| opendevreview | Merged openstack/neutron-tempest-plugin master: Drop AddressScope.shared from API tests https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/1005715 | 00:39 |
|---|---|---|
| opendevreview | Helen Chen proposed openstack/neutron master: EVPN router always redistributes static routes https://review.opendev.org/c/openstack/neutron/+/1006133 | 00:54 |
| opendevreview | Helen Chen proposed openstack/neutron master: EVPN router always redistributes static routes https://review.opendev.org/c/openstack/neutron/+/1006133 | 00:57 |
| *** ykarel__ is now known as ykarel | 06:23 | |
| opendevreview | Kyuyeong Lee proposed openstack/neutron-lib master: port-group: Add API definitions https://review.opendev.org/c/openstack/neutron-lib/+/1006159 | 07:26 |
| zigo | Question: with the Hibiscus release, is ovn-bgp-agent obsolete? | 07:33 |
| LarsErikP | haleyb: thanks! | 07:52 |
| opendevreview | Kyuyeong Lee proposed openstack/neutron master: Open the 2027.1 (Indri) DB branch https://review.opendev.org/c/openstack/neutron/+/1006161 | 08:04 |
| opendevreview | Kyuyeong Lee proposed openstack/neutron-lib master: port-group: Add API definitions https://review.opendev.org/c/openstack/neutron-lib/+/1006159 | 08:08 |
| opendevreview | Kyuyeong Lee proposed openstack/neutron master: Open the 2027.1 (Indri) DB branch https://review.opendev.org/c/openstack/neutron/+/1006161 | 08:10 |
| opendevreview | Rodolfo Alonso proposed openstack/ovsdbapp master: ovn-sb: Set default ``hostname`` in ``chassis_add`` https://review.opendev.org/c/openstack/ovsdbapp/+/1005207 | 08:50 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron-lib master: api: fix validate_ip_address() to handle None input https://review.opendev.org/c/openstack/neutron-lib/+/1005893 | 08:55 |
| opendevreview | Kyuyeong Lee proposed openstack/neutron master: db: Add port group models https://review.opendev.org/c/openstack/neutron/+/1006171 | 08:56 |
| opendevreview | Kyuyeong Lee proposed openstack/neutron master: objects: Add port group objects https://review.opendev.org/c/openstack/neutron/+/1006172 | 08:56 |
| opendevreview | Kyuyeong Lee proposed openstack/neutron master: api: Add port group CRUD https://review.opendev.org/c/openstack/neutron/+/1006173 | 08:56 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron-lib master: api: fix convert_to_boolean_if_not_none() to handle conversion failures https://review.opendev.org/c/openstack/neutron-lib/+/1005905 | 09:11 |
| opendevreview | Piotr Milewski proposed openstack/neutron-vpnaas master: Preserve system timezone in strongSwan namespace https://review.opendev.org/c/openstack/neutron-vpnaas/+/1006182 | 10:03 |
| opendevreview | Bharath M V proposed openstack/neutron-tempest-plugin master: [TaaS] Add MTU test to Tap-as-a-Service. https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/1006184 | 10:08 |
| opendevreview | Merged openstack/neutron-lib master: db: fix is_retriable() exception handling https://review.opendev.org/c/openstack/neutron-lib/+/1005891 | 10:54 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron-lib master: api-ref: fix typos on default SGs listing https://review.opendev.org/c/openstack/neutron-lib/+/1001816 | 11:04 |
| opendevreview | Eduardo Olivares proposed openstack/neutron-tempest-plugin master: evpn: Fix disk_over_commit handling in live migration test https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/1006006 | 11:15 |
| opendevreview | Piotr Milewski proposed openstack/neutron-vpnaas master: Preserve system timezone in strongSwan namespace https://review.opendev.org/c/openstack/neutron-vpnaas/+/1006182 | 11:40 |
| opendevreview | yatin proposed x/whitebox-neutron-tempest-plugin master: Make db sync work with master podified deployment https://review.opendev.org/c/x/whitebox-neutron-tempest-plugin/+/1006197 | 11:48 |
| opendevreview | Eduardo Olivares proposed openstack/neutron-tempest-plugin master: evpn: Reuse ip.IPCommand and shell.execute in EVPN provisioner https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/1005922 | 12:43 |
| haleyb | #startmeeting neutron_drivers | 13:00 |
| opendevmeet | Meeting started Fri Sep 18 13:00:22 2026 UTC and is due to finish in 60 minutes. The chair is haleyb. Information about MeetBot at http://wiki.debian.org/MeetBot. | 13:00 |
| opendevmeet | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 13:00 |
| opendevmeet | The meeting name has been set to 'neutron_drivers' | 13:00 |
| haleyb | Ping list: ykarel, mlavalle, mtomaska, slaweq, tobias-urdin, lajoskatona, haleyb, ralonsoh, cardoe | 13:00 |
| mlavalle | \o | 13:00 |
| ralonsoh | hello | 13:00 |
| lajoskatona | o/ | 13:00 |
| haleyb | i'm not sure slaweq will be here since he added one for next week, but there is one item on the agenda | 13:01 |
| haleyb | o/ btw, i'm not all business :) | 13:01 |
| haleyb | ichen: are you here? you had the one item carried over from last week | 13:02 |
| mlavalle | slaweq is on a business trip, so he might not join | 13:02 |
| haleyb | #link https://bugs.launchpad.net/neutron/+bug/2166027 | 13:02 |
| haleyb | [RFE] BGP EVPN Type-2 Route support for Neutron ML2/OVN backend | 13:02 |
| ichen | Yes, I’m here. | 13:02 |
| ichen | I wrote the EVPN Type-2 spec. I also submitted the spec. | 13:03 |
| lajoskatona | https://review.opendev.org/c/openstack/neutron-specs/+/1004908 | 13:04 |
| haleyb | so since i don't use evpn every day, type-2 is basically per-host? like a /32 or /128 ? | 13:04 |
| ichen | Type-2 is always distributed, meaning that only the Mac/ip routes local to a chassis is advertised by the chassis and doesn’t require the chassis routers like for type-5 distributed routing. | 13:04 |
| haleyb | or per-IP | 13:04 |
| ichen | Type-2 is Mac routes. | 13:05 |
| ichen | It’s a BGP router saying that to reach MAC1, send to me. | 13:05 |
| haleyb | alright, makes sense now | 13:06 |
| ichen | So, instead of querying for the MAC, the MAC is advertised. | 13:06 |
| haleyb | so the downside is a lot more entries? | 13:06 |
| ichen | Yes. | 13:07 |
| haleyb | i had no more questions, have not read the spec though | 13:08 |
| lajoskatona | but those are in FRR, or am I wrong? | 13:08 |
| ichen | I don’t know if OVN has the option to do it now or later, but OVN implemented a new option called dynamic-routing-advertise, which can be set to false, if a user needs to cherry-pick to not advertise. Right now, this option is only available for IP routes, such as static routes and host routes. | 13:09 |
| ichen | lajoskatona: what are in FRR? | 13:09 |
| haleyb | don't know if lajoskatona is still here? | 13:11 |
| ichen | Do you mean that MACs to be advertised? OVN relays the MAC addresses to FRR/BGP and FRR/BGP advertises them. The remote site does the same thing and OVN listens for the MAC addresses that FRR/BGP adds into the kernel and creates something (open flow rules or Mac bindings) based on MAC addresses learned from remote site. | 13:11 |
| lajoskatona | sorry, seems like I was dropped, or my network has issues, but back | 13:12 |
| ichen | I also chatted with cardoe separately to understand his request. I’ll add an option to the spec. | 13:13 |
| lajoskatona | +1 for spec update | 13:14 |
| ralonsoh | Can we have both type 2 and type 5 working at the same time? Are these 2 different extensions? | 13:14 |
| ichen | It is possible. But the spec is for pure type-2. | 13:15 |
| ralonsoh | yes, I know | 13:16 |
| ichen | Pure type-2 means one single ethernet segment. | 13:16 |
| ralonsoh | but is possible to have a network configured with evpn type 2 and another network+router with type 5? | 13:16 |
| ralonsoh | is that something possible? | 13:16 |
| ichen | A mix of type-2 and type-5 means that the network has multiple ethernet segments and there’s a router to route traffic from one ethernet segment to nother. | 13:16 |
| ichen | ralonsoh: yes, that is possible. | 13:16 |
| ralonsoh | perfect | 13:17 |
| mlavalle | kike a routed network in evpn | 13:17 |
| ralonsoh | I have no more questions. I know there are 2 RFEs (distributed and type 2) presented | 13:19 |
| ralonsoh | so I don't know what will be the first feature to be implemented | 13:19 |
| ralonsoh | (just to be aware if there is an order) | 13:19 |
| ichen | Type-2 is easier to implement now. | 13:19 |
| ralonsoh | cool | 13:19 |
| haleyb | ok, should we vote? | 13:20 |
| ralonsoh | +1 for me (the spec is already there) | 13:20 |
| lajoskatona | +1 from me | 13:20 |
| haleyb | +1 from me, can continue discussion in the spec | 13:20 |
| mlavalle | it seems a reasonable next step to me and besides, we can discuss further in the spec | 13:20 |
| mlavalle | so +1 | 13:20 |
| ralonsoh | thanks ichen! | 13:21 |
| haleyb | ok, i'll mark it approved | 13:21 |
| ichen | Thanks, everyone! | 13:21 |
| haleyb | are there any other on-demand topics? | 13:21 |
| lajoskatona | not from me , thanks | 13:22 |
| haleyb | ok, then i guess we are done | 13:22 |
| haleyb | thanks for attending, have a nice weekend! | 13:22 |
| haleyb | #endmeeting | 13:22 |
| opendevmeet | Meeting ended Fri Sep 18 13:22:58 2026 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 13:22 |
| opendevmeet | Minutes: https://meetings.opendev.org/meetings/neutron_drivers/2026/neutron_drivers.2026-09-18-13.00.html | 13:22 |
| opendevmeet | Minutes (text): https://meetings.opendev.org/meetings/neutron_drivers/2026/neutron_drivers.2026-09-18-13.00.txt | 13:22 |
| opendevmeet | Log: https://meetings.opendev.org/meetings/neutron_drivers/2026/neutron_drivers.2026-09-18-13.00.log.html | 13:22 |
| lajoskatona | o/, have a nice weekend | 13:23 |
| ralonsoh | bye! | 13:23 |
| opendevreview | Mohsen Sepandar proposed openstack/neutron-fwaas master: Enable fwaas_v2 driver in DVR mode https://review.opendev.org/c/openstack/neutron-fwaas/+/989857 | 14:43 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron stable/2026.2: dhcp: only use local subnets for dnsmasq --dhcp-range https://review.opendev.org/c/openstack/neutron/+/1006224 | 14:55 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron stable/2026.1: dhcp: only use local subnets for dnsmasq --dhcp-range https://review.opendev.org/c/openstack/neutron/+/1006225 | 14:55 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron stable/2025.2: dhcp: only use local subnets for dnsmasq --dhcp-range https://review.opendev.org/c/openstack/neutron/+/1006226 | 14:55 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron stable/2025.1: dhcp: only use local subnets for dnsmasq --dhcp-range https://review.opendev.org/c/openstack/neutron/+/1006227 | 14:55 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron stable/2026.2: ovo: Fix ``qos_network_policy_id`` dirty flag after DB load https://review.opendev.org/c/openstack/neutron/+/1006228 | 14:56 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron stable/2026.1: ovo: Fix ``qos_network_policy_id`` dirty flag after DB load https://review.opendev.org/c/openstack/neutron/+/1006229 | 14:57 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron stable/2025.2: ovo: Fix ``qos_network_policy_id`` dirty flag after DB load https://review.opendev.org/c/openstack/neutron/+/1006230 | 14:57 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron stable/2025.1: ovo: Fix ``qos_network_policy_id`` dirty flag after DB load https://review.opendev.org/c/openstack/neutron/+/1006231 | 14:57 |
| opendevreview | Rodolfo Alonso proposed openstack/ovsdbapp master: ovn-sb: Set default ``hostname`` in ``chassis_add`` https://review.opendev.org/c/openstack/ovsdbapp/+/1005207 | 15:05 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron-vpnaas master: vpnaas: Add swanctl (VICI protocol) support for strongSwan https://review.opendev.org/c/openstack/neutron-vpnaas/+/1003801 | 15:07 |
| opendevreview | Merged openstack/neutron master: Open the 2027.1 (Indri) DB branch https://review.opendev.org/c/openstack/neutron/+/1006161 | 15:08 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron-vpnaas master: conf: Move VPNaaS configuration options to neutron_vpnaas/conf https://review.opendev.org/c/openstack/neutron-vpnaas/+/1004571 | 15:09 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron-vpnaas master: zuul: add OVN VPNaaS tempest job with swanctl legacy mode https://review.opendev.org/c/openstack/neutron-vpnaas/+/1003820 | 15:09 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron-vpnaas master: DNM == Test ``neutron-tempest-plugin-vpnaas-ovn-sswan-swanctl`` 20 times https://review.opendev.org/c/openstack/neutron-vpnaas/+/1004432 | 15:10 |
| opendevreview | Merged openstack/neutron-lib master: doc: Document URL path hyphen and underscore equivalence https://review.opendev.org/c/openstack/neutron-lib/+/1005851 | 15:13 |
| stephenfin | ralonsoh: haleyb: There are a lot more commits on neutron-lib `master` than in `stable/2026.2`. Is there a reason we don't want to backport a lot of the "Add ``is_filter`` to API definition" patches? | 15:15 |
| stephenfin | iiuc, without those you will see HTTP 400 errors if you attempt to use one of the affected params as a filter, which feels like a regression? | 15:16 |
| stephenfin | I ask because I still see failures on https://review.opendev.org/c/openstack/openstacksdk/+/1005450 even after https://review.opendev.org/c/openstack/neutron-lib/+/1005527 merged. | 15:17 |
| stephenfin | I'm putting that down to the library not being installed correctly for neutron (I'm debugging currently) but it's got me wondering about ☝️ | 15:17 |
| ralonsoh | stephenfin, but all we need is to release a new version of n-lib, right? | 15:18 |
| ralonsoh | let me check if openstacksdk-functional-devstack-networking-ext depends on n-lib | 15:18 |
| stephenfin | n-lib has branched though. To fix this in hibiscus you presumably need to cut a release from stable/2026.2? | 15:19 |
| opendevreview | Lajos Katona proposed openstack/neutron-vpnaas master: Rally scenario plugin API changed https://review.opendev.org/c/openstack/neutron-vpnaas/+/1000417 | 15:19 |
| ralonsoh | yes, I would need to backport and release a new 2026.2 n-lib version | 15:20 |
| stephenfin | Okay. I just wanted to confirm that that was the plan :) | 15:21 |
| ralonsoh | stephenfin, the job openstacksdk-functional-devstack-networking-ext is not installing the neutron-lib patch, so we'll need a new version in Indri too | 15:21 |
| stephenfin | yup, I've just seen `Downloading neutron_lib-5.0.1-py3-none-any.whl.metadata (2.4 kB)` in `job-output.txt` | 15:22 |
| ralonsoh | let me check the releases patch that opens the new branches | 15:22 |
| stephenfin | ralonsoh: you must not have been reading openstack-discuss recently https://lists.openstack.org/archives/list/openstack-discuss@lists.openstack.org/thread/65GQWCGJRQNCLIA3L3A5AXLVTOPGNZQ7/ | 15:23 |
| stephenfin | we can't release from `master` until requirements branch, which I think doesn't happen until GA in 2 weeks :( | 15:24 |
| ralonsoh | stephenfin, so we can still create a new 2026.2 n-lib release? | 15:24 |
| ralonsoh | no | 15:25 |
| stephenfin | yes, but you'll need a requirements freeze exception | 15:25 |
| stephenfin | given this is a regression (right?) that seems an easy one to grant | 15:25 |
| ralonsoh | stephenfin, that wasn't a regression, it was a bug since the first implementation | 15:25 |
| stephenfin | oh, isn't the `is_filter` check a new thing in hibiscus? | 15:26 |
| stephenfin | at least enforcement of it | 15:26 |
| ralonsoh | ok, yes, in the Neutron API | 15:27 |
| ralonsoh | one sec | 15:27 |
| ralonsoh | I mean the definition was wrong | 15:27 |
| ralonsoh | and we made this mandatory in the API | 15:27 |
| stephenfin | right, that's what I mean | 15:27 |
| ralonsoh | so yes, that makes both a regression | 15:27 |
| haleyb | we are well past library freeze (sorry in another meeting) | 15:27 |
| ralonsoh | so when can I backport to n-lib and make a new release? | 15:28 |
| stephenfin | haleyb: Yep, a requirements freeze exception would be needed here | 15:28 |
| ralonsoh | and how should I request it? mail? | 15:28 |
| stephenfin | but as I said above, I believe this is a regression and would warrant one | 15:28 |
| stephenfin | yep. here's an example from last cycle https://lists.openstack.org/archives/list/openstack-discuss@lists.openstack.org/thread/H67NORC72ZELFUPBGFWUXVAKGMCK6TUU/#I2LMBBNY2DOM7YFNNVHS7H36MR75UASC | 15:29 |
| stephenfin | maybe elodilles hasn't gone home yet and could confirm?? | 15:30 |
| ralonsoh | ok, I think we need to backport a couple of patches (TaaS and VPNaaS) | 15:30 |
| ralonsoh | https://storage.bhs.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_9ab/openstack/9ab3ef4d943b4ea58e43e90221d3cabe/testr_results.html | 15:31 |
| stephenfin | elodilles: if you're still about, we'd like to confirm if a requirememts freeze exception email is required to address a regression in neutron(-lib) | 15:31 |
| stephenfin | err, s/to address a regression/to release a fix that addresses a regression/ | 15:31 |
| ralonsoh | stephenfin, we also need the fix in CLI: https://review.opendev.org/c/openstack/python-openstackclient/+/1005525 | 15:39 |
| ralonsoh | so do we need to backport ^ this patch too? | 15:39 |
| opendevreview | Stephen Finucane proposed openstack/neutron-lib stable/2026.2: placement: Raise ``PlacementResourceProviderNotFound`` on delete https://review.opendev.org/c/openstack/neutron-lib/+/1006249 | 15:39 |
| opendevreview | Stephen Finucane proposed openstack/neutron-lib stable/2026.2: taas: Add ``is_filter`` to all TaaS API definitions https://review.opendev.org/c/openstack/neutron-lib/+/1006250 | 15:39 |
| opendevreview | Stephen Finucane proposed openstack/neutron-lib stable/2026.2: floatingip-pools: Add ``is_filter`` to API definition https://review.opendev.org/c/openstack/neutron-lib/+/1006251 | 15:39 |
| opendevreview | Stephen Finucane proposed openstack/neutron-lib stable/2026.2: network-segment-range: Add ``is_filter`` to API definition https://review.opendev.org/c/openstack/neutron-lib/+/1006252 | 15:39 |
| opendevreview | Stephen Finucane proposed openstack/neutron-lib stable/2026.2: auto-allocated-topology: Add ``is_filter`` to API definition https://review.opendev.org/c/openstack/neutron-lib/+/1006253 | 15:39 |
| opendevreview | Stephen Finucane proposed openstack/neutron-lib stable/2026.2: metering: Add ``is_filter`` to API definition https://review.opendev.org/c/openstack/neutron-lib/+/1006254 | 15:39 |
| opendevreview | Stephen Finucane proposed openstack/neutron-lib stable/2026.2: agent: Add ``is_filter`` to API definition https://review.opendev.org/c/openstack/neutron-lib/+/1006255 | 15:39 |
| opendevreview | Stephen Finucane proposed openstack/neutron-lib stable/2026.2: flavors: Add ``is_filter`` to API definition https://review.opendev.org/c/openstack/neutron-lib/+/1006256 | 15:39 |
| opendevreview | Stephen Finucane proposed openstack/neutron-lib stable/2026.2: logging: Add ``is_filter`` to API definition https://review.opendev.org/c/openstack/neutron-lib/+/1006257 | 15:40 |
| opendevreview | Stephen Finucane proposed openstack/neutron-lib stable/2026.2: network-ip-availability: Add ``is_filter`` to API definition https://review.opendev.org/c/openstack/neutron-lib/+/1006258 | 15:40 |
| opendevreview | Stephen Finucane proposed openstack/neutron-lib stable/2026.2: qos: Add ``is_filter`` to API definition https://review.opendev.org/c/openstack/neutron-lib/+/1006259 | 15:40 |
| opendevreview | Stephen Finucane proposed openstack/neutron-lib stable/2026.2: segment: Add ``is_filter`` to API definition https://review.opendev.org/c/openstack/neutron-lib/+/1006260 | 15:40 |
| opendevreview | Stephen Finucane proposed openstack/neutron-lib stable/2026.2: doc: Document URL path hyphen and underscore equivalence https://review.opendev.org/c/openstack/neutron-lib/+/1006261 | 15:40 |
| opendevreview | Stephen Finucane proposed openstack/neutron-lib stable/2026.2: db: fix is_retriable() exception handling https://review.opendev.org/c/openstack/neutron-lib/+/1006262 | 15:40 |
| ralonsoh | stephenfin, I don't think we need all these patches | 15:40 |
| ralonsoh | at least for this problem | 15:40 |
| ralonsoh | what about the openstackclient patch? | 15:40 |
| stephenfin | agreed, but they're all potential regressions, right? | 15:41 |
| stephenfin | we need the OSC fix also, yes. If you want to propose that I can merge it | 15:41 |
| ralonsoh | these are improvements, nothing was using it | 15:41 |
| ralonsoh | I'll propose the backport | 15:41 |
| ralonsoh | and I'll send the mail for both projects | 15:41 |
| ralonsoh | haleyb, stephenfin mail sent, waiting for the link | 15:53 |
| elodilles | hi stephenfin ralonsoh . we are quite close to coordinated release (1.5 half weeks from now) so i'd rather not release any library at this point. we are too late for that :/ if the bug is a release critical, then neutron+requirements+release team might need assess the issue. otherwise i'd say if it's "only" regression, then let's wait with the release after the coordinated release and then do | 15:53 |
| elodilles | a stable release. | 15:53 |
| elodilles | i don't really want any last minute churning because a new lib release causing errors in service deliverables :/ | 15:54 |
| ralonsoh | elodilles, none of the bugs are critical: | 15:54 |
| ralonsoh | * the neutron-lib one just prevents from using ?name in the API | 15:54 |
| ralonsoh | * the openstack-client can be overriden by using the correct ike policy | 15:54 |
| ralonsoh | so no, none of them are critical | 15:55 |
| ralonsoh | --> from using ?name in the API **in TaaS** | 15:55 |
| elodilles | ralonsoh: thanks, so i'd really suggest a stable release AFTER the official coordinated 2026.2 Hibiscus release | 15:55 |
| ralonsoh | elodilles, can we update the stable/requirements file? just asking | 15:56 |
| ralonsoh | after the release, of course | 15:56 |
| elodilles | you mean if we release neutron-lib, then the upper-constraints can be bumped? | 15:56 |
| ralonsoh | yes, in the stable 2026.2 requirements branches | 15:56 |
| elodilles | yes, that is the usual process | 15:57 |
| ralonsoh | perfect, I didn't know about stable branches in requirements, I thought that was static once released | 15:57 |
| ralonsoh | thanks for the information! | 15:57 |
| elodilles | we do a stable release -> it's upper-constraints is bumped on stable/<series> | 15:57 |
| ralonsoh | perfect, we can continue with the backports and release in October | 15:58 |
| elodilles | ralonsoh: static for 3rd parties, but the new stable releases have to be consumed somehow ;) | 15:58 |
| elodilles | * the new openstack component's stable releases | 15:58 |
| *** ralonsoh is now known as ralonsoh_ooo | 16:02 | |
| opendevreview | Ivan Anfimov proposed openstack/neutron master: Fix typo https://review.opendev.org/c/openstack/neutron/+/1006298 | 17:27 |
| opendevreview | Ivan Anfimov proposed openstack/neutron master: Fix typo https://review.opendev.org/c/openstack/neutron/+/1006298 | 17:29 |
| opendevreview | Merged openstack/neutron master: dhcp: only use local subnets for dnsmasq --dhcp-range https://review.opendev.org/c/openstack/neutron/+/1005863 | 17:44 |
| opendevreview | Merged openstack/neutron-lib stable/2026.2: placement: Raise ``PlacementResourceProviderNotFound`` on delete https://review.opendev.org/c/openstack/neutron-lib/+/1006249 | 18:00 |
| opendevreview | Merged openstack/neutron-lib stable/2026.2: taas: Add ``is_filter`` to all TaaS API definitions https://review.opendev.org/c/openstack/neutron-lib/+/1006250 | 18:00 |
| opendevreview | Merged openstack/neutron-lib stable/2026.2: floatingip-pools: Add ``is_filter`` to API definition https://review.opendev.org/c/openstack/neutron-lib/+/1006251 | 18:00 |
| opendevreview | Merged openstack/neutron-lib stable/2026.2: network-segment-range: Add ``is_filter`` to API definition https://review.opendev.org/c/openstack/neutron-lib/+/1006252 | 18:00 |
| opendevreview | Merged openstack/neutron-lib stable/2026.2: auto-allocated-topology: Add ``is_filter`` to API definition https://review.opendev.org/c/openstack/neutron-lib/+/1006253 | 18:00 |
| opendevreview | Merged openstack/neutron-lib stable/2026.2: metering: Add ``is_filter`` to API definition https://review.opendev.org/c/openstack/neutron-lib/+/1006254 | 18:00 |
| opendevreview | Merged openstack/neutron-lib stable/2026.2: agent: Add ``is_filter`` to API definition https://review.opendev.org/c/openstack/neutron-lib/+/1006255 | 18:00 |
| opendevreview | Merged openstack/neutron-lib stable/2026.2: flavors: Add ``is_filter`` to API definition https://review.opendev.org/c/openstack/neutron-lib/+/1006256 | 18:00 |
| opendevreview | Merged openstack/neutron-lib stable/2026.2: logging: Add ``is_filter`` to API definition https://review.opendev.org/c/openstack/neutron-lib/+/1006257 | 18:00 |
| opendevreview | Merged openstack/neutron-lib stable/2026.2: network-ip-availability: Add ``is_filter`` to API definition https://review.opendev.org/c/openstack/neutron-lib/+/1006258 | 18:01 |
| opendevreview | Merged openstack/neutron-lib stable/2026.2: qos: Add ``is_filter`` to API definition https://review.opendev.org/c/openstack/neutron-lib/+/1006259 | 18:01 |
| opendevreview | Merged openstack/neutron-lib stable/2026.2: segment: Add ``is_filter`` to API definition https://review.opendev.org/c/openstack/neutron-lib/+/1006260 | 18:06 |
| opendevreview | Merged openstack/neutron-lib stable/2026.2: doc: Document URL path hyphen and underscore equivalence https://review.opendev.org/c/openstack/neutron-lib/+/1006261 | 18:06 |
| opendevreview | Merged openstack/neutron-lib stable/2026.2: db: fix is_retriable() exception handling https://review.opendev.org/c/openstack/neutron-lib/+/1006262 | 18:07 |
| haleyb | jlibosva: not sure if you're around, was trying to ping Terry about https://review.opendev.org/c/openstack/neutron/+/818850 | 20:12 |
| haleyb | Also have to re-work the other issue you commented on | 20:12 |
| jlibosva | haleyb: I am around :) | 20:22 |
| jlibosva | I can ping him | 20:22 |
| haleyb | jlibosva: great thanks. And regarding the other change, https://review.opendev.org/c/openstack/neutron/+/1005970 - the original bug that Lucas filed mentioned using external_ids to store cidrs. Is that a better option here? | 20:28 |
| jlibosva | haleyb: that's what I meant with my comment. Except we are not interested in the cidrs for the subnet itself but only the gateway IP. With the traditional Neutron (L2) the gateway IP is outside of OpenStack - it'll be some DC GW. The only place the IP is stored nowadays is the DHCP_Options (which corresponds to the Neutron subnet resource), because the DHCP configures the | 20:31 |
| jlibosva | default egress route for the clients. | 20:31 |
| jlibosva | thus the limitation the external network needs to have the DHCP enabled - it doesn't even make sense to not have the DHCP unless the whole spine'n leaf is configured with some DHCP relays, I think | 20:32 |
| jlibosva | but that said - if we can store the gw IP elsewhere - I think the Logical_Switch external_ids - then we won't need the dependency on the subnet dhcp enabled and we won't need to call the plugin object in the code too | 20:33 |
| jlibosva | but if there is a better way, we can find it :) | 20:33 |
| haleyb | jlibosva: it seems most of the code would live in that file, we'd have to see where it fills DHCP_Options and add something over there. So the BGP code only cares about the gateway_ip it seems? | 20:42 |
| jlibosva | haleyb: yes, only gw ip | 21:02 |
| haleyb | jlibosva: ack, i've sent Claude off with some new instructions, seems to be headed in the correct direction... | 21:05 |
| jlibosva | haleyb: good :) btw Terry replied on the patch | 21:10 |
| jlibosva | just fyi :) | 21:10 |
| opendevreview | Brian Haley proposed openstack/neutron master: Populate external subnet routing info when dhcp disabled https://review.opendev.org/c/openstack/neutron/+/1005970 | 21:11 |
| haleyb | jlibosva: thanks again, and otherwiseguy too! | 21:13 |
| opendevreview | Terry Wilson proposed openstack/neutron master: Use Neutron DB for OVN agent health https://review.opendev.org/c/openstack/neutron/+/818850 | 21:46 |
| opendevreview | Terry Wilson proposed openstack/neutron master: Move OVN agent health monitoring to maintenance worker https://review.opendev.org/c/openstack/neutron/+/997924 | 21:48 |
| opendevreview | Jakub Libosvar proposed openstack/neutron master: evpn: Schedule VRF hosting https://review.opendev.org/c/openstack/neutron/+/1006348 | 23:17 |
| opendevreview | Merged openstack/neutron-lib master: api: fix convert_to_int() to handle None input https://review.opendev.org/c/openstack/neutron-lib/+/1005894 | 23:39 |
Generated by irclog2html.py 4.1.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!