| opendevreview | Tyler Stachecki proposed openstack/neutron-lib master: model_query: Improve perf with large # of RBAC https://review.opendev.org/c/openstack/neutron-lib/+/817459 | 02:23 |
|---|---|---|
| opendevreview | Miro Tomaska proposed openstack/neutron master: Add workaround for eventlet bug[1] in get_hypervisor_hostname https://review.opendev.org/c/openstack/neutron/+/849122 | 03:09 |
| opendevreview | Takashi Kajinami proposed openstack/neutron-vpnaas-dashboard master: Replace deprecated ugettext_lazy and ungettext_lazy https://review.opendev.org/c/openstack/neutron-vpnaas-dashboard/+/849257 | 04:33 |
| opendevreview | Takashi Kajinami proposed openstack/neutron-vpnaas-dashboard master: Replace deprecated django.conf.urls.url() https://review.opendev.org/c/openstack/neutron-vpnaas-dashboard/+/849258 | 04:34 |
| opendevreview | Takashi Kajinami proposed openstack/neutron-fwaas-dashboard master: Replace deprecated ugettext_lazy and ungettext_lazy https://review.opendev.org/c/openstack/neutron-fwaas-dashboard/+/849259 | 04:35 |
| opendevreview | Takashi Kajinami proposed openstack/neutron-vpnaas-dashboard master: Replace deprecated ugettext_lazy and ungettext_lazy https://review.opendev.org/c/openstack/neutron-vpnaas-dashboard/+/849257 | 04:51 |
| opendevreview | Takashi Kajinami proposed openstack/neutron-vpnaas-dashboard master: Replace deprecated django.conf.urls.url() https://review.opendev.org/c/openstack/neutron-vpnaas-dashboard/+/849258 | 04:51 |
| opendevreview | Takashi Kajinami proposed openstack/neutron-vpnaas-dashboard master: Update python testing as per zed cycle testing runtime https://review.opendev.org/c/openstack/neutron-vpnaas-dashboard/+/849266 | 05:06 |
| opendevreview | Takashi Kajinami proposed openstack/neutron-fwaas-dashboard master: Replace deprecated ugettext_lazy and ungettext_lazy https://review.opendev.org/c/openstack/neutron-fwaas-dashboard/+/849259 | 05:42 |
| opendevreview | Sahid Orentino Ferdjaoui proposed openstack/neutron master: ovs: make vlanmanager to handle more vlan mapping per network https://review.opendev.org/c/openstack/neutron/+/839479 | 07:30 |
| opendevreview | Sahid Orentino Ferdjaoui proposed openstack/neutron master: ovs: use a local vlan per network/segmentation https://review.opendev.org/c/openstack/neutron/+/840415 | 07:30 |
| opendevreview | Sahid Orentino Ferdjaoui proposed openstack/neutron master: dhcp/rpc: retrieve network details with segments https://review.opendev.org/c/openstack/neutron/+/840416 | 07:30 |
| opendevreview | Sahid Orentino Ferdjaoui proposed openstack/neutron master: utils: add option to generate dhcp device id per segmentation https://review.opendev.org/c/openstack/neutron/+/840417 | 07:30 |
| opendevreview | Sahid Orentino Ferdjaoui proposed openstack/neutron master: ml2/mech: bind port on segment that subnet belongs https://review.opendev.org/c/openstack/neutron/+/840418 | 07:30 |
| opendevreview | Sahid Orentino Ferdjaoui proposed openstack/neutron master: dhcp: make device manager to clean only not used devices https://review.opendev.org/c/openstack/neutron/+/840419 | 07:30 |
| opendevreview | Sahid Orentino Ferdjaoui proposed openstack/neutron master: dhcp: add/use cleanup stale devices API https://review.opendev.org/c/openstack/neutron/+/840420 | 07:30 |
| opendevreview | Sahid Orentino Ferdjaoui proposed openstack/neutron master: dhcp: support multiple segmentations per network https://review.opendev.org/c/openstack/neutron/+/840421 | 07:30 |
| opendevreview | Sahid Orentino Ferdjaoui proposed openstack/neutron master: db: add segment_index to the unique constraint https://review.opendev.org/c/openstack/neutron/+/841459 | 07:30 |
| opendevreview | Sahid Orentino Ferdjaoui proposed openstack/neutron master: segment: enable multisegments support for host https://review.opendev.org/c/openstack/neutron/+/841838 | 07:30 |
| lajoskatona | obondarev: Hi, pushed a rally-openstack patch to limit docutils for it to <0.19, can you check it it please when you have few mintues? | 08:09 |
| lajoskatona | obondarev: the link, sorry : https://review.opendev.org/c/openstack/rally-openstack/+/849281 | 08:10 |
| obondarev | lajoskatona: sure, will do | 08:11 |
| lajoskatona | obondarev: I think the final solution can be done when in docutils the parser will be ready, but perhaps I can send them a mail to ask about how they plan, because now OptionParser is deprecated but no replacement for it, just a hint | 08:12 |
| obondarev | lajoskatona: strange that tests fail due to deprecation, let me check | 08:15 |
| lajoskatona | obondarev: that can be good direction also, I havent found it :-) | 08:15 |
| opendevreview | Tyler Stachecki proposed openstack/neutron master: external_net_db: Adhere to model_query changes https://review.opendev.org/c/openstack/neutron/+/817462 | 09:55 |
| sahid | Hello guys, do we know whehter we have issue with CI running DVR jobs? | 11:57 |
| sahid | https://zuul.opendev.org/t/openstack/build/0adaf40a86574a75a8d3a7a6a4c098d9 and https://zuul.opendev.org/t/openstack/build/94705b6010874e07bd782bf4a4bea466 | 11:58 |
| opendevreview | Merged openstack/neutron-dynamic-routing master: [sqlalchemy-20] Add missing DB context decorator https://review.opendev.org/c/openstack/neutron-dynamic-routing/+/848645 | 12:20 |
| lajoskatona | sahid: Hi, thanks, I saw some failures this morning, I check if it is general or just occasional | 12:21 |
| lajoskatona | sahid: the job itself looks ok: https://zuul.openstack.org/builds?job_name=neutron-ovs-tempest-dvr-ha-multinode-full in general | 12:25 |
| sahid | lajoskatona: Hello, thanks to have took time looking at it, does that ring a bell https://zuul.opendev.org/t/openstack/build/0adaf40a86574a75a8d3a7a6a4c098d9/log/compute1/logs/screen-q-agt.txt?severity=4 ? | 12:26 |
| sahid | yes... looks like all failing are related to me | 12:27 |
| lajoskatona | sahid: yes :-( | 12:30 |
| lajoskatona | sahid: the log you pointed previously is due to that q-svc is restarted at that time (grenade most probably doing upgrade at that time of the services): https://428b85e7030640e63a25-64321574d333a1d78009697cef7af9d1.ssl.cf2.rackcdn.com/839479/14/check/neutron-ovs-grenade-dvr-multinode/0adaf40/controller/logs/screen-q-svc.txt | 12:31 |
| lajoskatona | sahid: 10:27:15.550058 - 10:31:18.348196 (the timestamps for the restart of the service based on q-svc log) | 12:32 |
| lajoskatona | sahid: by-the-way: You are the "owner" of the multiple routed provider net feature? | 12:43 |
| lajoskatona | sahid: shall I ask if you can clean the specs? https://review.opendev.org/c/openstack/neutron-specs/+/823823 & https://review.opendev.org/c/openstack/neutron-specs/+/824464 & https://review.opendev.org/c/openstack/neutron-specs/+/824465 ? | 12:43 |
| lajoskatona | sahid: if you need help just ping me (i.e.: to abandon some of these three if you have all the content in one of them....) | 12:44 |
| *** damiandabrowski[m] is now known as damiandabrowski | 12:45 | |
| opendevreview | Sahid Orentino Ferdjaoui proposed openstack/neutron master: ovs: make vlanmanager to handle more vlan mapping per network https://review.opendev.org/c/openstack/neutron/+/839479 | 12:48 |
| opendevreview | Sahid Orentino Ferdjaoui proposed openstack/neutron master: ovs: use a local vlan per network/segmentation https://review.opendev.org/c/openstack/neutron/+/840415 | 12:48 |
| opendevreview | Sahid Orentino Ferdjaoui proposed openstack/neutron master: dhcp/rpc: retrieve network details with segments https://review.opendev.org/c/openstack/neutron/+/840416 | 12:48 |
| opendevreview | Sahid Orentino Ferdjaoui proposed openstack/neutron master: utils: add option to generate dhcp device id per segmentation https://review.opendev.org/c/openstack/neutron/+/840417 | 12:48 |
| opendevreview | Sahid Orentino Ferdjaoui proposed openstack/neutron master: ml2/mech: bind port on segment that subnet belongs https://review.opendev.org/c/openstack/neutron/+/840418 | 12:48 |
| opendevreview | Sahid Orentino Ferdjaoui proposed openstack/neutron master: dhcp: make device manager to clean only not used devices https://review.opendev.org/c/openstack/neutron/+/840419 | 12:48 |
| opendevreview | Sahid Orentino Ferdjaoui proposed openstack/neutron master: dhcp: add/use cleanup stale devices API https://review.opendev.org/c/openstack/neutron/+/840420 | 12:48 |
| opendevreview | Sahid Orentino Ferdjaoui proposed openstack/neutron master: dhcp: support multiple segmentations per network https://review.opendev.org/c/openstack/neutron/+/840421 | 12:48 |
| opendevreview | Sahid Orentino Ferdjaoui proposed openstack/neutron master: db: add segment_index to the unique constraint https://review.opendev.org/c/openstack/neutron/+/841459 | 12:48 |
| opendevreview | Sahid Orentino Ferdjaoui proposed openstack/neutron master: segment: enable multisegments support for host https://review.opendev.org/c/openstack/neutron/+/841838 | 12:48 |
| sahid | lajoskatona: could you abandon: https://review.opendev.org/c/openstack/neutron-specs/+/824464 and https://review.opendev.org/c/openstack/neutron-specs/+/824465 ? | 12:50 |
| sahid | yes I'm not able to clean them, i'm the new owner | 12:51 |
| sahid | thanks a lot | 12:51 |
| lajoskatona | sahid: done, thanks for checking | 12:52 |
| sahid | I may have find the CI issue, I have https://review.opendev.org/c/openstack/neutron/+/839479/14/neutron/plugins/ml2/drivers/openvswitch/agent/ovs_neutron_agent.py#1953, I removed the "else" | 12:52 |
| lajoskatona | sahid: I will check it | 13:01 |
| dulek | slaweq: Hi! We're working on an OpenShift feature and came across this problem. So technically number of IPs per port is unlimited in Neutron, but I bet there's some practical limit. At what value would you put a safe maximum here? | 13:35 |
| dulek | We're asking because e.g. in AWS that number is limited to 16. | 13:35 |
| *** dasm|off is now known as dasm | 13:51 | |
| sahid | s/CI issue/change in my patch that makes CI failing | 13:59 |
| lajoskatona | skoech[m]: Hi, regarding https://review.opendev.org/q/topic:cascade_delete_extension | 15:10 |
| lajoskatona | skoech[m]: thanks for uploading, I think you have to rebase to master as both patches have merge conflict | 15:10 |
| sahid | o/ dulek have you considered looking at how we map ips per port in database? | 15:28 |
| dulek | sahid: I just assumed it's a regular one-to-many DB relationship? | 15:32 |
| dulek | sahid: It seems so, right: https://github.com/openstack/neutron/blob/182044d628314290a38440bab203d27fb9db755d/neutron/db/models_v2.py#L71-L85 ? | 15:37 |
| sahid | yes looks like it is https://github.com/openstack/neutron/blob/182044d628314290a38440bab203d27fb9db755d/neutron/db/models_v2.py#L110 | 15:49 |
| sahid | does not look to be right to do not have limitation... | 15:49 |
| sahid | have you check whehter we don't have quota for that specific case? | 15:49 |
| sahid | lajoskatona: do you have pointers for me to build locally an env to execute neutron-ovs-grenade-dvr-multinode ? | 15:54 |
| lajoskatona | sahid: grenade is an interesting question, I never run grenade locally, but I would start here: https://github.com/openstack/grenade/blob/master/README.rst | 16:10 |
| lajoskatona | sahid: and the job definition is here: https://opendev.org/openstack/neutron/src/branch/master/zuul.d/grenade.yaml#L77 , not sure if you can reproduce it locally based on that | 16:11 |
| lajoskatona | dulek: Hi, slaweq, is on PTO, and to tell the truth I don't think we have specific hardcoded or configured limit for IPs for port | 16:12 |
| lajoskatona | dulek: I would expect 2 limiting factors: * performance after some and * backend limitation perhaps | 16:12 |
| lajoskatona | dulek: but now we have no such test to see if we have such (at least I don't know about it) | 16:13 |
| lajoskatona | dulek: tomorrow we have team meeting at 1400UTC, would be good to discuss your usecase if you think | 16:14 |
| sahid | lajoskatona: can you give me access to one of the job failing< | 16:14 |
| sahid | ? | 16:14 |
| dulek | lajoskatona: Thanks for help! I've discussed this a bit with ralonsoh and apparently it was tested up to 200 adresses. We'll probably just cap it at 64, which should be more than enough for our use case, while sounds like a safe limit for Neutron. | 16:15 |
| sahid | dulek: there is no quotas for this so? or you havnt checked? | 16:33 |
| dulek | sahid: Sorry, I don't think I understand. There doesn't seem to be a DB-induced limit as IPs are a separate table with as relationship to a port. | 16:36 |
| lajoskatona | sahid: I can't give you acces to these nodes, for that some infra people can help, try on #openstack-infra | 16:39 |
| lajoskatona | dulek: ok, good to know | 16:40 |
| sahid | dulek: https://docs.openstack.org/neutron/latest/admin/ops-quotas.html | 17:08 |
| sahid | lajoskatona: ack I think I have found the problem | 17:09 |
| lajoskatona | sahid: +1 | 17:09 |
| opendevreview | Sharon Koech proposed openstack/neutron master: Network Cascade Deletion Extension https://review.opendev.org/c/openstack/neutron/+/849039 | 17:28 |
| skoech[m] | lajoskatona: Yeah, I forgot to rebase. On it now, hope the merge conflicts are solved. | 17:31 |
| opendevreview | Sharon Koech proposed openstack/neutron-lib master: Network Cascade Deletion Extension https://review.opendev.org/c/openstack/neutron-lib/+/849046 | 17:34 |
| opendevreview | Brian Haley proposed openstack/neutron master: Remove import of 'imp' module https://review.opendev.org/c/openstack/neutron/+/842450 | 18:21 |
| turgle | Hi everyone, i have an issue with ovn. One of my compute node lost all vlan network | 18:34 |
| turgle | i Check no port patch-provnet-*-to-br-int create in br-ex | 18:34 |
| turgle | I deploy openstack victoria at 2021 and ovn 20.03 | 18:35 |
| turgle | anyone have idea about this isssue? | 18:35 |
| turgle | ovs-ofctl dump-flows br-int table=0 |grep vlan don't show anything | 18:44 |
| turgle | another compute run fine have flow about vlan | 18:45 |
| lajoskatona | skoech[m]: +1, good luck with it :-) | 19:15 |
| skoech[m] | Thanks, I need it. :D | 19:17 |
| skoech[m] | s/Thanks/lajoskatona: thanks/ | 19:18 |
| lajoskatona | skoech[m]: I still see mergeconflicts, if you have problems with it tomorrow morning we can try to push it in a video call | 19:19 |
| skoech[m] | lajoskatona: yeah, that sounds good. What time works for you? | 19:21 |
| *** dasm is now known as dasm|off | 22:18 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!