Friday, 2024-10-04

opendevreviewRico Lin proposed openstack/ovn-octavia-provider master: fix: specify endpoint INFO for neutron client  https://review.opendev.org/c/openstack/ovn-octavia-provider/+/93114103:37
opendevreviewRodolfo Alonso proposed openstack/neutron master: Remove the sleep calls in the ``create_or_update_agent`` method  https://review.opendev.org/c/openstack/neutron/+/93124906:55
opendevreviewRodolfo Alonso proposed openstack/neutron master: WIP - Replace ``greenthread.sleep`` with ``time.sleep``  https://review.opendev.org/c/openstack/neutron/+/93125106:56
opendevreviewRodolfo Alonso proposed openstack/neutron master: Replace ``greenthread.sleep`` with ``time.sleep``  https://review.opendev.org/c/openstack/neutron/+/93125106:56
opendevreviewRodolfo Alonso proposed openstack/neutron master: WIP - DNM - Test ML2/OVN plugin without eventlet  https://review.opendev.org/c/openstack/neutron/+/93125207:18
slaweqhaleyb hi, I proposed new neutron-lib release, first for Expoxy: https://review.opendev.org/c/openstack/releases/+/93136607:45
ralonsohslaweq, do you have PTL permissions??08:02
ralonsohhehehe the patch is ptl-approved now08:02
slaweqralonsoh yes, because I am release liaison08:02
ralonsohahhh right!08:02
slaweqso in terms of the releases I have same "power" as haleyb 08:02
ralonsohslaweq, can we have https://review.opendev.org/c/openstack/neutron-lib/+/927820 too?08:03
ralonsohif we merge it today08:03
slaweqralonsoh sure, we can wait even more than just today08:11
ralonsohslaweq, thanks!08:11
slaweqactually now it still needs to wait for the Epoxy placeholder to be merged first08:11
ralonsohyeah, I saw it08:11
slaweqand it it not urgent for me at all :)08:11
slaweqI just wanted to have it there to not forget and to be able to finish some leftovers which I have in my personal backlog but that can definitely wait few more days08:12
opendevreviewRodolfo Alonso proposed openstack/neutron master: Ensure the OVSDB and ovn-northd process stop  https://review.opendev.org/c/openstack/neutron/+/93101508:34
ralonsohlajoskatona, ^ this is related to the FT issue08:34
ralonsohif you have some minutes, please check it08:34
opendevreviewMerged openstack/ovn-bgp-agent master: Make in_port consistent type  https://review.opendev.org/c/openstack/ovn-bgp-agent/+/93119108:41
opendevreviewRodolfo Alonso proposed openstack/neutron master: DNM: test functional jobs  https://review.opendev.org/c/openstack/neutron/+/93101608:48
opendevreviewMerged openstack/neutron master: mypy: Use tuple type for SUPPORTED_LOGGING_TYPES  https://review.opendev.org/c/openstack/neutron/+/92986508:55
lajoskatonaralonsoh: ack, yesterday I seen it as I remember and run through it09:05
ralonsohcool09:05
opendevreviewMichel Nederlof proposed openstack/ovn-bgp-agent master: Fix race condition where an event could unexpose a network during sync  https://review.opendev.org/c/openstack/ovn-bgp-agent/+/92779109:22
andreykurilinHi folks! Can anyone suggest the direction for debugging security groups list API slowness? We have an OpenStack 2024.1 cluster. ~9k VMs , 6.4k Security Groups (122k rules in total). I'm receiving Timout error from Nova server list API (API call takes more than 60s to retrieve detailed view of 1k servers). Further debugging showed that Nova retrieves security groups from Neutron and it takes 90+% of user request duration. 09:54
andreykurilinNova makes multiple calls, grouping servers into 150-item chunks. The single call to Neutron looks simple - get security groups with passing 150 IDs and `fields=['id', 'name']` filter. It takes ~9s to complete. In comparison, retrieving the same amount of ports without any filters takes <0.4s. Our version of neutron API includes https://review.opendev.org/c/openstack/neutron/+/929967, which improved the case but has not 09:54
andreykurilinresolved it.09:54
ralonsohandreykurilin, please report a LP bug with the information of (1) the Nova query (what is the current API call to Neutron) and (2) your query10:02
opendevreviewMerged openstack/neutron-lib master: Deprecate get_notifier(service=None)  https://review.opendev.org/c/openstack/neutron-lib/+/91139210:18
andreykurilinralonsoh: https://bugs.launchpad.net/neutron/+bug/2083682 done.10:23
andreykurilinralonsoh: should I close a new bug report in favor of closed https://bugs.launchpad.net/neutron/+bug/2081087 ?10:36
ralonsohandreykurilin, why?10:37
andreykurilin2081087 is also about slowness of security groups api :)10:40
ralonsohyeah but is solved, at least reduced10:41
ralonsohyour case presents another use case10:41
opendevreviewLajos Katona proposed openstack/neutron-tempest-plugin master: Check if previous key is present for pagination  https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/93087710:43
andreykurilinOk, thank you10:45
slaweqykarel lajoskatona hi, can you take a look at https://review.opendev.org/c/openstack/neutron-lib/+/927820 ? Thx in advance10:58
opendevreviewRodolfo Alonso proposed openstack/neutron-tempest-plugin master: [WSGI] Move all OVN jobs to use WSGI API module  https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/93074311:11
noonedeadpunkhey folks! I'm trying to investigate here a bit around https://bugs.launchpad.net/neutron/+bug/2083527 11:44
noonedeadpunkand was looking for some confirmation if I understand things correctly and my assumptions are right11:44
noonedeadpunkthat for nested router NAT rule should not have a GATEWAY_PORT defined in OVN, as LRP is not pinned to any chassis11:45
noonedeadpunkthough LR is instead11:45
opendevreviewTakashi Natsume proposed openstack/os-vif master: Remove Python 3.8 support  https://review.opendev.org/c/openstack/os-vif/+/93148012:00
opendevreviewTakashi Natsume proposed openstack/os-vif master: Remove Python 3.8 support  https://review.opendev.org/c/openstack/os-vif/+/93148012:02
opendevreviewMichel Nederlof proposed openstack/ovn-bgp-agent master: Fix cleanup of rules per evpn device  https://review.opendev.org/c/openstack/ovn-bgp-agent/+/92781612:10
lajoskatonaslaweq: Hi, with your TC hat on do you expect VMWare related community discussions? I mean like some coordinated community effort covering questions, requests from ex-vmware users?12:12
sean-k-mooneyslaweq: https://review.opendev.org/c/openstack/neutron/+/899956 the expection is that each fo the core teams will implemtn that them selves 12:41
sean-k-mooneyslaweq: that is neede to supprot centos 10 stream or installing devstack in a global venv on c9s12:42
slaweqsean-k-mooney I abandoned it automatically with script which cleans some old patches, but if this is needed, I just restored it for you12:48
slaweqsorry :)12:48
slaweqlajoskatona I don't know about any effort like that so far12:48
slaweqI guess it can be addressed through the ML/IRC on demand, no?12:49
slaweqwhat do you mean by the coordinated effort exactly? Some WG for that or what?12:49
sean-k-mooneyso we likely shoud have an offical popup team12:50
sean-k-mooneythat work across projects12:50
sean-k-mooneyfor vmware related things upstream12:50
sean-k-mooneya ptg sessionon that would be good too12:50
sean-k-mooneybut i dont know if that is being cordinatied by any one specificly12:51
lajoskatonaslaweq: not sure about at the moment for WG, but perhaps to have some more coordinated response to such needs or wider door for incoming requests :-) 12:51
sean-k-mooneyost of the vmware topic so far were at teh board level not tc or in the comunity that im aware of12:51
lajoskatonasean-k-mooney, slaweq: thanks, I just checked some of the etherpads already there for the PTG and at this time can't see any mention of this on them so I thought I ask around for it12:52
opendevreviewJakub Libosvar proposed openstack/ovn-bgp-agent master: Introduce LSP address column parsing functions  https://review.opendev.org/c/openstack/ovn-bgp-agent/+/92580113:10
haleybslaweq: hi, so should we -1 the neutron-lib release change until the other change merges?13:17
haleybi -1'd it13:22
slaweqhaleyb yes, please13:23
lajoskatonahaleyb:Hi,  do we have drivers meeting today? I am in the office and have to commute home, so I have to miss it if we have :-(13:23
opendevreviewJakub Libosvar proposed openstack/neutron master: Set distributed flag to NB_Global  https://review.opendev.org/c/openstack/neutron/+/93106713:24
opendevreviewJakub Libosvar proposed openstack/neutron master: Set distributed flag to NB_Global  https://review.opendev.org/c/openstack/neutron/+/93106713:25
haleyblajoskatona: there was one RFE to discuss so was going to have a meeting. Don't know if we'll have quorom or the submittor, but no worries13:26
lajoskatonahaleyb: ack, thanks13:29
ykarelhaleyb, i will also not be able to join today13:48
haleybykarel: thanks for letting me know, will just move to next week if we don't have quorom13:51
opendevreviewMichel Nederlof proposed openstack/ovn-bgp-agent master: Fix cleanup of rules per evpn device  https://review.opendev.org/c/openstack/ovn-bgp-agent/+/92781613:57
opendevreviewMichel Nederlof proposed openstack/ovn-bgp-agent master: Fix cleanup of rules per evpn device  https://review.opendev.org/c/openstack/ovn-bgp-agent/+/92781614:00
haleyb#startmeeting neutron_drivers14:01
opendevmeetMeeting started Fri Oct  4 14:01:22 2024 UTC and is due to finish in 60 minutes.  The chair is haleyb. Information about MeetBot at http://wiki.debian.org/MeetBot.14:01
opendevmeetUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.14:01
opendevmeetThe meeting name has been set to 'neutron_drivers'14:01
mlavalle\o14:01
haleybnot sure if we'll have quorum14:01
cbuggyo/14:02
haleybPing list: ykarel, mlavalle, mtomaska, slaweq, obondarev, tobias-urdin, lajoskatona, amotoki, haleyb, ralonsoh14:02
mlavalle\o14:02
haleybmlavalle: hi again :)14:02
mlavallelol14:02
slaweqo/14:02
haleybso 3 does not make quorum, might have to just do this next week14:05
mlavalleack14:05
haleybif anyone is bored the agenda was only one item14:06
haleyb#link https://bugs.launchpad.net/neutron/+bug/208321414:06
haleybbut we will discuss next week14:06
haleybthanks for coming, will give you :54 back to fix our bugz :)14:07
mlavallewill take a look ahead of time so I'm ready for next Friday14:07
haleyb#endmeeting14:07
opendevmeetMeeting ended Fri Oct  4 14:07:22 2024 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)14:07
opendevmeetMinutes:        https://meetings.opendev.org/meetings/neutron_drivers/2024/neutron_drivers.2024-10-04-14.01.html14:07
opendevmeetMinutes (text): https://meetings.opendev.org/meetings/neutron_drivers/2024/neutron_drivers.2024-10-04-14.01.txt14:07
opendevmeetLog:            https://meetings.opendev.org/meetings/neutron_drivers/2024/neutron_drivers.2024-10-04-14.01.log.html14:07
vprokofevThis one's mine and I'm here btw it you have any questions.14:07
mlavallehave a nive weekend14:07
vprokofevOk, next week I guess 14:07
slaweqhave a good weekend14:08
mlavallethanks vprokofev. will reach out to you in case I have any questions14:08
haleybvprokofev: thanks for attending, we just didn't have enough people to discuss/vote14:08
vprokofevNo worries, there's always next week :)14:09
mlavallearen't we lucky?14:09
ralonsohhaleyb, I thought there were no CI meeting14:18
haleybralonsoh: i had not canceled drivers meeting, but we did not have quorum either way14:26
ralonsohsorry, I understood that from the IRC log14:26
ralonsoh(and I went to prepare a coffee)14:27
haleybnp14:27
pabloclsnHi guys, I'm using OVN on the master branch on a brand new openstack cluster I want to achieve a form of routing with two virtual router in two separated openstack cluster connected with a network type VLAN. I can see that the router on connected to the same VLAN, because I can see arp flowing from A to B. But the problem is that a router seems to never learn the ARP/IP of the other router. The port security is disab14:30
ralonsohpabloclsn, how are the routers configured? I guess you are defining this VLAN network as external15:10
opendevreviewBrian Haley proposed openstack/neutron master: Optionally configure IPv6 metadata address  https://review.opendev.org/c/openstack/neutron/+/92649715:17
opendevreviewDmitriy Rabotyagov proposed openstack/neutron master: [OVN] Do not supply gateway_port if it's not binded to chassis  https://review.opendev.org/c/openstack/neutron/+/93149515:17
noonedeadpunkralonsoh: do I missunedrstand complexity of the whole thing re ^ ?15:18
ralonsohnoonedeadpunk, I would need to review the code and test it. In any case, your user case is, by default, not tested15:20
ralonsohthis deserves at least a neutron-tempest-plugin test15:20
noonedeadpunkI think I can work on this one15:21
noonedeadpunkas well as some unit/functional tests15:22
noonedeadpunkand yeah, for sure I get that it's smth not tested - otherwise it won't appear "broken". we're just assesing migration from OVS and that appeared as a blocker for some of our workloads15:24
opendevreviewLajos Katona proposed openstack/neutron-fwaas master: Update jobs based on testing runtime for 2025.1  https://review.opendev.org/c/openstack/neutron-fwaas/+/93150215:52
opendevreviewMerged openstack/neutron-lib master: Add new API extension ``uplink-status-propagation-updatable``  https://review.opendev.org/c/openstack/neutron-lib/+/92782015:53
opendevreviewLajos Katona proposed openstack/neutron-dynamic-routing master: Update jobs based on testing runtime for 2025.1  https://review.opendev.org/c/openstack/neutron-dynamic-routing/+/93150315:56
opendevreviewLajos Katona proposed openstack/neutron-vpnaas master: Update jobs based on testing runtime for 2025.1  https://review.opendev.org/c/openstack/neutron-vpnaas/+/93150415:59
opendevreviewLajos Katona proposed openstack/neutron-tempest-plugin master: Update python version in setup.cfg  https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/93150616:03
opendevreviewLajos Katona proposed openstack/neutron-lib master: Update python version in setup.cfg  https://review.opendev.org/c/openstack/neutron-lib/+/93150716:04
haleyblajoskatona: thanks for those changes ^^ looks like neutron setup.cfg needs a change as well :)16:20
haleybslaweq: can you update the release change for neutron-lib? changes have merged16:23
opendevreviewMerged openstack/neutron master: Get rid of pkg_resources  https://review.opendev.org/c/openstack/neutron/+/93127116:40
opendevreviewJakub Libosvar proposed openstack/neutron master: Use set_nb_global_options in maintenance task  https://review.opendev.org/c/openstack/neutron/+/93106817:29
opendevreviewTakashi Kajinami proposed openstack/neutron master: Drop usage of importlib_metadata  https://review.opendev.org/c/openstack/neutron/+/93151517:57
tkajinam^^^ fungi this might be a better approach17:58
tkajinam(still running local tests17:58
fungilooking17:59
fungialso, for the record, i have nothing against importing a third-party compatibility module on older interpreters in order to approximate newer ones17:59
fungiit's why they exist18:00
tkajinamI agree18:01
fungitkajinam: but yeah, 931515 is closer to what we've done in other repos, though i think you have a mismatch in underscores between the two?18:02
tkajinamyes... I'm fixing that and minor pep8 error18:02
tkajinamI pushed that too early before the tests complete...18:02
fungicool. and i don't mean to be pedantic, i've just run into a lot of mistakes in the past stemming from assumptions about what a specific version number means. versions can lie18:03
tkajinamyeah18:04
tkajinamone good example was FIPS support18:04
fungiversions are useful clues for humans, but not reliable indicators for code18:04
opendevreviewFernando Royo proposed openstack/ovn-octavia-provider master: Add LB sync logic  https://review.opendev.org/c/openstack/ovn-octavia-provider/+/92532418:06
opendevreviewFernando Royo proposed openstack/ovn-octavia-provider master: Add Listener sync logic  https://review.opendev.org/c/openstack/ovn-octavia-provider/+/93125018:06
opendevreviewFernando Royo proposed openstack/ovn-octavia-provider master: Add Pool sync logic  https://review.opendev.org/c/openstack/ovn-octavia-provider/+/93126618:06
opendevreviewFernando Royo proposed openstack/ovn-octavia-provider master: Add Member sync logic  https://review.opendev.org/c/openstack/ovn-octavia-provider/+/93126718:06
opendevreviewFernando Royo proposed openstack/ovn-octavia-provider master: Add Health monitor sync logic  https://review.opendev.org/c/openstack/ovn-octavia-provider/+/93128818:06
opendevreviewFernando Royo proposed openstack/ovn-octavia-provider master: Add sync floating IP support  https://review.opendev.org/c/openstack/ovn-octavia-provider/+/92903918:06
opendevreviewJakub Libosvar proposed openstack/neutron master: Set distributed flag to NB_Global  https://review.opendev.org/c/openstack/neutron/+/93106718:06
opendevreviewFernando Royo proposed openstack/ovn-octavia-provider master: Add LB sync logic  https://review.opendev.org/c/openstack/ovn-octavia-provider/+/92532418:27
opendevreviewFernando Royo proposed openstack/ovn-octavia-provider master: Add Listener sync logic  https://review.opendev.org/c/openstack/ovn-octavia-provider/+/93125018:27
opendevreviewFernando Royo proposed openstack/ovn-octavia-provider master: Add Pool sync logic  https://review.opendev.org/c/openstack/ovn-octavia-provider/+/93126618:27
opendevreviewFernando Royo proposed openstack/ovn-octavia-provider master: Add Member sync logic  https://review.opendev.org/c/openstack/ovn-octavia-provider/+/93126718:27
opendevreviewFernando Royo proposed openstack/ovn-octavia-provider master: Add Health monitor sync logic  https://review.opendev.org/c/openstack/ovn-octavia-provider/+/93128818:27
opendevreviewFernando Royo proposed openstack/ovn-octavia-provider master: Add sync floating IP support  https://review.opendev.org/c/openstack/ovn-octavia-provider/+/92903918:27
opendevreviewJakub Libosvar proposed openstack/ovn-bgp-agent master: Expose FIP if external_mac was set  https://review.opendev.org/c/openstack/ovn-bgp-agent/+/93078418:37
opendevreviewFernando Royo proposed openstack/ovn-octavia-provider master: Add LB sync logic  https://review.opendev.org/c/openstack/ovn-octavia-provider/+/92532418:58
opendevreviewFernando Royo proposed openstack/ovn-octavia-provider master: Add Listener sync logic  https://review.opendev.org/c/openstack/ovn-octavia-provider/+/93125018:58
opendevreviewFernando Royo proposed openstack/ovn-octavia-provider master: Add Pool sync logic  https://review.opendev.org/c/openstack/ovn-octavia-provider/+/93126618:58
opendevreviewFernando Royo proposed openstack/ovn-octavia-provider master: Add Member sync logic  https://review.opendev.org/c/openstack/ovn-octavia-provider/+/93126718:58
opendevreviewFernando Royo proposed openstack/ovn-octavia-provider master: Add Health monitor sync logic  https://review.opendev.org/c/openstack/ovn-octavia-provider/+/93128818:58
opendevreviewFernando Royo proposed openstack/ovn-octavia-provider master: Add sync floating IP support  https://review.opendev.org/c/openstack/ovn-octavia-provider/+/92903918:58
opendevreviewDmitriy Rabotyagov proposed openstack/ovn-bgp-agent master: Handle trimming of vlan interface namings  https://review.opendev.org/c/openstack/ovn-bgp-agent/+/91065421:57
opendevreviewDmitriy Rabotyagov proposed openstack/ovn-bgp-agent master: Handle trimming of vlan interface namings  https://review.opendev.org/c/openstack/ovn-bgp-agent/+/91065421:58
opendevreviewTakashi Kajinami proposed openstack/neutron master: Drop usage of importlib_metadata  https://review.opendev.org/c/openstack/neutron/+/93151522:09
opendevreviewIhar Hrachyshka proposed openstack/neutron master: refactor: remove do_notify argument for disassociate_floatingips  https://review.opendev.org/c/openstack/neutron/+/93153223:21
opendevreviewIhar Hrachyshka proposed openstack/neutron master: WIP refactor: replace metaclass=ABCMeta with base ABC class  https://review.opendev.org/c/openstack/neutron/+/93153323:50
opendevreviewIhar Hrachyshka proposed openstack/neutron master: WIP: refactor: Don't inherit from object  https://review.opendev.org/c/openstack/neutron/+/93153423:50

Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!