| opendevreview | Brian Haley proposed openstack/neutron master: Fix used-before-assignment warnings https://review.opendev.org/c/openstack/neutron/+/909149 | 00:21 |
|---|---|---|
| opendevreview | Roberto Acosta proposed openstack/neutron master: WIP: Add coexistence support for OVN externally managed resources. https://review.opendev.org/c/openstack/neutron/+/909808 | 01:27 |
| opendevreview | Roberto Acosta proposed openstack/neutron master: WIP: Add coexistence support for OVN externally managed resources. https://review.opendev.org/c/openstack/neutron/+/909808 | 01:41 |
| opendevreview | Merged openstack/networking-bagpipe master: Add python3.10 & 3.11 support in testing runtime https://review.opendev.org/c/openstack/networking-bagpipe/+/902547 | 06:57 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: [OVN] Identify the LR GW port with "external_ids:neutron:is_ext_gw" https://review.opendev.org/c/openstack/neutron/+/909311 | 07:59 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: [OVN] Remove OVN_GATEWAY_INVALID_CHASSIS artifact https://review.opendev.org/c/openstack/neutron/+/909305 | 07:59 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: [OVN] "Logical_Router" pinned to chassis, OVN L3 scheduler https://review.opendev.org/c/openstack/neutron/+/909194 | 08:00 |
| *** mklejn_ is now known as mklejn | 08:34 | |
| slaweq | ralonsoh lajoskatona ykarel hi, qq about devstack. When I'm trying to run devstack locally in the same way as I was doing it since "forever" I have error like in https://paste.opendev.org/show/bFtXUL26S3nwIn1N8wv7/ - did You maybe saw something like that already? | 08:34 |
| ralonsoh | let me check | 08:34 |
| slaweq | I think it's something on my side as gates are working fine but I don't know what can I have wrong :/ | 08:34 |
| ralonsoh | slaweq, did you checkout the correct versions? | 08:35 |
| ralonsoh | OVN_BRANCH="v21.06.0" | 08:35 |
| ralonsoh | OVS_BRANCH="a4b04276ab5934d087669ff2d191a23931335c87" | 08:35 |
| ralonsoh | this is what I have in my local.conf | 08:35 |
| ralonsoh | but there are newer compatible versions | 08:36 |
| ralonsoh | OVN_BRANCH: main OVS_BRANCH: branch-3.3 | 08:36 |
| slaweq | I don't have those vars set in my local.conf at all | 08:36 |
| slaweq | so IIRC ovs should be installed from packages | 08:36 |
| slaweq | and ovn too | 08:36 |
| slaweq | I can try with installation from sources but it is failing on building python ovs package AFAIU | 08:37 |
| ralonsoh | https://mail.openvswitch.org/pipermail/ovs-discuss/2019-June/048856.html | 08:39 |
| ralonsoh | slaweq, do a devstack/clean.sh before executing stack again | 08:40 |
| ralonsoh | and manually delete any ovn/openvswitch module | 08:40 |
| slaweq | I will try | 08:41 |
| slaweq | thx | 08:41 |
| opendevreview | Michel Nederlof proposed openstack/ovn-bgp-agent master: Disable exposing remote_ips, when only the lrp prefix is sufficient https://review.opendev.org/c/openstack/ovn-bgp-agent/+/907057 | 08:53 |
| lajoskatona | slaweq, ralonsoh: OVN_BUILD_FROM_SOURCE should be False otherwise the OVN and OVS version should be set as ralonsoh suggested | 09:13 |
| ykarel | slaweq, /me not seen that issue. but /me have not deployed devstack locally from quite sometime | 09:16 |
| ykarel | but /me normally deploy with source | 09:20 |
| opendevreview | Slawek Kaplonski proposed openstack/neutron master: Ensure that haproxy spawned by the metadata agents is active https://review.opendev.org/c/openstack/neutron/+/909848 | 09:24 |
| slaweq | I have disabled installation of OVS and OVN from sources, the issue happens when I try to install "ovs" package using pip in the /opt/stack/data/venv virtualenv | 09:29 |
| slaweq | but if I install ovs package without virtual env - it works fine | 09:30 |
| opendevreview | Michel Nederlof proposed openstack/ovn-bgp-agent master: Disable exposing remote_ips, when only the lrp prefix is sufficient https://review.opendev.org/c/openstack/ovn-bgp-agent/+/907057 | 09:32 |
| ralonsoh | slaweq, why don't you remove /stack/data before stacking? | 09:35 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: [OVN] Remove ``create_lrouter`` and ``delete_lrouter`` implementation https://review.opendev.org/c/openstack/neutron/+/909649 | 09:36 |
| slaweq | ralonsoh I have this error on first run of stack.sh on fresh vm created with vagrant | 09:37 |
| ralonsoh | upsss | 09:37 |
| slaweq | but my ansible playbook installed openvswitch packages on it before stack.sh was run | 09:37 |
| ralonsoh | ah no | 09:37 |
| slaweq | I will try to not install ovs before and see if that will help | 09:37 |
| ralonsoh | slaweq, stop this then | 09:37 |
| opendevreview | Christian Rohmann proposed openstack/neutron-vpnaas master: Improve the reconciling for VPNaaS https://review.opendev.org/c/openstack/neutron-vpnaas/+/875745 | 09:48 |
| opendevreview | Michel Nederlof proposed openstack/ovn-bgp-agent master: Fix event handling for LSP and prefer the options.requested-chassis info https://review.opendev.org/c/openstack/ovn-bgp-agent/+/906111 | 10:45 |
| opendevreview | Michel Nederlof proposed openstack/ovn-bgp-agent master: Fix event handling for LSP and prefer the options.requested-chassis info https://review.opendev.org/c/openstack/ovn-bgp-agent/+/906111 | 10:47 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: [OVN] Identify the LR GW port with "external_ids:neutron:is_ext_gw" https://review.opendev.org/c/openstack/neutron/+/909311 | 10:49 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: [OVN] Remove OVN_GATEWAY_INVALID_CHASSIS artifact https://review.opendev.org/c/openstack/neutron/+/909305 | 10:49 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: [OVN] "Logical_Router" pinned to chassis, OVN L3 scheduler https://review.opendev.org/c/openstack/neutron/+/909194 | 10:49 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: [OVN] Provide HA functionality to "Logical_Router" chassis pinning https://review.opendev.org/c/openstack/neutron/+/909437 | 10:49 |
| ralonsoh | slaweq, how is going? | 10:50 |
| opendevreview | Michel Nederlof proposed openstack/ovn-bgp-agent master: Update NB driver to re-use new methods and make code cleaner https://review.opendev.org/c/openstack/ovn-bgp-agent/+/907073 | 10:55 |
| opendevreview | Merged openstack/ovn-bgp-agent master: Disable exposing remote_ips, when only the lrp prefix is sufficient https://review.opendev.org/c/openstack/ovn-bgp-agent/+/907057 | 11:42 |
| ralonsoh | slaweq, https://review.opendev.org/c/openstack/neutron/+/898238 replied. Once we have the functionality merged in Neutron, I will be able to migrate this job to n-t-p | 11:45 |
| ralonsoh | for now I can't run it there | 11:45 |
| slaweq | ralonsoh devstack is still failing for me, I have no idea why | 11:50 |
| slaweq | I will need to dive deeper into this probably but not now | 11:51 |
| slaweq | regarding Your patch, I just replied there | 11:51 |
| slaweq | ralonsoh lajoskatona ykarel please check https://review.opendev.org/c/openstack/neutron/+/909848 when You will have few minutes | 11:53 |
| opendevreview | Merged openstack/ovn-bgp-agent master: Trimm interface name consistently https://review.opendev.org/c/openstack/ovn-bgp-agent/+/909788 | 12:00 |
| lajoskatona | slaweq: added to my pile | 12:05 |
| sean-k-mooney | slaweq: regarding https://bugs.launchpad.net/nova/+bug/2052937 and https://review.opendev.org/c/openstack/neutron/+/909075 | 12:13 |
| sean-k-mooney | slaweq: do you have a corresponding patch to enable trusted vfs | 12:13 |
| sean-k-mooney | your not wrong to require the servie role for messing wiht the binding_profile | 12:15 |
| sean-k-mooney | it is intened to be used only be services | 12:15 |
| sean-k-mooney | but trusted_vf used to rely on custom policy and injecting stuff into it | 12:15 |
| sean-k-mooney | without a change to nova to make https://github.com/openstack/nova/blob/3209f6551652cff7bef0b9d9719ab940dd05a0f8/nova/network/neutron.py#L2186-L2196 look at whatever extention or other filed ye are not using for this | 12:20 |
| sean-k-mooney | this will break that feautre | 12:20 |
| sean-k-mooney | for the record i hate that feature so that is ok with me | 12:20 |
| sean-k-mooney | but proably not for users using it | 12:20 |
| sean-k-mooney | lajoskatona: ralonsoh ^ | 12:21 |
| ralonsoh | let me check | 12:21 |
| sean-k-mooney | https://github.com/openstack/neutron/blob/ebb7ec1eedd1f6702e747729a501156038a28bf1/doc/source/admin/config-sriov.rst?plain=1#L264-L290 | 12:22 |
| sean-k-mooney | the neutron docs still talk about updating the bidning profile | 12:22 |
| sean-k-mooney | https://github.com/openstack/neutron/blob/ebb7ec1eedd1f6702e747729a501156038a28bf1/doc/source/admin/config-sriov.rst?plain=1#L463-L467 | 12:23 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron-tempest-plugin master: [OVN] Add CI job "neutron-tempest-plugin-ovn-with-ovn-agent" https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/909860 | 12:23 |
| sean-k-mooney | ralonsoh: rather then revert the SRBAC change my suggestion woudl be to add a new port hint for tursted mode | 12:24 |
| sean-k-mooney | https://github.com/openstack/neutron-lib/blob/master/neutron_lib/api/definitions/port_hints.py | 12:24 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: [OVN] Implement OVN agent metadata extension https://review.opendev.org/c/openstack/neutron/+/898238 | 12:24 |
| sean-k-mooney | port hints are also admin only i belvie by default so it would have the same security profile as before | 12:26 |
| ralonsoh | sean-k-mooney, so this is the same as the HW offloaded feature: we have created an extension to define a HW offloaded port | 12:26 |
| ralonsoh | instead of writing in the port PB | 12:26 |
| ralonsoh | this is a technical debt in Neutron | 12:26 |
| sean-k-mooney | ralonsoh: no for hardware offload you don tneed to do anythng on the neutron side | 12:26 |
| ralonsoh | is an example | 12:27 |
| ralonsoh | you do yes | 12:27 |
| sean-k-mooney | no you dont | 12:27 |
| sean-k-mooney | nova sets the value for you now | 12:27 |
| ralonsoh | one second | 12:27 |
| sean-k-mooney | https://github.com/openstack/neutron-lib/blob/master/neutron_lib/api/definitions/port_hardware_offload_type.py is proably a mistake | 12:28 |
| sean-k-mooney | ralonsoh: https://review.opendev.org/c/openstack/nova/+/884439 fixes hardware offload | 12:29 |
| opendevreview | Merged openstack/os-vif master: Drop wrong stacklevel https://review.opendev.org/c/openstack/os-vif/+/909682 | 12:29 |
| sean-k-mooney | we have backporated it to all supported branches | 12:29 |
| ralonsoh | sean-k-mooney, what means "fixes"? | 12:30 |
| ralonsoh | sean-k-mooney, why is a mistake? | 12:30 |
| ralonsoh | if you change something in Nova, you should report that here, I think | 12:30 |
| sean-k-mooney | you nolonger need in the binding:profile which was alwasy broken by desgin | 12:30 |
| ralonsoh | for now, we need to define the binding profile | 12:30 |
| sean-k-mooney | ralonsoh: we agree on how to do this in the ptg | 12:30 |
| ralonsoh | no | 12:30 |
| ralonsoh | we didn't | 12:30 |
| sean-k-mooney | yes we did | 12:31 |
| sean-k-mooney | this has been in flint for a few releases now | 12:31 |
| ralonsoh | actually I implemented this feature without touching anything on the nova side | 12:31 |
| ralonsoh | it is on the spec | 12:31 |
| ralonsoh | https://review.opendev.org/c/openstack/neutron-specs/+/882272/7/specs/2023.2/hardware-offload-port-extension.rst#58 | 12:31 |
| ralonsoh | if that has changed, please report a Neutron bug | 12:31 |
| ralonsoh | just to have this information | 12:32 |
| ralonsoh | openstack port create --vnic-type direct \ | 12:32 |
| ralonsoh | --binding-profile '{"capabilities": ["switchdev"]}' port_hwol | 12:32 |
| ralonsoh | this is how we create a HW offloaded port right now | 12:32 |
| ralonsoh | if that is not the case anymore, please report that in Neutron LP to have this info | 12:32 |
| sean-k-mooney | that is not required anymore | 12:32 |
| sean-k-mooney | openstack port create --vnic-type direct | 12:33 |
| sean-k-mooney | is all that is needed | 12:33 |
| ralonsoh | sean-k-mooney, so please report a LP bug for Neutron | 12:33 |
| ralonsoh | to have this info | 12:33 |
| ralonsoh | if we don't sync between each other, this is not going to work | 12:33 |
| sean-k-mooney | i tougt hwe had a neutron docs patch as part of htis | 12:33 |
| sean-k-mooney | ralonsoh: lajoskatona was on the review | 12:33 |
| ralonsoh | slaweq, lajoskatona ^^ | 12:33 |
| ralonsoh | where? | 12:33 |
| sean-k-mooney | https://review.opendev.org/c/openstack/neutron/+/904273 | 12:33 |
| sean-k-mooney | https://review.opendev.org/c/openstack/neutron/+/898556 | 12:34 |
| sean-k-mooney | that the master version | 12:34 |
| sean-k-mooney | we finally fixed this bug 4 months ago and backported it to all supported stable branches | 12:34 |
| ralonsoh | to be honest, sometimes I think I work for noithing | 12:35 |
| ralonsoh | when this patch was pushed, I was in parental leave | 12:36 |
| ralonsoh | but in any case, please, where that was discussed in the PTG? | 12:36 |
| ralonsoh | sean-k-mooney, ^ | 12:36 |
| sean-k-mooney | i can go find the discussion but it was disucss 2 ptgs ago i think in the nova neutron session | 12:36 |
| sean-k-mooney | on sec ill find the notes | 12:37 |
| ralonsoh | if that was in the last ptg, I was in parentl leave | 12:38 |
| ralonsoh | anyway, I worked for nothing again | 12:38 |
| ralonsoh | sean-k-mooney, ok, I'll open a LP bug for Neutron with the reference to https://bugs.launchpad.net/nova/+bug/2008238 | 12:43 |
| ralonsoh | or just add Neutron to this bug | 12:43 |
| sean-k-mooney | so for trusted_vf we need an extenion (i would just use port hints) to allow the user(with admin rights) to request enabling trusted_vf | 12:43 |
| sean-k-mooney | and idealy then nova would be upstaed ot check for that prot hint | 12:43 |
| sean-k-mooney | i can do the later quickly | 12:44 |
| sean-k-mooney | or we just need to update the documation for the custom policy for trusted_vf | 12:44 |
| sean-k-mooney | to document how to remove the requirement for the service role | 12:44 |
| ralonsoh | -> https://bugs.launchpad.net/nova/+bug/2008238/comments/18 | 12:45 |
| ralonsoh | we'll triage it next week, on tuesday | 12:45 |
| sean-k-mooney | ralonsoh: https://etherpad.opendev.org/p/neutron-antelope-ptg#L341 | 12:46 |
| ralonsoh | exactly, and this is what I did: to add an extension for HWoffload | 12:47 |
| ralonsoh | and now it not needed | 12:47 |
| sean-k-mooney | we agreed not to make a change in neutorn for this in antelope | 12:47 |
| sean-k-mooney | and to defer if it was a bug or feature in nova to the code review | 12:47 |
| sean-k-mooney | we converted https://blueprints.launchpad.net/nova/+spec/non-admin-hw-offloaded-ovs to a bug | 12:48 |
| sean-k-mooney | because of the secuiryt implciations of the custom policy | 12:48 |
| ralonsoh | because we had no feedback, https://etherpad.opendev.org/p/neutron-bobcat-ptg#L356 | 12:48 |
| ralonsoh | we implemented it | 12:48 |
| sean-k-mooney | so we endup not talking about that in the corss project session | 12:49 |
| sean-k-mooney | and i dont think i was in the neutron operator hour | 12:50 |
| sean-k-mooney | but we did discuss it in the nova rooom i think | 12:50 |
| sean-k-mooney | alex started workign on the bugfix in antelope but it just didnt get done until 4 months ago since they are part time on nova | 12:51 |
| sean-k-mooney | anyway sorry i agree we proably could have been in sync more but we asked for neutorn review and i had hoped lajoskatona had spread the awareness | 12:52 |
| sean-k-mooney | i can see how "Translate VF network capabilities to port binding" does not scream | 12:53 |
| sean-k-mooney | this fixes hardware offloaded ovs to work securely | 12:53 |
| sean-k-mooney | or correctly | 12:53 |
| sean-k-mooney | even if the commit message mentions it | 12:54 |
| opendevreview | Vasyl Saienko proposed openstack/neutron-vpnaas master: Enable VPN for both primary and backup routers https://review.opendev.org/c/openstack/neutron-vpnaas/+/868307 | 12:59 |
| opendevreview | Maor Blaustein proposed openstack/neutron-tempest-plugin master: Skip HA check meant for OVS setup https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/909862 | 13:00 |
| opendevreview | Maor Blaustein proposed openstack/neutron-tempest-plugin master: Skip HA check meant for OVS setup https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/909862 | 13:02 |
| lajoskatona | sean-k-mooney:, ralonsoh: the doc patch is to document in Neutron doc the situation we have in Nova, sorry for not pushing this to some wider audience (drivers meeting?) | 13:31 |
| ykarel | ralonsoh, until https://review.opendev.org/c/openstack/neutron/+/908325/ if there was no chassis with enable-chassis-as-gw explicitly set(bridge mappings set), router scheduling used to work for external networks(with provider:physical_network set) but not now | 13:37 |
| ralonsoh | ykarel, do you have an example? | 13:39 |
| ralonsoh | if CI failing I mean | 13:39 |
| ralonsoh | https://review.opendev.org/c/openstack/neutron/+/908325/4/neutron/plugins/ml2/drivers/ovn/mech_driver/ovsdb/ovn_client.py#b1555 was incorrect | 13:39 |
| ralonsoh | if there is no mapping of a physical network in a chassis, we should not use it | 13:39 |
| ykarel | ralonsoh, https://review.rdoproject.org/zuul/builds?job_name=weirdo-caracal-promote-packstack-scenario001-centos-stream-9&project=rdoinfo are the impactd job | 13:44 |
| ykarel | neutron server logs https://logserver.rdoproject.org/openstack-periodic-daily/review.rdoproject.org/rdoinfo/master/weirdo-caracal-promote-packstack-scenario001-centos-stream-9/64b339f/logs/weirdo-project/logs/neutron/server.log.txt.gz | 13:44 |
| ykarel | public network is physnet=extnet, ovn bridge mappings=extnet:br-ex ovn-cms-options:"" | 13:45 |
| ralonsoh | ykarel, but why don't change the job? This configuration is incorrect | 13:48 |
| ralonsoh | what we can't do is to use a chassis without the proper mappings | 13:49 |
| ralonsoh | as commented, https://review.opendev.org/c/openstack/neutron/+/908325/4/neutron/plugins/ml2/drivers/ovn/mech_driver/ovsdb/ovn_client.py#b1555 is incorrect | 13:49 |
| ralonsoh | and the code before | 13:49 |
| ykarel | ralonsoh, but bridge mappings is set correctly, you mean ovn-cms-options not having enable-chassis-as-gw set? | 13:49 |
| ralonsoh | ykarel, both: chassis should be GW and have the mappings | 13:51 |
| ykarel | yes job will be changed to set cms-options=enable-chassis-as-gw, i am just trying to understand expected config with that cnahge | 13:51 |
| ykarel | ok thx for confirming, will ask the job to be updated | 13:51 |
| opendevreview | Michel Nederlof proposed openstack/ovn-bgp-agent master: Fix backward compatibility for setups that export subnets per host https://review.opendev.org/c/openstack/ovn-bgp-agent/+/909869 | 14:00 |
| opendevreview | Michel Nederlof proposed openstack/ovn-bgp-agent master: Fix backward compatibility for setups that export subnets per host https://review.opendev.org/c/openstack/ovn-bgp-agent/+/909869 | 14:12 |
| sean-k-mooney | lajoskatona: the real discussion for the drivers meeting is if the new extsion has any use and if it should be deprecated/removed if not | 14:15 |
| sean-k-mooney | lajoskatona: there has never been any supprot for requesting hardware offloaded ovs via neutron | 14:16 |
| sean-k-mooney | thjat was proposed orgianblly but never implemented | 14:16 |
| sean-k-mooney | so if you woudl like the hardware offload exteion to be used for that we could but it need nova/placement work to enable that | 14:16 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron-tempest-plugin master: [OVN] Use the OVN agent in "neutron-tempest-plugin-ovn" https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/909860 | 14:19 |
| sean-k-mooney | lajoskatona: just some other histroy for context. hardware offloaded ovs orgianly did not require anythign to be set on the prot other then vnic type when it was first added | 14:20 |
| sean-k-mooney | the bidnign profile stuff was added to "fix" a bug that was not really a bug and was an incorrect fix | 14:21 |
| sean-k-mooney | it was an incorrect fix because it was based on the premis that functionatliy that was not implemtned in nova existed | 14:21 |
| lajoskatona | sean-k-mooney: thanks, we can have a chat tomorrow on drivers meeting or later. | 14:25 |
| opendevreview | Michel Nederlof proposed openstack/ovn-bgp-agent master: Fix address scope test and add address scope unit tests https://review.opendev.org/c/openstack/ovn-bgp-agent/+/907187 | 14:41 |
| ralonsoh | lajoskatona, RH people won't be here tomorrow, we have a "recharge day" | 14:42 |
| ralonsoh | haleyb|out, ^ | 14:42 |
| ralonsoh | so maybe we can push it to the team meeting next Tuesday | 14:42 |
| *** haleyb|out is now known as haleyb | 14:52 | |
| haleyb | ralonsoh: sounds good, i will not be here either | 14:52 |
| opendevreview | Michel Nederlof proposed openstack/ovn-bgp-agent master: Fix backward compatibility for setups that export subnets per host https://review.opendev.org/c/openstack/ovn-bgp-agent/+/909869 | 14:52 |
| frickler | if nobody is around tomorrow, what the status for the neutron-lib release? https://review.opendev.org/c/openstack/releases/+/909397 | 15:17 |
| frickler | +is | 15:17 |
| frickler | ralonsoh: haleyb: please also have another look regarding the changed version bump for the os-ken release https://review.opendev.org/c/openstack/releases/+/909401 | 15:19 |
| frickler | it would be good to keep the number of requirements updates that need to happen next week (or later) to a minimum | 15:20 |
| ralonsoh | frickler, apart from asking other people to review the patches, I can't do anything else | 15:21 |
| ralonsoh | of course if there are some patches pending in n-lib, they won't be included (and 3 of them are mine) | 15:22 |
| ralonsoh | mlavalle, lajoskatona haleyb slaweq ykarel obondarev please check the n-lib pending patches | 15:24 |
| ralonsoh | https://review.opendev.org/q/project:openstack/neutron-lib+status:open | 15:24 |
| ralonsoh | we are running out of time | 15:24 |
| ralonsoh | tomorrow morning I'll push an update in the releases patch, regardless of any pending patch under review | 15:24 |
| frickler | ralonsoh: I just wasn't sure if there are still open reviews that are meant to be included in this cycle. if we can still expect some progress, I'm fine with waiting | 15:26 |
| ralonsoh | frickler, there are, yes | 15:26 |
| haleyb | ralonsoh: yes, i finally got a response to the charm-octavia review, but it might be enough to just get your's merged for now, i don't think that will break anything | 15:26 |
| ralonsoh | frickler, when the releases patch should be updated? | 15:26 |
| ralonsoh | haleyb, ok but this should not happen again | 15:27 |
| frickler | ralonsoh: tomorrow would be good, early next week might still be ok-ish, though I'm not sure what elodilles or other release managers think | 15:28 |
| ralonsoh | I'll update the patch tomorrow | 15:29 |
| haleyb | ralonsoh: it is rare to remove constants as well | 15:29 |
| ralonsoh | but please folks, check the pending patches | 15:29 |
| ralonsoh | some of them are ready to be merged | 15:29 |
| opendevreview | Merged openstack/ovn-bgp-agent master: Fix backward compatibility for setups that export subnets per host https://review.opendev.org/c/openstack/ovn-bgp-agent/+/909869 | 16:03 |
| elodilles | ralonsoh frickler : tomorrow is fine, after that a Feature Freeze Exception can be discussed and we can wait some more days next week if needed | 16:04 |
| ralonsoh | perfect | 16:05 |
| haleyb | slaweq: regarding your metadata change, can we merge https://review.opendev.org/c/openstack/neutron/+/894399 first? that way i don't have to rebase again | 16:32 |
| opendevreview | Vasyl Saienko proposed openstack/neutron-vpnaas master: Do not spawn strongswan ipsec for admin down connections https://review.opendev.org/c/openstack/neutron-vpnaas/+/868501 | 17:44 |
| mnaser | It seems that stable/zed is broken for the grenade jobs: `ImportError: cannot import name 'port_mac_address_override' from 'neutron_lib.api.definitions' (/usr/local/lib/python3.8/dist-packages/neutron_lib/api/definitions/__init__.py)` | 18:05 |
| mnaser | job: https://5d8dc30f05602a44926d-137d8ab1f6e795378a0b5ffa2abe6af6.ssl.cf1.rackcdn.com/909098/2/check/nova-grenade-multinode/59b64a2/controller/logs/screen-q-svc.txt | 18:06 |
| mnaser | Is it possible that it's installing a neutron-lib that's too new? | 18:06 |
| mnaser | Or something was forgotten to be backported to neutron-lib .. or we need to cut a neutron-lib release? | 18:07 |
| mnaser | port_mac_address_override references seem to be around for 2yrs | 18:07 |
| opendevreview | Merged openstack/neutron-lib master: Add "socket" NUMA affinity policy https://review.opendev.org/c/openstack/neutron-lib/+/903531 | 18:21 |
| opendevreview | Merged openstack/neutron-lib master: Remove constants for LBaaS(v2) https://review.opendev.org/c/openstack/neutron-lib/+/902133 | 18:21 |
| opendevreview | Roberto Acosta proposed openstack/neutron master: [ML2/OVN] Add external_ids key for Static Routes https://review.opendev.org/c/openstack/neutron/+/907489 | 18:31 |
| haleyb | mnaser: https://bugs.launchpad.net/neutron/+bug/2052915 - related to unmaintained/yoga change i believe | 19:32 |
| mnaser | ah right, affecting nova too, there we go | 19:33 |
| haleyb | mlavalle, slaweq: this seems to be the only neutron-lib change in the queue, https://review.opendev.org/c/openstack/neutron-lib/+/907949 - if you have cycles today | 19:53 |
| opendevreview | Brian Haley proposed openstack/neutron-lib master: Fix iptables mapping of 'ipip' protocol https://review.opendev.org/c/openstack/neutron-lib/+/909942 | 23:37 |
| opendevreview | Brian Haley proposed openstack/neutron master: Test iptables mapping of 'ipip' protocol https://review.opendev.org/c/openstack/neutron/+/909943 | 23:37 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!