| *** oklhost_ is now known as oklhost | 07:16 | |
| opendevreview | Slawek Kaplonski proposed openstack/neutron master: Pass host parameter to the get_network_info method https://review.opendev.org/c/openstack/neutron/+/827101 | 07:23 |
|---|---|---|
| opendevreview | Slawek Kaplonski proposed openstack/neutron master: Pass host parameter to the get_network_info method https://review.opendev.org/c/openstack/neutron/+/827101 | 07:32 |
| *** kleini_ is now known as kleini | 08:13 | |
| slaweq | lajoskatona: ralonsoh ykarel hi, I don't have time to look at it now but please check failure https://36d827c3b1c315f941c5-25d2473a8435e86d2da970876b6bac73.ssl.cf1.rackcdn.com/827101/3/check/neutron-ovs-tempest-multinode-full/8f28733/testr_results.html if You will have some time | 09:37 |
| slaweq | it seems for me that it may be some new bug in some tempest test (or maybe in job definition, IDK) | 09:37 |
| slaweq | thx in advance | 09:37 |
| * slaweq needs to go afk now | 09:37 | |
| ykarel | slaweq, ack will check after lunch | 09:38 |
| lajoskatona | slaweq: thanks for highlight | 09:38 |
| ykarel | lajoskatona, slaweq caused by https://review.opendev.org/c/openstack/tempest/+/826946 | 09:42 |
| ykarel | merged today | 09:42 |
| lajoskatona | ykarel: was quick, I just started to blame :-) | 09:42 |
| ykarel | lajoskatona, you proposing fix or shall i send? | 09:45 |
| lajoskatona | ykarel: please, it's your finding :-) | 09:45 |
| ykarel | ok /me sends | 09:46 |
| ykarel | lajoskatona, slaweq ralonsoh https://review.opendev.org/c/openstack/tempest/+/827258 | 09:53 |
| opendevreview | Fernando Royo proposed openstack/ovn-octavia-provider stable/ussuri: Fix race condition retrieving logical router rows https://review.opendev.org/c/openstack/ovn-octavia-provider/+/827077 | 10:02 |
| ykarel | lajoskatona, can you +W https://review.opendev.org/c/openstack/neutron/+/826811 | 10:14 |
| ykarel | xena gates are clear | 10:14 |
| lajoskatona | ykarel: done | 10:21 |
| ykarel | Thanks lajoskatona | 10:25 |
| opendevreview | Lajos Katona proposed openstack/tap-as-a-service master: Use ovs TUNNEL_ constants from new location https://review.opendev.org/c/openstack/tap-as-a-service/+/827263 | 10:51 |
| frickler | lajoskatona: bcafarel: am I right in my understanding that no progress has been made yet in terms of getting n-d-r to work with OVN? the topic just came up again internally and I wonder how you folks deploy IPv6 connectivity for tenants at all without it | 10:53 |
| lajoskatona | frickler: I have seen no progress, but perhaps RH folks have more insights on this topic | 10:55 |
| opendevreview | Lajos Katona proposed openstack/neutron master: OVN TestNBDbResources wait for NB_Global table to be present https://review.opendev.org/c/openstack/neutron/+/825530 | 11:10 |
| frickler | seems I never created a bug report after adding it to the OVN gap list. I just did that and look what a nice number it received https://bugs.launchpad.net/neutron/+bug/1959666 | 11:10 |
| ralonsoh | lajoskatona, sorry, I think you didn't see my last comment https://review.opendev.org/c/openstack/neutron/+/825530/7/neutron/tests/functional/plugins/ml2/drivers/ovn/mech_driver/ovsdb/test_ovn_db_resources.py | 11:25 |
| opendevreview | Frode Nordahl proposed openstack/neutron master: [OVN] Off-path SmartNIC DPU Port Binding with OVN https://review.opendev.org/c/openstack/neutron/+/808961 | 11:47 |
| lajoskatona | ralonsoh: checking | 11:57 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: [OVS] Add IPv6 ICMP RA to the default ingress rules https://review.opendev.org/c/openstack/neutron/+/827159 | 12:13 |
| opendevreview | Frode Nordahl proposed openstack/neutron master: [OVN] Off-path SmartNIC DPU Port Binding with OVN https://review.opendev.org/c/openstack/neutron/+/808961 | 12:37 |
| *** dasm|off is now known as dasm | 13:02 | |
| *** dasm is now known as dasm|rover | 13:02 | |
| marc-vorwerk676009243582062483 | Hello, we disabled today the neutron fwaas driver in our openstack environment. Is it safe to delete now all fwaas database tables? Or are these tables used for other things? | 13:25 |
| opendevreview | Frode Nordahl proposed openstack/neutron master: [OVN] Off-path SmartNIC DPU Port Binding with OVN https://review.opendev.org/c/openstack/neutron/+/808961 | 13:26 |
| ralonsoh | marc-vorwerk676009243582062483, unused tables won't affect and I don't know if fwaas did some other change to the neutron tables referring to the fwaas ones | 13:38 |
| ralonsoh | if you delete them, first make a DB backup | 13:38 |
| lajoskatona | all: I will be perhaps a few minutes late from the team meeting (have to fetch my son from the school) | 13:39 |
| opendevreview | Lajos Katona proposed openstack/neutron master: OVN TestNBDbResources wait for NB_Global table to be present https://review.opendev.org/c/openstack/neutron/+/825530 | 13:40 |
| marc-vorwerk676009243582062483 | thank you ralonsoh. Do you think its also safe to delete the neutron lbaas tables? | 13:50 |
| *** marlinc is now known as Guest1393 | 13:51 | |
| ralonsoh | same answer here, first make a backup. Those tables should be used only for lbaas, thus there should not be any problem | 13:54 |
| opendevreview | Maor Blaustein proposed openstack/neutron-tempest-plugin master: Fix dependencies and PEP 8 new comments. https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/827292 | 14:01 |
| lajoskatona | #startmeeting networking | 14:07 |
| opendevmeet | Meeting started Tue Feb 1 14:07:02 2022 UTC and is due to finish in 60 minutes. The chair is lajoskatona. Information about MeetBot at http://wiki.debian.org/MeetBot. | 14:07 |
| opendevmeet | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 14:07 |
| opendevmeet | The meeting name has been set to 'networking' | 14:07 |
| lajoskatona | Hi, sorry for being late.... | 14:07 |
| mlavalle | o/ | 14:07 |
| ralonsoh | hi | 14:07 |
| * mlavalle will be multitasking with a company internal meeting | 14:07 | |
| slaweq | hi | 14:07 |
| rubasov | o/ | 14:07 |
| obondarev | hi | 14:08 |
| isabek | Hi | 14:08 |
| amotoki | hi | 14:08 |
| lajoskatona | ok, let's start | 14:08 |
| lajoskatona | #topic Announcements | 14:08 |
| lajoskatona | the usual Yoga cycle calendar: https://releases.openstack.org/yoga/schedule.html | 14:09 |
| lajoskatona | We are in week R-8, so I collected some patches that hang in the client libraries | 14:09 |
| lajoskatona | we are good actually as nothing in non-client libraries, like n-lib or ovsdbapp | 14:10 |
| lajoskatona | and we had recently release of them | 14:10 |
| lajoskatona | but we have few things in client facing libs | 14:10 |
| lajoskatona | ndp proxy: https://review.opendev.org/q/topic:bug%252F1877301 | 14:10 |
| lajoskatona | the series looks quite active, so if you have time please help with reviews :-) | 14:11 |
| lajoskatona | qos-minimum-guaranteed-packet-rate: https://review.opendev.org/q/topic:bp/qos-minimum-guaranteed-packet-rate+status:open | 14:11 |
| opendevreview | Fernando Royo proposed openstack/ovn-octavia-provider stable/victoria: Return UnsupportedOptionError() on loadbalancer failover https://review.opendev.org/c/openstack/ovn-octavia-provider/+/827296 | 14:12 |
| lajoskatona | I took over from przemek, and try to push python-openstackclient patch | 14:12 |
| lajoskatona | another one which is in quite good shape and has great progress: Local IP: https://review.opendev.org/q/topic:bug/1930200+status:open | 14:13 |
| lajoskatona | smart-NIC: https://review.opendev.org/q/topic:bug%252F1932154 | 14:14 |
| lajoskatona | and the last open topic which has client dependency: Check quota limits: https://review.opendev.org/c/openstack/python-openstackclient/+/806016 | 14:15 |
| lajoskatona | If I left out anything please tell me :-) | 14:15 |
| slaweq | lajoskatona: I will add them to my review list and will try to go through them this week | 14:16 |
| lajoskatona | If we have to push something to be in time we can increase the priority also | 14:16 |
| lajoskatona | slaweq: thanks, I try to do the same, and they are really complex and interesting features so hard work even to review them, but can learn a lot :-) | 14:17 |
| lajoskatona | do you have comments or questions for the announcements/ release schedule or to any of the above patch series? | 14:19 |
| ralonsoh | no thanks | 14:20 |
| lajoskatona | #topic Bugs | 14:20 |
| lajoskatona | I reschufled the bug deputy list: https://wiki.openstack.org/wiki/Network/Meetings#Bug_deputy | 14:21 |
| lajoskatona | please check it and if it is not ok for you tell me | 14:21 |
| lajoskatona | We have one report from obondarev: http://lists.openstack.org/pipermail/openstack-discuss/2022-January/026897.html | 14:22 |
| lajoskatona | obondarev do you have something to highlight? | 14:23 |
| obondarev | maybe just one Critical - https://bugs.launchpad.net/neutron/+bug/1958363 | 14:23 |
| obondarev | not sure what's the failure rate for this now | 14:24 |
| slaweq | I didn't saw it this week personally | 14:24 |
| lajoskatona | me neither, so perhaps not that frequent | 14:24 |
| slaweq | but basically it may be hard to track e.g. in the logstash | 14:25 |
| lajoskatona | sadly logstash is not that useful, at least for me this week... | 14:25 |
| slaweq | as the visible issue is that many tests fails with instances failed to build | 14:25 |
| slaweq | then you can check in neutron server logs about that disabled notification to nova in one of the workers | 14:26 |
| slaweq | that's the issue :) | 14:26 |
| lajoskatona | recently I started to use this tool from gibi: https://github.com/gibizer/zuul-log-search | 14:26 |
| lajoskatona | it can look into q-svc or any logs, but donwload them to locally, so you need net bandwidth and storage space | 14:27 |
| * gibi happy to take questions about that tool in any forms | 14:27 | |
| lajoskatona | gibi: thanks :-) | 14:27 |
| ralonsoh | gibi thanks! | 14:27 |
| frickler | there also the issue with system scope testing being broken after recent policy changes | 14:28 |
| lajoskatona | thanks slaweq for adding logs to this nova notifier issue | 14:28 |
| slaweq | yw. I really want to understand what happens there that this is disabled permanently for one of the workers | 14:29 |
| lajoskatona | frickler: you mean this one: https://bugs.launchpad.net/neutron/+bug/1959196 ? | 14:30 |
| lajoskatona | New Secure RBAC policies broke devstack-enforce-scope job | 14:30 |
| frickler | yes | 14:30 |
| frickler | would be good to include that job in neutron, too, I think, once it is repaired | 14:31 |
| slaweq | maybe we can include it in periodic queue of neutron? | 14:31 |
| frickler | good idea | 14:32 |
| ralonsoh | that would be better not to increase the check queue | 14:32 |
| lajoskatona | +1 | 14:32 |
| slaweq | I will add it to periodic queue | 14:33 |
| lajoskatona | slaweq: thanks, I can do it if you have no time this week | 14:33 |
| slaweq | lajoskatona: no worries, I will send patch right now | 14:34 |
| lajoskatona | :-) | 14:34 |
| lajoskatona | last week's report from isabek: http://lists.openstack.org/pipermail/openstack-discuss/2022-January/026972.html | 14:34 |
| lajoskatona | I see only a few unasigned bugs: | 14:35 |
| lajoskatona | https://bugs.launchpad.net/neutron/+bug/1959176 neutron fullstack/functional jobs TIMED_OUT randomly | 14:36 |
| opendevreview | Slawek Kaplonski proposed openstack/neutron master: Add devstack-enforce-scope job to our periodic queue https://review.opendev.org/c/openstack/neutron/+/827302 | 14:37 |
| lajoskatona | for this I have a question, see: #link https://bugs.launchpad.net/neutron/+bug/1959176/comments/2 | 14:37 |
| lajoskatona | currently OS_TEST_TIMEOUT is 600: https://opendev.org/openstack/neutron/src/branch/master/tox.ini#L73 | 14:37 |
| ralonsoh | Agree: this timeout is not needed anymore | 14:38 |
| ralonsoh | we solved the issues with SQL testing | 14:38 |
| lajoskatona | based on comment it was changed to 600 due to an old bug | 14:38 |
| lajoskatona | that was my understanding also, I will propose a patch to change that, thanks | 14:38 |
| slaweq | ++ | 14:38 |
| ykarel | +1 | 14:38 |
| lajoskatona | another unassigned from last week: https://bugs.launchpad.net/neutron/+bug/1959098 [ovn] metadata route missing on the guest | 14:39 |
| slaweq | I think that ykarel is on it | 14:39 |
| ralonsoh | still waiting for more info on this | 14:39 |
| slaweq | at least trying to get some more info | 14:39 |
| ralonsoh | this is very specific to a single network | 14:39 |
| ykarel | yes rodolfo and i requested some more info from reporter | 14:40 |
| lajoskatona | slaweq, ralonaoh, ykarel: thanks good to know | 14:40 |
| lajoskatona | And that's it from me for the bugs | 14:41 |
| lajoskatona | This week lajoskatona is the deputy, and next week bcafarel will be. | 14:41 |
| lajoskatona | I have to check with bcafarel | 14:41 |
| ralonsoh | he is aware | 14:41 |
| lajoskatona | ralonsoh: cool, thanks | 14:42 |
| lajoskatona | and actually that's all what I have for today | 14:42 |
| lajoskatona | Do you have anything to discuss? | 14:43 |
| ralonsoh | nothing from me, thanks | 14:43 |
| slaweq | not from me | 14:43 |
| ralonsoh | (today we don't have CI meeting) | 14:43 |
| obondarev | ack | 14:43 |
| slaweq | exactly :) | 14:43 |
| mlavalle | nothing from me either | 14:43 |
| ykarel | nothing from me too | 14:44 |
| lajoskatona | ok, than let's close the meeting, thanks for participating | 14:44 |
| lajoskatona | #endmeeting | 14:44 |
| opendevmeet | Meeting ended Tue Feb 1 14:44:34 2022 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 14:44 |
| opendevmeet | Minutes: https://meetings.opendev.org/meetings/networking/2022/networking.2022-02-01-14.07.html | 14:44 |
| opendevmeet | Minutes (text): https://meetings.opendev.org/meetings/networking/2022/networking.2022-02-01-14.07.txt | 14:44 |
| opendevmeet | Log: https://meetings.opendev.org/meetings/networking/2022/networking.2022-02-01-14.07.log.html | 14:44 |
| mlavalle | o/ | 14:44 |
| lajoskatona | o/ | 14:44 |
| ralonsoh | bye | 14:44 |
| slaweq | o/ | 14:44 |
| obondarev | o/ | 14:44 |
| isabek | bye | 14:44 |
| ykarel | bye | 14:44 |
| opendevreview | yatin proposed openstack/neutron-tempest-plugin master: Re-enable IPv6Test in OVN scenario job https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/827303 | 14:44 |
| amotoki | o/ | 14:44 |
| frickler | is there a CI bug for ovn failing to start in devstack like here? https://ef7f9dfdac3d80d7fe06-334c29a1db3b5ea30b4f88988d44f729.ssl.cf1.rackcdn.com/827155/4/check/devstack-ipv6/48cd888/job-output.txt | 14:47 |
| ralonsoh | frickler, if I'm not wrong, this is because OVN or OVS, one was installed from source and the other one not | 14:48 |
| ralonsoh | and the directories and different | 14:49 |
| ralonsoh | but let me check | 14:49 |
| frickler | ralonsoh: it seems to be some kind of race condition, only happens like an estimated 10% of all times, I usually just recheck and it is gone | 14:49 |
| ralonsoh | oh, that's something different then | 14:50 |
| ralonsoh | I'll open a bug then | 14:50 |
| ralonsoh | that's happening in devstack-ipv6, right? | 14:50 |
| frickler | ralonsoh: no, it happens for all devstack jobs that deploy OVN, that is just the latest example I had at hand | 14:50 |
| ralonsoh | ok | 14:51 |
| frickler | ykarel: nice coincidence, I was just looking at that IPv6 bug earlier today and wondered if it was still present | 14:51 |
| ykarel | frickler, yeah slawek asked me to check that today | 14:52 |
| seba | Should a user be able to allocate a floatingip outside of the fip-network's allocation pool? Especially for ips like the gateway_ip of the subnet? | 15:02 |
| seba | or to say it in code: I would have expected this test to fail: https://paste.debian.net/hidden/de2e633c/ | 15:18 |
| seba | it seems like the only validation on a user specified floating-ip-address is done by the ipam module, i.e. "is the ip in use". as the gateways on our external networks don't have ports it seems that ipam deems them as being free. | 15:19 |
| seba | so, is this a bug/problem for others? If so I'd create a proper bug report for this. Also, if this is a bug, I'd be curious if we should restrict fip allocation to the subnets allocation pools or just deny using the gateway ip. | 15:21 |
| ralonsoh | seba, a FIP should belong to the range of the subnet of the external network | 15:22 |
| seba | yes. in my example this would be 11.0.0.0/24 with 11.0.0.1 as gw and an allocation pool 11.0.0.2-11.0.0.254. 11.0.0.1 is inside the range of the subnet, but in my case it is a bad idea for the user to be able to allocate it | 15:24 |
| opendevreview | Maor Blaustein proposed openstack/neutron-tempest-plugin master: Fix dependencies and PEP 8 new comments. https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/827292 | 15:26 |
| ralonsoh | seba, let me check that case | 15:26 |
| opendevreview | Frode Nordahl proposed openstack/neutron master: [OVN] Off-path SmartNIC DPU Port Binding with OVN https://review.opendev.org/c/openstack/neutron/+/808961 | 15:36 |
| ralonsoh | seba, right, we don't check if this is the GW IP. This is because we don't assign this IP address to any Neutron port | 15:41 |
| seba | yeah exactly | 15:43 |
| seba | I don't think this gateway ip on an external network should be allocatable for a FIP. Maybe not even for an internal OpenStack port? | 15:44 |
| seba | at least it could make sense to do a check on create_floatingip() to check if the user is currently trying to allocate the external subnet's gateway ip and throw an error if they do | 15:45 |
| ralonsoh | seba, why not? the GW could be another port | 15:45 |
| ralonsoh | and you can redirect the traffic from there | 15:45 |
| ralonsoh | you can build your own GW VM | 15:45 |
| seba | that's right, but does that also work for external networks? | 15:46 |
| ralonsoh | why not? | 15:46 |
| seba | I thought external also means "router external" i.e. "gateway not managed by openstack" | 15:46 |
| seba | (external as in "external networks") | 15:46 |
| ralonsoh | I wouldn't limit that but add a warning message in the logs | 15:47 |
| ralonsoh | that's all | 15:47 |
| ralonsoh | but, of course, that could be considered | 15:47 |
| ralonsoh | please, open a bug to discuss this | 15:47 |
| ralonsoh | you can also propose a patch to be reviewed | 15:47 |
| seba | okay, will do | 15:48 |
| opendevreview | Oleg Bondarev proposed openstack/neutron master: Make sure "dead vlan" ports cannot transmit packets https://review.opendev.org/c/openstack/neutron/+/827315 | 16:09 |
| opendevreview | Fernando Royo proposed openstack/ovn-octavia-provider stable/ussuri: Return UnsupportedOptionError() on loadbalancer failover https://review.opendev.org/c/openstack/ovn-octavia-provider/+/827317 | 16:21 |
| opendevreview | Fernando Royo proposed openstack/ovn-octavia-provider stable/ussuri: Return UnsupportedOptionError() on loadbalancer failover https://review.opendev.org/c/openstack/ovn-octavia-provider/+/827317 | 16:22 |
| opendevreview | Fernando Royo proposed openstack/ovn-octavia-provider stable/ussuri: Return UnsupportedOptionError() on loadbalancer failover https://review.opendev.org/c/openstack/ovn-octavia-provider/+/827324 | 16:56 |
| opendevreview | Frode Nordahl proposed openstack/neutron master: [OVN] Off-path SmartNIC DPU Port Binding with OVN https://review.opendev.org/c/openstack/neutron/+/808961 | 17:05 |
| seba | bug created: https://bugs.launchpad.net/neutron/+bug/1959699 | 17:37 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: Use a thread local variable to store the Nova Notifier enable flag https://review.opendev.org/c/openstack/neutron/+/827331 | 17:47 |
| opendevreview | Frode Nordahl proposed openstack/neutron master: [OVN] Off-path SmartNIC DPU Port Binding with OVN https://review.opendev.org/c/openstack/neutron/+/808961 | 18:15 |
| opendevreview | Fernando Royo proposed openstack/networking-ovn stable/train: Return UnsupportedOptionError() on loadbalancer failover https://review.opendev.org/c/openstack/networking-ovn/+/827332 | 18:17 |
| opendevreview | Oleg Bondarev proposed openstack/neutron master: Make sure "dead vlan" ports cannot transmit packets https://review.opendev.org/c/openstack/neutron/+/827315 | 18:56 |
| opendevreview | Lajos Katona proposed openstack/tap-as-a-service master: Do not try to call status setting methods in case of periodic task https://review.opendev.org/c/openstack/tap-as-a-service/+/825299 | 19:41 |
| opendevreview | Lajos Katona proposed openstack/tap-as-a-service master: WIP: Make ovs-taas start in VLAN only env https://review.opendev.org/c/openstack/tap-as-a-service/+/817449 | 19:41 |
| *** dasm|rover is now known as dasm|off | 22:23 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!