opendevreview | Merged openstack/neutron-fwaas-dashboard master: Imported Translations from Zanata https://review.opendev.org/c/openstack/neutron-fwaas-dashboard/+/847117 | 03:10 |
---|---|---|
opendevreview | OpenStack Proposal Bot proposed openstack/neutron-fwaas-dashboard master: Imported Translations from Zanata https://review.opendev.org/c/openstack/neutron-fwaas-dashboard/+/866098 | 03:49 |
opendevreview | Takashi Kajinami proposed openstack/os-vif master: Fix how deprecated_reason of ovsdb_interface is logged https://review.opendev.org/c/openstack/os-vif/+/866102 | 05:13 |
labedz__ | hello, do you think it is good idea to backport this 'improvement' https://review.opendev.org/c/openstack/neutron/+/860471 to older releases? | 09:12 |
*** labedz__ is now known as labedz | 09:12 | |
ralonsoh | labedz, actually I don't know why this patch is merged | 09:13 |
ralonsoh | I would prefer Terry Wilson to check it first | 09:13 |
ralonsoh | I'm not sure about it | 09:13 |
labedz | sure | 09:13 |
opendevreview | Sahid Orentino Ferdjaoui proposed openstack/neutron master: dhcp: fix issue when network is already removed https://review.opendev.org/c/openstack/neutron/+/865823 | 09:28 |
sahid | o/ | 09:28 |
sahid | ralonsoh: ^ I fixed the patch based on your suggestion | 09:28 |
ralonsoh | ok, let me check | 09:29 |
marlinc | Hey there, last year this patch was introduced that makes a 'parent' interface for VXLAN interfaces mandatory https://opendev.org/openstack/neutron/commit/16a793af19a2dc13e3d121b36db52d6b9ec25278, it refers to this bug in Launchpad which mentions the kernel documentation as reason for it https://bugs.launchpad.net/neutron/+bug/1954316 | 09:48 |
marlinc | This is however not a requirement in the kernel and actually causes problems for EVPN based VXLAN deployments | 09:49 |
ralonsoh | marlinc, do you have a doc reference? | 09:49 |
marlinc | This is the documentation you mention, it refers to no requirement for a parent interface https://www.kernel.org/doc/Documentation/networking/vxlan.txt | 09:50 |
ralonsoh | marlinc, this is the document of the bug | 09:50 |
ralonsoh | first command: | 09:50 |
ralonsoh | 1. Create vxlan device | 09:50 |
ralonsoh | # ip link add vxlan0 type vxlan id 42 group 239.1.1.1 dev eth1 dstport 4789 | 09:50 |
ralonsoh | --> dev eth1 | 09:50 |
marlinc | Yes but it is not a requirement, in fact when you look in practice at for example EVPN-VXLAN deployments or even tools like CIlium on Kubernetes they don't define this | 09:51 |
ralonsoh | marlinc, | 09:51 |
ralonsoh | root@u20ovn:~# ip link add vxlan0 type vxlan id 50 group 239.1.1.1 dstport 4789 | 09:51 |
ralonsoh | vxlan: 'group' requires 'dev' to be specified | 09:51 |
ralonsoh | if you don't specify a parent port, the vxlan port can't be created | 09:52 |
marlinc | Yes if you use group which is only necessary in case of multicast based deployments | 09:52 |
marlinc | VXLAN however also can be used in unicast settings where neither group nor dev are required | 09:52 |
ralonsoh | marlinc, you are right | 09:53 |
ralonsoh | please open a bug | 09:53 |
marlinc | Will do! | 09:53 |
marlinc | We recently discovered this while upgrading our stable xena release to the latest commit and discovered this was backported so I was just wondering what existing feature this was breaking | 09:54 |
ralonsoh | marlinc, because we always use vxlan devices in multicast mode, that was mandatory | 09:55 |
ralonsoh | and pyroute2 was failing because of this | 09:55 |
marlinc | Ah cool | 09:56 |
marlinc | Might be good time to upstream our VXLAN EVPN patch for the nolearning option | 10:03 |
ralonsoh | slaweq, https://bugs.launchpad.net/neutron/+bug/1998235. Is this related to this patch you were implementing? | 10:10 |
opendevreview | Rodolfo Alonso proposed openstack/neutron master: [WIP] Allow shared resources between physical and tunnelled networks https://review.opendev.org/c/openstack/neutron/+/866127 | 10:11 |
ralonsoh | froyo, hi, do you mind checking https://bugs.launchpad.net/neutron/+bug/1998226? | 10:12 |
froyo | ralonsoh: sure!! | 10:13 |
opendevreview | Rodolfo Alonso proposed openstack/networking-ovn stable/train: Remove "networking-ovn-dsvm-functional-py27" job https://review.opendev.org/c/openstack/networking-ovn/+/866022 | 10:49 |
opendevreview | Roberto Bartzen Acosta proposed openstack/neutron-dynamic-routing master: Fix address_scope calculation https://review.opendev.org/c/openstack/neutron-dynamic-routing/+/863708 | 10:59 |
*** dasm|off is now known as dasm | 12:43 | |
opendevreview | Lajos Katona proposed openstack/neutron master: Fix bulk create without mac https://review.opendev.org/c/openstack/neutron/+/863881 | 12:44 |
opendevreview | Takashi Kajinami proposed openstack/os-vif master: Fix how deprecated_reason of ovsdb_interface is logged https://review.opendev.org/c/openstack/os-vif/+/866102 | 13:01 |
bcafarel | ralonsoh: for https://review.opendev.org/c/openstack/networking-ovn/+/866022 you can also remove playbooks/legacy/networking-ovn-dsvm-functional-py27/ directory (promised after that I will stop finding files to remove) | 13:57 |
ralonsoh | bcafarel, sure | 13:57 |
slaweq | ralonsoh sorry, I missed Your ping earlier | 14:08 |
slaweq | yes I marked this LP as duplicate as it's the same, old and known issue | 14:08 |
ralonsoh | slaweq, thanks | 14:08 |
slaweq | ralonsoh: yw | 14:08 |
opendevreview | Rodolfo Alonso proposed openstack/networking-ovn stable/train: Remove "networking-ovn-dsvm-functional-py27" job https://review.opendev.org/c/openstack/networking-ovn/+/866022 | 15:06 |
opendevreview | Anton Kurbatov proposed openstack/neutron master: Do not emit ssl warnings in the test_ssl_connection test https://review.opendev.org/c/openstack/neutron/+/866178 | 15:20 |
opendevreview | Felix Huettner proposed openstack/neutron master: Fix handling the restart of ovn-controllers https://review.opendev.org/c/openstack/neutron/+/865697 | 15:23 |
lajoskatona | seems fullstack is broken: https://zuul.opendev.org/t/openstack/builds?job_name=neutron-fullstack-with-uwsgi&branch=master&skip=0 | 15:51 |
lajoskatona | a lot of failures with test_packet_rate_limit_qos_policy_rule_lifecycle(ingress) & test_packet_rate_limit_qos_policy_rule_lifecycle(egress) | 15:52 |
lajoskatona | the bug for it: https://bugs.launchpad.net/neutron/+bug/1998353 | 15:56 |
ralonsoh | lajoskatona, checking now, most probably due to https://review.opendev.org/c/openstack/neutron/+/856411 | 16:18 |
ralonsoh | but why now?? | 16:18 |
lajoskatona | ralonsoh: ok, thanks, I checked more recent patches | 16:20 |
lajoskatona | ralonsoh: perhaps somehow merged effect with https://review.opendev.org/c/openstack/neutron/+/837780 ? | 16:20 |
lajoskatona | just a tip, I have to check the details | 16:21 |
ralonsoh | lajoskatona, yes because of this https://review.opendev.org/c/openstack/neutron/+/837780/18/neutron/services/trunk/drivers/openvswitch/driver.py#96 | 16:22 |
ralonsoh | or at least is what I think | 16:22 |
ralonsoh | I'll check it | 16:22 |
*** labedz is now known as labedz_ | 16:49 | |
opendevreview | Merged openstack/ovn-octavia-provider stable/yoga: Make clear distinction between health check and health monitor https://review.opendev.org/c/openstack/ovn-octavia-provider/+/865907 | 16:49 |
opendevreview | Rodolfo Alonso proposed openstack/neutron master: Load OVS Trunk extension if service plugin configured https://review.opendev.org/c/openstack/neutron/+/866210 | 20:22 |
*** Guest305 is now known as atmark | 21:07 | |
*** dasm is now known as dasm|off | 22:23 | |
opendevreview | Brian Haley proposed openstack/neutron master: Fix some pylint indentation warnings https://review.opendev.org/c/openstack/neutron/+/866225 | 22:41 |
opendevreview | Brian Haley proposed openstack/neutron master: Update .pylintrc https://review.opendev.org/c/openstack/neutron/+/866226 | 22:41 |
opendevreview | Brian Haley proposed openstack/neutron master: Fix some pylint indentation warnings https://review.opendev.org/c/openstack/neutron/+/863928 | 23:03 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!