| opendevreview | liuyulong proposed openstack/neutron master: Refactor for meter ID Generator https://review.opendev.org/c/openstack/neutron/+/860765 | 03:05 |
|---|---|---|
| opendevreview | liuyulong proposed openstack/neutron master: Refactor for ovs qos driver meter limit features https://review.opendev.org/c/openstack/neutron/+/860766 | 03:05 |
| opendevreview | liuyulong proposed openstack/neutron master: Add meter bandwidth bandwidth support https://review.opendev.org/c/openstack/neutron/+/860767 | 03:05 |
| opendevreview | Luis Tomas Bolivar proposed openstack/ovn-octavia-provider master: Ensure OVN-LB is properly configured upon LS removal from LR https://review.opendev.org/c/openstack/ovn-octavia-provider/+/860781 | 06:29 |
| *** ramishra_ is now known as ramishra | 06:38 | |
| opendevreview | Slawek Kaplonski proposed openstack/neutron stable/zed: Disable in-band management for bridges before setting up controllers https://review.opendev.org/c/openstack/neutron/+/861622 | 07:03 |
| opendevreview | Slawek Kaplonski proposed openstack/neutron stable/yoga: Disable in-band management for bridges before setting up controllers https://review.opendev.org/c/openstack/neutron/+/861623 | 07:03 |
| opendevreview | Slawek Kaplonski proposed openstack/neutron stable/xena: Disable in-band management for bridges before setting up controllers https://review.opendev.org/c/openstack/neutron/+/861624 | 07:04 |
| opendevreview | Slawek Kaplonski proposed openstack/neutron stable/wallaby: Disable in-band management for bridges before setting up controllers https://review.opendev.org/c/openstack/neutron/+/861625 | 07:04 |
| ralonsoh | slaweq, hey, I have one small question regarding to https://review.opendev.org/c/openstack/neutron-specs/+/857858/2/specs/2023.1/configurable-default-sg-rules.rst | 07:08 |
| ralonsoh | it is related to the default default SG rules | 07:08 |
| ralonsoh | if those are registers in the DB | 07:09 |
| ralonsoh | how are we going to define/create the default ones? | 07:09 |
| ralonsoh | I'm saying this because, when installed, the table will be empty | 07:10 |
| opendevreview | Slawek Kaplonski proposed openstack/neutron stable/victoria: Disable in-band management for bridges before setting up controllers https://review.opendev.org/c/openstack/neutron/+/861705 | 07:10 |
| opendevreview | Slawek Kaplonski proposed openstack/neutron stable/ussuri: Disable in-band management for bridges before setting up controllers https://review.opendev.org/c/openstack/neutron/+/861706 | 07:12 |
| opendevreview | Slawek Kaplonski proposed openstack/neutron stable/train: Disable in-band management for bridges before setting up controllers https://review.opendev.org/c/openstack/neutron/+/861707 | 07:15 |
| opendevreview | Slawek Kaplonski proposed openstack/neutron stable/stein: Disable in-band management for bridges before setting up controllers https://review.opendev.org/c/openstack/neutron/+/861708 | 07:18 |
| opendevreview | Slawek Kaplonski proposed openstack/neutron stable/rocky: Disable in-band management for bridges before setting up controllers https://review.opendev.org/c/openstack/neutron/+/861626 | 07:19 |
| opendevreview | Slawek Kaplonski proposed openstack/neutron stable/queens: Disable in-band management for bridges before setting up controllers https://review.opendev.org/c/openstack/neutron/+/861627 | 07:19 |
| slaweq | ralonsoh lajoskatona bcafarel hi, please add to Your review queue backports https://review.opendev.org/q/Ibca81eb59fbfad71f223832228f408fb248c5dfa and check them when CI results will be ready. Thx in advance | 07:21 |
| ralonsoh | slaweq, sure | 07:21 |
| bcafarel | ralonsoh: sure thing, CI should be done by the time I check my review queue :) | 07:22 |
| crohmann | mnaser__: We spoke months ago about issues with VPNaaS on linuxbridge not coming up and reporting duplicate iptables even. I raised https://bugs.launchpad.net/neutron/+bug/1943449. As mentioned there I have the issue reproduced the issue. Could you or anybody take a look at my finding and maybe give me a hint on what else I should collect to help finding the cause for this? | 08:06 |
| frickler | ralonsoh: others: we discussed in the QA session about having OVS or OVN as default, for us it would seem preferable to keep OVS as default for openstack-wide testing, but I assume you might disagree. do you want to add that topic to neutron PTG somewhere or is your schedule already filled? today or tomorrow would work best for me. https://etherpad.opendev.org/p/qa-antelope-ptg#L120 | 08:12 |
| frickler | kopecmartin: ^^ | 08:12 |
| lajoskatona | frickler: Hi, Wednesday we have a session for devstack lib/neutron, anyway, what do you think that topic needs 30mins? | 08:25 |
| lajoskatona | frickler: https://etherpad.opendev.org/p/neutron-antelope-ptg#L124 | 08:25 |
| ralonsoh | frickler, the agenda is pretty tight | 08:27 |
| ralonsoh | I'll check when it could be possible | 08:27 |
| ralonsoh | frickler, we can try today at 16UTC | 08:29 |
| frickler | lajoskatona: depends on whether someone volunteers to do the work, that would make it pretty fast ;) | 08:29 |
| ralonsoh | frickler, what is the rationale to keep OVS | 08:29 |
| ralonsoh | ? | 08:29 |
| lajoskatona | friskler: :D | 08:29 |
| frickler | ralonsoh: I wanted to keep that for the discussion, in short OVS is the default in grenade and we also want to try to support Y->A upgrade. also IMHO OVN isn't production-ready because of gaps | 08:31 |
| ralonsoh | frickler, what gaps? | 08:32 |
| frickler | n-d-r + DNS for isolated networks | 08:33 |
| ralonsoh | this is something you are looking for for a long time, I know | 08:33 |
| ralonsoh | but this just a small feature | 08:33 |
| frickler | but very important for some deployments. plus potentially stateless firewalling | 08:34 |
| opendevreview | Luca Czesla proposed openstack/neutron master: Add the address scope to the OVN DBs https://review.opendev.org/c/openstack/neutron/+/861719 | 08:34 |
| ralonsoh | and where do we have stateless FW in OVS? | 08:34 |
| frickler | with the hybrib driver | 08:34 |
| frickler | hybrid | 08:34 |
| ralonsoh | it uses conntrack | 08:34 |
| ralonsoh | so it is not stateless | 08:35 |
| ralonsoh | in any case, I've created a slot today at 16UTC | 08:35 |
| frickler | the usecase behind it is direct server return. it used to work with that, but haven't checked in quite some time (was on stein or train) | 08:36 |
| crohmann | frickler: Just as an observation as a "user" or installation in regards to OVN: We have a switch away from Linuxbridge planned and where under the impression it's full force ahead for OVN. Devstack and recommended drivers for production. Yes there are some gaps (e.g. VPNaaS - https://review.opendev.org/c/openstack/neutron-vpnaas/+/765353) but those are received strong attention and a being | 08:38 |
| crohmann | actively worked on. | 08:39 |
| lajoskatona | crohmann: Hi, just wanted to answer your original question regarding https://bugs.launchpad.net/neutron/+bug/1943449 | 08:44 |
| frickler | crohmann: that seems to be what a lot of developers want people to think, I'm not sure operators agree. and I haven't seen any reports from working large scale deployments yet, but that may be my ignorance | 08:44 |
| lajoskatona | crohmann: in zed we added a statement that linuxbridge is exmerimental, and we slowly start decreasing the test coverage (as things start to fail) as nobody is really maintaining that code (https://docs.openstack.org/releasenotes/neutron/zed.html#prelude) | 08:46 |
| crohmann | frickler: I know of cloud provider in Germany who uses OVN on Yoga in PROD. | 08:46 |
| crohmann | lajoskatona: Thanks for the link and this statement. As with the silent degradation of the test coverage or fixing of bugs for PostgreSQL database which use used way to long ... please even word this stronger at some point. It's ok to deprecate and consequently remove things - for the sake of quality. You cannot move forward by just adding more and more (complexity). | 08:50 |
| crohmann | "which we used in our setup way too long" ... was what I meant to write | 08:51 |
| lajoskatona | crohmann: we really apreciate such feedback, if you have time please join the operators session on Friday 1400UTC, Junot room (see on https://ptg.opendev.org/ptg.html ) | 08:52 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: [OVN] Allow only one physical network per bridge https://review.opendev.org/c/openstack/neutron/+/858863 | 08:55 |
| crohmann | lajoskatona: regarding https://bugs.launchpad.net/neutron/+bug/1943449, mnaser__ did respond to my bug quite quickly and I am really sorry that I did not provdide more debug data sooner. I first wanted to have a more recent OS release (>=Xena) to nag about this. But is there anybody who could take a peek at this, even though it's linuxbridge? Or am I out of luck there? | 08:58 |
| lajoskatona | crohmann: good question, if it really happens with linuxbridge, recently nobody arrived to work on similar bugs | 09:13 |
| amorin | hey neutron team, I came accross a change in kolla: https://review.opendev.org/c/openstack/kolla/+/749601 | 09:13 |
| amorin | they tried to enable wsgi at that time, but abandonned | 09:14 |
| amorin | is neutron known to not work yet using wsgi? | 09:14 |
| crohmann | frickler: Regarding the identified gaps or missing features in OVN - is there a list somewhere? *If* there was the agreement to have OVN be the de-facto standard and to name this a goal, this would relate all those little things. What I am saying is that a little feature someone does not know does not imply this need to be done for a certain release. But if the release goal is OVN feature | 09:16 |
| crohmann | parity then this implies priorities and causes even little things implemented to be worth the time. | 09:16 |
| ralonsoh | amorin, no, this is still pending for OVN: https://bugs.launchpad.net/neutron/+bug/1912359 | 09:18 |
| amorin | ok, and what about OVS? | 09:19 |
| ralonsoh | amorin, it is working | 09:21 |
| amorin | ok | 09:21 |
| amorin | thanks | 09:21 |
| amorin | ok, so, we are using neutron behind apache2+mod_wsgi, if I disable lazy loading, some RPC messages are lost | 09:25 |
| amorin | is there anyone aware of this? | 09:25 |
| ralonsoh | I don't remember any related bug | 09:28 |
| ralonsoh | please, report it in Launchpad | 09:28 |
| amorin | ralonsoh yes, will do, I just want to make sure this is not related to my deployment first | 09:29 |
| frickler | crohmann: https://docs.openstack.org/neutron/latest/ovn/gaps.html | 09:54 |
| frickler | actually might be good to add wsgi support to that | 09:58 |
| opendevreview | Merged openstack/neutron master: [OVN] Allow to execute ``MetadataProxyHandler`` in a local thread https://review.opendev.org/c/openstack/neutron/+/861649 | 10:01 |
| opendevreview | Merged openstack/ovn-octavia-provider stable/ussuri: Ensure LB are removed from LS not connected to the LB LR https://review.opendev.org/c/openstack/ovn-octavia-provider/+/861330 | 10:28 |
| opendevreview | Merged openstack/ovn-octavia-provider stable/victoria: Ensure LB are removed from LS not connected to the LB LR https://review.opendev.org/c/openstack/ovn-octavia-provider/+/861297 | 10:28 |
| opendevreview | Merged openstack/ovn-octavia-provider stable/wallaby: Ensure LB are removed from LS not connected to the LB LR https://review.opendev.org/c/openstack/ovn-octavia-provider/+/861296 | 10:30 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron stable/zed: [OVN] Allow to execute ``MetadataProxyHandler`` in a local thread https://review.opendev.org/c/openstack/neutron/+/861733 | 10:53 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron stable/yoga: [OVN] Allow to execute ``MetadataProxyHandler`` in a local thread https://review.opendev.org/c/openstack/neutron/+/861734 | 10:53 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron stable/xena: [OVN] Allow to execute ``MetadataProxyHandler`` in a local thread https://review.opendev.org/c/openstack/neutron/+/861735 | 10:54 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron stable/wallaby: [OVN] Allow to execute ``MetadataProxyHandler`` in a local thread https://review.opendev.org/c/openstack/neutron/+/861736 | 10:55 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: Remove OVN gap minimum bandwidth support https://review.opendev.org/c/openstack/neutron/+/861739 | 11:06 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron stable/zed: Remove OVN gap minimum bandwidth support https://review.opendev.org/c/openstack/neutron/+/861740 | 11:07 |
| opendevreview | Merged openstack/networking-ovn stable/train: Ensure LB are removed from LS not connected to the LB LR https://review.opendev.org/c/openstack/networking-ovn/+/861331 | 11:07 |
| opendevreview | Merged openstack/neutron stable/zed: Disable in-band management for bridges before setting up controllers https://review.opendev.org/c/openstack/neutron/+/861622 | 12:00 |
| opendevreview | Merged openstack/ovn-octavia-provider stable/xena: Ensure LB are removed from LS not connected to the LB LR https://review.opendev.org/c/openstack/ovn-octavia-provider/+/861295 | 12:06 |
| opendevreview | Merged openstack/ovn-octavia-provider stable/yoga: Ensure LB are removed from LS not connected to the LB LR https://review.opendev.org/c/openstack/ovn-octavia-provider/+/861294 | 12:33 |
| opendevreview | Merged openstack/neutron stable/wallaby: Disable in-band management for bridges before setting up controllers https://review.opendev.org/c/openstack/neutron/+/861625 | 12:34 |
| opendevreview | Merged openstack/neutron stable/victoria: Disable in-band management for bridges before setting up controllers https://review.opendev.org/c/openstack/neutron/+/861705 | 12:38 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: [OVN] Set the default OVN metadata worker number to 0 https://review.opendev.org/c/openstack/neutron/+/861751 | 13:00 |
| bcafarel | neutron PTG session started https://www.openinfra.dev/ptg/rooms/mitaka | 13:08 |
| *** dasm|off is now known as dasm | 13:49 | |
| frickler | ralonsoh: can we move the backend topic to tomorrow 16 UTC? I missed an overlap with kolla operator hour | 13:51 |
| opendevreview | Luca Czesla proposed openstack/neutron master: Add the address scope to the OVN DBs https://review.opendev.org/c/openstack/neutron/+/861719 | 14:00 |
| crohmann | frickler: thanks for the pointer to the OVN gaps ... as you said there are things missing and also some things are already implemented ... see e.g. https://review.opendev.org/c/openstack/neutron/+/861739. Does it make sense to have a more dynamic list somewhere? Something like a topic / tag for bugs? | 14:48 |
| crohmann | frickler: thanks for the pointer to the OVN gaps ... as you said there are things missing and also some things are already implemented ... see e.g. https://review.opendev.org/c/openstack/neutron/+/861739. Does it make sense to have a more dynamic list somewhere? Something like a topic / tag for bugs? | 14:48 |
| crohmann | frickler: thanks for the pointer to the OVN gaps ... as you said there are things missing and also some things are already implemented ... see e.g. https://review.opendev.org/c/openstack/neutron/+/861739. Does it make sense to have a more dynamic list somewhere? Something like a topic / tag for bugs? | 14:50 |
| opendevreview | Arnaud Morin proposed openstack/neutron master: Allow restoration of tun_ofports on agent restart https://review.opendev.org/c/openstack/neutron/+/860270 | 14:51 |
| frickler | crohmann: for bugs there is a tag, makes it to the top of the tag list even https://bugs.launchpad.net/neutron/+bugs?field.tag=ovn | 14:55 |
| opendevreview | Arnaud Morin proposed openstack/neutron master: Discard port with ofport -1 in _get_ofport_moves https://review.opendev.org/c/openstack/neutron/+/860649 | 14:59 |
| opendevreview | Merged openstack/neutron stable/ussuri: Disable in-band management for bridges before setting up controllers https://review.opendev.org/c/openstack/neutron/+/861706 | 16:27 |
| *** haleyb_ is now known as haleyb | 17:49 | |
| opendevreview | Merged openstack/ovn-octavia-provider stable/zed: Ensure LB are removed from LS not connected to the LB LR https://review.opendev.org/c/openstack/ovn-octavia-provider/+/861293 | 18:42 |
| *** dasm is now known as dasm|off | 22:15 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!