | |
| *** sbelous has quit IRC | 12:09 | |
| pc_m | ihrachyshka: In the HEADS file, is there a specific ordering? Does it only include the expand scripts or both expand and contract? | 12:09 |
| *** sbelous has joined #openstack-neutron | 12:10 | |
| ihrachyshka | pc_m, it's latest heads for all branches (those are three now but will be more in the future, 2 per cycle) | 12:11 |
| ihrachyshka | pc_m, it's sorted | 12:11 |
| *** ramanjaneya_ has joined #openstack-neutron | 12:12 | |
| pc_m | ihrachyshka: by string value or by sequence (newest to oldest)? | 12:12 |
| ihrachyshka | pc_m, string | 12:15 |
| pc_m | ihrachyshka: thanks! | 12:15 |
| ihrachyshka | pc_m, the file is not involved in actual migration, it's just a nice way to catch a conflict on merging two scripts in parallel. | 12:16 |
| *** fnordahl has joined #openstack-neutron | 12:16 | |
| *** shaohe_robot has joined #openstack-neutron | 12:16 | |
| *** ByPasS has joined #openstack-neutron | 12:16 | |
| ihrachyshka | pc_m, I plan to move it back to HEAD file per branch | 12:16 |
| ihrachyshka | it would be easier to understand where each string belongs | 12:17 |
| *** shaohe_robot1 has quit IRC | 12:17 | |
| pc_m | ihrachyshka: I'm seeing a failure on check_migration | 12:17 |
| ihrachyshka | pc_m, show me | 12:17 |
| *** coolsvap is now known as coolsvap|away | 12:17 | |
| *** kiran-r has quit IRC | 12:18 | |
| pc_m | It says that HEADS doesn't match migration timeline heads. Expected 24f... (new expand), 333...(new contract), kilo. In the file I have 24f..., 300...(old contract), 333..., 568...(old expand), kilo. | 12:19 |
| pc_m | correction... 300... is old expand, 568... is old contract. | 12:21 |
| *** salv-orl_ has joined #openstack-neutron | 12:22 | |
| *** oomichi has joined #openstack-neutron | 12:23 | |
| *** xek_ is now known as xek | 12:23 | |
| *** salv-orlando has quit IRC | 12:25 | |
| *** salv-orlando has joined #openstack-neutron | 12:26 | |
| pc_m | ihrachyshka: Is HEADS supposed to now have ALL migrations, or just the latest of each branch/type? | 12:26 |
| *** salv-orl_ has quit IRC | 12:26 | |
| pc_m | ihrachyshka: I guess I have the former and it should be the latter? | 12:26 |
| ihrachyshka | pc_m, latest. revision command should have updated the file for you | 12:27 |
| *** edmondsw has joined #openstack-neutron | 12:27 | |
| ihrachyshka | have you used the command? | 12:27 |
| *** salv-orl_ has joined #openstack-neutron | 12:27 | |
| pc_m | ihrachyshka: no, I was having trouble doing autogenerate last week, so I manually created the contract file, and then I split my old migration file into expand/contract parts. | 12:28 |
| ihrachyshka | pc_m, you can still avoid autogenerate and use it just to create files and update HEADS | 12:28 |
| ihrachyshka | pc_m, autogenerate is broken until alembic 0.8.0 (and a patch to consume it properly) | 12:29 |
| pc_m | ihrachyshka: autogenerate? I was having problems with it. | 12:29 |
| pc_m | ihrachyshka: Looks like check_migration is working now. | 12:29 |
| ihrachyshka | pc_m, revision command should work. if not, report a bug. | 12:29 |
| *** salv-orlando has quit IRC | 12:30 | |
| ihrachyshka | pc_m, also, afaik neutron-db-manage says what it wants to see in HEADS file on failure | 12:30 |
| *** Guest100_ has quit IRC | 12:31 | |
| *** trinaths has joined #openstack-neutron | 12:31 | |
| pc_m | ihrachyshka: I just tried upgrade using head. Should I say "upgrade head" or "upgrade liberty_expand@head" to test my new migration files? | 12:32 |
| *** ekuris has quit IRC | 12:32 | |
| *** mkolesni__ has quit IRC | 12:32 | |
| ihrachyshka | pc_m, you can try both. bare 'head' will apply all branches. | 12:32 |
| ihrachyshka | liberty_expand@head will apply expand only | 12:33 |
| *** johnbelamaric has quit IRC | 12:33 | |
| ihrachyshka | you can also try to trigger liberty_contract@head and see the order is correct (first deps from depends_on applied, then contract) | 12:33 |
| pc_m | ihrachyshka: It seems like it is running the contract migration (only) from 568... -> 333... which fails because table columns in expand migration file is not running. | 12:33 |
| *** salv-orlando has joined #openstack-neutron | 12:35 | |
| ihrachyshka | pc_m, well, it probably means there are issues with deps | 12:35 |
| *** numan has quit IRC | 12:35 | |
| ihrachyshka | depends_on missing in a contract script? | 12:35 |
| *** caspinol has quit IRC | 12:35 | |
| ihrachyshka | the one that would point to corresponding expand script | 12:35 |
| scheuran | neiljerr`: hi | 12:36 |
| pc_m | ihrachyshka: Ah, so contract will have down_revision pointing to previous contract, and depends_on pointing to corresponding expand. | 12:36 |
| scheuran | neiljerr`, in yesterdays neutron meeting, you mentioned, that with your dhcp agent you might be in a similar position like I'm with the macvtap stuff that has been discussed. | 12:36 |
| ihrachyshka | pc_m, yeap | 12:37 |
| scheuran | neiljerr`, could you please elaborate on that a little bit or/and send me a pointer to your code? | 12:37 |
| *** salv-orl_ has quit IRC | 12:38 | |
| *** regXboi has joined #openstack-neutron | 12:40 | |
| pc_m | ihrachyshka: That works now. Thanks! | 12:41 |
| *** azbiswas has joined #openstack-neutron | 12:42 | |
| ihrachyshka | pc_m, I am here to serve :) | 12:43 |
| pc_m | :) | 12:44 |
| *** salv-orl_ has joined #openstack-neutron | 12:44 | |
| openstackgerrit | Paul Michali proposed openstack/neutron-vpnaas: VPNaaS: Store local side's tunnel IP for later retrieval https://review.openstack.org/199670 | 12:46 |
| *** azbiswas has quit IRC | 12:47 | |
| *** salv-orlando has quit IRC | 12:47 | |
| *** salv-orl_ has quit IRC | 12:47 | |
| *** coolsvap|away is now known as coolsvap | 12:48 | |
| *** salv-orlando has joined #openstack-neutron | 12:48 | |
| jschwarz | ihrachyshka, lets discuss 'list bandwidth rules' | 12:48 |
| *** caspinol has joined #openstack-neutron | 12:49 | |
| ihrachyshka | jschwarz, in which context? | 12:49 |
| ihrachyshka | jschwarz, fyi I'm working on transforming it into generic rules list | 12:49 |
| *** yamamoto has joined #openstack-neutron | 12:49 | |
| jschwarz | ihrachyshka, currently the plugin signature is: 'def get_bandwidth_limit_rules(self, context, policy_id, **kwargs)' | 12:49 |
| ihrachyshka | ah, api side. ok | 12:50 |
| jschwarz | however the actual function doesn't filter by policy_id | 12:50 |
| jschwarz | implying that *all* bw limit rules are returned | 12:50 |
| jschwarz | this is... meh. | 12:50 |
| ihrachyshka | meh indeed | 12:50 |
| *** ChanServ changes topic to "Discussion of OpenStack Networking || for support join #openstack" | 12:50 | |
| -openstackstatus- NOTICE: zuul's disks were at capacity. Space has been freed up and jobs are being re-queued. | 12:50 | |
| jschwarz | ihrachyshka, I'll fix it into actually filtering by policy_id some time soon | 12:51 |
| jschwarz | ihrachyshka, what are you working on? is it on the api server side? | 12:51 |
| ihrachyshka | jschwarz, for the least, get_objects there should recieve qos_profile_id=policy_id | 12:51 |
| jschwarz | yeah | 12:51 |
| ihrachyshka | jschwarz, it's no object side for now. but I'll update api tests. from now on, policies won't have per type lists but a single list with all rules | 12:52 |
| jschwarz | oh uhm | 12:52 |
| jschwarz | uhmmm | 12:52 |
| ihrachyshka | per type lists don't make sense since there is a single rule of a type per policy should be allowed | 12:52 |
| jschwarz | uuhhm | 12:52 |
| *** baoli has joined #openstack-neutron | 12:52 | |
| ihrachyshka | don't ask me how it got thru spec review. | 12:53 |
| jschwarz | well that changes things for what i'm doing : | 12:53 |
| jschwarz | :\ | 12:53 |
| ihrachyshka | well maybe, but we must go this route | 12:53 |
| * HenryG reads backscroll and spots plenty of devref material in ihrachyshka's replies to pc_m | 12:53 | |
| jschwarz | ihrachyshka, eta on patch for review? | 12:53 |
| ihrachyshka | HenryG, meh. like the fact revision should be used to generate scripts? | 12:54 |
| ihrachyshka | jschwarz, an hour. at least code wise (not tests wise) | 12:54 |
| *** pal has joined #openstack-neutron | 12:54 | |
| ihrachyshka | maybe less | 12:55 |
| jschwarz | ihrachyshka, ok. do send a patch before the tests so I can see what it really effects | 12:55 |
| ihrachyshka | but I'm pessimistic these days | 12:55 |
| scheuran | sc68cal, hi | 12:55 |
| jschwarz | ihrachyshka, because the API changes i'm doing are pretty invasive | 12:55 |
| jschwarz | ihrachyshka, is there even logic for listing all bw rules for a specific policy_id? | 12:57 |
| *** dims has quit IRC | 12:57 | |
| jschwarz | is that even something we want to support? | 12:57 |
| HenryG | ihrachyshka: I was thinking you had some nice clarifications there for devs who don't use alembic every day. Maybe we could start a FAQ section. | 12:57 |
| jschwarz | because... uhm... we could just 'show policy' and that returns all the rules it has anyway. | 12:57 |
| *** salv-orl_ has joined #openstack-neutron | 12:57 | |
| *** annegentle has joined #openstack-neutron | 12:57 | |
| openstackgerrit | Neil Jerram proposed openstack/neutron: DHCP agent: clarify logic of setup_dhcp_port https://review.openstack.org/205181 | 12:57 |
| *** dims has joined #openstack-neutron | 12:57 | |
| *** nmagnezi_ has joined #openstack-neutron | 12:57 | |
| *** roeyc has joined #openstack-neutron | 12:58 | |
| *** veena has quit IRC | 12:58 | |
| ihrachyshka | jschwarz, there is. cls.get_objects(qos_policy_id=id) | 12:59 |
| ihrachyshka | you can pass any kwargs there | 12:59 |
| ihrachyshka | that would match with db | 12:59 |
| ihrachyshka | HenryG, yeah, I am generally for documenting it, maybe out of scope of the feature. | 12:59 |
| ihrachyshka | jschwarz, well, I think that's at least to be consistent. | 13:00 |
| *** salv-orlando has quit IRC | 13:00 | |
| *** salv-or__ has joined #openstack-neutron | 13:00 | |
| *** zhipeng has joined #openstack-neutron | 13:00 | |
| sc68cal | scheuran: hello | 13:01 |
| scheuran | sc68cal, hi. I wanted to hear your opinion on the macvtap integration that was in yesterdays neutron meeting | 13:01 |
| scheuran | sc68cal, so my main use case why we discussed that is reusing code | 13:02 |
| *** scheuran has quit IRC | 13:02 | |
| *** scheuran has joined #openstack-neutron | 13:02 | |
| *** trinaths has quit IRC | 13:02 | |
| jschwarz | ihrachyshka, ok.. so I'll try to implement something along the lines of 'http://10.35.6.13:9696/v2.0/bandwidth-limit-rules?policy_id=61467e66-983f-49a3-8ff5-319dcd3b4e5b' | 13:03 |
| jschwarz | no promises though ;p | 13:03 |
| scheuran | sc68cal: Do you think reusing code is a reason for integrating it into linuxbridge? | 13:03 |
| *** salv-orl_ has quit IRC | 13:04 | |
| scheuran | sc68cal, the code that is being reused is the main loop an the mechanism to detect new devices and so on (about 500 lines) | 13:04 |
| sc68cal | scheuran: I haven't reviewed any of the macvtap stuff, but in general I think it's better to re-use than re-implement whole new agents | 13:05 |
| *** wei-wang_ has quit IRC | 13:05 | |
| sc68cal | scheuran: is this the related spec? https://review.openstack.org/#/c/117505/ | 13:05 |
| *** shwetaap has joined #openstack-neutron | 13:06 | |
| scheuran | sc68cal, no my inital plan was to start in a separate repo, spec is on github, let me paste you the link | 13:06 |
| *** shwetaap1 has joined #openstack-neutron | 13:07 | |
| scheuran | sc68cal, https://github.com/scheuran/networking-macvtap/commit/bc239419398ec039da35ec949a14691280733bf6 | 13:07 |
| *** oomichi has quit IRC | 13:08 | |
| scheuran | sc68cal, but this still assumes that everything lands in a separate repo | 13:08 |
| scheuran | sc68cal, the discussion regarding linuxbridge came up, on the request of the networking-macvtap repo | 13:08 |
| scheuran | sc68cal, this is a prototype which integrates macvtap support into github: https://github.com/scheuran/networking-macvtap/commit/36a068cf3d3d6930ab9330efb099cd95a84ca785 | 13:09 |
| *** salv-or__ has quit IRC | 13:09 | |
| scheuran | sc68cal, the macvtap part is fully functional - haven't tested the lb part of it yet - but it should run, too (maybe some tiny fixes) | 13:09 |
| *** lindycoder has joined #openstack-neutron | 13:10 | |
| scheuran | sc68cal, probably there would be a config option to either run the agent in macvtap mode or linuxbridge mode | 13:11 |
| *** salv-orlando has joined #openstack-neutron | 13:11 | |
| *** shwetaap has quit IRC | 13:11 | |
| *** phschwartz has quit IRC | 13:11 | |
| *** phschwartz has joined #openstack-neutron | 13:12 | |
| *** dboik has joined #openstack-neutron | 13:12 | |
| *** dave-mccowan has joined #openstack-neutron | 13:13 | |
| *** nyechiel_ has quit IRC | 13:14 | |
| *** coolsvap is now known as coolsvap|away | 13:14 | |
| sc68cal | scheuran: I would prefer to see the macvtap agent subclass the linux bridge agent instead of being a direct copy - although that may not be possible currently. There has already been some discussion about a common agent framework since there is overlap in the code for the OVS agent and the LB agent - I think the SR-IOV agent, or the DPDK agent (can't remember which) was a copy of the OVS agent c | 13:14 |
| sc68cal | ode already | 13:14 |
| *** salv-orl_ has joined #openstack-neutron | 13:14 | |
| sc68cal | so we've got a lot of agents floating around that are descendants of the OVS and LB agents | 13:14 |
| openstackgerrit | Roey Chen proposed openstack/neutron: Don't use autonested_transaction when creating default security-group https://review.openstack.org/207034 | 13:16 |
| *** salv-orlando has quit IRC | 13:17 | |
| scheuran | sc68cal, yes, all agents have a very similar base logic | 13:17 |
| *** dane_leblanc has joined #openstack-neutron | 13:18 | |
| *** salv-orlando has joined #openstack-neutron | 13:18 | |
| scheuran | sc68cal, so you would prefer to integrate it into lb instead of having a own repo? | 13:18 |
| scheuran | sc68cal, or more having a own agent but only subclassing linuxbridge? | 13:19 |
| sc68cal | scheuran: I would, however I don't know if we've given you good enough tools and interfaces to make that possible - since the modular l2 work is still in discussion currently. | 13:19 |
| *** lucas__ has joined #openstack-neutron | 13:19 | |
| scheuran | sc68cal, yes that's the problem | 13:20 |
| sc68cal | scheuran: so, I see three choices | 13:20 |
| sc68cal | scheuran: 1) continue with your agent for the time being | 13:21 |
| sc68cal | scheuran: 2) Work to try and refactor the LB agent to make macvtap able to be a subclass | 13:21 |
| sc68cal | scheuran: 3) Wait for modular L2 to be completed, and use #1 for the time being | 13:21 |
| *** salv-orl_ has quit IRC | 13:21 | |
| sc68cal | scheuran: #2 would actually help map out what work would need to be done for modular l2 agent | 13:22 |
| sc68cal | scheuran: and could form the basis for a modular agent framework | 13:22 |
| scheuran | sc68cal, to 1) I see to flavors: a) having it in neutron tree as own agent b) having it in a separate networking-macvtap repo | 13:24 |
| scheuran | sc68cal, to 2) this is actually what I did in the prototype. But I'm not sure if it is going into the direction of the ml2 agent that is proposed | 13:25 |
| *** ArchiFleKs has quit IRC | 13:25 | |
| scheuran | sc68cal, the ml2 agent concept seems to be much more complex | 13:25 |
| *** helenxu has joined #openstack-neutron | 13:26 | |
| scheuran | sc68cal, to 3) is the one that would get me started quickly | 13:26 |
| -openstackstatus- NOTICE: zuul jobs after about 07:00 UTC may need a 'recheck' to enter the queue. Look if your change is in http://status.openstack.org/zuul/ and recheck if not. | 13:26 | |
| sc68cal | scheuran: my main concern is that with your own agent you're going to accumulate technical debt - since you'll not get fixes we make to the lb agent in tree | 13:27 |
| scheuran | sc68cal, if we go into the direction of modular l2 agent, I would prefer to start on a green field | 13:27 |
| jschwarz | ihrachyshka, great success. | 13:27 |
| scheuran | sc68cal, just to get things forward (that's my personal opinion) | 13:28 |
| scheuran | sc68cal, I totally share your concern | 13:28 |
| ihrachyshka | jschwarz, tell me. I'm about to push the rules patch too | 13:28 |
| jschwarz | ihrachyshka, managed to convert all the bw rules API URIs to /qos/bandwidth_rules_limit%s | 13:29 |
| jschwarz | ihrachyshka, and 'list bw rules' is now /qos/bandwidth_rules_limit?policy_id=%s | 13:29 |
| jschwarz | so that's nice | 13:29 |
| sc68cal | scheuran: I'm going to grab some breakfast, should we continue on the ML? | 13:29 |
| *** yapeng has joined #openstack-neutron | 13:30 | |
| *** yfried_ has joined #openstack-neutron | 13:30 | |
| scheuran | sc68cal, ok. If you think this is the right place, sure I'll be there | 13:31 |
| scheuran | sc68cal, I'll put an item on the agenda with all the relevant information | 13:31 |
| scheuran | sc68cal, thanks a lot! | 13:31 |
| *** yfried has quit IRC | 13:31 | |
| *** nmagnezi_ has quit IRC | 13:32 | |
| jschwarz | ihrachyshka, AND all the API tests are successful :) | 13:32 |
| jschwarz | so today was a good day | 13:32 |
| ihrachyshka | /qos/bandwidth_rules_limit?policy_id=%s ?? | 13:33 |
| jschwarz | yeah | 13:33 |
| ihrachyshka | why limit in the end? | 13:33 |
| *** kbringard has joined #openstack-neutron | 13:33 | |
| jschwarz | ihrachyshka, ah sorry | 13:33 |
| ihrachyshka | :) | 13:34 |
| jschwarz | ihrachyshka, /qos/bandwidth_limit_rules?policy_id=%s | 13:34 |
| *** rpothier has joined #openstack-neutron | 13:34 | |
| ihrachyshka | that's better for my taste. I suppose it makes things easier on client side? | 13:34 |
| jschwarz | ihrachyshka, we're about to see | 13:35 |
| ihrachyshka | jschwarz, https://review.openstack.org/#/c/207043/ may you check API changes whether they make sense? | 13:35 |
| jschwarz | ihrachyshka, need to do a bit of coding there to adjust to the new URIs | 13:35 |
| ihrachyshka | jschwarz, I'm lost in how API manages those URIs, you are probably the expert | 13:35 |
| jschwarz | 'expert' is a big word, but thanks ;-) | 13:35 |
| ihrachyshka | heh | 13:35 |
| ihrachyshka | jlibosva, https://review.openstack.org/#/c/207043/ please look too | 13:36 |
| jschwarz | ihrachyshka, https://review.openstack.org/#/c/207043/1/neutron/extensions/qos.py -- ouch | 13:36 |
| ihrachyshka | jschwarz, well, that's the point - having rules instead of per type rules lists | 13:36 |
| jschwarz | ihrachyshka, did you run the API tests? | 13:36 |
| ihrachyshka | jschwarz, no, it's for sissies! | 13:36 |
| jschwarz | ihrachyshka, lol | 13:36 |
| *** annegentle has quit IRC | 13:36 | |
| jschwarz | ihrachyshka, so yeah, you basically changed half of the neutronclient's second patch just now. | 13:36 |
| ihrachyshka | jschwarz, http://bbuniverse.net/wp-content/uploads/2013/09/Im_Sorry.gif | 13:37 |
| ihrachyshka | moshele, https://review.openstack.org/#/c/207043/ please | 13:37 |
| jschwarz | ihrachyshka, so wait, you were serious about not running those API tests? | 13:38 |
| ihrachyshka | jschwarz, I am | 13:38 |
| ihrachyshka | I don't believe it will work | 13:38 |
| ihrachyshka | posted just for reference | 13:38 |
| ihrachyshka | unit tests work, that's all | 13:38 |
| *** lucas__ has quit IRC | 13:38 | |
| *** lucas__ has joined #openstack-neutron | 13:39 | |
| moshele | ihrachyshka: maybe we should wait until ajo is back with this change? | 13:39 |
| jschwarz | ihrachyshka, ;p; | 13:39 |
| jschwarz | ihrachyshka, lol | 13:39 |
| *** johnbelamaric has joined #openstack-neutron | 13:40 | |
| ihrachyshka | moshele, I don't believe so. unless we want to miss L | 13:40 |
| *** johnbelamaric has quit IRC | 13:40 | |
| jschwarz | ihrachyshka, so following that patch we need to rewrite also a large part of the api tests, the api server and the neutronclient. | 13:40 |
| *** johnbelamaric has joined #openstack-neutron | 13:40 | |
| jschwarz | or do you? | 13:40 |
| ihrachyshka | jschwarz, why the large part? | 13:40 |
| ihrachyshka | jschwarz, well, I wonder whether you can handle it while I'm on some irrelevant stuff ;) you have all in place, right? | 13:41 |
| jschwarz | ihrachyshka, erm | 13:41 |
| jschwarz | ihrachyshka, let me just run the API tests for a second | 13:41 |
| ihrachyshka | I am not sure I have env to run API ones just now. | 13:42 |
| ihrachyshka | it will take me some time to set auth for keystone that is broken in my devstack | 13:42 |
| jschwarz | ihrachyshka, lol | 13:43 |
| jschwarz | ihrachyshka, ahahahah | 13:43 |
| jschwarz | ihrachyshka, $ nosetests neutron.tests.api.test_qos -s | 13:43 |
| jschwarz | FFFF.FFFFFFFF | 13:43 |
| jschwarz | XD | 13:43 |
| jschwarz | nmagnezi, ^ | 13:43 |
| nmagnezi | jschwarz, looks like your keyboard is jammed :) | 13:44 |
| *** markvoelker has joined #openstack-neutron | 13:44 | |
| jschwarz | ihrachyshka, OK, so the API tests pass (had to re-create the database) | 13:45 |
| jschwarz | So the change is basically renaming the 'bandwidth_limit_rules' to 'rules' in the policy object? | 13:46 |
| * jlibosva looking | 13:47 | |
| *** sridhar_ram has joined #openstack-neutron | 13:47 | |
| jschwarz | alright, so forget about the 'rewrite large parts of...' part ;p | 13:47 |
| ihrachyshka | jschwarz, passed with no changes/ | 13:48 |
| ihrachyshka | ? | 13:48 |
| jschwarz | ihrachyshka, yes | 13:48 |
| jschwarz | :) well done | 13:48 |
| ihrachyshka | huh see! | 13:48 |
| jschwarz | ! | 13:48 |
| ihrachyshka | I haven't even executed anything out of neutron.tests.unit.objects | 13:48 |
| jlibosva | ihrachyshka: wait - I'm confused | 13:48 |
| ihrachyshka | jschwarz, yes, that's the change - rename | 13:48 |
| *** jasondotstar has joined #openstack-neutron | 13:48 | |
| jlibosva | ihrachyshka: about the review you sent | 13:49 |
| jlibosva | ihrachyshka: how are we gonna create a specific rule with type via API then? | 13:49 |
| jschwarz | ihrachyshka, let me post my changes | 13:49 |
| ihrachyshka | jlibosva, which one :) I have like 30+ of them now :) | 13:49 |
| jlibosva | https://review.openstack.org/#/c/207043/2 | 13:49 |
| jlibosva | ihrachyshka: ^^ | 13:49 |
| jlibosva | disregard | 13:49 |
| jlibosva | I'm blind | 13:49 |
| ihrachyshka | jlibosva, create is different. there is still /qos/bandwidth_limit_rule API entrypoint | 13:50 |
| ihrachyshka | it only renames policy fields | 13:50 |
| jlibosva | yeah, I got confused by the resource attribute map | 13:50 |
| jlibosva | ok, I'm confused again :) | 13:50 |
| jschwarz | ihrachyshka, https://review.openstack.org/#/c/207053/ | 13:51 |
| *** roeyc has quit IRC | 13:51 | |
| *** roeyc has joined #openstack-neutron | 13:51 | |
| *** salv-orlando has quit IRC | 13:52 | |
| *** jasondot_ has quit IRC | 13:52 | |
| ihrachyshka | jlibosva, it may be the case that the one confused is me | 13:52 |
| ihrachyshka | I am bad at that API layer | 13:52 |
| jlibosva | ihrachyshka: https://review.openstack.org/#/c/203963/12/neutron/extensions/qos.py - I remember that the API didn't work without this addition | 13:53 |
| jlibosva | ihrachyshka: I mean the bandwidth limit rules weren't paired with policy | 13:53 |
| *** fawadkhaliq has quit IRC | 13:53 | |
| ihrachyshka | jlibosva, I don't believe so, but anyway, now we have api tests to validate :) | 13:54 |
| jschwarz | ihrachyshka, jlibosva, the path Ihar wrote earlier is correct. | 13:54 |
| jlibosva | so when we want to create a limit rule, we still should use POST qos/policies/<policy_id>/bandwidth_limit_rules ? | 13:54 |
| jschwarz | jlibosva, no. POST /qos/bandwidth_limit_rules | 13:54 |
| *** LukeHinds has joined #openstack-neutron | 13:54 | |
| jschwarz | policy_id goes in the json data | 13:55 |
| jlibosva | whaaat? | 13:55 |
| jlibosva | then the resource map is completely wrong? | 13:55 |
| jschwarz | jlibosva, see: https://review.openstack.org/#/c/207053/1/neutron/tests/tempest/services/network/json/network_client.py line 657 | 13:55 |
| jlibosva | jschwarz: so what about https://review.openstack.org/#/c/207043/2/neutron/extensions/qos.py L69? | 13:56 |
| jschwarz | jlibosva, that URI is correct only following patch 207053 | 13:56 |
| jlibosva | jschwarz: aaaah | 13:56 |
| *** fawadkhaliq has joined #openstack-neutron | 13:56 | |
| jschwarz | jlibosva, that's ok - it's a different attribute that's relevant only to /qos/policies/ based URIs | 13:57 |
| jlibosva | jschwarz: I didn't know about your patch that removes the parent | 13:57 |
| jschwarz | so 'show me a policy' will now give you a list of rules instead of a bandwidth_limit_rules | 13:57 |
| jlibosva | now I'm getting the picture | 13:57 |
| jschwarz | XD | 13:57 |
| *** yamahata has joined #openstack-neutron | 13:57 | |
| *** mwagner_lap has joined #openstack-neutron | 13:58 | |
| jschwarz | right, I need to dive into the neutronclient code | 13:58 |
| jlibosva | so...that changes api A LOT | 13:58 |
| jschwarz | ping me if I'm needed | 13:58 |
| *** jckasper has quit IRC | 13:58 | |
| jschwarz | jlibosva, yes | 13:58 |
| *** ferest has joined #openstack-neutron | 13:58 | |
| *** evrardjp has quit IRC | 13:59 | |
| *** ferest has quit IRC | 13:59 | |
| *** lukasa has quit IRC | 13:59 | |
| *** mragupat has joined #openstack-neutron | 13:59 | |
| *** mragupat has quit IRC | 13:59 | |
| *** breitz has joined #openstack-neutron | 13:59 | |
| *** fawadkhaliq has quit IRC | 14:00 | |
| *** mragupat has joined #openstack-neutron | 14:00 | |
| *** fawadkhaliq has joined #openstack-neutron | 14:01 | |
| jschwarz | ihrachyshka, do we have a weekly QoS meeting? | 14:02 |
| moshele | ihrachyshka: do we have qos meeting? | 14:02 |
| jschwarz | jlibosva, can you ping ihrachyshka re weekly qos meeting? | 14:03 |
| *** markvoelker has quit IRC | 14:04 | |
| *** q3k has quit IRC | 14:05 | |
| *** fouxm has quit IRC | 14:05 | |
| *** dr_gogeta86 has quit IRC | 14:05 | |
| *** EmilienM has quit IRC | 14:05 | |
| *** evrardjp has joined #openstack-neutron | 14:06 | |
| *** dr_gogeta86 has joined #openstack-neutron | 14:06 | |
| *** dr_gogeta86 has quit IRC | 14:06 | |
| *** dr_gogeta86 has joined #openstack-neutron | 14:06 | |
| ihrachyshka | ouch | 14:06 |
| *** fouxm has joined #openstack-neutron | 14:07 | |
| jlibosva | my apologies, I bugged ihrachyshka with complains :) | 14:07 |
| *** y_sawai has joined #openstack-neutron | 14:07 | |
| *** arajagopal has joined #openstack-neutron | 14:07 | |
| *** zhipeng has quit IRC | 14:08 | |
| *** garyk has quit IRC | 14:09 | |
| *** tfukushima has joined #openstack-neutron | 14:09 | |
| *** EmilienM has joined #openstack-neutron | 14:09 | |
| *** MentalRay_ has joined #openstack-neutron | 14:10 | |
| *** caspinol has quit IRC | 14:10 | |
| *** nyechiel_ has joined #openstack-neutron | 14:11 | |
| *** yfried_ has quit IRC | 14:12 | |
| *** q3k has joined #openstack-neutron | 14:13 | |
| *** yfried_ has joined #openstack-neutron | 14:13 | |
| *** annegentle has joined #openstack-neutron | 14:14 | |
| *** gsagie_ has joined #openstack-neutron | 14:15 | |
| openstackgerrit | Cedric Brandily proposed openstack/neutron: DO NOT MERGE: search failing py34 tests https://review.openstack.org/207066 | 14:15 |
| openstackgerrit | Merged openstack/neutron: Allow overriding of the neutron endpoint URL in metadata agent https://review.openstack.org/192949 | 14:16 |
| *** eckelcu has joined #openstack-neutron | 14:16 | |
| *** padkrish has joined #openstack-neutron | 14:17 | |
| *** Miouge has joined #openstack-neutron | 14:18 | |
| *** l6unchpad has quit IRC | 14:19 | |
| *** arcimboldo has joined #openstack-neutron | 14:19 | |
| *** eckelcu has quit IRC | 14:20 | |
| *** markvoelker_ has joined #openstack-neutron | 14:21 | |
| *** eckelcu has joined #openstack-neutron | 14:22 | |
| *** Murali has quit IRC | 14:22 | |
| *** padkrish has quit IRC | 14:23 | |
| *** l6unchpad has joined #openstack-neutron | 14:24 | |
| openstackgerrit | Cedric Brandily proposed openstack/neutron: Python 3: set webob.Response.text instead of webob.Response.body https://review.openstack.org/204791 | 14:24 |
| *** absubram has joined #openstack-neutron | 14:25 | |
| openstackgerrit | Neil Jerram proposed openstack/neutron: DHCP agent: let interface driver control dnsmasq bind options https://review.openstack.org/206077 | 14:25 |
| *** davidjc has joined #openstack-neutron | 14:25 | |
| openstackgerrit | Cedric Brandily proposed openstack/neutron: DO NOT MERGE: search failing py34 tests https://review.openstack.org/207066 | 14:26 |
| *** r-daneel has joined #openstack-neutron | 14:26 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/neutron: Updated from global requirements https://review.openstack.org/206825 | 14:27 |
| *** padkrish has joined #openstack-neutron | 14:28 | |
| *** yamamoto has quit IRC | 14:29 | |
| *** yamamoto has joined #openstack-neutron | 14:29 | |
| *** baoli has quit IRC | 14:31 | |
| openstackgerrit | Sean M. Collins proposed openstack/neutron: Linux Bridge Agent: Import utils via neutron.agent.common https://review.openstack.org/207071 | 14:31 |
| *** baoli has joined #openstack-neutron | 14:32 | |
| *** padkrish has quit IRC | 14:32 | |
| *** padkrish has joined #openstack-neutron | 14:33 | |
| *** caspinol has joined #openstack-neutron | 14:34 | |
| *** lajos-katona1 has left #openstack-neutron | 14:36 | |
| *** coolsvap|away is now known as coolsvap | 14:37 | |
| *** rossella_s has quit IRC | 14:37 | |
| *** yamamoto has quit IRC | 14:38 | |
| *** rossella_s has joined #openstack-neutron | 14:38 | |
| *** markvoelker_ has quit IRC | 14:38 | |
| *** dkehn has quit IRC | 14:39 | |
| *** markvoelker has joined #openstack-neutron | 14:39 | |
| *** yamahata has quit IRC | 14:41 | |
| *** yapeng has quit IRC | 14:41 | |
| openstackgerrit | Cedric Brandily proposed openstack/neutron: Python 3: use a hash to sort dictionaries https://review.openstack.org/204792 | 14:45 |
| arcimboldo | hi all: I have L3 neutron in HA (vrrp). When I create the external network, I see 2 ports ACTIVE, but after a while one of the two becomes BUILD and on neutron I have DBDuplicateEntry | 14:45 |
| *** nlahouti has joined #openstack-neutron | 14:45 | |
| arcimboldo | it seems that neutron is trying to update the status from BUILD to ACTIVE but its' complaining that the id of the port is duplicate, but it's not | 14:45 |
| *** pal has quit IRC | 14:45 | |
| openstackgerrit | Cedric Brandily proposed openstack/neutron: Python 3: use a hash to sort dictionaries https://review.openstack.org/204792 | 14:45 |
| *** nmagnezi has quit IRC | 14:46 | |
| *** nlahouti has quit IRC | 14:46 | |
| *** pal has joined #openstack-neutron | 14:46 | |
| amuller | arcimboldo: are you talking about the port of the router on the external network? why do you have two? | 14:46 |
| ihrachyshka | ddepaoli, hi | 14:46 |
| ddepaoli | hi | 14:47 |
| *** markvoelker has quit IRC | 14:47 | |
| ihrachyshka | ddepaoli, so you can try some pieces, but agent side is still not there (though we have patches in fly that need to go into gate once unblocked) | 14:47 |
| ihrachyshka | ddepaoli, to check it you would configure devstack with NEUTRON_BRANCH=feature/qos | 14:47 |
| *** markvoelker has joined #openstack-neutron | 14:47 | |
| ihrachyshka | ddepaoli, also, qos service_plugin should be on | 14:47 |
| *** gcossu has joined #openstack-neutron | 14:48 | |
| *** ZZelle has quit IRC | 14:48 | |
| ihrachyshka | ddepaoli, see how we configure it in gate: https://review.openstack.org/#/c/206317/2/jenkins/jobs/devstack-gate.yaml | 14:48 |
| ddepaoli | so enable_service qos? | 14:48 |
| arcimboldo | amuller, it's a HA setup, so neutorn it's creating 2 ports, for the 2 neutron routers created on the network nodes | 14:48 |
| ihrachyshka | ddepaoli, no, Q_SERVICE_PLUGIN_CLASSES=qos | 14:48 |
| ihrachyshka | ddepaoli, it's not a separate service, it's a plugin for neutron | 14:48 |
| *** crose has joined #openstack-neutron | 14:48 | |
| *** vivek-ebay has joined #openstack-neutron | 14:49 | |
| amuller | arcimboldo: sorry when you do neutron port-list --network-id=<external network> --device-id=<id of router> do you see 1 port or 2? | 14:49 |
| ihrachyshka | ddepaoli, what is missing to test it is that patch: https://review.openstack.org/#/c/206525/ | 14:49 |
| amuller | arcimboldo: or are you talking about the 'qg' device in the router namespace? | 14:49 |
| *** johnbelamaric has quit IRC | 14:49 | |
| ihrachyshka | ddepaoli, and even with that, your agents won't see updates to qos policies unless you restart the agent | 14:49 |
| ihrachyshka | ddepaoli, for client side, you can check with jschwarz on what's the best way | 14:49 |
| arcimboldo | amuller, sorry: I'm mixing router and dhcp... | 14:50 |
| arcimboldo | rewind: two ports are created, one for each dhcp agent | 14:50 |
| jschwarz | ddepaoli, client side isn't ready yet so I'd hold on that for a few more days | 14:50 |
| amuller | arcimboldo: yeah for L3 HA it's different | 14:50 |
| arcimboldo | (it's an external network, so no router) | 14:50 |
| ddepaoli | ok, thank you <ihrachyshka> | 14:50 |
| *** pal has quit IRC | 14:50 | |
| amuller | arcimboldo: ok so the ports with the DBDuplicateEntry error, is that a DHCP port or a router port? | 14:51 |
| arcimboldo | dhcp port | 14:51 |
| ddepaoli | ihrachysha, is this link https://etherpad.openstack.org/p/neutron-liberty-qos-code-sprint a useful starting point? | 14:51 |
| arcimboldo | (sorry for the confusion) | 14:51 |
| arcimboldo | however, it happens also with ports related to VM I start | 14:52 |
| amuller | arcimboldo: you mean when you spawn a VM (And thus a new Neutron port is created) you're also seeing that traceback in the neutron server log? | 14:52 |
| amuller | is it for every VM or only some of them? | 14:52 |
| arcimboldo | amuller, it happens when I start *many* vms, ~100 at once | 14:53 |
| arcimboldo | only some of them | 14:53 |
| *** u_glide2 has joined #openstack-neutron | 14:54 | |
| arcimboldo | I'm testing with Rally, I'm usually get ~97% success when starting 100 vms at once, 300 vms total | 14:54 |
| ihrachyshka | ddepaoli, I don't believe so. what are you interested in there, specifically? | 14:54 |
| *** yamamoto has joined #openstack-neutron | 14:54 | |
| amuller | arcimboldo: what error do you get in that case? what's the CPU usage of the neutron-server process? are you using api or rpc workers? | 14:54 |
| ihrachyshka | ddepaoli, do you want to play with it or debug/code it? | 14:54 |
| gcossu | jschwarz: so, do you think next week it's possible to test/review the neutronclient? | 14:54 |
| openstackgerrit | Cedric Brandily proposed openstack/neutron: Python 3: hmac requires bytes key/msg https://review.openstack.org/206993 | 14:54 |
| *** vivek-ebay has quit IRC | 14:55 | |
| u_glide2 | mestery, iwamoto: ping | 14:55 |
| openstackgerrit | Cedric Brandily proposed openstack/neutron: Python 3: hmac requires bytes key/msg https://review.openstack.org/206993 | 14:55 |
| *** sridhar_ram has quit IRC | 14:55 | |
| jschwarz | gcossu, yes | 14:55 |
| *** absubram has quit IRC | 14:55 | |
| u_glide2 | mestery, iwamoto: hello folks, Do you plan to merge AZ support in liberty? | 14:55 |
| arcimboldo | from the log, for instance, I see 6561 DuplicateEntry errors on a port that was a dhcp port | 14:56 |
| arcimboldo | then I deleted and re-created the network | 14:56 |
| gcossu | jschwarz: ;) | 14:56 |
| amuller | arcimboldo: you're mixing up different issues I think | 14:57 |
| ddepaoli | ihrachyshka, I'm interested in testing and coding | 14:57 |
| *** dsneddon has quit IRC | 14:57 | |
| *** dsneddon has joined #openstack-neutron | 14:58 | |
| *** mickeys has joined #openstack-neutron | 14:58 | |
| ihrachyshka | ddepaoli, oh so we have new candidates! cool. ok, so for the devstack start, the thing I told should be enough. for the time being, you may use curl to access API, or play with neutronclient patches that are in review (though they are broken, at least on rules side) | 14:58 |
| ddepaoli | ihrachyshka, I developed a very simple qos for neutron last year https://github.com/SmartInfrastructures/neutron/tree/qos-dev | 14:58 |
| ihrachyshka | ddepaoli, there are a lot of TODO(QoS) in the tree, you may try to choose one and kill it :) | 14:59 |
| *** absubram has joined #openstack-neutron | 14:59 | |
| *** tfukushima has quit IRC | 14:59 | |
| *** jasondotstar has quit IRC | 14:59 | |
| arcimboldo | amuller, what do you mean? I don't understand if there are two different issues or just one, that's why I'm here :) | 14:59 |
| *** slaweq has quit IRC | 14:59 | |
| ihrachyshka | ddepaoli, also, take a look at: http://git.openstack.org/cgit/openstack/neutron/tree/doc/source/devref/quality_of_service.rst?h=feature/qos | 15:00 |
| amuller | arcimboldo: when do you get the duplicate entry error? only on DHCP ports? after what operation / how do you trigger it? | 15:00 |
| openstackgerrit | Sridhar Gaddam proposed openstack/neutron: Update dhcp agent cache for network:dhcp ports https://review.openstack.org/205888 | 15:00 |
| *** jistr has quit IRC | 15:00 | |
| *** ByPasS has quit IRC | 15:01 | |
| *** slaweq has joined #openstack-neutron | 15:01 | |
| *** mragupat has quit IRC | 15:01 | |
| *** edwarnicke has quit IRC | 15:02 | |
| *** rex_lee has quit IRC | 15:02 | |
| *** sweston has quit IRC | 15:02 | |
| *** dougwig has quit IRC | 15:02 | |
| *** mestery has joined #openstack-neutron | 15:02 | |
| *** carl_baldwin has joined #openstack-neutron | 15:02 | |
| arcimboldo | amuller, I still need to re-do some tests, but I know I have got the duplicate entry for DHCP ports | 15:02 |
| *** jistr has joined #openstack-neutron | 15:02 | |
| *** ByPasS has joined #openstack-neutron | 15:02 | |
| *** dsanders has joined #openstack-neutron | 15:03 | |
| ddepaoli | ok ihrachyshka , I'll start with testing and debugging and then I should start to complete some TODOs | 15:04 |
| *** phil_h has quit IRC | 15:04 | |
| ihrachyshka | ddepaoli, cool, thanks a lot. ask anything you need, I am around :) | 15:04 |
| ihrachyshka | I'm in UTC+1 TZ btw | 15:04 |
| *** alex_klimov has quit IRC | 15:04 | |
| *** itzikb has quit IRC | 15:05 | |
| *** phil_h has joined #openstack-neutron | 15:05 | |
| arcimboldo | amuller, pardon me... I think I've confused two IPs... it was actually a L3 HA port... | 15:06 |
| arcimboldo | 2015-07-29 13:48:19.113 85940 INFO neutron.plugins.openvswitch.agent.ovs_neutron_agent [req-46fddd63-9530-4dcf-aa92-c665e547045b ] Port 11443d24-3548-4f0f-acd4-6132a00279f7 updated. Details: {u'profile': {}, u'allowed_address_pairs': [], u'admin_state_up': True, u'network_id': u'a5c39294-b098-41d1-8771-78ccad8530a2', u'segmentation_id': 842, u'device_owner': u'network:router_gateway', u'physical_network': u'vlannet', | 15:06 |
| arcimboldo | u'mac_address': u'fa:16:3e:28:34:c4', u'device': u'11443d24-3548-4f0f-acd4-6132a00279f7', u'port_security_enabled': True, u'port_id': u'11443d24-3548-4f0f-acd4-6132a00279f7', u'fixed_ips': [{u'subnet_id': u'20f6ceaf-9f55-4f46-a8f6-a6ec0a0eddef', u'ip_address': u'172.23.2.2'}], u'network_type': u'vlan'} | 15:06 |
| arcimboldo | so it has to be the port of the router connected to the external network | 15:07 |
| ddepaoli | ihrachyshka, ok! thank you very much | 15:07 |
| *** johnbelamaric has joined #openstack-neutron | 15:07 | |
| *** eckelcu has quit IRC | 15:09 | |
| *** cdl has quit IRC | 15:09 | |
| *** cdl has joined #openstack-neutron | 15:10 | |
| amuller | arcimboldo: I think you need to double down and investigate some more, and when you've identified a specific issue come back here or on ask.openstack.org | 15:10 |
| *** eckelcu has joined #openstack-neutron | 15:10 | |
| amuller | arcimboldo: because you're talking about DHCP ports, L3 HA ports and VMs failing to spawn, it's difficult =D | 15:10 |
| *** roeyc has quit IRC | 15:11 | |
| *** alex_klimov has joined #openstack-neutron | 15:14 | |
| *** jckasper has joined #openstack-neutron | 15:15 | |
| *** padkrish has quit IRC | 15:15 | |
| *** josecastroleon has quit IRC | 15:15 | |
| *** jprovazn has quit IRC | 15:16 | |
| *** gcossu has quit IRC | 15:16 | |
| *** phil_h has quit IRC | 15:17 | |
| *** carl_baldwin has quit IRC | 15:18 | |
| *** SridharG has left #openstack-neutron | 15:18 | |
| arcimboldo | amuller, I've just replicated the issue. | 15:19 |
| arcimboldo | amuller, I have a Duplicate Entry error on a DHCP port, the port is in DOWN | 15:19 |
| arcimboldo | I have dhcp_agents_per_network = 2 in neutron.conf | 15:19 |
| *** carl_baldwin has joined #openstack-neutron | 15:19 | |
| arcimboldo | the port was in ACTIVE before running a rally benchmark | 15:20 |
| *** iymht has joined #openstack-neutron | 15:20 | |
| arcimboldo | so, to reacp: I have this duplicate error for ports DHCP, L3 _and_ VMs | 15:20 |
| *** ronis has quit IRC | 15:20 | |
| *** fzdarsky has joined #openstack-neutron | 15:23 | |
| amuller | arcimboldo: that's definitely separate issues | 15:23 |
| *** carl_baldwin has quit IRC | 15:23 | |
| *** azbiswas has joined #openstack-neutron | 15:24 | |
| arcimboldo | amuller, good, so how can I debug these issue? | 15:24 |
| amuller | arcimboldo: try to reproduce manually without rally, as simply as possible | 15:24 |
| amuller | imo | 15:24 |
| amuller | find a minimal reproducer | 15:24 |
| *** Miouge has quit IRC | 15:25 | |
| *** ddepaoli has quit IRC | 15:25 | |
| arcimboldo | amuller, :( I'm having problems only when I start many VMs concurrently | 15:27 |
| amuller | arcimboldo: can you paste the trace you're hitting? | 15:29 |
| amuller | arcimboldo: on pastebin? | 15:29 |
| *** azbiswas has quit IRC | 15:29 | |
| *** ijw has joined #openstack-neutron | 15:29 | |
| *** lucas__ has quit IRC | 15:30 | |
| arcimboldo | amuller, sure: http://pastebin.com/aksNderA | 15:31 |
| *** baoli has quit IRC | 15:31 | |
| *** banix has joined #openstack-neutron | 15:31 | |
| *** lucas__ has joined #openstack-neutron | 15:32 | |
| arcimboldo | the trace is relative to one of the 2 DHCP port for a vlan network | 15:32 |
| *** mlavalle has joined #openstack-neutron | 15:33 | |
| *** cdl has quit IRC | 15:33 | |
| *** alexpilotti has quit IRC | 15:34 | |
| *** lucas__ has quit IRC | 15:37 | |
| *** baoli has joined #openstack-neutron | 15:39 | |
| *** lucas____ has joined #openstack-neutron | 15:41 | |
| *** vishwanathj has quit IRC | 15:41 | |
| *** dannywilson has joined #openstack-neutron | 15:42 | |
| arcimboldo | btw: how can I fix it now, that I have the port DOWN and neutron is unable to set it to ACTIVE? | 15:44 |
| amuller | arcimboldo: that's a weird trace | 15:45 |
| *** devvesa has joined #openstack-neutron | 15:45 | |
| arcimboldo | amuller, why? | 15:45 |
| *** lucas____ has quit IRC | 15:46 | |
| arcimboldo | is it relevant saying that we are running galera? | 15:46 |
| amuller | arcimboldo: perhaps | 15:46 |
| arcimboldo | btw: what's strange to me is that it seems it's updating an entry, so the primary key shouldn't be updated | 15:46 |
| amuller | arcimboldo: still that trace is saying you have two records in that table with the same ID, which, is quite odd... | 15:46 |
| amuller | arcimboldo: how does neutron port-list looking? | 15:47 |
| amuller | is* | 15:47 |
| *** Sukhdev_ has joined #openstack-neutron | 15:47 | |
| arcimboldo | $ neutron port-list -c id -c fixed_ips -c status | 15:47 |
| arcimboldo | +--------------------------------------+-----------------------------------------------------------------------------------+--------+ | 15:47 |
| arcimboldo | | id | fixed_ips | status | | 15:47 |
| arcimboldo | +--------------------------------------+-----------------------------------------------------------------------------------+--------+ | 15:47 |
| arcimboldo | | cafda7d0-5c38-4f46-b59f-87ff22db1b3c | {"subnet_id": "0b646402-f3aa-43b2-aa91-f33435d69a8b", "ip_address": "172.23.2.0"} | BUILD | | 15:47 |
| *** fzdarsky has quit IRC | 15:47 | |
| arcimboldo | | dbc6bfe1-533c-4b47-a612-a26289781b51 | {"subnet_id": "0b646402-f3aa-43b2-aa91-f33435d69a8b", "ip_address": "172.23.2.1"} | ACTIVE | | 15:47 |
| arcimboldo | +--------------------------------------+-----------------------------------------------------------------------------------+--------+ | 15:47 |
| arcimboldo | the DB is fine, I only have one entry with that id | 15:47 |
| *** padkrish has joined #openstack-neutron | 15:47 | |
| arcimboldo | mysql> select id,status,device_id,device_owner from ports; | 15:48 |
| arcimboldo | +--------------------------------------+--------+-------------------------------------------------------------------------------+--------------+ | 15:48 |
| arcimboldo | | id | status | device_id | device_owner | | 15:48 |
| arcimboldo | +--------------------------------------+--------+-------------------------------------------------------------------------------+--------------+ | 15:48 |
| arcimboldo | | cafda7d0-5c38-4f46-b59f-87ff22db1b3c | BUILD | dhcp2307f9ac-d2b7-50cc-baad-c69a48b83ef3-4ecbb0a0-09f2-4762-bca1-ebdc1b8ecb25 | network:dhcp | | 15:48 |
| arcimboldo | | dbc6bfe1-533c-4b47-a612-a26289781b51 | ACTIVE | dhcp940eea95-cecb-503c-8340-3569e77d02f4-4ecbb0a0-09f2-4762-bca1-ebdc1b8ecb25 | network:dhcp | | 15:48 |
| arcimboldo | +--------------------------------------+--------+-------------------------------------------------------------------------------+--------------+ | 15:48 |
| *** ijw has quit IRC | 15:50 | |
| *** l6unchpad has quit IRC | 15:50 | |
| ihrachyshka | arcimboldo, please don't paste that way in the channel. use pastebin if needed. thanks. | 15:50 |
| amuller | arcimboldo: I'm at a loss, I think you should report a bug with your reproducer (What rally test are you running?) and details about your environment (DHCP HA, galera, etc) | 15:51 |
| *** TrevorV has joined #openstack-neutron | 15:51 | |
| ihrachyshka | jlibosva, https://review.openstack.org/#/c/207108/1/neutron/tests/unit/objects/test_base.py | 15:51 |
| *** alex_klimov has quit IRC | 15:51 | |
| *** nyechiel has joined #openstack-neutron | 15:51 | |
| *** nyechiel_ has quit IRC | 15:51 | |
| arcimboldo | ihrachyshka, sorry, I thought it was small enough | 15:51 |
| ihrachyshka | gsagie, around? | 15:52 |
| arcimboldo | amuller, thanx for your help, will do it soon | 15:52 |
| *** patrickeast has joined #openstack-neutron | 15:53 | |
| gsagie_ | ihrachyshka: yes but only for a little while :) | 15:54 |
| *** fawadkhaliq has quit IRC | 15:54 | |
| *** garyk has joined #openstack-neutron | 15:54 | |
| ihrachyshka | gsagie, I wonder whether you have smth for vsctl ovsdb in qos | 15:54 |
| ihrachyshka | gsagie, that may be a blocker to merge back | 15:54 |
| gsagie_ | no | 15:55 |
| gsagie_ | everything was implemented both for the native and vsctl | 15:55 |
| jlibosva | ihrachyshka: cool! | 15:55 |
| ihrachyshka | gsagie, oh really? I though it was native only. that's cool then. | 15:55 |
| gsagie_ | nope, vsctl works as well | 15:55 |
| *** sthillma has joined #openstack-neutron | 15:55 | |
| ihrachyshka | gsagie, great. we may also need linuxbridge, not sure | 15:56 |
| *** shashank_hegde has joined #openstack-neutron | 15:56 | |
| amuller | gsagie_: do you think you can update the devref for qos about what happens in the agent? | 15:56 |
| ihrachyshka | gsagie, ml2 rule_types algorithm depends on both mechanisms tested in gate to have qos rules supported. | 15:56 |
| gsagie_ | amuller : yes, i have it on my todo list | 15:56 |
| ihrachyshka | amuller++ yeah, I guess that part is limited there | 15:56 |
| amuller | ihrachyshka: it doesn't exist =D | 15:56 |
| *** nyechiel_ has joined #openstack-neutron | 15:57 | |
| ihrachyshka | gsagie, we will have amuller looking into it closely so it would be great to have it for him to ramp up | 15:57 |
| *** padkrish has quit IRC | 15:57 | |
| ihrachyshka | amuller, well... there are TODOs! | 15:57 |
| *** nyechiel has quit IRC | 15:57 | |
| *** gampel has quit IRC | 15:57 | |
| *** jlibosva has quit IRC | 15:58 | |
| *** matrohon has quit IRC | 15:58 | |
| *** ub has quit IRC | 15:58 | |
| *** nlahouti has joined #openstack-neutron | 15:58 | |
| *** sthillma_ has joined #openstack-neutron | 15:59 | |
| *** doude has quit IRC | 15:59 | |
| *** rotbeard has quit IRC | 15:59 | |
| *** iymht has quit IRC | 16:00 | |
| *** jlanoux has quit IRC | 16:01 | |
| *** baoli has quit IRC | 16:01 | |
| *** sthillma has quit IRC | 16:01 | |
| *** sthillma_ is now known as sthillma | 16:01 | |
| *** cdl has joined #openstack-neutron | 16:02 | |
| *** s_jain has joined #openstack-neutron | 16:02 | |
| *** baoli_ has joined #openstack-neutron | 16:03 | |
| *** jbell8 has joined #openstack-neutron | 16:03 | |
| *** lucas____ has joined #openstack-neutron | 16:04 | |
| *** dane_leblanc has quit IRC | 16:04 | |
| *** y_sawai_ has joined #openstack-neutron | 16:05 | |
| *** moshele has quit IRC | 16:06 | |
| *** fzdarsky has joined #openstack-neutron | 16:06 | |
| *** cbouch has joined #openstack-neutron | 16:06 | |
| *** jistr has quit IRC | 16:08 | |
| *** ygbo has quit IRC | 16:08 | |
| *** y_sawai has quit IRC | 16:08 | |
| *** sdake has joined #openstack-neutron | 16:09 | |
| *** ihrachyshka has quit IRC | 16:09 | |
| *** daneyon has joined #openstack-neutron | 16:11 | |
| *** dane_leblanc has joined #openstack-neutron | 16:12 | |
| *** jaypipes has joined #openstack-neutron | 16:12 | |
| *** fzdarsky has quit IRC | 16:12 | |
| *** coolsvap is now known as coolsvap|away | 16:12 | |
| *** ctracey has quit IRC | 16:13 | |
| *** briancurtin has quit IRC | 16:13 | |
| *** zhiyan has quit IRC | 16:13 | |
| *** LukeHinds has quit IRC | 16:13 | |
| *** sballe has quit IRC | 16:13 | |
| *** nimrodsun has quit IRC | 16:13 | |
| *** devvesa has quit IRC | 16:13 | |
| *** Piet has joined #openstack-neutron | 16:13 | |
| openstackgerrit | Cedric Brandily proposed openstack/neutron: Python 3: set webob.Response.text instead of webob.Response.body https://review.openstack.org/204791 | 16:13 |
| *** lucas____ has quit IRC | 16:15 | |
| *** lucas__ has joined #openstack-neutron | 16:15 | |
| *** edwarnicke has joined #openstack-neutron | 16:16 | |
| *** kawa2014 has quit IRC | 16:16 | |
| openstackgerrit | Merged openstack/neutron-vpnaas: py34: Fix usage of gettext.install https://review.openstack.org/206612 | 16:17 |
| *** david-lyle has quit IRC | 16:18 | |
| openstackgerrit | Thomas Morin proposed openstack/neutron-specs: Neutron Extension for BGPVPN https://review.openstack.org/177740 | 16:21 |
| *** ip0353 has quit IRC | 16:22 | |
| *** ivar-lazzaro has joined #openstack-neutron | 16:23 | |
| openstackgerrit | Neil Jerram proposed openstack/neutron: DHCP agent: clarify logic of setup_dhcp_port https://review.openstack.org/205181 | 16:23 |
| *** yfried_ has quit IRC | 16:25 | |
| *** abregman has quit IRC | 16:25 | |
| *** Marga_ has quit IRC | 16:26 | |
| *** tmorin has quit IRC | 16:27 | |
| *** pritesh has joined #openstack-neutron | 16:27 | |
| *** afazekas has quit IRC | 16:27 | |
| *** Marga_ has joined #openstack-neutron | 16:28 | |
| *** dougwig has joined #openstack-neutron | 16:29 | |
| *** carl_baldwin has joined #openstack-neutron | 16:31 | |
| *** ihrachyshka has joined #openstack-neutron | 16:32 | |
| *** sweston has joined #openstack-neutron | 16:33 | |
| *** lindycoder has quit IRC | 16:35 | |
| *** shashank_hegde has quit IRC | 16:35 | |
| *** nimrodsun has joined #openstack-neutron | 16:35 | |
| *** sdake has quit IRC | 16:37 | |
| *** ctracey has joined #openstack-neutron | 16:37 | |
| *** shashank_hegde has joined #openstack-neutron | 16:38 | |
| *** coolsvap|away is now known as coolsvap | 16:39 | |
| *** anilvenkata has quit IRC | 16:39 | |
| *** madhu_ak has joined #openstack-neutron | 16:42 | |
| *** david-lyle has joined #openstack-neutron | 16:42 | |
| *** ub has joined #openstack-neutron | 16:43 | |
| *** vivek-ebay has joined #openstack-neutron | 16:44 | |
| *** lindycoder has joined #openstack-neutron | 16:44 | |
| *** alexpilotti has joined #openstack-neutron | 16:45 | |
| *** fzdarsky has joined #openstack-neutron | 16:45 | |
| *** zhiyan has joined #openstack-neutron | 16:46 | |
| *** eckelcu has quit IRC | 16:46 | |
| *** Marga_ has quit IRC | 16:47 | |
| *** baoli_ has quit IRC | 16:48 | |
| *** MentalRay_ has quit IRC | 16:48 | |
| *** carl_baldwin has quit IRC | 16:49 | |
| *** baoli_ has joined #openstack-neutron | 16:49 | |
| *** eckelcu has joined #openstack-neutron | 16:49 | |
| *** sballe has joined #openstack-neutron | 16:50 | |
| *** crose has quit IRC | 16:50 | |
| *** ronis has joined #openstack-neutron | 16:51 | |
| *** fzdarsky_ has joined #openstack-neutron | 16:51 | |
| *** eckelcu has quit IRC | 16:51 | |
| *** hitalia has joined #openstack-neutron | 16:51 | |
| *** eckelcu has joined #openstack-neutron | 16:52 | |
| *** shivrao has joined #openstack-neutron | 16:53 | |
| *** sdake has joined #openstack-neutron | 16:53 | |
| *** _cjones_ has joined #openstack-neutron | 16:53 | |
| *** _cjones_ has quit IRC | 16:53 | |
| *** _cjones_ has joined #openstack-neutron | 16:53 | |
| *** pritesh has quit IRC | 16:53 | |
| *** padkrish has joined #openstack-neutron | 16:54 | |
| *** gsagie_ has quit IRC | 16:54 | |
| *** pritesh has joined #openstack-neutron | 16:54 | |
| *** mohankumar has quit IRC | 16:54 | |
| *** Ramanjaneya has quit IRC | 16:55 | |
| *** shivrao_ has joined #openstack-neutron | 16:56 | |
| *** vivek-ebay has quit IRC | 16:57 | |
| *** nlahouti has quit IRC | 16:57 | |
| *** shivrao has quit IRC | 16:57 | |
| *** briancurtin has joined #openstack-neutron | 16:57 | |
| *** shivrao_ is now known as shivrao | 16:57 | |
| *** padkrish has quit IRC | 16:57 | |
| *** coolsvap is now known as coolsvap|away | 16:58 | |
| *** padkrish has joined #openstack-neutron | 16:58 | |
| *** coolsvap|away is now known as coolsvap | 16:58 | |
| jschwarz | ihrachyshka, ramanjaneya_, http://pastebin.com/ed5usr2f | 16:58 |
| jschwarz | what do you think? :) | 16:59 |
| *** cdl has quit IRC | 16:59 | |
| *** eckelcu has quit IRC | 16:59 | |
| *** nlahouti has joined #openstack-neutron | 16:59 | |
| *** padkrish has quit IRC | 17:00 | |
| *** amuller is now known as amuller_afk | 17:00 | |
| *** LukeHinds has joined #openstack-neutron | 17:00 | |
| *** padkrish has joined #openstack-neutron | 17:00 | |
| *** arcimboldo has quit IRC | 17:00 | |
| *** nlahouti has quit IRC | 17:01 | |
| *** pal has joined #openstack-neutron | 17:02 | |
| ihrachyshka | jschwarz, (y) | 17:03 |
| jschwarz | ihrachyshka, :) posted a comment on one of your patches | 17:03 |
| jschwarz | I'm going home now.. I'll be back tomorrow after lunch and will post the neutronclient changes then | 17:03 |
| jschwarz | too tired to do it now :) | 17:04 |
| ihrachyshka | jschwarz, enjoy, you deserve rest | 17:04 |
| jschwarz | ihrachyshka, ty.. today was fun all in all ^_^ | 17:05 |
| jschwarz | you should get some rest as well! | 17:05 |
| ihrachyshka | any day on qos is a roller coaster | 17:05 |
| *** rex_lee has joined #openstack-neutron | 17:05 | |
| jschwarz | roller coasters are fun | 17:06 |
| jschwarz | ciao | 17:06 |
| *** cbouch has quit IRC | 17:06 | |
| *** nlahouti has joined #openstack-neutron | 17:08 | |
| *** Marga_ has joined #openstack-neutron | 17:08 | |
| *** nyechiel_ has quit IRC | 17:09 | |
| *** john-davidge has quit IRC | 17:09 | |
| *** fzdarsky has quit IRC | 17:10 | |
| *** scheuran has quit IRC | 17:10 | |
| *** fzdarsky_ has quit IRC | 17:10 | |
| *** jschwarz has quit IRC | 17:11 | |
| *** slaweq has quit IRC | 17:14 | |
| *** salv-orlando has joined #openstack-neutron | 17:14 | |
| *** alexpilotti has quit IRC | 17:14 | |
| *** cbouch has joined #openstack-neutron | 17:15 | |
| *** akamyshnikova_ has joined #openstack-neutron | 17:15 | |
| *** ajmiller has quit IRC | 17:18 | |
| *** ub has quit IRC | 17:18 | |
| *** s_jain has quit IRC | 17:18 | |
| *** ajmiller has joined #openstack-neutron | 17:19 | |
| *** slaweq has joined #openstack-neutron | 17:19 | |
| *** dboik has quit IRC | 17:19 | |
| *** ub has joined #openstack-neutron | 17:20 | |
| *** salv-orlando has quit IRC | 17:20 | |
| *** moshele has joined #openstack-neutron | 17:20 | |
| *** salv-orlando has joined #openstack-neutron | 17:22 | |
| openstackgerrit | Merged openstack/neutron-vpnaas: Remove quantum untracked files from .gitignore https://review.openstack.org/205120 | 17:23 |
| openstackgerrit | Merged openstack/neutron-vpnaas: Register alembic_migrations at install time https://review.openstack.org/198523 | 17:23 |
| *** patrickeast has quit IRC | 17:23 | |
| *** alexpilotti has joined #openstack-neutron | 17:23 | |
| *** ub has quit IRC | 17:25 | |
| *** ihrachyshka has quit IRC | 17:25 | |
| *** amuller_afk is now known as amuller | 17:25 | |
| *** achanda has joined #openstack-neutron | 17:27 | |
| *** dsneddon has quit IRC | 17:28 | |
| *** dsneddon has joined #openstack-neutron | 17:28 | |
| *** aranjan has joined #openstack-neutron | 17:29 | |
| *** sthillma has quit IRC | 17:29 | |
| *** vivek-ebay has joined #openstack-neutron | 17:30 | |
| *** mmdurrant has joined #openstack-neutron | 17:30 | |
| *** aix has quit IRC | 17:30 | |
| *** sridhar_ram has joined #openstack-neutron | 17:30 | |
| *** ramanjaneya_ has quit IRC | 17:31 | |
| *** Marga_ has quit IRC | 17:32 | |
| *** Marga_ has joined #openstack-neutron | 17:32 | |
| *** shaohe_robot has quit IRC | 17:32 | |
| *** eckelcu has joined #openstack-neutron | 17:33 | |
| *** shaohe_robot has joined #openstack-neutron | 17:33 | |
| *** vivek-ebay has quit IRC | 17:35 | |
| *** saksham has joined #openstack-neutron | 17:36 | |
| *** armax has joined #openstack-neutron | 17:36 | |
| *** jobewan has joined #openstack-neutron | 17:37 | |
| openstackgerrit | Merged openstack/neutron: Correct fcntl.flock use in Pidfile.unlock https://review.openstack.org/199623 | 17:38 |
| *** patrickeast has joined #openstack-neutron | 17:38 | |
| *** ihrachyshka has joined #openstack-neutron | 17:39 | |
| *** minwang2 has joined #openstack-neutron | 17:39 | |
| *** jaypipes has quit IRC | 17:40 | |
| *** dsneddon has quit IRC | 17:43 | |
| *** mickeys has quit IRC | 17:43 | |
| *** r-daneel has quit IRC | 17:44 | |
| openstackgerrit | Merged openstack/neutron: Moving out cisco n1kv extensions https://review.openstack.org/201333 | 17:44 |
| *** ByPasS has quit IRC | 17:45 | |
| *** dannywilson has quit IRC | 17:46 | |
| *** Sukhdev_ has quit IRC | 17:47 | |
| *** nlahouti has quit IRC | 17:47 | |
| *** padkrish has quit IRC | 17:48 | |
| *** ByPasS has joined #openstack-neutron | 17:48 | |
| *** padkrish_ has joined #openstack-neutron | 17:50 | |
| *** ub has joined #openstack-neutron | 17:51 | |
| *** dsneddon has joined #openstack-neutron | 17:51 | |
| *** Marga_ has quit IRC | 17:51 | |
| *** zhhuabj has quit IRC | 17:51 | |
| *** Marga_ has joined #openstack-neutron | 17:52 | |
| *** nlahouti has joined #openstack-neutron | 17:52 | |
| *** moshele has quit IRC | 17:53 | |
| *** zhhuabj has joined #openstack-neutron | 17:54 | |
| *** jasondotstar has joined #openstack-neutron | 17:54 | |
| *** s3wong has joined #openstack-neutron | 17:57 | |
| *** lucas__ has quit IRC | 17:58 | |
| *** mickeys has joined #openstack-neutron | 18:00 | |
| openstackgerrit | Roman Bogorodskiy proposed openstack/neutron: sriov: implement spoofchecking configuration https://review.openstack.org/192065 | 18:02 |
| *** prad has quit IRC | 18:04 | |
| *** Ramanjaneya has joined #openstack-neutron | 18:05 | |
| *** mohankumar has joined #openstack-neutron | 18:05 | |
| *** lucas__ has joined #openstack-neutron | 18:05 | |
| *** azbiswas has joined #openstack-neutron | 18:06 | |
| *** adam_g has joined #openstack-neutron | 18:07 | |
| *** adam_g has quit IRC | 18:07 | |
| *** adam_g has joined #openstack-neutron | 18:07 | |
| *** davidjc has quit IRC | 18:08 | |
| *** fzdarsky has joined #openstack-neutron | 18:08 | |
| *** tflynn has joined #openstack-neutron | 18:08 | |
| *** fzdarsky_ has joined #openstack-neutron | 18:08 | |
| *** tflynn has joined #openstack-neutron | 18:09 | |
| openstackgerrit | Assaf Muller proposed openstack/neutron: Enable fullstack multinode tests, add L3 HA test exemplar https://review.openstack.org/188221 | 18:09 |
| *** lucas__ has quit IRC | 18:09 | |
| *** lucas__ has joined #openstack-neutron | 18:10 | |
| *** tflynn has quit IRC | 18:10 | |
| *** azbiswas has quit IRC | 18:11 | |
| *** tflynn has joined #openstack-neutron | 18:11 | |
| *** ByPasS has quit IRC | 18:13 | |
| *** ub2 has joined #openstack-neutron | 18:14 | |
| *** ub has quit IRC | 18:16 | |
| *** tiswanso has joined #openstack-neutron | 18:16 | |
| *** dannywilson has joined #openstack-neutron | 18:18 | |
| *** coolsvap is now known as coolsvap|brb | 18:18 | |
| *** ByPasS has joined #openstack-neutron | 18:19 | |
| *** prad has joined #openstack-neutron | 18:19 | |
| *** crc32 has joined #openstack-neutron | 18:20 | |
| *** sridhar_ram has quit IRC | 18:20 | |
| *** abregman has joined #openstack-neutron | 18:21 | |
| *** jasondotstar has quit IRC | 18:23 | |
| *** dims_ has joined #openstack-neutron | 18:24 | |
| *** prad has quit IRC | 18:24 | |
| *** dboik has joined #openstack-neutron | 18:25 | |
| *** annegentle has quit IRC | 18:26 | |
| *** pal has quit IRC | 18:26 | |
| *** annegentle has joined #openstack-neutron | 18:26 | |
| *** prad has joined #openstack-neutron | 18:27 | |
| *** dims has quit IRC | 18:28 | |
| *** josecastroleon has joined #openstack-neutron | 18:28 | |
| *** amickus has joined #openstack-neutron | 18:30 | |
| *** MentalRay_ has joined #openstack-neutron | 18:30 | |
| *** vivek-ebay has joined #openstack-neutron | 18:31 | |
| *** openstackgerrit has quit IRC | 18:31 | |
| *** openstackgerrit has joined #openstack-neutron | 18:32 | |
| *** armax has quit IRC | 18:32 | |
| *** nlahouti has quit IRC | 18:32 | |
| *** dannywilson has quit IRC | 18:34 | |
| *** dannywilson has joined #openstack-neutron | 18:34 | |
| *** jaypipes has joined #openstack-neutron | 18:34 | |
| *** ByPasS has quit IRC | 18:34 | |
| *** padkrish_ has quit IRC | 18:35 | |
| *** padkrish has joined #openstack-neutron | 18:35 | |
| *** vivek-ebay has quit IRC | 18:35 | |
| *** sthillma has joined #openstack-neutron | 18:36 | |
| *** thorst has quit IRC | 18:36 | |
| *** ByPasS has joined #openstack-neutron | 18:36 | |
| *** safchain has quit IRC | 18:37 | |
| *** shashank_hegde has quit IRC | 18:38 | |
| *** minwang2 has quit IRC | 18:39 | |
| *** vishwanathj_ has joined #openstack-neutron | 18:39 | |
| *** ByPasS has quit IRC | 18:40 | |
| *** salv-orlando has quit IRC | 18:40 | |
| *** vishwanathj has joined #openstack-neutron | 18:42 | |
| *** ByPasS has joined #openstack-neutron | 18:43 | |
| *** sdake has quit IRC | 18:43 | |
| *** bharath_ has joined #openstack-neutron | 18:45 | |
| *** y_sawai has joined #openstack-neutron | 18:45 | |
| *** fawadkhaliq has joined #openstack-neutron | 18:45 | |
| *** sdake has joined #openstack-neutron | 18:46 | |
| *** vishwanathj has quit IRC | 18:46 | |
| *** padkrish has quit IRC | 18:47 | |
| *** y_sawai_ has quit IRC | 18:48 | |
| *** eckelcu has quit IRC | 18:48 | |
| *** nlahouti has joined #openstack-neutron | 18:50 | |
| *** padkrish_ has joined #openstack-neutron | 18:50 | |
| *** thorst has joined #openstack-neutron | 18:51 | |
| *** aranjan has quit IRC | 18:52 | |
| *** aranjan_ has joined #openstack-neutron | 18:52 | |
| ihrachyshka | amuller, will you have cycles today for https://review.openstack.org/#/c/205762/ ? it unbreaks a lot in qos. | 18:53 |
| *** nlahouti has quit IRC | 18:54 | |
| *** shaohe_robot1 has joined #openstack-neutron | 18:55 | |
| *** Sukhdev_ has joined #openstack-neutron | 18:55 | |
| *** fawadk has joined #openstack-neutron | 18:55 | |
| *** padkrish_ has quit IRC | 18:56 | |
| *** padkrish has joined #openstack-neutron | 18:56 | |
| *** fawadkhaliq has quit IRC | 18:56 | |
| *** shaohe_robot has quit IRC | 18:57 | |
| *** pal has joined #openstack-neutron | 18:57 | |
| amuller | ihrachyshka: I'll take a look | 18:58 |
| *** josecastroleon has quit IRC | 18:58 | |
| amuller | ihrachyshka: so now the inheritance is between the DB model classes, but every rule type will have only 1 table, and there will be no shared table, yes? | 18:59 |
| ihrachyshka | amuller, there is no real inheritance in db models now. and yes, there is a table per type. | 19:00 |
| *** padkrish has quit IRC | 19:00 | |
| ihrachyshka | no qos_rules table in between qos policies and type specific rules | 19:00 |
| amuller | ihrachyshka: QosBandwidthLimitRule(QosRuleColumns, model_base.BASEV2)? | 19:00 |
| ihrachyshka | that was... a bad decision | 19:00 |
| ihrachyshka | amuller, well, it only melds in common fields (qos_policy_id that is) | 19:01 |
| *** clayton has joined #openstack-neutron | 19:01 | |
| ihrachyshka | there is no real table for that melded class | 19:01 |
| amuller | ihrachyshka: yes | 19:01 |
| amuller | a common table seems like a bad idea | 19:01 |
| ihrachyshka | but yeah, inherited... | 19:01 |
| *** salv-orlando has joined #openstack-neutron | 19:01 | |
| ihrachyshka | it is. you can see how much buggy code we needed to handle it | 19:01 |
| ihrachyshka | and it still didn't work in some cases. | 19:02 |
| clayton | I've been experimenting with setting port settings that are exposed via an neutron extension. I managed to make "neutron --debug port-update 26c75566-febd-48ca-bc48-28ce1f0d8619 --allowed_address_pairs list=true type=dict ip_address=10.10.3.124" do what I wanted, but it seems like there is some sort of generic facility in neutron client for adding generic information to the api call. Is this documented somewhere? | 19:02 |
| ihrachyshka | mestery, amuller: one thing I wonder about upcoming qos merge into master is what we do with API tests that depend on the service being enabled in gate via local.conf... they will fail without gate changes, and gate changes cannot go before we merge back... chicken and egg | 19:03 |
| ihrachyshka | we may need to disable those tests just before merge-back... | 19:03 |
| amuller | ihrachyshka: we could skip the qos api tests conditionally | 19:04 |
| ihrachyshka | or maybe we can conditionalize them somehow?.. like check extensions list and skip if it's not there?.. | 19:04 |
| amuller | ihrachyshka: depending if the service plugin is loaded or not | 19:04 |
| *** krtaylor has quit IRC | 19:04 | |
| ihrachyshka | oh ok, we need to look into it | 19:04 |
| amuller | or the extension is exposed, that's better | 19:05 |
| ihrachyshka | amuller, it's also depending on whether ml2 extension is loaded | 19:05 |
| amuller | ihrachyshka: not enough to just check the qos extension? | 19:05 |
| ihrachyshka | amuller, exposed? you mean /qos/ replied? | 19:05 |
| amuller | ihrachyshka: I mean if qos appears in neutron ext-list | 19:05 |
| *** padkrish has joined #openstack-neutron | 19:05 | |
| ihrachyshka | amuller, hm, maybe it is enough actually. we will enable the ml2 extension with service plugin in one go in gate I guess | 19:05 |
| ihrachyshka | ok cool, I didn't know about conditionals | 19:06 |
| *** padkrish has quit IRC | 19:06 | |
| amuller | ihrachyshka: a lot of tempest has skipUnless sort of decorators | 19:06 |
| ihrachyshka | I will need jschwarz to play with it :) | 19:06 |
| amuller | we have some of those in tests/api for sure | 19:06 |
| amuller | ihrachyshka: the only issue with this approach is that if the gate turns off the qos service plugin by mistake, all tests will pass, and you'll kind of have no idea the tests aren't running unless you do notice manually | 19:07 |
| amuller | ihrachyshka: so we kind of try to avoid conditional skips | 19:07 |
| ihrachyshka | amuller, another testing thing is that one of our API endpoints depend on ml2 drivers loaded: rule types reply with supported rules based on what drivers provide | 19:07 |
| ihrachyshka | amuller, and in gate, we have lb + ovs | 19:07 |
| ihrachyshka | for ovs, we have support | 19:07 |
| *** lucas__ has quit IRC | 19:07 | |
| ihrachyshka | but I believe we won't for lb | 19:07 |
| amuller | ihrachyshka: now that I think about it, tests/api should pass with other core plugins that don't support qos | 19:08 |
| ihrachyshka | that complicates things since our tests assume all loaded drivers support it | 19:08 |
| *** padkrish has joined #openstack-neutron | 19:08 | |
| amuller | so we need the conditional skip anyway | 19:08 |
| *** nlahouti has joined #openstack-neutron | 19:08 | |
| ihrachyshka | for the time being, I added a hack in lb that makes lb driver | 19:09 |
| ihrachyshka | 'support' qos, but it should be solved in better way | 19:09 |
| amuller | ihrachyshka: tempest can check the ml2 conf somehow to see what drivers are loaded | 19:09 |
| ihrachyshka | amuller, yeah, but at the same time, what would you do when lb gets qos support? | 19:09 |
| amuller | change the condition, it's in our own tree | 19:10 |
| *** coolsvap|brb is now known as coolsvap|away | 19:10 | |
| ihrachyshka | ah right, I always forget about that own tree thing... | 19:10 |
| *** Piet has quit IRC | 19:10 | |
| ihrachyshka | :) | 19:10 |
| ihrachyshka | cool | 19:10 |
| *** pritesh has quit IRC | 19:10 | |
| *** shashank_hegde has joined #openstack-neutron | 19:12 | |
| ihrachyshka | amuller, I hope we won't loose ref implementation before that ;) | 19:13 |
| *** ZZelle has joined #openstack-neutron | 19:13 | |
| *** jamiec has joined #openstack-neutron | 19:13 | |
| amuller | ihrachyshka: meh I'm trying to repress that | 19:14 |
| *** padkrish has quit IRC | 19:14 | |
| *** nlahouti has quit IRC | 19:14 | |
| openstackgerrit | Saksham Varma proposed openstack/neutron: Moved out cisco n1kv mech driver and db models https://review.openstack.org/205383 | 19:14 |
| ihrachyshka | amuller, good luck with that | 19:14 |
| amuller | ihrachyshka: I didn't see any traffic on that since the spec was merged | 19:15 |
| amuller | maybe if we keep quiet it will be forgotten ;-) | 19:15 |
| *** fawadkhaliq has joined #openstack-neutron | 19:16 | |
| ihrachyshka | amuller, except file moves, not really | 19:17 |
| ihrachyshka | which were good | 19:17 |
| *** krtaylor has joined #openstack-neutron | 19:17 | |
| amuller | ihrachyshka: be back later | 19:18 |
| amuller | ihrachyshka: gotta pack and stuff | 19:18 |
| amuller | ihrachyshka: I'll try to look at that patch more later today | 19:18 |
| *** jasondotstar has joined #openstack-neutron | 19:18 | |
| ihrachyshka | amuller, aye, see you in proper timezone | 19:18 |
| amuller | hah | 19:18 |
| *** jamiec has quit IRC | 19:19 | |
| *** fawadk has quit IRC | 19:19 | |
| openstackgerrit | Ritesh Anand proposed openstack/neutron: Fixes a typo phys_brs in place of phys_br https://review.openstack.org/195791 | 19:19 |
| amuller | ihrachyshka: hmm when looking at neutron/tests/api/base.py, it looks like policies are deleted before rules | 19:20 |
| amuller | ihrachyshka: does that make sense? Can I delete a policy with rules still defined on it? | 19:20 |
| ihrachyshka | amuller, well actually rules will be killed by cascade | 19:21 |
| ihrachyshka | and yeah, the order should probably be vice versa there to make sense. | 19:21 |
| ihrachyshka | but it's nice to have, it will still cleanup all when killing policies | 19:22 |
| ihrachyshka | so rule cleanup is worthless | 19:22 |
| *** lucas__ has joined #openstack-neutron | 19:22 | |
| amuller | oh now I know what I meant, the policies are deleted before ports and networks | 19:22 |
| amuller | that doesn't seem right | 19:22 |
| amuller | you could try to delete a policy still attached to ports/networks and fail the clean up | 19:23 |
| *** Apsu has quit IRC | 19:24 | |
| *** yamamoto has quit IRC | 19:24 | |
| ihrachyshka | amuller, why? policy drop, as well as port/network one, will only kill binding between them | 19:24 |
| openstackgerrit | Ritesh Anand proposed openstack/neutron: Fixes a typo phys_brs in place of phys_br https://review.openstack.org/195791 | 19:24 |
| ihrachyshka | objects themselves are intact | 19:24 |
| *** fawadk has joined #openstack-neutron | 19:24 | |
| *** Apsu has joined #openstack-neutron | 19:24 | |
| ihrachyshka | amuller, note there is no qos_policy_id field in net/port models really. | 19:25 |
| *** gampel has joined #openstack-neutron | 19:25 | |
| *** fawadkhaliq has quit IRC | 19:26 | |
| amuller | ihrachyshka: if a SG is attached to a port and I try to delete a SG it yells that it's in use | 19:26 |
| amuller | ihrachyshka: it's not the case with a QoS policy? | 19:26 |
| *** dims_ has quit IRC | 19:27 | |
| *** dims has joined #openstack-neutron | 19:27 | |
| ihrachyshka | amuller, not that I know of plans like that. | 19:29 |
| ihrachyshka | amuller, we can obviously make it like that with ease. | 19:30 |
| amuller | ihrachyshka: imagine a tenant happily applying a bandwidth guarantee policy (in the future), and an evil admin deletes the policy itself, and without the tenant having any indication all of a sudden his bandwidth is not guaranteed | 19:30 |
| *** eckelcu has joined #openstack-neutron | 19:31 | |
| *** lucas__ has quit IRC | 19:32 | |
| *** lucas__ has joined #openstack-neutron | 19:33 | |
| ihrachyshka | amuller, I understand the use case, it's just that evil admin sounds... unrealistic. I will check whether we had anything about it in spec. and will add in todo to consider. | 19:33 |
| amuller | ihrachyshka: evil admin can mean an accident, not a malicious admin (Which I agree is unrealistic) | 19:34 |
| *** jlibosva has joined #openstack-neutron | 19:35 | |
| *** vishwanathj has joined #openstack-neutron | 19:35 | |
| *** salv-orlando has quit IRC | 19:35 | |
| *** ByPasS has quit IRC | 19:35 | |
| *** ByPasS has joined #openstack-neutron | 19:35 | |
| ihrachyshka | amuller, "By default QoS policies and rules will be managed by the cloud administrator, that makes the tenant unable to create specific qos rules, or attaching specific ports to policies." | 19:36 |
| *** mickeys has quit IRC | 19:37 | |
| *** hitalia has quit IRC | 19:37 | |
| *** mickeys has joined #openstack-neutron | 19:37 | |
| ihrachyshka | seems like the topic is not touched there at all | 19:37 |
| *** hitalia has joined #openstack-neutron | 19:38 | |
| *** hitalia has quit IRC | 19:39 | |
| *** txomon|fon has quit IRC | 19:39 | |
| *** hitalia has joined #openstack-neutron | 19:39 | |
| *** minwang2 has joined #openstack-neutron | 19:39 | |
| *** neophy has joined #openstack-neutron | 19:40 | |
| *** phil_h has joined #openstack-neutron | 19:42 | |
| *** azbiswas has joined #openstack-neutron | 19:42 | |
| *** jlibosva has quit IRC | 19:45 | |
| *** jlibosva has joined #openstack-neutron | 19:46 | |
| *** pal has quit IRC | 19:46 | |
| *** achanda has quit IRC | 19:46 | |
| *** krtaylor has quit IRC | 19:46 | |
| *** fzdarsky has quit IRC | 19:46 | |
| *** fzdarsky_ has quit IRC | 19:47 | |
| *** ByPasS has quit IRC | 19:47 | |
| *** HenryG has quit IRC | 19:49 | |
| *** pal has joined #openstack-neutron | 19:49 | |
| openstackgerrit | Cedric Brandily proposed openstack/neutron: Python 3: set webob.Response.text instead of webob.Response.body https://review.openstack.org/204791 | 19:50 |
| *** padkrish has joined #openstack-neutron | 19:50 | |
| amuller | ihrachyshka: ahh so only an admin can attach a policy to a port/network then | 19:50 |
| *** salv-orlando has joined #openstack-neutron | 19:51 | |
| *** HenryG has joined #openstack-neutron | 19:52 | |
| *** ByPasS has joined #openstack-neutron | 19:52 | |
| *** vishwanathj has quit IRC | 19:57 | |
| *** crc32 has quit IRC | 19:57 | |
| *** ajmiller has quit IRC | 19:58 | |
| *** akamyshnikova_ has quit IRC | 19:59 | |
| *** ajmiller has joined #openstack-neutron | 19:59 | |
| openstackgerrit | Kyle Mestery proposed openstack/neutron: thirdparty-ci: Don't allow voting -1 https://review.openstack.org/207198 | 19:59 |
| *** krtaylor has joined #openstack-neutron | 19:59 | |
| *** edmondsw has quit IRC | 20:01 | |
| *** Aish has joined #openstack-neutron | 20:02 | |
| *** fawadk has quit IRC | 20:02 | |
| *** shashank_hegde has quit IRC | 20:03 | |
| *** mickeys has quit IRC | 20:04 | |
| ihrachyshka | amuller, in default policy.json conf, yes | 20:05 |
| *** crc32 has joined #openstack-neutron | 20:05 | |
| *** shashank_hegde has joined #openstack-neutron | 20:06 | |
| *** abhiraut has joined #openstack-neutron | 20:06 | |
| *** busterswt has joined #openstack-neutron | 20:06 | |
| amuller | ihrachyshka: I'm looking at oslo versioned objects, am I correct to understand that its main bit of functionality is the ability to up or down the version of the object *for the purposes of RPC serialization*? | 20:06 |
| *** tiswanso has quit IRC | 20:06 | |
| amuller | ihrachyshka: so this has nothing to do with DB upgrades | 20:07 |
| amuller | ihrachyshka: also the only way to use this functionality (versioning across RPC) is to call methods on the versioned objects, where these methods have been wrapped by remotable_classmethod? | 20:09 |
| amuller | ihrachyshka: what happens if I send this object via the regular oslo messaging API? | 20:09 |
| *** ByPasS has quit IRC | 20:09 | |
| *** ByPasS has joined #openstack-neutron | 20:09 | |
| *** jobewan has quit IRC | 20:10 | |
| *** vishwanathj has joined #openstack-neutron | 20:10 | |
| *** baoli_ has quit IRC | 20:12 | |
| *** padkrish has quit IRC | 20:14 | |
| *** padkrish has joined #openstack-neutron | 20:15 | |
| ihrachyshka | amuller, well... it's not ONLY versioning | 20:15 |
| ihrachyshka | amuller, it also gives a nice abstraction layer handling fields and all | 20:16 |
| *** openstackgerrit has quit IRC | 20:16 | |
| ihrachyshka | amuller, for db upgrades, it's not directly related, but still: if we want to do data migrations in runtime, they may help since you can hide schema under the object properties | 20:16 |
| *** openstackgerrit has joined #openstack-neutron | 20:17 | |
| *** annegentle has quit IRC | 20:17 | |
| ihrachyshka | amuller, afaik versioning is not directly related to remotable_classmethod actually. remotable is to be able to call e.g. create() on agent side and get it propagated in some form to server thru proxy. while the same method on server side would avoid proxies and do it locally. | 20:18 |
| amuller | ihrachyshka: so when are the versioning methods used then? | 20:18 |
| ihrachyshka | amuller, as for sending it via regular oslo.messaging API... you cannot send an object, you can send a dict only. and for dicts, if you send obj_to_primitive() result, it will be versioned, and if you only need fields, you do smth like we do in to_dict() | 20:18 |
| *** jasondotstar has quit IRC | 20:19 | |
| *** padkrish has quit IRC | 20:19 | |
| ihrachyshka | amuller, they are used in case you want to send an older version of an object thru RPC for whatever reason. then versioned objects allow you to define rules on how to backport it to that older version | 20:19 |
| ihrachyshka | amuller, in qos, we'll use it for upgrade (see target_version= we set in get_info) | 20:20 |
| *** vishwanathj has quit IRC | 20:20 | |
| ihrachyshka | so an agent requests an object of version X, and gets it, even though server already knows later Y version. | 20:20 |
| *** u_glide2 has quit IRC | 20:21 | |
| ihrachyshka | amuller, afaik in nova they have object versions pinned on upgrade so that server always pushes older object versions on the wire. | 20:21 |
| ihrachyshka | also note that in neutron, we probably don't control RPC payload as tight as they do in nova. | 20:22 |
| amuller | ihrachyshka: they use a server side configuration value to pin to juno or kilo right? | 20:22 |
| amuller | so if the conf value is on the server will respond with that version's objects | 20:22 |
| amuller | ihrachyshka: each object needs to map a global version to its own version? | 20:22 |
| ihrachyshka | amuller, yeah, smth like that. I haven't looked at details, but that's mentioned in e.g. http://superuser.openstack.org/articles/upgrading-nova-to-kilo-with-minimal-downtime see step 2 | 20:23 |
| amuller | so global version kilo is DB object X version 1, and DB object Y version 2, while global version liberty is DB object X version 2 and DJ object Y version 3? | 20:23 |
| *** rkukura_ has joined #openstack-neutron | 20:23 | |
| ihrachyshka | amuller, yeah, there should be a global map. | 20:23 |
| *** gampel has quit IRC | 20:23 | |
| ihrachyshka | yeah, exactly. | 20:23 |
| ihrachyshka | I think there are also ways to automate the increment | 20:24 |
| *** padkrish has joined #openstack-neutron | 20:24 | |
| openstackgerrit | Cedric Brandily proposed openstack/neutron: Ensure non-overlapping cidrs in subnetpools with galera https://review.openstack.org/187985 | 20:24 |
| *** nlahouti has joined #openstack-neutron | 20:24 | |
| ihrachyshka | also, server may detect the oldest version in the cluster and use it (assuming agents report it on report_state) | 20:24 |
| *** mickeys has joined #openstack-neutron | 20:24 | |
| *** Aish has quit IRC | 20:24 | |
| *** yamamoto has joined #openstack-neutron | 20:24 | |
| ihrachyshka | in qos, when pushing updates to agents, we'll send multiple versions of objects in parallel versioned topics based on a pin | 20:25 |
| *** achanda has joined #openstack-neutron | 20:25 | |
| ihrachyshka | the logic is not there since it's not strictly needed for initial version | 20:25 |
| ihrachyshka | we will need to work it out in M though | 20:25 |
| ihrachyshka | for now agents just listen on a topic of latest object version | 20:26 |
| *** nmagnezi_ has joined #openstack-neutron | 20:26 | |
| *** rkukura has quit IRC | 20:26 | |
| *** rkukura_ is now known as rkukura | 20:26 | |
| amuller | ihrachyshka: we'll need to use versioned objects across the entire code base to properly support this =/ | 20:26 |
| ihrachyshka | btw that usage of versioned objects is not really usual, it's smth we learn for the first time I guess | 20:26 |
| amuller | ihrachyshka: how on earth did they accomplish this in Nova | 20:26 |
| ZZelle | amuller, only between server/agents? | 20:26 |
| *** nmagnezi_ has quit IRC | 20:27 | |
| amuller | ZZelle: any sort of information you'd send over the wire | 20:27 |
| amuller | this includes weird stuff like l2pop | 20:27 |
| *** phil_h has quit IRC | 20:27 | |
| amuller | which sends data that is not mapped to any DB model | 20:27 |
| ihrachyshka | amuller, afaik they have agents replying with 'unsupported version' and then conductor backporting | 20:27 |
| ihrachyshka | amuller, does not matter whether it maps to db. my NeutronObject is not really bound to db | 20:28 |
| ihrachyshka | for db, there is NeutronDbObject | 20:28 |
| *** shashank_hegde has quit IRC | 20:28 | |
| *** Aish has joined #openstack-neutron | 20:28 | |
| amuller | ihrachyshka: I understand that, just saying it's going to be alot of those object definitions | 20:28 |
| *** padkrish has quit IRC | 20:28 | |
| *** mriedem has joined #openstack-neutron | 20:28 | |
| ihrachyshka | amuller, I don't believe they actually send desired version numbers on the wire | 20:28 |
| ZZelle | exactly nova-conductor is able to talk with computes using older versions | 20:28 |
| *** lucas__ has quit IRC | 20:28 | |
| ihrachyshka | amuller, they just complain with unsupported version if it seems not applicable (or so I understand it) | 20:28 |
| mriedem | mestery: i'm trying to figure out what's going on with this networking-macvtap thing, there is a nova change to add plugging support to the libvirt driver https://review.openstack.org/#/c/182283/ | 20:29 |
| *** padkrish has joined #openstack-neutron | 20:29 | |
| mriedem | mestery: but that's dependent on some project placement discussions with neutron that aren't sorted out yet, | 20:29 |
| ihrachyshka | we are probably the first ones to request desired version explicitly, for one specific RPC interface. we'll see whether it will fly for other pushes to agents | 20:29 |
| mestery | mriedem: Right, we discussed this at the Neutron meeting. | 20:29 |
| mriedem | so i'm finding it hard to approve a thing in nova that doesn't even have a project in neutron land to provide vif details to it | 20:29 |
| *** yamamoto has quit IRC | 20:29 | |
| mestery | mriedem: I'm with you | 20:29 |
| ihrachyshka | amuller, that's a lot of work. are you scared? ;) | 20:29 |
| ihrachyshka | amuller, it makes interfaces clean though, for the least. | 20:30 |
| mestery | mriedem: The discussion has apparently landed on the fact that the macvtap folks do not think they can use some extensions to the Linuxbridge agent in Neutron ,and need their own agent. | 20:30 |
| mestery | The sticking point is why do they need a new repo just for an ew agent. | 20:30 |
| ihrachyshka | so even outside versioning it's a good thing | 20:30 |
| mestery | mriedem: We'll resolve this at the next Neutron meeting at this point, I think they put it back on the agenda. | 20:30 |
| mestery | mriedem: Hope that helps. | 20:30 |
| *** y_sawai has quit IRC | 20:30 | |
| mriedem | mestery: ok, nova FF is tomorrow, when is the next neutron meeting? | 20:30 |
| *** lucas__ has joined #openstack-neutron | 20:31 | |
| mestery | mriedem: It's next week, sadly. Let me circle back with those folks (IBM folks I believe) and try to resolve it earlier. I think they're in Germany though, so meh. | 20:31 |
| *** jasondotstar has joined #openstack-neutron | 20:32 | |
| mriedem | mestery: yeah andreas is done for the day | 20:32 |
| mriedem | ffs, i found the networking-macvtap in our internal github | 20:32 |
| *** tiswanso has joined #openstack-neutron | 20:32 | |
| mriedem | he should have had that in public github at least | 20:32 |
| mriedem | stackforge, something | 20:32 |
| mriedem | jckasper: ^ in case you talk to these people | 20:33 |
| *** padkrish has quit IRC | 20:33 | |
| *** tiswanso has quit IRC | 20:33 | |
| *** tiswanso has joined #openstack-neutron | 20:33 | |
| *** padkrish has joined #openstack-neutron | 20:33 | |
| mestery | mriedem: Just replied in the rvview | 20:33 |
| mestery | mriedem: Thanks for bringing to my (very short and pulled in 10000 directions) attention | 20:33 |
| *** tiswanso has quit IRC | 20:34 | |
| *** achanda has quit IRC | 20:34 | |
| *** achanda has joined #openstack-neutron | 20:35 | |
| *** Ramanjaneya has quit IRC | 20:35 | |
| *** lucas__ has quit IRC | 20:35 | |
| *** salv-orlando has quit IRC | 20:36 | |
| *** Ramanjaneya has joined #openstack-neutron | 20:38 | |
| *** Aish has quit IRC | 20:38 | |
| *** armax has joined #openstack-neutron | 20:39 | |
| *** armax has quit IRC | 20:39 | |
| *** eckelcu has quit IRC | 20:39 | |
| *** jlibosva has quit IRC | 20:40 | |
| *** Aish has joined #openstack-neutron | 20:40 | |
| *** ronis has quit IRC | 20:41 | |
| *** achanda has quit IRC | 20:41 | |
| mriedem | mestery: np, thanks - fwiw we have FFEs | 20:41 |
| mriedem | i wish the guy would have at least has his code in public github | 20:42 |
| mestery | mriedem: Awesome | 20:42 |
| mestery | +1000 | 20:42 |
| *** phil_h has joined #openstack-neutron | 20:42 | |
| *** fzdarsky_ has joined #openstack-neutron | 20:43 | |
| *** fzdarsky has joined #openstack-neutron | 20:43 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/neutron: Updated from global requirements https://review.openstack.org/206825 | 20:44 |
| *** mriedem has left #openstack-neutron | 20:44 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/neutron-vpnaas: Updated from global requirements https://review.openstack.org/206826 | 20:44 |
| *** abhiraut has quit IRC | 20:45 | |
| *** abhiraut has joined #openstack-neutron | 20:45 | |
| *** salv-orlando has joined #openstack-neutron | 20:46 | |
| banix | apuimedo: Hi Tony. Are you around? | 20:46 |
| *** cbouch has quit IRC | 20:47 | |
| apuimedo | banix: Nice to see you aroung Mohammad :-) | 20:47 |
| *** shashank_hegde has joined #openstack-neutron | 20:48 | |
| banix | apuimedo: same here :) Just saw the email by Gal; took me a while to get his irc nick name: gsagie | 20:48 |
| *** lucas__ has joined #openstack-neutron | 20:48 | |
| banix | noticed the meeting email doesn’t have an IRC channel | 20:48 |
| apuimedo | ;-) | 20:49 |
| banix | looking at current meeting the only channel available is openstack-meeting-4 | 20:49 |
| banix | should i go ahead and book that? | 20:49 |
| apuimedo | yes, that's the one he requested, IIRC | 20:49 |
| apuimedo | let me see if I find it | 20:49 |
| *** eckelcu has joined #openstack-neutron | 20:49 | |
| *** yapeng has joined #openstack-neutron | 20:50 | |
| apuimedo | banix: https://review.openstack.org/#/c/206324/ | 20:50 |
| *** edmondsw has joined #openstack-neutron | 20:50 | |
| apuimedo | is that change enough, or something extra needs to be done? | 20:51 |
| banix | apuimedo: gsagie cool! just need to send an email when approved | 20:51 |
| apuimedo | :-) | 20:51 |
| *** lucas__ has quit IRC | 20:51 | |
| *** lucas__ has joined #openstack-neutron | 20:51 | |
| apuimedo | Hopefully before Monday | 20:51 |
| apuimedo | so we can have the first | 20:51 |
| banix | sounds good! | 20:52 |
| apuimedo | :-) | 20:53 |
| *** pal has quit IRC | 20:54 | |
| banix | @apuimedo do we have a wiki page for Kuryr? | 20:55 |
| *** Sukhdev_ has quit IRC | 20:55 | |
| *** neophy has quit IRC | 20:55 | |
| *** ByPasS has quit IRC | 20:55 | |
| *** lucas__ has quit IRC | 20:55 | |
| banix | where the agenda can be posted as requested in the guideline | 20:56 |
| *** ByPasS has joined #openstack-neutron | 20:56 | |
| apuimedo | not yet | 20:56 |
| *** rmcall has joined #openstack-neutron | 20:56 | |
| mestery | banix apuimedo: Did gsagie not create a meeting wiki page for the agenda? | 20:56 |
| *** yapeng has quit IRC | 20:56 | |
| mestery | If not, feel free to do that, put a skeleton there, and send email to ML please | 20:56 |
| banix | mestery: not in the patch submitted so infra may not like it | 20:56 |
| apuimedo | I did not see it | 20:57 |
| banix | apuimedo: let me do that and update the patch if that’s ok; you can update as you see fit later | 20:57 |
| apuimedo | it's okay for me | 20:57 |
| *** lucas__ has joined #openstack-neutron | 20:57 | |
| apuimedo | and I appreciate it ;-) | 20:58 |
| mestery | banix: I commented on the review about that | 20:58 |
| apuimedo | thanks mestery | 20:58 |
| apuimedo | just saw it | 20:58 |
| banix | mestery: makes sense. we’ll fix it shortly | 20:59 |
| *** helenxu has quit IRC | 20:59 | |
| mestery | awesome thanks! | 20:59 |
| *** tflynn_ has joined #openstack-neutron | 20:59 | |
| *** roeyc has joined #openstack-neutron | 20:59 | |
| *** crc32 has quit IRC | 21:01 | |
| *** abhiraut has quit IRC | 21:01 | |
| *** lucas__ has quit IRC | 21:01 | |
| *** nagyz_ has quit IRC | 21:02 | |
| ZZelle | mestery, hi | 21:02 |
| *** tflynn has quit IRC | 21:02 | |
| mestery | ZZelle: Howdy! How are you? | 21:02 |
| *** nagyz_ has joined #openstack-neutron | 21:03 | |
| ZZelle | mestery, fine!, J-2 before holidays! | 21:03 |
| mestery | :) | 21:03 |
| *** absubram has quit IRC | 21:04 | |
| ZZelle | mestery, should we abandon old specs? typically https://review.openstack.org/152769 | 21:04 |
| *** Aish has quit IRC | 21:04 | |
| mestery | ZZelle: Let me run the auto-abandon script and see if this one goes away. | 21:04 |
| Sam-I-Am | mestery: btw, havent had a chance to audit the devdocs :/ | 21:04 |
| mestery | Sam-I-Am: It's fine, no worries | 21:05 |
| ZZelle | mestery, someone required some updates on the change, so it won't go away | 21:05 |
| mestery | ZZelle: I may have to abandon that one by hand, it hasn't gone anywhere anyways | 21:05 |
| ZZelle | mestery, exactly anywhere O_o | 21:05 |
| mestery | ZZelle: Gone :) | 21:06 |
| ZZelle | mestery, byebye | 21:07 |
| mestery | ZZelle: later :) | 21:07 |
| mestery | lol | 21:07 |
| *** jckasper has quit IRC | 21:07 | |
| *** ajo_ has joined #openstack-neutron | 21:11 | |
| *** ajo_ has quit IRC | 21:12 | |
| *** yamamoto has joined #openstack-neutron | 21:12 | |
| *** ajo_ has joined #openstack-neutron | 21:12 | |
| *** ajo_ has quit IRC | 21:13 | |
| *** ByPasS has quit IRC | 21:13 | |
| ajo | ihrachyshka ping ;) | 21:13 |
| *** yamamoto has quit IRC | 21:14 | |
| ihrachyshka | ajo, pong. here you go. | 21:15 |
| ihrachyshka | ajo, so, first thing first, we lack reviews, so you can help with it. do you want to go one by one? or will you check the queue yourself? | 21:16 |
| ajo | I'm checking queue now, I see there are lot's of +A not merged yet | 21:16 |
| *** tflynn_ has quit IRC | 21:16 | |
| *** ekarlso has quit IRC | 21:16 | |
| ihrachyshka | ajo, yeah, those are gampel's message queue notification driver + api tests (with some fixes) | 21:17 |
| banix | mestery: apuimedo: fyi, updated the gal’s infra patch with the meeting url | 21:17 |
| mestery | banix: +1 | 21:17 |
| *** tflynn has joined #openstack-neutron | 21:17 | |
| *** phil_h has quit IRC | 21:17 | |
| *** ByPasS has joined #openstack-neutron | 21:18 | |
| apuimedo | banix: I have to agree with Kyle about the capitalization of "Kuryr" | 21:18 |
| *** rpothier has quit IRC | 21:18 | |
| ihrachyshka | ajo, tell me when there are questions or you are done for first review iteration thru the queue | 21:19 |
| ajo | ihrachyshka, ack, I'm re-reviewing the db-rule contraction to one table | 21:19 |
| ihrachyshka | ajo, btw jschwarz made a huge progress on clinet | 21:20 |
| ajo | clinet? | 21:20 |
| ihrachyshka | I think he has all done and will post it tomorrow | 21:20 |
| ajo | client? :) | 21:20 |
| ihrachyshka | client, yes | 21:20 |
| ajo | ahhh | 21:20 |
| ajo | niiice | 21:20 |
| *** tflynn_ has joined #openstack-neutron | 21:21 | |
| banix | @apuimedo @mestery Done. Capitalized Kuryr | 21:21 |
| apuimedo | banix: just +1ed it ;-) | 21:21 |
| mestery | awesome | 21:22 |
| mestery | :) | 21:22 |
| banix | at least i got the spelling right :) | 21:22 |
| apuimedo | ;-) | 21:23 |
| ihrachyshka | ajo, it's bad there is a bug in underlying patch with API tests that I just spotted so unification won't go just now. | 21:23 |
| *** tflynn has quit IRC | 21:23 | |
| apuimedo | banix: well, the spelling is wrong by default, it should be Kurýr | 21:23 |
| apuimedo | :P | 21:23 |
| ajo | ihrachyshka :/ yikes | 21:24 |
| banix | apuimedo: hahaha | 21:24 |
| ihrachyshka | ajo, but that's ok, just review other stuff. in most cases votes will remain | 21:24 |
| *** eckelcu has quit IRC | 21:25 | |
| *** banix has quit IRC | 21:25 | |
| *** tflynn_ has quit IRC | 21:25 | |
| ihrachyshka | ajo, it may take some time for you. let's discuss critical stuff and then I leave you with it and go to sleep :) | 21:26 |
| *** tflynn has joined #openstack-neutron | 21:26 | |
| ajo | ihrachyshka, yes, | 21:26 |
| ajo | let's sync and then I can go on | 21:27 |
| openstackgerrit | Rossella Sblendido proposed openstack/neutron: Introduce get_ports_attributes in OVSBridge https://review.openstack.org/200279 | 21:27 |
| ihrachyshka | ajo, so generally, we have those pieces missing: 1) func tests 2) fullstack 3) client 4) rpc updates. 3 is under control; 1 and 2 kinda yes; as for 4, we don't have anything reviewable. do you think we'll have smth tomorrow? | 21:27 |
| *** spandhe has joined #openstack-neutron | 21:28 | |
| ajo | ihrachyshka, my energy is limited, may be we prefer me to work on that before reviewing? | 21:28 |
| *** tflynn has quit IRC | 21:28 | |
| ihrachyshka | ajo, meh, reviewers are also limited hugely | 21:28 |
| ajo | I charged myself with caffeine, but it will last what it will last... | 21:28 |
| ihrachyshka | ajo, we can't merge anything lately | 21:28 |
| *** tflynn has joined #openstack-neutron | 21:29 | |
| ajo | ack | 21:29 |
| ihrachyshka | partly due to gate but not only | 21:29 |
| ajo | ihrachyshka, did deadlocks go away completely or other stuff? | 21:29 |
| ihrachyshka | so let's make reviews prio where they are more or less quick and then switch to updates. | 21:29 |
| ihrachyshka | ajo, no deadlocks, all clear | 21:30 |
| ihrachyshka | pretty stable I would say. gate wise | 21:30 |
| *** TrevorV has quit IRC | 21:30 | |
| ihrachyshka | ajo, one thing to mention is John was trying to push https://review.openstack.org/207053 today | 21:30 |
| ajo | ack | 21:30 |
| ihrachyshka | I think we should not go that route as long as he has client up though | 21:31 |
| *** abregman is now known as abregman|ZzZz | 21:31 | |
| ajo | no, we may not | 21:31 |
| ajo | that makes the API inconsistent | 21:31 |
| ajo | btw, I've seen some services (I think it was Lbaas or metering) doing something similar, exposing objects via two URIs | 21:32 |
| ihrachyshka | ajo, not really if it's done properly, but my reasoning is a bit different - we should not switch horses if we can avoid it. | 21:32 |
| ajo | so you could access then via client without the top-object | 21:32 |
| ajo | yeah | 21:32 |
| ajo | that too | 21:32 |
| ihrachyshka | ajo, that's bad I would say, but whatever. as long as client is up, I am good | 21:32 |
| ihrachyshka | ajo, another thing to note is https://review.openstack.org/207043 | 21:32 |
| ihrachyshka | comments on that one? | 21:33 |
| *** eckelcu has joined #openstack-neutron | 21:33 | |
| *** achanda has joined #openstack-neutron | 21:34 | |
| ajo | ihrachyshka, that's good, it was my initial intention, but VOs support for that was in a version that wasn't released weeks ago | 21:34 |
| ajo | I didn't check how you did it yet btw | 21:34 |
| *** dane_leblanc has quit IRC | 21:34 | |
| ajo | 'rules': obj_fields.ListOfObjectsField('QosRule', subclasses=True) | 21:34 |
| ajo | ahh | 21:34 |
| ajo | cool | 21:34 |
| ajo | is it supported now? :) | 21:34 |
| ihrachyshka | seems like yes | 21:34 |
| ihrachyshka | it works in unit and API tests so... | 21:34 |
| ajo | ihrachyshka, what do we show when we GET a policy now? | 21:35 |
| ajo | can we distinguish which kind of rule each is ? | 21:35 |
| ihrachyshka | ajo, a dict with 'rules' attr that is a list of dicts per rule (each rule has a 'type') | 21:35 |
| ihrachyshka | ajo, see https://review.openstack.org/#/c/207043/6/neutron/objects/qos/rule.py | 21:35 |
| ajo | hmmm | 21:36 |
| ihrachyshka | specifically, to_dict | 21:36 |
| ajo | dict_['type'] = self.rule_type | 21:36 |
| ajo | I see :) | 21:36 |
| ihrachyshka | yeap | 21:36 |
| *** johnbelamaric has quit IRC | 21:36 | |
| ajo | ok, so we create the 'type' field in runtime from classes | 21:36 |
| ajo | makes sense | 21:36 |
| ihrachyshka | yeap. it would be weird to have type as object fields | 21:37 |
| ajo | ok, I will review, but looks good to me. | 21:37 |
| ihrachyshka | since object is already type specific | 21:37 |
| *** tflynn_ has joined #openstack-neutron | 21:37 | |
| *** tflynn has quit IRC | 21:37 | |
| ajo | yeah yeah :) | 21:37 |
| ajo | good work | 21:37 |
| ihrachyshka | and here I enforce single rule type per policy: https://review.openstack.org/207170 | 21:37 |
| *** achanda has quit IRC | 21:39 | |
| ihrachyshka | ajo, the plan is btw to merge before pecan, and some time in 2 weeks from now or so | 21:39 |
| ajo | ihrachyshka, ack, that makes sense for now, yes, that was missing in my todo analysis | 21:39 |
| ihrachyshka | the next week should be decisive | 21:39 |
| ajo | understood | 21:39 |
| ajo | mestery, ihrachyshka ++ ;) ^ | 21:39 |
| ihrachyshka | ajo, one thing that amuller mentioned today is that we may need to block policy kill if there are bindings to it | 21:40 |
| ihrachyshka | the way we do for sec groups | 21:40 |
| ihrachyshka | do you think we should do it? | 21:40 |
| ajo | ihrachyshka, in the future we may want to create some mechanism to let rules decide if they play well with each other, etc.. | 21:40 |
| *** tflynn has joined #openstack-neutron | 21:40 | |
| ihrachyshka | that is definitely in the future :) | 21:40 |
| *** tflynn has quit IRC | 21:40 | |
| ajo | yes | 21:40 |
| ajo | about blocking vs deleting | 21:40 |
| ajo | I guess we may want to block, as that's the general neutron behavior | 21:41 |
| *** vishwanathj_ has quit IRC | 21:41 | |
| *** rmcall has quit IRC | 21:41 | |
| *** tflynn has joined #openstack-neutron | 21:41 | |
| openstackgerrit | Paul Michali proposed openstack/neutron-vpnaas: VPNaaS: Correcting method name for functional test https://review.openstack.org/207230 | 21:41 |
| ajo | When other objects are depending on the one you're deleting, we return an error | 21:41 |
| ihrachyshka | ajo, ok. I suppose it's done in API layer, not thru some db constraints | 21:41 |
| *** jasondotstar has quit IRC | 21:41 | |
| amuller | ihrachyshka: I think for most objects it's done in the plugin layer | 21:42 |
| ajo | yes..., I haven't look at how SGs do it, but it's a possiblity | 21:42 |
| ajo | possibility | 21:42 |
| ajo | hi amuller ;) | 21:42 |
| amuller | ajo: hey man :) | 21:42 |
| ihrachyshka | amuller, so qos_plugin.py in our case. | 21:42 |
| *** tiswanso has joined #openstack-neutron | 21:43 | |
| ihrachyshka | ajo, so apart from testing that hopefully is under control, and client - same, we need update notifications | 21:43 |
| ajo | yes, at the plugin level we may want to check if in use -> return error/throw exception that can be correctly understood by the API layer, | 21:43 |
| ajo | ihrachyshka, ack, I will go on with reviews | 21:43 |
| ajo | and jump into that. | 21:43 |
| *** tflynn_ has quit IRC | 21:43 | |
| ihrachyshka | ajo, ok, I will leave you here | 21:43 |
| ihrachyshka | cu man! | 21:43 |
| ajo | I think it's under control now that I found the right way to do it | 21:43 |
| *** ekarlso has joined #openstack-neutron | 21:44 | |
| ihrachyshka | ajo, great to hear it. | 21:44 |
| ajo | I just need to finish coding it | 21:44 |
| ihrachyshka | let's target Fri, with tests ;) | 21:44 |
| ihrachyshka | well, ok, no tests | 21:44 |
| ajo | ack :) | 21:44 |
| ajo | let's target something working for friday | 21:44 |
| ajo | and an update tonight | 21:44 |
| ihrachyshka | ah, ok, I saw your ack! It means Fri! No way back lol. | 21:44 |
| ajo | addressing the comments and advancing toward completion :) | 21:44 |
| ajo | lol | 21:45 |
| ajo | it's more realistic what you said afterwards, let's face it | 21:45 |
| ajo | ;) | 21:45 |
| ihrachyshka | good good. I will let you run on caffeine for a while. gn. | 21:45 |
| *** pc_m has quit IRC | 21:45 | |
| ajo | gn ihrachyshka ;) | 21:45 |
| ihrachyshka | ajo, nah, you should set unrealistic schedule and then meet it | 21:45 |
| ihrachyshka | that's qos way | 21:45 |
| *** abhiraut has joined #openstack-neutron | 21:46 | |
| *** nmagnezi has joined #openstack-neutron | 21:46 | |
| *** dboik has quit IRC | 21:46 | |
| *** tflynn has quit IRC | 21:47 | |
| *** tflynn has joined #openstack-neutron | 21:48 | |
| *** nmagnezi has quit IRC | 21:49 | |
| *** ByPasS has quit IRC | 21:51 | |
| *** arajagopal has quit IRC | 21:51 | |
| *** Aish has joined #openstack-neutron | 21:52 | |
| *** Aish has left #openstack-neutron | 21:52 | |
| *** roeyc has quit IRC | 21:53 | |
| *** ByPasS has joined #openstack-neutron | 21:53 | |
| openstackgerrit | Henry Gessau proposed openstack/neutron: Support for independent alembic branches in sub-projects https://review.openstack.org/198524 | 21:54 |
| *** eckelcu has quit IRC | 21:54 | |
| *** tflynn_ has joined #openstack-neutron | 21:55 | |
| *** tflynn has quit IRC | 21:55 | |
| *** roeyc has joined #openstack-neutron | 21:56 | |
| *** ihrachyshka has quit IRC | 21:56 | |
| *** eckelcu has joined #openstack-neutron | 21:57 | |
| *** tflynn_ has quit IRC | 21:57 | |
| openstackgerrit | Henry Gessau proposed openstack/neutron: Support for independent alembic branches in sub-projects https://review.openstack.org/198524 | 21:58 |
| *** tflynn has joined #openstack-neutron | 21:58 | |
| *** thorst has quit IRC | 21:58 | |
| *** dboik has joined #openstack-neutron | 21:59 | |
| *** prad has quit IRC | 21:59 | |
| *** jbell8 has quit IRC | 21:59 | |
| *** markvoelker has quit IRC | 22:02 | |
| *** Kennan2 has joined #openstack-neutron | 22:02 | |
| *** Kennan has quit IRC | 22:03 | |
| *** vivek-ebay has joined #openstack-neutron | 22:04 | |
| *** spandhe has quit IRC | 22:06 | |
| *** achanda has joined #openstack-neutron | 22:11 | |
| *** daneyon has quit IRC | 22:11 | |
| *** busterswt has quit IRC | 22:12 | |
| *** tiswanso has quit IRC | 22:13 | |
| *** txomon|fon has joined #openstack-neutron | 22:13 | |
| *** LukeHinds has quit IRC | 22:14 | |
| *** yamamoto has joined #openstack-neutron | 22:15 | |
| *** hitalia has quit IRC | 22:15 | |
| *** vivek-ebay has quit IRC | 22:15 | |
| *** tflynn_ has joined #openstack-neutron | 22:17 | |
| *** tflynn_ has quit IRC | 22:17 | |
| *** vivek-ebay has joined #openstack-neutron | 22:18 | |
| *** Sukhdev_ has joined #openstack-neutron | 22:19 | |
| *** yamamoto has quit IRC | 22:20 | |
| *** tflynn has quit IRC | 22:20 | |
| openstackgerrit | Merged openstack/neutron: Create packages for quota modules https://review.openstack.org/188618 | 22:21 |
| *** minwang2 has quit IRC | 22:23 | |
| *** dboik has quit IRC | 22:25 | |
| *** banix has joined #openstack-neutron | 22:25 | |
| *** jasondotstar has joined #openstack-neutron | 22:28 | |
| *** minwang2 has joined #openstack-neutron | 22:28 | |
| *** armax has joined #openstack-neutron | 22:29 | |
| *** tflynn has joined #openstack-neutron | 22:32 | |
| *** mickeys has quit IRC | 22:32 | |
| *** mickeys has joined #openstack-neutron | 22:33 | |
| *** ByPasS has quit IRC | 22:35 | |
| *** armax has quit IRC | 22:36 | |
| *** lindycoder has quit IRC | 22:36 | |
| *** tflynn has quit IRC | 22:37 | |
| *** rossella_s has quit IRC | 22:37 | |
| *** ByPasS has joined #openstack-neutron | 22:37 | |
| *** rossella_s has joined #openstack-neutron | 22:38 | |
| *** mwagner_lap has quit IRC | 22:38 | |
| *** sdake has quit IRC | 22:40 | |
| *** sdake has joined #openstack-neutron | 22:40 | |
| *** padkrish has quit IRC | 22:42 | |
| *** sdake has quit IRC | 22:42 | |
| *** padkrish has joined #openstack-neutron | 22:42 | |
| *** ByPasS has quit IRC | 22:43 | |
| *** fzdarsky has quit IRC | 22:43 | |
| *** fzdarsky_ has quit IRC | 22:43 | |
| *** jaypipes has quit IRC | 22:43 | |
| *** ByPasS has joined #openstack-neutron | 22:43 | |
| *** sridhar_ram has joined #openstack-neutron | 22:44 | |
| *** asdf has joined #openstack-neutron | 22:45 | |
| *** thorst has joined #openstack-neutron | 22:45 | |
| *** padkrish_ has joined #openstack-neutron | 22:45 | |
| *** padkrish has quit IRC | 22:46 | |
| *** eckelcu has quit IRC | 22:48 | |
| openstackgerrit | Assaf Muller proposed openstack/neutron: Enable fullstack multinode tests, add L3 HA test exemplar https://review.openstack.org/188221 | 22:50 |
| *** abehl has quit IRC | 22:50 | |
| *** ByPasS has quit IRC | 22:50 | |
| *** ByPasS has joined #openstack-neutron | 22:52 | |
| *** vishwanathj has joined #openstack-neutron | 22:53 | |
| *** amuller has quit IRC | 22:53 | |
| *** eckelcu has joined #openstack-neutron | 22:53 | |
| *** markvoelker_ has joined #openstack-neutron | 22:57 | |
| *** dsneddon has quit IRC | 22:57 | |
| *** ByPasS has quit IRC | 22:59 | |
| *** asdf has quit IRC | 23:01 | |
| *** gsagie has quit IRC | 23:03 | |
| *** ByPasS has joined #openstack-neutron | 23:04 | |
| *** alexpilotti has quit IRC | 23:05 | |
| *** moshele has joined #openstack-neutron | 23:06 | |
| *** ZZelle has quit IRC | 23:07 | |
| *** daneyon has joined #openstack-neutron | 23:09 | |
| *** daneyon_ has joined #openstack-neutron | 23:10 | |
| *** daneyon has quit IRC | 23:13 | |
| *** thorst has quit IRC | 23:19 | |
| *** vishwanathj has quit IRC | 23:20 | |
| *** ivar-lazzaro has quit IRC | 23:23 | |
| *** moshele has quit IRC | 23:24 | |
| *** annegentle has joined #openstack-neutron | 23:28 | |
| *** daneyon_ has quit IRC | 23:29 | |
| *** regXboi has quit IRC | 23:30 | |
| *** oomichi has joined #openstack-neutron | 23:31 | |
| *** zhenguo has joined #openstack-neutron | 23:31 | |
| *** ByPasS has quit IRC | 23:32 | |
| *** jasondotstar has quit IRC | 23:33 | |
| *** ByPasS has joined #openstack-neutron | 23:34 | |
| *** hichihara has joined #openstack-neutron | 23:36 | |
| *** mwagner_lap has joined #openstack-neutron | 23:37 | |
| *** mickeys has quit IRC | 23:38 | |
| *** obondarev has quit IRC | 23:38 | |
| *** fzdarsky has joined #openstack-neutron | 23:39 | |
| *** obondarev has joined #openstack-neutron | 23:39 | |
| *** fzdarsky_ has joined #openstack-neutron | 23:39 | |
| *** jamielennox is now known as jamielennox|away | 23:39 | |
| *** eckelcu has quit IRC | 23:43 | |
| openstackgerrit | Roey Chen proposed openstack/neutron: Don't use autonested_transaction when creating default security-group https://review.openstack.org/207034 | 23:44 |
| openstackgerrit | Kevin Benton proposed openstack/neutron: Check that VXLAN is not in use in LB VXLAN check https://review.openstack.org/207256 | 23:44 |
| *** jasondotstar has joined #openstack-neutron | 23:45 | |
| *** armax has joined #openstack-neutron | 23:47 | |
| *** rmcall has joined #openstack-neutron | 23:47 | |
| *** Kennan2 is now known as Kennan | 23:53 | |
| *** daneyon has joined #openstack-neutron | 23:53 | |
| *** bharath_ has quit IRC | 23:53 | |
| *** miyagishi_t has joined #openstack-neutron | 23:53 | |
| *** shwetaap1 has quit IRC | 23:54 | |
| *** amickus has quit IRC | 23:54 | |
| *** zhhuabj has quit IRC | 23:54 | |
| *** ajmiller has quit IRC | 23:56 | |
| *** mlavalle has quit IRC | 23:56 | |
| *** zhhuabj has joined #openstack-neutron | 23:58 | |
| *** azbiswas has quit IRC | 23:58 | |
| *** eckelcu has joined #openstack-neutron | 23:59 | |
| *** tflynn has joined #openstack-neutron | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
