| opendevreview | Merged openstack/neutron master: ovn: don't log about rev number bumped when it was not bumped https://review.opendev.org/c/openstack/neutron/+/919536 | 00:53 |
|---|---|---|
| opendevreview | Merged openstack/neutron master: tests: Remove unnecessary hasattr() by always defining attribute https://review.opendev.org/c/openstack/neutron/+/938850 | 01:34 |
| opendevreview | Rico Lin proposed openstack/ovn-octavia-provider master: Add LB sync logic https://review.opendev.org/c/openstack/ovn-octavia-provider/+/925324 | 06:29 |
| opendevreview | Rico Lin proposed openstack/ovn-octavia-provider master: Add Listener sync logic https://review.opendev.org/c/openstack/ovn-octavia-provider/+/931250 | 06:29 |
| opendevreview | Rico Lin proposed openstack/ovn-octavia-provider master: Add Pool sync logic https://review.opendev.org/c/openstack/ovn-octavia-provider/+/931266 | 06:29 |
| opendevreview | Rico Lin proposed openstack/ovn-octavia-provider master: Add Member sync logic https://review.opendev.org/c/openstack/ovn-octavia-provider/+/931267 | 06:29 |
| opendevreview | Rico Lin proposed openstack/ovn-octavia-provider master: Add Health monitor sync logic https://review.opendev.org/c/openstack/ovn-octavia-provider/+/931288 | 06:29 |
| opendevreview | Rico Lin proposed openstack/ovn-octavia-provider master: Add sync floating IP support https://review.opendev.org/c/openstack/ovn-octavia-provider/+/929039 | 06:29 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: DNM - Test "neutron-ovn-tempest-ipv6-only-ovs*" with WSGI https://review.opendev.org/c/openstack/neutron/+/932601 | 07:52 |
| opendevreview | Dr. Jens Harbott proposed openstack/neutron master: Add to setup.cfg the long_description_content_type field https://review.opendev.org/c/openstack/neutron/+/939806 | 08:06 |
| vsaienko | hello neutron team, please review https://review.opendev.org/c/openstack/neutron-lib/+/939108 patch to api-docs that removes tenant_id from resources as this field is not longer returned by API due to switching to keyston v3 | 08:18 |
| ralonsoh | vsaienko, but this patch is not correct | 08:28 |
| ralonsoh | we never finished the migration to v3 and we are still returning tenant_id in the API calls | 08:28 |
| ralonsoh | an example: https://paste.opendev.org/show/bPRgRypjqb7oX4wBw6fq/ | 08:29 |
| slaweq | but we definitely should finaly do this :) | 08:30 |
| ralonsoh | slaweq, yes, of course | 08:31 |
| ralonsoh | I spent some months 2 years ago on this | 08:31 |
| ralonsoh | and there are a few places yet to cover | 08:31 |
| ralonsoh | the problem are also the other projects | 08:31 |
| ralonsoh | that are still expecting tenant_id | 08:31 |
| vsaienko | oh... I was checking this for ports/subnets and segment ranges, but looks like it is present for some resources indeed | 08:37 |
| vsaienko | maybe for network its related to https://review.opendev.org/c/openstack/neutron/+/939624 | 08:39 |
| opendevreview | Dr. Jens Harbott proposed openstack/neutron master: Add to setup.cfg the long_description_content_type field https://review.opendev.org/c/openstack/neutron/+/939806 | 08:44 |
| vsaienko | ralonsoh: yeah definitely for network https://review.opendev.org/c/openstack/neutron/+/939624 this patch is related, if look into logs from it https://6e1e1c69332cbc52bf1e-63e0619ac9734be8bebe39a119169a2b.ssl.cf1.rackcdn.com/939624/3/check/neutron-ovs-tempest-dvr-ha-multinode-full/fbf2731/job-output.txt you can see that network create returns network without tenant_id | 08:50 |
| ralonsoh | vsaienko, this patch is not merged yet | 08:55 |
| ralonsoh | network (and any other resource) is still returning tenant_id | 08:55 |
| vsaienko | outputs taken from https://review.opendev.org/c/openstack/neutron/+/934283 https://6e1e1c69332cbc52bf1e-63e0619ac9734be8bebe39a119169a2b.ssl.cf1.rackcdn.com/939624/3/check/neutron-ovs-tempest-dvr-ha-multinode-full/fbf2731/job-output.txt https://paste.openstack.org/show/bG2umz5WhboUXG0lPZSk/ | 08:59 |
| vsaienko | tenant_id is not present neither for network create or for subnet create | 09:00 |
| ralonsoh | this is the output of the CLI, not the API | 09:00 |
| ralonsoh | I've pasted you an example of the output of the API and is still returning tenant_id | 09:00 |
| vsaienko | oh.. yes you are right, I've missed that openstackclient hides it | 09:02 |
| sahid | o/ any chance to get review on https://review.opendev.org/c/openstack/neutron/+/939348 https://review.opendev.org/c/openstack/neutron/+/939627 it's related to the eventlet removal, I think we are good with them | 09:03 |
| sahid | lajoskatona: haleyb: ralonsoh: :-) | 09:03 |
| ralonsoh | sure, give me some time | 09:03 |
| sahid | if need i can adress comments shorlty | 09:03 |
| sahid | ralonsoh: sure | 09:04 |
| opendevreview | Vasyl Saienko proposed openstack/neutron master: Document gaps related to VLAN and Flat networks https://review.opendev.org/c/openstack/neutron/+/939946 | 10:27 |
| vsaienko | ralonsoh: I've found that we have a lot know issues with OVN and VLAN networking, but at same time we do not have any jobs with this configuration (or I'm missing something and we have them?) if not do you think it makes sense to add a job with VLAN tenant networks and OVN? | 10:35 |
| ralonsoh | vsaienko, why a job for this only? You can create networks of any type in the tests | 10:38 |
| vsaienko | you mean to have both geneve and vlan tenant networks enabled and in test explicitly ask tenant network type when create it? | 10:39 |
| ralonsoh | vsaienko, yes, this is something you can actually do in the tests | 10:43 |
| ralonsoh | what do you think is missing? | 10:43 |
| vsaienko | to be more precise this scenario https://github.com/openstack/octavia-tempest-plugin/blob/master/octavia_tempest_plugin/tests/scenario/v2/test_traffic_ops.py#L937 is failing on ovn + vlan + dvr, while its passing on ovn + vlan + nondvr. By looking into the code I've found there are a lot of places where we try to handle VLAN/Flat networks differently. So I thought it would be good to have a job where we can cover such scenarios b | 10:46 |
| vsaienko | y tests. | 10:46 |
| ralonsoh | vsaienko, this is something that can be added to the current jobs | 10:47 |
| ralonsoh | and yes, there are documented limitations to dvr and vlan with OVN | 10:47 |
| vsaienko | they should be here https://docs.openstack.org/neutron/latest/ovn/gaps.html right? | 10:52 |
| ralonsoh | there is a refence already: "Floating IP Port Forwarding in provider networks and with distributed routing" | 10:52 |
| vsaienko | thanks, I see it now, this limitation is not only related to port forwarding extension, it applies to whole FIP functionality with ``external`` sriov/baremetal and octavia VIP ports | 11:01 |
| vsaienko_ | ralonsoh: I'm sorry was digging into CI, so OVN + DVR exists only as periodic job, it will mean that we need to. 1. Allow VLAN segmentation in one of jobs. 2 write functional/tempest test that will reproduce octavia scenario with VIP port is this how you see it? | 12:12 |
| ralonsoh | vsaienko_, fine with this, of course | 12:14 |
| opendevreview | Vasyl Saienko proposed openstack/neutron master: Update OVN installation guide with tunings for VLAN + DVR https://review.opendev.org/c/openstack/neutron/+/939961 | 13:08 |
| opendevreview | Vasyl Saienko proposed openstack/neutron master: Add link to Octavia and SRIOV limitations from generic OVN Gaps page https://review.opendev.org/c/openstack/neutron/+/939946 | 13:09 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: [eventlet-removal] Reimplement ``common.utils.spawn_n`` https://review.opendev.org/c/openstack/neutron/+/939095 | 14:09 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: [eventlet-removal] Remove ``subprocess_popen`` https://review.opendev.org/c/openstack/neutron/+/939097 | 14:10 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: [eventlet-removal] Remove eventlet from ``TestNovaNotify`` https://review.opendev.org/c/openstack/neutron/+/939210 | 14:10 |
| ralonsoh | haleyb, ^ please check these 3 small patches | 14:10 |
| opendevreview | Anton Kurbatov proposed openstack/neutron master: Fix DHCP agent events throttling malfunction https://review.opendev.org/c/openstack/neutron/+/939970 | 14:20 |
| haleyb | ralonsoh: ack, and i guess the twine issue is almost resolved :-/ | 14:44 |
| ralonsoh | haleyb, yes, there is a patch blocking the 6.1.0 version | 14:44 |
| * haleyb was more laughing that such a small patch required a large investment | 14:46 | |
| zigo | Hi. After we had a rabbitmq cluster down time (during an upgrade), we're experiencing a shit-storm of ml2 fdb_add + fdb_add_tun on all our network nodes. It's been ongoing for like 12 hours already, and still not finished. | 14:52 |
| zigo | Is it expeted ? | 14:52 |
| zigo | Is this due to l2_population ? | 14:52 |
| zigo | Note: we use Victoria in a non-DVR mode, with 12 network nodes. | 14:53 |
| zigo | Oh, and each time I restart ovs-agent on a compute, I get the same thing in that compute... | 14:53 |
| zigo | Does anyone understand what's going on? | 14:53 |
| ralonsoh | zigo, that should stop at sometime, when you have populated all ARP tables in the nodes | 14:56 |
| zigo | ralonsoh: Is there a way to avoid all of this? | 14:56 |
| ralonsoh | I guess you are using tunnelled networks between nodes | 14:56 |
| zigo | This will make it impossible (or very hard) to handle upgrades. | 14:57 |
| zigo | Yes, we are (vxlan). | 14:57 |
| ralonsoh | to be honest, is a long time since a look at this code in particular | 14:57 |
| ralonsoh | are you sure these are update messages after the restart or messages from new VMs? | 14:58 |
| ralonsoh | do you have logs from before the downtime? | 14:58 |
| zigo | It should be since we restarted the control plane, not the agents. | 15:00 |
| zigo | The agents didn't restart. | 15:00 |
| zigo | Maybe they are doing a full-resync after the 1min timeout? | 15:00 |
| ralonsoh | is that happening only in the networker nodes OVS agent? | 15:05 |
| zigo | It's happening: | 15:07 |
| zigo | 1/ In all our network nodes since the control plane restart | 15:07 |
| zigo | 2/ In all compute nodes when restarting ovs-agent (so I didn't re-enable puppet on all compute nodes since our config management update, waiting for it to complete). | 15:07 |
| ralonsoh | the problem is that there is no way to handle the priority of these messages | 15:10 |
| ralonsoh | so they will interfere in the initial sync process | 15:11 |
| zigo | Well, everything is working, so I'm not stressed that much, it's just that in the case where I'll need to restart ovs-agent on all 100+ compute nodes (to do an OpenStack upgrade), I will not want all of them to resync their OVS state from scratch. | 15:15 |
| zigo | This, for sure, would bring down rabbit ... | 15:15 |
| zigo | I'm not supposed to run a different version of Neutron in my control plane and in the network and compute nodes too, right? | 15:16 |
| ralonsoh | you can but 1 version jump only | 15:25 |
| ralonsoh | and this is only for upgrades only | 15:25 |
| ralonsoh | and, of course, controller should have the updated version | 15:25 |
| zigo | Ok. | 15:25 |
| zigo | So maybe we'll just slow-down upgrades, to let agents process messages slowly. | 15:26 |
| opendevreview | Slawek Kaplonski proposed openstack/neutron master: Add .mypy_cache directory to the gitignore file https://review.opendev.org/c/openstack/neutron/+/938123 | 15:31 |
| opendevreview | Slawek Kaplonski proposed x/whitebox-neutron-tempest-plugin master: Wait a bit longer for tcpdump to store captures to the file https://review.opendev.org/c/x/whitebox-neutron-tempest-plugin/+/939229 | 15:31 |
| sahid | guys, any chance to make progressses on https://review.opendev.org/c/openstack/neutron/+/939627/4 https://review.opendev.org/c/openstack/neutron/+/939348/9 | 15:38 |
| sahid | sorry for that but I will start to ask for reviews on the ones that I think they are not so bad and well tested :-) | 15:39 |
| sahid | We have a lot to review aswell for osken | 15:39 |
| opendevreview | Sahid Orentino Ferdjaoui proposed openstack/neutron master: async_process: fix potential race condition with respawn https://review.opendev.org/c/openstack/neutron/+/939627 | 15:42 |
| opendevreview | Sahid Orentino Ferdjaoui proposed openstack/neutron master: async_process: remove usage of eventlet for AsyncProcess https://review.opendev.org/c/openstack/neutron/+/939348 | 15:42 |
| opendevreview | Sahid Orentino Ferdjaoui proposed openstack/neutron master: common: fix wait_until_true to support native thread https://review.opendev.org/c/openstack/neutron/+/937843 | 15:42 |
| opendevreview | Sahid Orentino Ferdjaoui proposed openstack/neutron master: ovs: reimplement signals handling https://review.opendev.org/c/openstack/neutron/+/939321 | 15:42 |
| opendevreview | Sahid Orentino Ferdjaoui proposed openstack/neutron master: ovs: remove the usage of eventlet in the OVS agent https://review.opendev.org/c/openstack/neutron/+/937765 | 15:42 |
| sahid | https://review.opendev.org/c/openstack/neutron/+/937843 this one should be not so bad as well (I have reorganised order of the patches) | 15:42 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: DNM - Test "neutron-ovn-tempest-ipv6-only-ovs*" with WSGI, ThreadPoolExecutor https://review.opendev.org/c/openstack/neutron/+/939977 | 15:43 |
| opendevreview | Bodo Petermann proposed openstack/neutron-vpnaas master: Fix automatic rescheduling from dead VPN agents https://review.opendev.org/c/openstack/neutron-vpnaas/+/939978 | 15:51 |
| opendevreview | Merged openstack/neutron master: Fixup conntrackd support https://review.opendev.org/c/openstack/neutron/+/938800 | 17:51 |
| fungi | thanks for getting that merged! | 18:07 |
| opendevreview | Merged x/whitebox-neutron-tempest-plugin master: Wait a bit longer for tcpdump to store captures to the file https://review.opendev.org/c/x/whitebox-neutron-tempest-plugin/+/939229 | 19:59 |
| opendevreview | Merged openstack/neutron stable/2024.2: Make sure that policy enforcer is initialized before use https://review.opendev.org/c/openstack/neutron/+/938911 | 20:06 |
| opendevreview | Merged openstack/neutron master: Fix neutron-status upgrade check tool https://review.opendev.org/c/openstack/neutron/+/939248 | 20:34 |
| opendevreview | Jakub Libosvar proposed openstack/neutron master: Update NAT entry on FIP update https://review.opendev.org/c/openstack/neutron/+/939918 | 21:11 |
| opendevreview | Brian Haley proposed openstack/neutron master: Add to setup.cfg the long_description_content_type field https://review.opendev.org/c/openstack/neutron/+/939806 | 21:44 |
| -opendevstatus- NOTICE: The Gerrit service on review.opendev.org will be offline momentarily while we reboot for a patch version upgrade of the software, but should return again within a few minutes | 22:45 | |
| opendevreview | Merged openstack/neutron master: Don't change original target dict by the OwnerCheck policy rule https://review.opendev.org/c/openstack/neutron/+/939624 | 22:58 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!