Tuesday, 2020-04-07

*** mlavalle has quit IRC00:09
*** yamamoto has joined #openstack-neutron00:31
*** ociuhandu has joined #openstack-neutron01:04
*** happyhemant has quit IRC01:08
*** ociuhandu has quit IRC01:09
*** zhanglong has joined #openstack-neutron01:12
*** tbachman has quit IRC01:24
openstackgerritYi Yang proposed openstack/neutron master: Enable egress direct flow for iptables_hybrid  https://review.opendev.org/71264001:29
*** tbachman has joined #openstack-neutron01:30
*** yamamoto has quit IRC01:33
*** jamesdenton_wfh has joined #openstack-neutron01:37
*** tbachman has quit IRC01:40
*** tbachman has joined #openstack-neutron01:41
*** yamamoto has joined #openstack-neutron02:29
*** igordc has joined #openstack-neutron02:33
*** ociuhandu has joined #openstack-neutron02:34
*** jamesdenton_wfh has quit IRC02:37
*** ociuhandu has quit IRC02:39
*** dayou has quit IRC02:50
*** dayou has joined #openstack-neutron02:52
*** igordc has quit IRC03:16
*** spatel has joined #openstack-neutron03:24
*** spatel has quit IRC03:25
*** lajoskatona has joined #openstack-neutron04:15
*** ratailor has joined #openstack-neutron05:00
openstackgerritLajos Katona proposed openstack/networking-bagpipe master: DNM: Fix lower-constraints tox target  https://review.opendev.org/70290005:26
*** gthiemonge has quit IRC05:46
*** gthiemonge has joined #openstack-neutron05:47
*** ralonsoh has joined #openstack-neutron06:02
*** armax_ has joined #openstack-neutron06:05
*** armax has quit IRC06:06
*** armax_ is now known as armax06:06
*** ircuser-1 has quit IRC06:07
*** jlibosva has joined #openstack-neutron06:16
*** gcheresh has joined #openstack-neutron06:17
*** ratailor has quit IRC06:40
openstackgerritYi Yang proposed openstack/neutron master: Enable egress direct flow for iptables_hybrid  https://review.opendev.org/71264006:42
*** ratailor has joined #openstack-neutron06:47
*** rpittau|afk is now known as rpittau06:56
*** zhanglong has quit IRC06:58
*** slaweq has joined #openstack-neutron07:00
*** ccamposr__ has joined #openstack-neutron07:06
*** hjensas has quit IRC07:07
*** ccamposr has quit IRC07:09
*** tmorin has joined #openstack-neutron07:13
*** tesseract has joined #openstack-neutron07:16
openstackgerritMaciej Józefczyk proposed openstack/neutron master: [OVN] Enable qos service plugin in tempest jobs  https://review.opendev.org/70483307:19
*** hjensas has joined #openstack-neutron07:21
openstackgerritLajos Katona proposed openstack/networking-bagpipe master: Fix fullstack tests  https://review.opendev.org/70289507:25
*** gcheresh has quit IRC07:28
openstackgerritJakub Libosvar proposed openstack/networking-ovn stable/train: [ovn] Stricter matching on metadata port binding event  https://review.opendev.org/71448407:31
openstackgerritRadosław Piliszek proposed openstack/neutron master: Make OVN driver validate Geneve max_header_size  https://review.opendev.org/71471607:33
*** yamamoto has quit IRC07:36
*** ociuhandu has joined #openstack-neutron07:45
openstackgerritLajos Katona proposed openstack/networking-bagpipe master: DNM: Fix lower-constraints tox target  https://review.opendev.org/70290007:47
*** yamamoto has joined #openstack-neutron07:59
*** hjensas is now known as hjensas|afk08:04
*** ralonsoh has quit IRC08:05
openstackgerritRadosław Piliszek proposed openstack/neutron master: Make OVN driver validate Geneve max_header_size  https://review.opendev.org/71471608:12
*** gcheresh has joined #openstack-neutron08:13
*** tkajinam has quit IRC08:17
*** ralonsoh has joined #openstack-neutron08:17
*** lucasagomes has joined #openstack-neutron08:22
*** zhanglong has joined #openstack-neutron08:25
*** gcheresh has quit IRC08:34
*** gcheresh has joined #openstack-neutron08:40
*** ociuhandu has quit IRC08:50
*** ociuhandu has joined #openstack-neutron08:57
*** dtantsur|afk is now known as dtantsur09:01
*** priteau has joined #openstack-neutron09:01
*** zhanglong has quit IRC09:03
*** zhanglong has joined #openstack-neutron09:06
openstackgerritIvan Kolodyazhny proposed openstack/neutron-fwaas-dashboard master: flake8: Ensure local check in hacking 2.0.0  https://review.opendev.org/71800009:21
openstackgerritIvan Kolodyazhny proposed openstack/neutron-vpnaas-dashboard master: flake8: Ensure local check in hacking 2.0.0  https://review.opendev.org/71800109:23
lajoskatonabcafarel: Hi, I've seen your bug for tempest on stable/stein09:28
*** gcheresh has quit IRC09:34
lajoskatonabcafarel: Is it possible to cap tempest on stable branches to specific tempest tag?09:35
*** zhanglong has quit IRC09:35
lajoskatonabcafarel: I remember some discussion about it or patch, but can't find....09:35
*** zhanglong has joined #openstack-neutron09:38
bcafarellajoskatona: yes we probably can do it (setting a new job based on tempest-full and version), though I hope fix can be done in main job09:47
bcafarelrocky version does not have the issue and runs fine09:47
*** ccamposr has joined #openstack-neutron09:47
bcafarel(with same nodes I think and same job description)09:49
bcafarellajoskatona: that's why I also added tempest as affected project :)09:49
bcafarelgmann: btw https://bugs.launchpad.net/tempest/+bug/1871327 is the one we are talking about ^ (and mentioned yesterday)09:50
openstackLaunchpad bug 1871327 in tempest "stable/stein tempest-full job fails with "tempest requires Python '>=3.6' but the running Python is 2.7.17"" [Undecided,New]09:50
*** ccamposr__ has quit IRC09:50
lajoskatonabcafarel: ok,10:06
*** zhanglong has quit IRC10:08
openstackgerritLucas Alvares Gomes proposed openstack/neutron master: [OVN] Add IGMP snooping configuration guide  https://review.opendev.org/71777810:10
*** rpittau is now known as rpittau|bbl10:25
*** yamamoto has quit IRC11:02
*** yamamoto has joined #openstack-neutron11:04
openstackgerritMaciej Józefczyk proposed openstack/neutron master: [OVN] Bump up transaction timeout for functional tests  https://review.opendev.org/71770411:05
*** gcheresh has joined #openstack-neutron11:07
*** njohnston has quit IRC11:18
*** yamamoto has quit IRC11:23
*** jamesdenton_wfh has joined #openstack-neutron11:33
*** yamamoto has joined #openstack-neutron11:35
*** yamamoto has quit IRC11:40
*** ociuhandu has quit IRC11:41
*** ociuhandu has joined #openstack-neutron11:42
*** ociuhandu has quit IRC11:47
*** nweinber has joined #openstack-neutron11:47
*** tmorin has quit IRC11:51
*** tmorin has joined #openstack-neutron12:06
*** rpittau|bbl is now known as rpittau12:06
*** tkajinam has joined #openstack-neutron12:08
*** hjensas|afk is now known as hjensas12:10
openstackgerritLajos Katona proposed openstack/networking-bagpipe master: DNM: Fix lower-constraints tox target  https://review.opendev.org/70290012:13
*** beagles_afk is now known as beagles12:17
*** gcheresh has quit IRC12:19
*** igordc has joined #openstack-neutron12:34
*** ociuhandu has joined #openstack-neutron12:45
*** ociuhandu has quit IRC12:52
openstackgerritMerged openstack/neutron-tempest-plugin master: Added new API test_sg_rules_quota_values.  https://review.opendev.org/71385012:58
*** njohnston has joined #openstack-neutron12:59
*** ociuhandu has joined #openstack-neutron12:59
*** ociuhandu has quit IRC13:00
*** ociuhandu has joined #openstack-neutron13:01
openstackgerritLajos Katona proposed openstack/networking-bgpvpn master: Switch to hacking 2.0  https://review.opendev.org/71044713:07
openstackgerritLajos Katona proposed openstack/networking-bgpvpn master: Switch to hacking 3.0.0  https://review.opendev.org/71044713:07
*** Luzi has joined #openstack-neutron13:09
*** ratailor has quit IRC13:11
lajoskatonaslaweq: Hi, regarding https://review.opendev.org/70289513:28
gmannbcafarel: sorry, did not get chance to look due to m-3 things to finish. I will check that13:29
lajoskatonaslaweq: now the agents for bagpipe started to run, and now api_paste is in wrong place based on logs: https://aceef6c1526cbdb83fa6-ecaf6f24685cfabe89c3e9d8fac0dfc4.ssl.cf5.rackcdn.com/702895/16/check/networking-bagpipe-dsvm-fullstack/cfc2ec7/testr_results.html13:29
lajoskatonaslaweq: so for me it seems that this one https://review.opendev.org/712702 is needed13:30
lajoskatonaslaweq: I don't understand fully why now the runs are different then previously, as the difference is that some kernel modules are there....13:31
slaweqlajoskatona: ok, so let's restore it :)13:31
lajoskatonaslaweq: ok13:32
*** ociuhandu has quit IRC13:32
*** ociuhandu has joined #openstack-neutron13:33
*** gcheresh has joined #openstack-neutron13:41
bcafarelgmann: no problem, I looked on my side and not sure what can cause the difference - but at least this is summarized in LP now for better tracking13:48
gmannok.13:50
*** gcheresh has quit IRC13:53
*** mlavalle has joined #openstack-neutron13:58
*** dave-mccowan has joined #openstack-neutron13:59
*** tkajinam has quit IRC14:03
*** tkajinam has joined #openstack-neutron14:03
*** Luzi has quit IRC14:07
*** dave-mccowan has quit IRC14:10
*** gcheresh has joined #openstack-neutron14:14
*** dayou has quit IRC14:20
*** dayou has joined #openstack-neutron14:21
ralonsohmlavalle, https://review.opendev.org/#/c/432260/ for example, based on SDK patch https://review.opendev.org/#/c/430363/14:21
mlavalleralonsoh: gracias caballero14:22
ralonsohheheheh14:22
ralonsohmlavalle, btw, I can submit this patch if you want me14:22
ralonsohthat's an easy one14:22
mlavalleralonsoh: no, I'll do it, thanks. want to learn :-)14:22
*** njohnston has quit IRC14:24
gmannbcafarel: found the issue, fixing that. devstack is installing Tempest on system wide also at the end which try it on py2 and fail.14:25
gmannits failing all proejcts stein gate seems.14:26
bcafarelgmann++ nice!14:26
bcafarelYeah this is why I added tempest as affected project, there was no neutron-specific part, just normal tempest job14:27
*** njohnston_ has joined #openstack-neutron14:27
gmann+114:30
*** njohnston has joined #openstack-neutron14:34
*** njohnston_ has quit IRC14:35
*** priteau has quit IRC14:37
*** yamamoto has joined #openstack-neutron14:38
gmannbcafarel: here you go, you can mark neutron as invalid on bug - https://review.opendev.org/#/q/I60949fb735c82959fb2cfcb6aeef9e33fb0445b614:38
bcafarel:) I was reading your latest comment at same time14:39
bcafarelgmann: will do, guess that is yet another $BEVERAGE I owe you when we can travel again14:39
gmannhehe, or virutal cheers  :) for now14:40
bcafarelthat I can do \o/14:41
*** tkajinam has quit IRC14:42
amotokilajoskatona: I just +2'ed bagpipe zuulv3 patch https://review.opendev.org/#/c/703949/ . hope my understanding on horizon requirements is correct.14:55
lajoskatonaamotoki: thanks14:57
lajoskatonaamotoki: If once I will have time I plan to move bgpvpn dashboard code to a new repo :-)14:57
amotokilajoskatona: that would be best :)14:58
*** ircuser-1 has joined #openstack-neutron15:11
lajoskatonaamotoki, slaweq: do you have some information about Yamamoto? In tap-as-a-service repo the progress is zero as he is the last core: https://review.opendev.org/#/q/project:x/tap-as-a-service15:12
*** rcernin has quit IRC15:13
*** yamamoto has quit IRC15:20
*** dking_desktop has joined #openstack-neutron15:23
*** gcheresh has quit IRC15:31
openstackgerritJakub Libosvar proposed openstack/neutron master: [ovn]: Create neutron_pg_drop Port Group on init  https://review.opendev.org/71140415:45
zigoHi there !15:51
openstackgerritJakub Libosvar proposed openstack/neutron master: [ovn]: Create neutron_pg_drop Port Group on init  https://review.opendev.org/71140415:51
zigoamotoki: It's not possible normally that a port has 2 entries in ml2_port_binding_levels, right ?15:51
zigoIt looks like a bug to me, to see a port on multiple hosts in that table.15:52
amotokimlavalle: hi, do you have any information on the above question from zigo?15:53
openstackgerritJakub Libosvar proposed openstack/neutron master: [ovn]: Create neutron_pg_drop Port Group on init  https://review.opendev.org/71140415:53
openstackgerritTerry Wilson proposed openstack/ovsdbapp master: Accept a Row as a result in AddCommand  https://review.opendev.org/71815515:53
zigohttp://paste.openstack.org/show/791747/ <--- Here's what we see in our Rocky cluster.15:54
zigoThis has caused us major headakes, with VMs loosing connectivity.15:54
zigoThe first entry in the table is wrong ...15:54
openstackgerritMerged openstack/networking-bagpipe master: Switch fullstack job to Zuulv3 syntax  https://review.opendev.org/70394915:58
fricklerwasn't that a feature introduced in order to allow nova to do live-migration with less network outage?15:59
fricklerzigo: do you do live-migration and may those port be related to that?16:00
zigofrickler: That's possible yes.16:00
*** lajoskatona has quit IRC16:01
zigofrickler: At some point, my colleages thought it was due to some failed live-migrations.16:01
*** jlibosva has quit IRC16:01
zigofrickler: I had lots of "port binding failed, check neutron logs" errors in nova-compute ...16:01
zigoI didn't pay much attention to that, and selected another host for the live migration.16:03
*** rpittau is now known as rpittau|afk16:08
*** lucasagomes has quit IRC16:08
*** dtantsur is now known as dtantsur|afk16:08
*** ociuhandu has quit IRC16:18
*** ociuhandu has joined #openstack-neutron16:19
*** lpetrut has joined #openstack-neutron16:27
openstackgerritTerry Wilson proposed openstack/ovsdbapp master: Stop iterating on first matching QoS rule  https://review.opendev.org/71816616:28
*** tmorin has quit IRC16:37
*** ociuhandu has quit IRC16:42
*** ociuhandu has joined #openstack-neutron16:43
*** ociuhandu has quit IRC16:48
*** lpetrut has quit IRC16:56
*** tmorin has joined #openstack-neutron17:25
*** dking_desktop is now known as dking17:51
*** dking has quit IRC17:56
*** dking_desktop has joined #openstack-neutron17:57
*** dking_desktop has quit IRC17:58
*** gcheresh has joined #openstack-neutron18:09
openstackgerritFlavio Fernandes proposed openstack/ovsdbapp stable/queens: Unbreak mock-releated autospec issue  https://review.opendev.org/71726418:26
openstackgerritFlavio Fernandes proposed openstack/ovsdbapp stable/queens: Make Event logging more useful by default  https://review.opendev.org/71306918:37
*** maciejjozefczyk has quit IRC18:39
*** AJaeger has joined #openstack-neutron18:40
AJaegerI have some py27 cleanups, could you review and merge them, please? neutron-dynamic-routing : https://review.opendev.org/717565 ; neutron-vpnaas: https://review.opendev.org/717566 ; networking-bagpipe: https://review.opendev.org/717562 ; and midonet but that's failing and needs help to fix: https://review.opendev.org/71756318:42
openstackgerritTerry Wilson proposed openstack/ovsdbapp master: Accept a Row as a result in AddCommand  https://review.opendev.org/71815518:47
*** gcheresh has quit IRC18:51
openstackgerritBrian Haley proposed openstack/neutron master: Add Octavia file in devstack/lib  https://review.opendev.org/71786019:08
*** tmorin has quit IRC19:09
*** tesseract has quit IRC19:25
*** TomStappaerts has joined #openstack-neutron19:26
openstackgerritRodolfo Alonso Hernandez proposed openstack/neutron master: Default (shared) network segment range is not mandatory  https://review.opendev.org/71775419:29
*** TomStappaerts has quit IRC19:36
*** ralonsoh has quit IRC19:40
openstackgerritTerry Wilson proposed openstack/ovsdbapp master: Stop iterating on first matching QoS rule  https://review.opendev.org/71816619:40
*** slaweq has quit IRC19:44
openstackgerritMerged openstack/neutron master: Add fullstack tests for stateless security groups  https://review.opendev.org/71025219:49
*** slaweq has joined #openstack-neutron19:51
*** nweinber has quit IRC19:52
*** ociuhandu has joined #openstack-neutron19:54
*** ociuhandu has quit IRC20:00
*** prometheanfire has joined #openstack-neutron20:00
prometheanfireso, I just got told about LB possibly being demoted from being an in tree network backend.  Has any decision been made here?20:02
openstackgerritMerged openstack/neutron master: Migrate the OVN migration scripts  https://review.opendev.org/70224720:04
openstackgerritBrian Haley proposed openstack/ovn-octavia-provider master: Don't spawn long-living processes in the driver agent  https://review.opendev.org/71820320:04
openstackgerritNate Johnston proposed openstack/neutron master: For trunk parent ports set device_id to trunk id  https://review.opendev.org/71735020:09
openstackgerritSlawek Kaplonski proposed openstack/neutron stable/rocky: Check dnsmasq process is active when spawned  https://review.opendev.org/71820620:09
openstackgerritSlawek Kaplonski proposed openstack/neutron stable/queens: Check dnsmasq process is active when spawned  https://review.opendev.org/71820720:10
*** njohnston_ has joined #openstack-neutron20:14
openstackgerritSlawek Kaplonski proposed openstack/neutron stable/queens: Remove Floating IP DNS record upon associated port deletion  https://review.opendev.org/71820920:15
njohnston_prometheanfire: In the Shanghai PTG we talked about it, and we decided it would remain in-tree but would no longer be considered a first class backend; fixes would be on a best effort basis, IIRC20:16
openstackgerritSlawek Kaplonski proposed openstack/neutron stable/train: Increase waiting time for network rescheduling  https://review.opendev.org/71821020:18
openstackgerritSlawek Kaplonski proposed openstack/neutron stable/stein: Increase waiting time for network rescheduling  https://review.opendev.org/71821120:19
openstackgerritSlawek Kaplonski proposed openstack/neutron stable/rocky: Increase waiting time for network rescheduling  https://review.opendev.org/71821220:19
openstackgerritSlawek Kaplonski proposed openstack/neutron stable/queens: Increase waiting time for network rescheduling  https://review.opendev.org/71821320:19
*** njohnston_ has quit IRC20:21
openstackgerritSlawek Kaplonski proposed openstack/neutron stable/queens: Handle ports assigned to routers without routerports  https://review.opendev.org/71821620:22
openstackgerritSlawek Kaplonski proposed openstack/neutron stable/stein: Configure keepalived interfaces according to config file  https://review.opendev.org/71821720:24
prometheanfirenjohnston: in that case I don't think it's recommended to use it, which sucks, it's very ops friendly20:38
openstackgerritSlawek Kaplonski proposed openstack/neutron stable/stein: Kill all processes running in a namespace before deletion  https://review.opendev.org/71822620:41
openstackgerritSlawek Kaplonski proposed openstack/neutron stable/stein: Implement "list_ns_pids" inside Neutron  https://review.opendev.org/71822720:44
*** igordc has quit IRC20:47
*** igordc has joined #openstack-neutron20:48
eanderssonYea that sounds nuts20:50
eanderssonSo many larger orgs are running LB20:50
prometheanfiredevs like the new shiney :P20:52
slaweqeandersson: prometheanfire: I would be happy if this backend would be still "first class" driver but reality now is that there is almost none u/s developers interested in maintaining it20:52
slaweqso if You want to help with that, You are more than welcome :)20:53
eanderssonWe contribute a lot wherever we can, but never had to do anything for LB as it's been so stable.20:53
slaweqeandersson: so that is great, I'm just saying that we would like to have someone who will be kind of point of contact if there will be any issue with LB driver20:54
eanderssonFor sure. I personally know very little on network. So I would be a terrible point of contact. :p20:55
slaweqeandersson: :)20:56
slaweqeandersson: I just wanted to clarify what is the current status of this backend20:56
*** slaweq_ has joined #openstack-neutron21:04
*** slaweq has quit IRC21:04
openstackgerritMaciej Józefczyk proposed openstack/neutron master: [OVN] Bump up transaction timeout for functional tests  https://review.opendev.org/71770421:24
*** slaweq_ has quit IRC21:34
*** ociuhandu has joined #openstack-neutron21:34
*** slaweq_ has joined #openstack-neutron21:37
*** ociuhandu has quit IRC21:39
*** slaweq_ has quit IRC21:41
prometheanfireslaweq is gone :(22:03
prometheanfirefor new installs then, OVS is the recomendation going forward?22:04
*** rcernin has joined #openstack-neutron22:11
jamesdenton_wfhyes?22:14
prometheanfirethe questionmark is what worries me :P22:25
openstackgerritMerged openstack/neutron master: [fullstack] Add option for custom apipaste.ini  https://review.opendev.org/71270222:27
openstackgerritMerged openstack/neutron stable/queens: Optimize DVR related port DB query  https://review.opendev.org/71631022:27
njohnstonprometheanfire: yes, OVS is recommended, although we would like to shift the default to OVN in the not too distant future22:33
prometheanfireok, will do a deeper review then22:54
prometheanfiremay end up on sriov anyway22:55
*** tkajinam has joined #openstack-neutron22:56
*** jamesdenton_wfh has quit IRC23:05
*** mlavalle has quit IRC23:15
jamesdentonnjohnston does that mean OVS will go into a sort of "maintenance" mode once OVN becomes the default?23:18
* prometheanfire wonders how ovn does with high bandwidth low latency stuff23:40

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!