| opendevreview | Brian Haley proposed openstack/neutron master: Fix API workers forever loop in OVN _init_hash_ring https://review.opendev.org/c/openstack/neutron/+/1008632 | 00:31 |
|---|---|---|
| opendevreview | chanyeol yoon proposed openstack/neutron-lib master: Add the tap-mirror-lport API definition for TaaS https://review.opendev.org/c/openstack/neutron-lib/+/1009192 | 04:39 |
| opendevreview | yatin proposed openstack/neutron-tempest-plugin master: Skip L3AgentSchedulerTestJSON if no L3 agent is found https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/1009695 | 04:46 |
| opendevreview | Merged openstack/neutron-vpnaas master: Fix a race condition in swanctl OVN agent test https://review.opendev.org/c/openstack/neutron-vpnaas/+/1008878 | 06:20 |
| opendevreview | yatin proposed openstack/neutron-tempest-plugin master: Fix fwaas_v2_client using removed tempest.exceptions.TimeoutException https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/1009702 | 07:07 |
| opendevreview | yatin proposed openstack/neutron-tempest-plugin master: Skip L3AgentSchedulerTestJSON if no L3 agent is found https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/1009695 | 07:18 |
| opendevreview | Thibaut Démaret proposed openstack/neutron master: bgp: Allow VLAN and extra flat provider networks via ``provider_physnet`` https://review.opendev.org/c/openstack/neutron/+/1009441 | 07:44 |
| opendevreview | Arnaud Morin proposed openstack/neutron master: Restore duplicate subnet check in extra gateways overlap check https://review.opendev.org/c/openstack/neutron/+/1009710 | 08:03 |
| opendevreview | yatin proposed openstack/neutron-tempest-plugin master: Skip L3AgentSchedulerTestJSON if no L3 agent is found https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/1009695 | 08:21 |
| opendevreview | Arnaud Morin proposed openstack/neutron-specs master: RFE: make ovs distributed dhcp production ready https://review.opendev.org/c/openstack/neutron-specs/+/1009719 | 09:01 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron-vpnaas master: ci: add Rally OVN job and promote Rally jobs to check queue https://review.opendev.org/c/openstack/neutron-vpnaas/+/1002277 | 09:09 |
| opendevreview | Arnaud Morin proposed openstack/neutron master: Restrict port device_id and device_owner to admin and service roles https://review.opendev.org/c/openstack/neutron/+/861169 | 09:45 |
| opendevreview | Arnaud Morin proposed openstack/neutron master: dhcp-l2-ext: fix domain_search typo https://review.opendev.org/c/openstack/neutron/+/1009724 | 10:04 |
| opendevreview | Arnaud Morin proposed openstack/neutron master: dhcp-l2-ext: intercept DHCP per served port only https://review.opendev.org/c/openstack/neutron/+/1009725 | 10:04 |
| opendevreview | Arnaud Morin proposed openstack/neutron master: dhcp-l2-ext: remove the flows of a re-plugged port https://review.opendev.org/c/openstack/neutron/+/1009726 | 10:04 |
| opendevreview | Arnaud Morin proposed openstack/neutron master: dhcp-l2-ext: get DHCPv6 closer to RFC 9915 https://review.opendev.org/c/openstack/neutron/+/1009727 | 10:04 |
| opendevreview | Arnaud Morin proposed openstack/neutron master: dhcp-l2-ext: answer the DHCPv4 client states https://review.opendev.org/c/openstack/neutron/+/1009728 | 10:04 |
| opendevreview | Arnaud Morin proposed openstack/neutron master: dhcp-l2-ext: pick the IPv4 to serve as dnsmasq does https://review.opendev.org/c/openstack/neutron/+/1009729 | 10:04 |
| opendevreview | Arnaud Morin proposed openstack/neutron master: dhcp-l2-ext: split option 121 instead of overflowing a single byte https://review.opendev.org/c/openstack/neutron/+/1009730 | 10:04 |
| opendevreview | Arnaud Morin proposed openstack/neutron master: dhcp-l2-ext: advertise sibling subnets as on-link routes https://review.opendev.org/c/openstack/neutron/+/1009731 | 10:04 |
| opendevreview | Arnaud Morin proposed openstack/neutron master: dhcp-l2-ext: cap the reply to what the client and the link accept https://review.opendev.org/c/openstack/neutron/+/1009732 | 10:04 |
| opendevreview | Arnaud Morin proposed openstack/neutron master: dhcp-l2-ext: keep DHCPv4 working on a /31 and a /32 https://review.opendev.org/c/openstack/neutron/+/1009733 | 10:04 |
| opendevreview | Arnaud Morin proposed openstack/neutron master: dhcp-l2-ext: encode the DHCPv6 domain names as DNS names https://review.opendev.org/c/openstack/neutron/+/1009734 | 10:04 |
| opendevreview | Arnaud Morin proposed openstack/neutron master: dhcp-l2-ext: let each responder ignore the other's packets https://review.opendev.org/c/openstack/neutron/+/1009735 | 10:04 |
| opendevreview | Arnaud Morin proposed openstack/neutron master: dhcp-l2-ext: log the DHCP packet-in path at debug https://review.opendev.org/c/openstack/neutron/+/1009736 | 10:04 |
| opendevreview | Arnaud Morin proposed openstack/neutron master: dhcp-l2-ext: drop a packet os_ken cannot parse https://review.opendev.org/c/openstack/neutron/+/1009737 | 10:04 |
| opendevreview | Arnaud Morin proposed openstack/neutron master: dhcp-l2-ext: meter the DHCP packets punted to the controller https://review.opendev.org/c/openstack/neutron/+/1009738 | 10:04 |
| opendevreview | Arnaud Morin proposed openstack/neutron master: dhcp-l2-ext: charge each port for the DHCP it asks for https://review.opendev.org/c/openstack/neutron/+/1009739 | 10:04 |
| opendevreview | Arnaud Morin proposed openstack/neutron master: dhcp-l2-ext: log what a guest can trigger at debug https://review.opendev.org/c/openstack/neutron/+/1009740 | 10:04 |
| opendevreview | Arnaud Morin proposed openstack/neutron master: dhcp-l2-ext: announce no resolver for 0.0.0.0 and :: https://review.opendev.org/c/openstack/neutron/+/1009741 | 10:04 |
| opendevreview | Arnaud Morin proposed openstack/neutron master: dhcp-l2-ext: send the DHCPv4 host name without the domain https://review.opendev.org/c/openstack/neutron/+/1009742 | 10:04 |
| ralonsoh | amorin, just asking, did you think about moving to ML2/OVN? | 10:36 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron-vpnaas master: ci: add Rally OVN job and promote Rally jobs to check queue https://review.opendev.org/c/openstack/neutron-vpnaas/+/1002277 | 11:06 |
| amorin | ralonsoh: yes we did. But not recently. Our conclusions were that it was too much more effort at that moment and that it would take us too long time to migrate from ovs to ovn. So we decided to stay with ovs. I don't remember all details and maybe we should reconsider, but we are not in that move yet | 11:38 |
| amorin | I'd like to attend the PTG to talk about this (dhcp distributed) and also distributed metadata. Does it sounds a good idea? I added my name and topics at the end of the etherpad, should I do something else? | 12:35 |
| ralonsoh | amorin, sure, any topic is welcome. Just add the topics, Brian will make the schedule | 12:45 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron-vpnaas master: ci: add Rally OVN job and promote Rally jobs to check queue https://review.opendev.org/c/openstack/neutron-vpnaas/+/1002277 | 12:56 |
| opendevreview | Eduardo Olivares proposed openstack/neutron master: DNM == Stress the BGP job IPv6 DAD fix https://review.opendev.org/c/openstack/neutron/+/1004573 | 12:58 |
| ralonsoh | lajoskatona, hi! if you have a couple of mins: https://review.opendev.org/c/openstack/neutron/+/1008016 | 13:02 |
| ralonsoh | thanks! | 13:02 |
| ralonsoh | no rush, this is just a follow-up of the mentioned patch | 13:02 |
| *** ykarel__ is now known as ykarel | 13:08 | |
| ykarel | haleyb, Hi | 13:08 |
| ykarel | i see your email for drivers meeting cancelled today | 13:09 |
| opendevreview | Lajos Katona proposed openstack/neutron-tempest-plugin master: dns_forwarder extension: add extension to distributed dhcp job https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/958775 | 13:09 |
| ykarel | seems same applies to all next week meetings, CI, drivers? if yes can you send cancellation for those too? | 13:10 |
| opendevreview | Jakub Libosvar proposed openstack/neutron master: evpn: Use random scheduler for synced evpn routers https://review.opendev.org/c/openstack/neutron/+/1007011 | 13:22 |
| opendevreview | Helen Chen proposed openstack/neutron-specs master: Propose spec EVPN Type-2 route advertisement https://review.opendev.org/c/openstack/neutron-specs/+/1004908 | 13:35 |
| blanson[m] | Hello neutron people ! :) we're digging out an old issue of ours, which is that the openvswitch firewall_driver, using ml2/ovs, does not filter traffic for collocated instances on the same compute node, belonging to the same subnet ? I had a discussion about it here not too long ago, and essentially, it's by design, and disabling this breaks HA routers and other stuff. | 13:39 |
| blanson[m] | We noticed iptables_hybrid firewall driver does not exhibit the same behavior ? so My question is I guess, would you advise rolling back to iptables_hybrid fw driver ? or not ? and if yes, is it something that can be done live ? | 13:39 |
| blanson[m] | (migrating to ovn is not on the table yet for us unfortunately) | 13:39 |
| ralonsoh | blanson[m], I'm not aware of this issue that you are describing. That would be a critical bug | 13:45 |
| ralonsoh | correct me if I'm wrong: you are saying that OVS FW driver does not enforce any rule for VMs in the same node? | 13:45 |
| blanson[m] | so I had a discussion about it with someone here this year and I understood that this was known ? | 13:47 |
| blanson[m] | essentially, 2 VMs that are in the same subnet, and on the same node, can see each-other's traffic | 13:47 |
| blanson[m] | this is mostly invisible for people using private subnets and whatnot, but if you have a shared public network with direct routing of public IP addresses, people can see each other traffic | 13:48 |
| ralonsoh | what does it mean "see each other's traffic"? that you can tcpdump somehow in the interface and see other VM port traffic? | 13:48 |
| blanson[m] | I can spawn a VM on this network and intercept traffic from other VMs yes | 13:49 |
| blanson[m] | unicast type stuff | 13:49 |
| blanson[m] | it's not great | 13:49 |
| blanson[m] | assuming they areon my computye node and in my subnet | 13:49 |
| ralonsoh | can you provide a reproducer?? Try using https://bugs.launchpad.net/neutron and then setting this bug under embargo | 13:50 |
| blanson[m] | will do | 13:50 |
| ralonsoh | Only neutron core reviewers can see this bug | 13:50 |
| blanson[m] | the issue that I have is that I might not be able to reproduce on recent enough version of openstack, but I'll reproduce on the most recent cluster we have and then check code to see if it changed | 13:58 |
| slaweq | haleyb: hi, regarding PTG, can you schedule topics added by me on Tuesday or if Wednesday or Thursday then before 3pm UTC? I will need to leave around 3pm utc on those 2 days unfortunately | 13:59 |
| ralonsoh | I would also prefer to start earlier, if possible | 14:00 |
| ralonsoh | can we start 1 hour earlier? or is this too early for the US? | 14:00 |
| ralonsoh | for me is the same: on Wednesday and Tuesday I would need to leave at 3 UTC | 14:01 |
| ralonsoh | * Wednesday and Thursday (not Tuesday) | 14:01 |
| haleyb | ykarel: yes, we should cancel all next weeks meetings too | 14:09 |
| haleyb | slaweq: yes, can make sure to schedule things earlier - we might only have 2 days of topics anyway | 14:10 |
| haleyb | ralonsoh: 1 hour earlier would be too early (i think) for mlavalle, we can just end earlier | 14:11 |
| haleyb | i have to do the scheduling today | 14:12 |
| ralonsoh | cool then, maybe we can do more things on Tuesday | 14:12 |
| ralonsoh | thanks for preparing the schedule! | 14:12 |
| haleyb | and monday is a us holiday, i will not be around for any of the tc meetings | 14:12 |
| ralonsoh | and in Spain, is public holiday | 14:13 |
| ralonsoh | very bad week... | 14:13 |
| haleyb | in a number of other countries too, most of my team here is out | 14:13 |
| ralonsoh | hahahaah I don't know who reserved this week | 14:14 |
| * haleyb shrugs | 14:14 | |
| opendevreview | sean mooney proposed openstack/os-vif stable/2026.1: ovs: Pass ``multi_queue`` inside ``ifr`` dict for tuntap devices https://review.opendev.org/c/openstack/os-vif/+/1009778 | 14:16 |
| ykarel | ack haleyb , can you also send cancellation email for these | 14:35 |
| haleyb | ykarel: yes, can do later | 14:39 |
| opendevreview | Miro Tomaska proposed openstack/neutron master: Add EvpnExecutor for EVPN provisioning https://review.opendev.org/c/openstack/neutron/+/1006431 | 14:48 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron-vpnaas master: ci: add Rally OVN job and promote Rally jobs to check queue https://review.opendev.org/c/openstack/neutron-vpnaas/+/1002277 | 14:58 |
| haleyb | amorin: is 30 minutes enough time for a discussion on your topics? | 15:13 |
| amorin | oh yes, I believe that's enough | 15:14 |
| opendevreview | Merged openstack/neutron master: evpn: Schedule VRF hosting https://review.opendev.org/c/openstack/neutron/+/1006348 | 15:15 |
| opendevreview | Arnaud Morin proposed openstack/neutron master: RFE: make ovs distributed metadata production ready https://review.opendev.org/c/openstack/neutron/+/1009796 | 15:50 |
| opendevreview | Merged openstack/neutron-tempest-plugin master: Fix fwaas_v2_client using removed tempest.exceptions.TimeoutException https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/1009702 | 15:50 |
| opendevreview | Doug Goldstein proposed openstack/neutron-lib master: add initial typing to TypeDriver interface https://review.opendev.org/c/openstack/neutron-lib/+/987769 | 15:54 |
| blanson[m] | ralonsoh: do I just tick the security box in LP to submit a report under embargo ? | 15:56 |
| cardoe | haleyb: you prompted me to revive my attempt at typing the ML2 interfaces | 16:21 |
| haleyb | woohoo! :) | 17:06 |
| opendevreview | Stephen Finucane proposed openstack/neutron-lib master: pre-commit: Remove bandit https://review.opendev.org/c/openstack/neutron-lib/+/1009827 | 17:49 |
| opendevreview | Stephen Finucane proposed openstack/neutron-lib master: pre-commit: Use local neutron-lib hacking checks https://review.opendev.org/c/openstack/neutron-lib/+/1009828 | 17:49 |
| opendevreview | Stephen Finucane proposed openstack/neutron-lib master: tox: Use constraints option https://review.opendev.org/c/openstack/neutron-lib/+/1009829 | 17:49 |
| opendevreview | Stephen Finucane proposed openstack/neutron-lib master: mypy: Enable disallow_incomplete_defs https://review.opendev.org/c/openstack/neutron-lib/+/1009830 | 17:49 |
| opendevreview | Doug Goldstein proposed openstack/neutron master: ovn: Use the active binding in is_port_external() https://review.opendev.org/c/openstack/neutron/+/1009837 | 18:48 |
| opendevreview | Doug Goldstein proposed openstack/neutron-lib master: add initial typing to TypeDriver interface https://review.opendev.org/c/openstack/neutron-lib/+/987769 | 18:52 |
| opendevreview | Doug Goldstein proposed openstack/neutron master: l3: publish ROUTER_GATEWAY events for extra gateway ports https://review.opendev.org/c/openstack/neutron/+/1009843 | 20:13 |
| opendevreview | Doug Goldstein proposed openstack/neutron master: l3: publish ROUTER_GATEWAY events for extra gateway ports https://review.opendev.org/c/openstack/neutron/+/1009843 | 20:14 |
| opendevreview | Jakub Libosvar proposed openstack/neutron master: evpn: Use round-robin scheduler for synced evpn routers https://review.opendev.org/c/openstack/neutron/+/1007011 | 22:29 |
Generated by irclog2html.py 4.1.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!