Tuesday, 2021-06-08

« Monday, 2021-06-07 Index Wednesday, 2021-06-09 »
opendevreviewFlavio Fernandes proposed openstack/neutron master: Revert "use payloads for PORT AFTER_CREATE events"  https://review.opendev.org/c/openstack/neutron/+/79519401:05
*** TMM has quit IRC01:31
*** TMM has joined #openstack-neutron01:31
*** spatel has joined #openstack-neutron01:31
*** spatel has quit IRC01:31
*** spatel has joined #openstack-neutron03:00
*** spatel has quit IRC03:19
*** spatel has joined #openstack-neutron03:22
*** digitalsimboja has joined #openstack-neutron04:23
*** spatel has quit IRC04:26
*** opendevreview has quit IRC04:26
*** manub has joined #openstack-neutron04:34
*** lajoskatona has joined #openstack-neutron04:43
*** mo has joined #openstack-neutron04:49
*** opendevreview has joined #openstack-neutron05:01
opendevreviewLajos Katona proposed openstack/neutron master: Mark fullstack test_l2_agent_restart as unstable  https://review.opendev.org/c/openstack/neutron/+/79422805:01
*** hemanth_n has joined #openstack-neutron05:03
*** digitalsimboja has quit IRC05:17
*** digitalsimboja has joined #openstack-neutron05:55
*** ltomasbo has joined #openstack-neutron06:11
*** hjensas has quit IRC06:14
*** hjensas has joined #openstack-neutron06:19
*** ltomasbo has quit IRC06:22
*** ltomasbo has joined #openstack-neutron06:22
*** isabek has joined #openstack-neutron06:27
*** hjensas has quit IRC06:29
*** ralonsoh has joined #openstack-neutron06:34
*** manub has quit IRC06:52
*** gibi has quit IRC06:55
*** gibi has joined #openstack-neutron06:55
*** jlibosva has joined #openstack-neutron07:06
*** hjensas has joined #openstack-neutron07:14
*** digitalsimboja has quit IRC07:15
*** dulek has joined #openstack-neutron07:26
opendevreviewOpenStack Proposal Bot proposed openstack/neutron-vpnaas-dashboard stable/victoria: Imported Translations from Zanata  https://review.opendev.org/c/openstack/neutron-vpnaas-dashboard/+/79522707:27
*** rpittau|afk is now known as rpittau07:30
opendevreviewGaudenz Steinlin proposed openstack/neutron master: Copy existing IPv6 leases to generated lease file  https://review.opendev.org/c/openstack/neutron/+/76287407:30
*** digitalsimboja has joined #openstack-neutron07:45
opendevreviewSzymon Wróblewski proposed openstack/neutron master: Fix ObjectChangeHandler thread usage  https://review.opendev.org/c/openstack/neutron/+/78851007:48
*** elodilles is now known as elodilles_pto07:55
*** Luzi has joined #openstack-neutron07:55
dulekslaweq: Good morning! I got this question in the review and started to worry. Is this issue possible: https://review.opendev.org/c/openstack/kuryr-kubernetes/+/795163/2/kuryr_kubernetes/controller/drivers/nested_vlan_vif.py#140 ?07:58
slaweqdulek hi, checking08:00
*** digitalsimboja has quit IRC08:00
*** digitalsimboja has joined #openstack-neutron08:02
*** digitalsimboja has quit IRC08:02
*** digitalsimboja has joined #openstack-neutron08:03
*** elvira has joined #openstack-neutron08:05
slaweqdulek I replied in the comment, IMO it is possible what ltomasbo suggested08:05
dulekslaweq: Hm, I know that I can't make a subport twice, it's a question more about having two neutron-servers in an unsynced DB state. That's possible?08:09
slaweq@dulek, actually db state should be synced - but it's more question to the db and galera experts08:10
slaweqbut are those calls synchronous? so You will always make create_subport after delete_subport returns something?08:10
slaweqor can it be that delete is still in progress on one server and create is called already08:11
*** luksky has joined #openstack-neutron08:14
*** digitalsimboja has quit IRC08:14
*** lucasagomes has joined #openstack-neutron08:25
*** lucasagomes has quit IRC08:41
dulekslaweq: No, no, it's always synchronous, we can only have one kuryr-controller and only one thread can process on KuryrPort.08:51
slaweqdulek so in that case on neutron side it should be ok but, I'm still unsure how e.g. galera sync will work08:51
slaweqmaybe there will be rare race condition possible08:52
*** lucasagomes has joined #openstack-neutron08:53
*** lucasagomes has quit IRC09:03
ralonsohslaweq, good morning09:19
ralonsohdid you check the CI fullstack jobs>09:20
ralonsoh?09:20
ralonsohfor example09:20
ralonsohhttps://storage.bhs.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_98d/793141/4/check/neutron-fullstack-with-uwsgi/98daff1/testr_results.html09:20
*** isabek has quit IRC09:20
ralonsohneutron.tests.fullstack.test_l3_agent.TestLegacyL3Agent are failing a lot09:20
opendevreviewMamatisa Nurmatov proposed openstack/neutron master: Update handler for PORT AFTER_CREATE  https://review.opendev.org/c/openstack/neutron/+/79526009:21
*** isabek has joined #openstack-neutron09:22
opendevreviewBodo Petermann proposed openstack/neutron-vpnaas master: VPNaaS support for OVN  https://review.opendev.org/c/openstack/neutron-vpnaas/+/76535309:23
ralonsohlajoskatona, https://storage.bhs.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_98d/793141/4/check/neutron-fullstack-with-uwsgi/98daff1/controller/logs/dsvm-fullstack-logs/TestLegacyL3Agent.test_gateway_ip_changed/neutron-dhcp-agent--2021-06-08--07-26-27-307941_log.txt09:24
ralonsohis this the same issue we have with privsep?09:25
*** lucasagomes has joined #openstack-neutron09:25
slaweqralonsoh hi09:25
slaweqI only reported new LP for that https://bugs.launchpad.net/neutron/+bug/193121709:25
opendevmeetLaunchpad bug 1931217 in neutron "Fullstack Legacy L3 agent tests are failing very often recently" [Critical,Confirmed]09:25
slaweqbut didn't had time yet to check it09:25
ralonsohyeah09:25
ralonsohI think this could be related with the privsep problem09:26
slaweqit started failing like that around 5.0609:26
slaweqaccording to grafana at least09:26
ralonsohin the DHCP agent the privsep does not spawn09:26
ralonsohlet me check if we have new libraries09:26
slaweqk09:27
slaweqbut that would cause much more issues, no?09:27
slaweqand only those few tests from the Legacy L3 agents module are failing a lot09:27
ralonsohyeah and the LB problems too09:28
ralonsohwe see that a lot of times in the DHCP agent09:28
ralonsohfor example https://storage.bhs.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_98d/793141/4/check/neutron-fullstack-with-uwsgi/98daff1/controller/logs/dsvm-fullstack-logs/TestLegacyL3Agent.test_gateway_ip_changed/neutron-dhcp-agent--2021-06-08--07-26-27-307941_log.txt09:28
ralonsoh07:26:42.78509:28
ralonsohthis is the last timestamp09:28
ralonsohbut in the server is 07:28:13.51209:29
ralonsohso something is stuck there09:29
lajoskatonaralonsoh: as I see yes, I have seen a lot today...09:29
*** obondarev has joined #openstack-neutron09:30
*** dulek has quit IRC09:30
*** dulek has joined #openstack-neutron09:32
*** lucasagomes has quit IRC09:38
*** lucasagomes has joined #openstack-neutron09:39
*** obondarev_ has joined #openstack-neutron09:40
*** obondarev has quit IRC09:41
*** obondarev_ has quit IRC09:43
*** obondarev has joined #openstack-neutron09:44
*** hemanth_n has quit IRC09:54
*** hemanth_n has joined #openstack-neutron09:54
*** lucasagomes has quit IRC10:01
opendevreviewMamatisa Nurmatov proposed openstack/neutron master: Payloads for PORT: BEFORE_CREATE and PRECOMMIT_CREATE  https://review.opendev.org/c/openstack/neutron/+/79457910:10
opendevreviewSlawek Kaplonski proposed openstack/neutron-tempest-plugin master: New API test for subnet's service types  https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/78771810:18
*** lucasagomes has joined #openstack-neutron10:18
*** hemanth_n has quit IRC10:18
*** hemanth_n has joined #openstack-neutron10:19
*** obondarev has quit IRC10:29
*** lucasagomes has quit IRC10:29
*** yamamoto has joined #openstack-neutron10:32
*** obondarev has joined #openstack-neutron10:49
*** obondarev has quit IRC10:58
*** lucasagomes has joined #openstack-neutron11:00
*** lucasagomes has quit IRC11:08
opendevreviewyangjianfeng proposed openstack/neutron stable/wallaby: Make phynet paramter also is optional when network_segment_range enabled  https://review.opendev.org/c/openstack/neutron/+/79527711:25
opendevreviewyangjianfeng proposed openstack/neutron stable/victoria: Make phynet paramter also is optional when network_segment_range enabled  https://review.opendev.org/c/openstack/neutron/+/79527811:25
opendevreviewyangjianfeng proposed openstack/neutron stable/ussuri: Make phynet paramter also is optional when network_segment_range enabled  https://review.opendev.org/c/openstack/neutron/+/79527911:26
opendevreviewyangjianfeng proposed openstack/neutron stable/train: Make phynet paramter also is optional when network_segment_range enabled  https://review.opendev.org/c/openstack/neutron/+/79528011:26
*** mo has quit IRC11:30
opendevreviewyangjianfeng proposed openstack/neutron-specs master: L3 router support ndp proxy  https://review.opendev.org/c/openstack/neutron-specs/+/72862811:30
opendevreviewSlawek Kaplonski proposed openstack/neutron master: WIP DVR: Populate ARP entries of the allowed_address_pairs to the routers  https://review.opendev.org/c/openstack/neutron/+/60133611:31
*** yamamoto has quit IRC11:34
opendevreviewOleg Bondarev proposed openstack/neutron master: Use 2 dhcp agents in TestLegacyL3Agent  https://review.opendev.org/c/openstack/neutron/+/79530011:41
*** obondarev has joined #openstack-neutron11:42
obondarevslaweq, ralonsoh, lajoskatona: I uploaded a workaround to unblock CI while proper fix is developed. Please check https://review.opendev.org/c/openstack/neutron/+/79530011:43
ralonsohobondarev, let me check11:46
obondarevthe idea is that 2 dhcp agent will unlikely hang at the same time. For example TestHAL3Agent tests don't fail11:49
ralonsohbut we can't have two dhcp agents in the same node11:49
ralonsohwell, this is not the same node11:50
obondarevyep11:51
ralonsohbut are those dhcp agents creating a duplicated dnsmasq process for the same net?11:51
obondarevit's how it's supposed to work, no?11:51
obondarevHA for dhcp11:51
obondarevTestHAL3Agent has 2 dhcp agents11:52
obondarevand from logs can see that  sometimes one of those dhcp agent hangs (privsep) - but 2nd dhcp agent sends DHCP success and provisioning block is done11:53
ralonsohyes but not two dnsmasq processes at the same time11:53
ralonsohon the same network11:53
obondarevI guess isolation is provided by the framework11:54
obondarevI didn't check how, though11:54
obondarevbut I see TestHAL3Agent also has 2 hosts with dhcp agent on both hosts11:55
obondarevhttps://github.com/openstack/neutron/blob/master/neutron/tests/fullstack/test_l3_agent.py#L418-L42011:56
ralonsohyes because that is HA11:57
ralonsohone of the networks does not receive packets11:57
obondarevI was wondering why only TestLegacyL3Agent cases fail11:59
obondarevand TestHAL3Agent always passes11:59
obondarevand I saw that 2 dhcp agent have higher chances to handle the port (even is one agent hangs)12:00
obondarevthe fact that test is HA matters for l3 part (router) - but for DHCP it doesn't matter12:01
obondarevralonsoh: "two dnsmasq processes at the same time" - what do you mean? It always worked fine12:03
slaweqobondarev thx, for the patch12:10
slaweqI agree with You that it should works like that as we have doubled chance that dhcp agent will actually finish provisioning12:10
*** yamamoto has joined #openstack-neutron12:14
opendevreviewOleg Bondarev proposed openstack/neutron master: Use 2 dhcp agents in TestLegacyL3Agent  https://review.opendev.org/c/openstack/neutron/+/79530012:15
*** hemanth_n has quit IRC12:25
*** yamamoto has quit IRC12:28
*** yamamoto has joined #openstack-neutron12:32
*** spatel has joined #openstack-neutron12:32
*** dmitriis has joined #openstack-neutron12:37
*** dmitriis has quit IRC12:43
*** dmitriis has joined #openstack-neutron12:43
ralonsohobondarev, slaweq I really don't understand that12:45
ralonsohthis will spawn dhcp agents12:45
ralonsoheach one will spawn a dnsmasq process. If both are attending the same network, that means two dnsmasq processes will be running in the same network12:46
ralonsohthat should never happen12:46
slaweqdnsmasq processes will run in different namespaces12:47
slaweqwhich simulates different hosts12:48
ralonsohattending petitions in the same network12:48
slaweqralonsoh, please see https://github.com/openstack/neutron/blob/master/neutron/tests/fullstack/agents/dhcp_agent.py12:48
obondarevralonsoh: dnsmasq works fine in active-active HA mode12:48
ralonsohthis is not HA12:48
slaweqdhcp_agent used in the fullstack tests is a bit changed12:48
slaweqso it is configuring namespaces with some specific suffix for each "host"12:49
ralonsohI know12:49
ralonsohboth dhcp agents should be configured in the  same network12:49
ralonsohright?12:49
slaweqright12:49
slaweqbut in different namespaces, with different configs etc.12:50
ralonsohso you'll have two dnsmasq processes attending the same requests12:50
obondarevralonsoh: correct12:50
ralonsohso this is an error12:50
slaweqas You have in normal deployments, when You have e.g. 3 controllers12:50
slaweqand dhcp agent on each of them12:50
ralonsohin HA only one will receive those packets12:51
ralonsohnot all of them12:51
ralonsohthe processes will not compete12:51
slaweqralonsoh there is no HA/nonHA for dhcp agents12:51
slaweqall of the dhcp agents will have configured same leases12:51
slaweqDHCP agent is always "active" for each network12:51
ralonsohpackets will not reach the dnsmasq processes in backup state12:52
obondarevralonsoh: please check dhcp_agents_per_network options12:52
obondarevoption*12:52
ralonsohyes, in HA12:52
slaweqralonsoh there is no backup option for dhcp agent12:53
ralonsohanyway, with those reasons I won't approve this patch12:53
ralonsohI know that12:53
ralonsohwhat I'm saying is that in you can have more that one DHCP in HA mode12:53
ralonsohwith more that one controller per network12:53
obondarevralonsoh: dnsmasq works in active-active mode (with 2+ dnsmasq for the same network) for a long time already12:58
obondarevand https://review.opendev.org/c/openstack/neutron/+/795300 has nothing to do with that13:00
slaweqobondarev++ I agree :)13:00
ralonsohonly one agent will attend the request13:01
ralonsohonly one13:01
ralonsohif that agent is stuck, this patch won't do anything13:01
obondarevralonsoh: right, and second one will see it13:01
obondarevand update it's info13:01
obondarevlet me find some bugs..13:02
obondarevhttps://bugs.launchpad.net/neutron/+bug/145790013:05
opendevmeetLaunchpad bug 1457900 in neutron kilo "dhcp_agents_per_network > 1 cause conflicts (NACKs) from dnsmasqs (break networks)" [Undecided,New]13:05
obondarevhttps://review.opendev.org/c/openstack/neutron/+/185486/213:08
opendevreviewSlawek Kaplonski proposed openstack/neutron-tempest-plugin master: Add new scenario test for VIP address added as allowed addr pair  https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/79484113:11
opendevreviewSlawek Kaplonski proposed openstack/neutron master: WIP DVR: Populate ARP entries of the allowed_address_pairs to the routers  https://review.opendev.org/c/openstack/neutron/+/60133613:20
slaweqralonsoh obondarev  it isn't 100% ready yet, but You can review it, code should be more or less ready13:21
slaweqthere is also scenario test https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/794841 which tests such scenario13:21
obondarevslaweq: ack13:22
slaweqmaybe it will help You review all of that13:22
slaweqI still need to propose some new api tests and some docs13:22
*** lucasagomes has joined #openstack-neutron13:23
*** elodilles_pto has quit IRC13:41
*** elodilles has joined #openstack-neutron13:43
*** coreycb has joined #openstack-neutron13:43
opendevreviewKamil Sambor proposed openstack/neutron master: [WIP] Set trunk sub-port when bind profile is created  https://review.opendev.org/c/openstack/neutron/+/79533413:55
*** Luzi has quit IRC13:58
*** lucasagomes has quit IRC13:59
slaweq#startmeeting networking14:01
opendevmeetMeeting started Tue Jun  8 14:01:38 2021 UTC and is due to finish in 60 minutes.  The chair is slaweq. 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 'networking'14:01
slaweqhi everyone on first neutron meeting in the #openstack-neutron channel :)14:01
lajoskatonao/14:02
obondarevhi14:02
amotokihi14:02
rubasovo/14:02
ralonsohhi14:02
haleybhi14:02
slaweq#topic announcements14:02
slaweqXena cycle calendar https://releases.openstack.org/xena/schedule.html14:02
slaweqXena-2 milestone is in the week of the Jul 12th14:02
slaweqso we have few more weeks before that milestone14:03
slaweqnext one14:03
slaweqas You already noticed, all neutron meetings are now in our main channel Move to OTFC is done: http://lists.openstack.org/pipermail/openstack-discuss/2021-May/022724.html14:03
*** lucasagomes has joined #openstack-neutron14:03
slaweqso we will have in the neutron channel ci, l3 and drivers meetings as well14:03
slaweqtime of the meetings is not changed14:04
opendevreviewRodolfo Alonso proposed openstack/neutron-lib master: API "subnet-segmentid-writable" should inherit field definition  https://review.opendev.org/c/openstack/neutron-lib/+/79534014:04
ralonsoh(sorry)14:04
slaweqralonsoh np :)14:04
slaweqand last one14:05
slaweqstill there is few more days to propose names for Y release http://lists.openstack.org/pipermail/openstack-discuss/2021-May/022383.html14:05
slaweqmaybe You have some idea, so please follow the process from that email14:05
slaweqand that are all announcements/reminders from me for today14:05
slaweqdo You have anything else to share with the team?14:06
slaweqif not, let's move on14:07
slaweq#topic Blueprints14:07
slaweqNeutron Xena-2 BPs https://bugs.launchpad.net/neutron/+milestone/xena-214:07
slaweqas we discussed last week, I marked secure-rbac BP as completed14:07
slaweqnow, next closest one to be finished is https://blueprints.launchpad.net/neutron/+spec/distributed-dhcp-for-ml2-ovs14:08
slaweqpatches are available at https://review.opendev.org/q/topic:%22bp%252Fdistributed-dhcp-for-ml2-ovs%22+(status:open%20OR%20status:merged)14:08
slaweqthx Oleg for reviewing them14:08
slaweqDo You have any other updates/questions regarding BPs?14:09
slaweqok, I guess that this silence means "no" :)14:11
slaweqso I think we can quickly move to the next topic today14:12
slaweq#topic Bugs14:12
slaweqhongbin was bug deputy last week. Report http://lists.openstack.org/pipermail/openstack-discuss/2021-May/022776.html14:13
slaweqI found in his report 3 bugs which require some attention14:13
slaweqhttps://bugs.launchpad.net/neutron/+bug/1930402 - this is OVN related gate issue14:13
opendevmeetLaunchpad bug 1930402 in neutron "SSH timeouts happens very often in the ovn based CI jobs" [Critical,Confirmed]14:13
slaweqand https://bugs.launchpad.net/neutron/+bug/1930858 - also related to OVN14:13
opendevmeetLaunchpad bug 1930858 in neutron "OVN central service does not start properly" [Undecided,New]14:13
slaweqjlibosva lucasagomes can You help with those ^^ ?14:14
*** lucasagomes has quit IRC14:18
ralonsohI'll check second one14:19
slaweqralonsoh: thx14:19
ralonsohwe'll need logs from the serrvice14:19
slaweqthere is also one bug related to the Linuxbridge agent https://bugs.launchpad.net/neutron/+bug/1930443 - maybe someone would have some cycles to check it14:19
opendevmeetLaunchpad bug 1930443 in neutron "[LB] Linux Bridge agent always loads trunk extension, regardless of the loaded service plugins" [Medium,Confirmed]14:19
ralonsohI opened that one, yes14:20
ralonsohI can take a look at it14:20
slaweqralonsoh thx a lot, if You have time, that would be great14:20
slaweqbut maybe someone else could take a look at this one14:20
*** digitalsimboja has joined #openstack-neutron14:21
*** lucasagomes has joined #openstack-neutron14:21
slaweqother bugs from hongbin's summary are assigned and in most cases in progress already14:22
slaweqdo You have any other bugs You want to discuss today?14:22
slaweqI guess that this means "no" :)14:23
ralonsohno14:24
slaweqour bug deputy this week is haleyb14:24
slaweqand I think he is aware of it already14:24
haleybo/14:24
slaweqnext week will be lajoskatona's turn14:24
lajoskatonaYeah, I am aware of it14:24
slaweqgreat14:24
slaweqthx to both of You :)14:25
slaweqok, next topic14:25
slaweq#topic CLI/SDK14:25
slaweqneutronclient patch https://review.opendev.org/c/openstack/python-neutronclient/+/793366 is now merged14:25
slaweqand I proposed new release https://review.opendev.org/c/openstack/releases/+/795087 to include that new message there ASAP14:26
slaweqand I think that should be all regarding migration to OSC for now14:26
slaweqnext action item will be in the Z cycle :)14:26
amotokiis it better to add a release note in neutornclient to clarify when we will drop the CLI support?14:27
slaweqamotoki that's good idea, I totally forgot about release note14:27
slaweqdo You want to add it or should I send patch?14:28
amotokii can14:28
slaweqthx14:28
slaweqI will hold on release patch for now14:28
amotokiDo we want to add it before the release?14:28
slaweq++14:28
slaweqamotoki I just -W release patch14:29
amotokithanks14:29
slaweqto wait for new release note14:29
slaweqthank You14:29
slaweqand that was all what I had for today's meeting14:30
slaweqdo You have anything else You would like to discuss now?14:30
slaweqor if not, I will give You 30 minutes back14:30
amotokiI think we see no problem in having a meeting here :)14:31
slaweqamotoki for me it is fine so far14:31
slaweqbut if anyone has any other feedback, please share it with everyone so we can adjust things properly :)14:31
obondarevso far so good14:32
slaweqok, I think we can finish meeting earlier today as we don't have other topics to discuss14:33
lajoskatonasimpler to stay on the same channel14:33
ralonsohsee you in 25 mins14:33
lajoskatonaBye14:34
slaweqplease remember that we have ci meeting in same channel in 25 minutes :)14:34
slaweqthx for attending the meeting14:34
slaweqo/14:34
slaweq#endmeeting14:34
rubasovo/14:34
opendevmeetMeeting ended Tue Jun  8 14:34:21 2021 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)14:34
opendevmeetMinutes:        http://eavesdrop.openstack.org/meetings/networking/2021/networking.2021-06-08-14.01.html14:34
opendevmeetMinutes (text): http://eavesdrop.openstack.org/meetings/networking/2021/networking.2021-06-08-14.01.txt14:34
opendevmeetLog:            http://eavesdrop.openstack.org/meetings/networking/2021/networking.2021-06-08-14.01.log.html14:34
amotokio/14:35
*** elvira1 has joined #openstack-neutron14:37
*** yamamoto has quit IRC14:38
*** elvira has quit IRC14:39
opendevreviewGaudenz Steinlin proposed openstack/neutron master: Copy existing IPv6 leases to generated lease file  https://review.opendev.org/c/openstack/neutron/+/76287415:00
*** elvira1 is now known as elvira15:02
slaweq#startmeeting neutron_ci15:02
opendevmeetMeeting started Tue Jun  8 15:02:14 2021 UTC and is due to finish in 60 minutes.  The chair is slaweq. Information about MeetBot at http://wiki.debian.org/MeetBot.15:02
opendevmeetUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.15:02
opendevmeetThe meeting name has been set to 'neutron_ci'15:02
slaweqhi (again)15:02
opendevreviewSlawek Kaplonski proposed openstack/neutron-tempest-plugin master: Add new API tests for active field in the allowed_address_pairs  https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/79535815:04
slaweqanyone still here? :)15:04
lajoskatonaHi,15:04
ralonsohhi15:04
slaweqok, let's start :)15:05
obondarevhi15:05
slaweqGrafana dashboard: http://grafana.openstack.org/dashboard/db/neutron-failure-rate15:05
slaweqplease open to have it ready15:05
slaweqand we can move on15:05
slaweq#topic Actions from previous meetings15:05
slaweqslaweq to prepare plan of switch ovn <-> ovs jobs in neutron CI15:06
slaweqI did such table: https://etherpad.opendev.org/p/neutron-ci-ovs-to-ovn15:06
slaweqplease check it and let me know what do You think about it15:06
ralonsohgood stuff15:06
ralonsohthanks15:06
ralonsohwhat are we going to do with the naming?15:07
lajoskatonawe said last time to have the backend in the job name as I remember15:07
obondarevneutron-rally-task should be  ovs-ovs15:08
obondarevovn multinode seems missing15:08
slaweqfor neutron-rally-task that's good point - it can be renamed to neutron-ovs-rally-task15:09
lajoskatona+115:09
slaweqthat will be more consistent15:09
slaweqand good news is that I think that we can remove completly neutron-ovn-tempest-ovs-release  :)15:09
slaweqso one job less in our CI15:09
slaweqwdyt?15:09
ralonsohbut this one is using master ovs, right?15:10
slaweqno, released ovs15:10
ralonsohso what's the different then?15:10
lajoskatonaBy the way haleyb missed jobs last time: http://lists.openstack.org/pipermail/openstack-discuss/2021-June/022904.html15:11
slaweqralonsoh I think we have same coverage in other jobs which by default runs ovn now15:11
slaweqbut I will check that to be sure15:11
slaweqlajoskatona I think that this should be now fixed, right?15:12
haleybyes, that was reverted and i see the jobs again15:12
lajoskatonaok15:12
slaweqhaleyb++ thx for catching that15:13
haleybzuul passing in 30 minutes was suspicious :-o15:13
lajoskatonaneutron-ovn-tempest-slow   & neutron-tempest-slow-py3 can't we keep them and rename to show which one is ovn and ovs?15:13
ralonsohyeah15:14
*** yamamoto has joined #openstack-neutron15:14
slaweqlajoskatona yes, good point15:14
slaweqso basically we will have pairs like *-ovs and *ovn jobs15:16
slaweqand it will be for rally, tempest-slow, and ipv6-only jobs15:16
*** isabek has quit IRC15:16
slaweqanyone wants to push patches for that?15:17
ralonsohI can do it15:18
slaweqthx ralonsoh15:19
*** digitalsimboja has quit IRC15:19
slaweq#action ralonsoh to update our ci job's definitions15:20
*** digitalsimboja has joined #openstack-neutron15:20
slaweqok, let's move on15:20
slaweqralonsoh to propose tempest patch to use resolvectl to address https://bugs.launchpad.net/neutron/+bug/192952315:20
opendevmeetLaunchpad bug 1929523 in neutron "Test tempest.scenario.test_network_basic_ops.TestNetworkBasicOps.test_subnet_details is failing from time to time" [High,Confirmed]15:20
ralonsohone sec15:20
ralonsohhttps://review.opendev.org/c/openstack/tempest/+/79477615:20
slaweqthx ralonsoh15:21
slaweqI will review it later15:21
slaweqok, next one15:22
slaweqslaweq to reach out to boden about payload patches and x/vmware-nsx15:22
slaweqI tried to contact Adit from Vmware but he didn't reply to me15:22
slaweqbut I see that Salvadore approved that patch15:22
slaweqso it's merged now15:22
slaweqlajoskatona I think we are good with neutron patch now, right?15:23
*** yamamoto has quit IRC15:23
lajoskatonacool, than we didn't break their ci15:23
slaweqyeah :)15:23
slaweqok, next one15:23
slaweqralonsoh to check failing neutron-lib-from-master periodic job15:23
ralonsohI didn't check anything15:24
obondarevI think I saw a patch15:24
ralonsohreally?15:24
slaweqyeah, I saw it too :)15:24
obondarevyes :)15:24
slaweqhttps://review.opendev.org/c/openstack/neutron/+/79423615:25
slaweqthat one I think15:25
slaweqno?15:25
ralonsohahhh yes15:26
slaweqLOL15:26
ralonsohpffff my brain if melting15:26
slaweqralonsoh TooManyPatchesException() :P15:26
obondarev:D15:27
slaweqso thx for fixing that issue ralonsoh :)15:27
slaweqand let's move on15:27
slaweqlajoskatona to check fullstack failures https://bugs.launchpad.net/neutron/+bug/193040115:27
opendevmeetLaunchpad bug 1930401 in neutron "Fullstack l3 agent tests failing due to timeout waiting until port is active" [Critical,Confirmed] - Assigned to Lajos Katona (lajos-katona)15:27
lajoskatonathat is some ugly privsep issue15:28
slaweqahh, that one15:28
lajoskatonaI pushed one patch, but with ralonsoh and obondarev we agreed that is not enough15:28
slaweqthe famous one :)15:28
ralonsohI agree with the idea of having two decorators15:28
ralonsohone with timeout15:28
ralonsohand we can also add a default timeout in the context definition15:29
obondarev+115:29
lajoskatonayeah I started to check that and can produce terrible failures in my locan env already15:29
lajoskatonaso it's ongoing15:29
obondarevmeanwhile workaround is almost there: https://review.opendev.org/c/openstack/neutron/+/79530015:30
lajoskatonathis privsep looks like if I touch somewhere I have a break somewhere else and can't see the connection :-)15:30
obondarevlajoskatona: please add workaround revert in your patch15:30
lajoskatonaobondarev: sure15:30
slaweqthx lajoskatona and obondarev for working on that issue15:31
slaweqit is really killing our CI now15:31
obondarevto ralonsoh as well15:31
slaweqahh, right :)15:31
slaweqsorry ralonsoh that I missed You15:32
ralonsohhehehe15:32
slaweqok, last one from previous week15:32
slaweqlajoskatona to mark failing TestUninterruptedConnectivityOnL2AgentRestart fullstack tests as unstable temporary15:32
lajoskatonahttps://review.opendev.org/c/openstack/neutron/+/79422815:32
lajoskatonaohhh, it's more red than green now....15:33
lajoskatonalooks like some glitch in pip mirroring perhaps15:34
slaweqthx lajoskatona15:34
slaweqand thx obondarev and ralonsoh for reviewing it :)15:34
ralonsohyw15:34
slaweqok, let's move on to the other topics15:35
obondarevsure15:35
slaweq#topic Stadium projects15:35
slaweqlajoskatona any issues with stadium's ci?15:35
lajoskatonanothing as I know15:35
slaweqno news is good news in this case :)15:35
*** digitalsimboja has quit IRC15:36
lajoskatonaexactly15:36
slaweqI will skip today stable branches topic as there is no bcafarel here15:36
slaweqand I don't think we have any new issues there, I saw bunch of patches merged in stable branches recently15:36
slaweqfinally even rocky and queens are running more or less fine :)15:36
slaweq#topic Grafana15:36
slaweq#link http://grafana.openstack.org/dashboard/db/neutron-failure-rate15:37
slaweqthe biggest issue for now is with fullstack15:37
slaweqand either we will fix that privsep thing or we will need to mark some tests as unstable or make job non-voting temporary15:38
slaweqwdyt?15:38
obondarevslaweq: https://review.opendev.org/c/openstack/neutron/+/79530015:38
obondarevno?15:38
slaweqright15:38
slaweqwe talked about it few minutes ago :)15:38
ralonsohyes, I prefer not to mark all the job as non-voting15:38
slaweqI hope this will make it better15:38
slaweqralonsoh me neighter15:39
slaweqok, let's see how it will be with that patch15:39
slaweqanything else regarding grafana for today?15:39
slaweqok, I guess this means "no"15:41
slaweqregarding specific jobs' failures, I only found today fullstack issue15:42
slaweqI even reported new bug  https://bugs.launchpad.net/neutron/+bug/1931217 but now I see that this is dup of https://bugs.launchpad.net/neutron/+bug/193040115:42
opendevmeetLaunchpad bug 1931220 in neutron "duplicate for #1931217 _ObjectChangeHandler.handle_event failing on port after_create event" [Critical,In progress] - Assigned to Mamatisa Nurmatov (isabek)15:42
ralonsohyes, this is the main issue15:42
opendevmeetLaunchpad bug 1930401 in neutron "Fullstack l3 agent tests failing due to timeout waiting until port is active" [Critical,Confirmed] - Assigned to Lajos Katona (lajos-katona)15:42
slaweqso good news is that we don't have new ci issues, at least no that I would be aware of :)15:43
obondarevonly pip glitch :)15:43
slaweqdo You have anything else related to the CI You would like to discuss today?15:44
ralonsohno15:45
slaweqor if not, I think we can finish earlier today15:45
opendevreviewRodolfo Alonso proposed openstack/neutron-lib master: API "subnet-segmentid-writable" should inherit field definition  https://review.opendev.org/c/openstack/neutron-lib/+/79534015:45
ralonsoh(sorry)15:45
slaweqnp :)15:45
*** whoami-rajat has quit IRC15:46
slaweqok, thx for attending the meeting15:46
ralonsohbye!15:46
slaweqhave a great rest of the day and see You online15:46
slaweqo/15:46
obondarevo/15:46
slaweq#endmeeting15:46
opendevmeetMeeting ended Tue Jun  8 15:46:53 2021 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)15:46
opendevmeetMinutes:        http://eavesdrop.openstack.org/meetings/neutron_ci/2021/neutron_ci.2021-06-08-15.02.html15:46
opendevmeetMinutes (text): http://eavesdrop.openstack.org/meetings/neutron_ci/2021/neutron_ci.2021-06-08-15.02.txt15:46
opendevmeetLog:            http://eavesdrop.openstack.org/meetings/neutron_ci/2021/neutron_ci.2021-06-08-15.02.log.html15:46
opendevreviewRodolfo Alonso proposed openstack/neutron master: Add CONTEXT_READER decorator to "get_ha_router_port_bindings"  https://review.opendev.org/c/openstack/neutron/+/79423615:51
*** obondarev has quit IRC15:59
*** ltomasbo has quit IRC16:14
*** lucasagomes has quit IRC16:15
*** elvira1 has joined #openstack-neutron16:26
*** elvira has quit IRC16:27
*** elvira1 has quit IRC16:29
*** rpittau is now known as rpittau|afk16:31
opendevreviewMerged openstack/neutron stable/victoria: HA-non-DVR router don't need manually add static route  https://review.opendev.org/c/openstack/neutron/+/79287516:47
*** hemanth_n has joined #openstack-neutron16:51
*** hemanth_n has quit IRC16:52
*** ralonsoh has quit IRC17:01
opendevreviewMerged openstack/neutron master: Add "filter-validation" support for "OVNL3RouterPlugin"  https://review.opendev.org/c/openstack/neutron/+/79310917:36
*** tyldis has joined #openstack-neutron18:37
*** jlibosva has quit IRC18:42
*** digitalsimboja has joined #openstack-neutron18:53
opendevreviewMerged openstack/neutron master: [FT] Remove skip guard from SQL tests  https://review.opendev.org/c/openstack/neutron/+/79295319:33
*** lajoskatona has quit IRC19:54
*** spatel has quit IRC19:58
*** digitalsimboja has quit IRC20:14
*** TMM has quit IRC20:41
*** TMM has joined #openstack-neutron20:41
opendevreviewSlawek Kaplonski proposed openstack/neutron-tempest-plugin master: Add API tests for basic address_groups CRUD operations  https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/79541521:12
opendevreviewGaudenz Steinlin proposed openstack/neutron master: Copy existing IPv6 leases to generated lease file  https://review.opendev.org/c/openstack/neutron/+/76287421:51
opendevreviewVadim Ponomarev proposed openstack/neutron-lib master: Introduce rbac-bgpvpn api extension  https://review.opendev.org/c/openstack/neutron-lib/+/79542321:56
*** dmitriis has quit IRC22:02
opendevreviewMerged openstack/neutron master: Use 2 dhcp agents in TestLegacyL3Agent  https://review.opendev.org/c/openstack/neutron/+/79530022:05
opendevreviewMerged openstack/neutron master: Update handler for PORT AFTER_CREATE  https://review.opendev.org/c/openstack/neutron/+/79526022:05
*** luksky has quit IRC22:10
*** dmitriis has joined #openstack-neutron22:13
*** yamamoto has joined #openstack-neutron22:30
*** yamamoto has quit IRC22:38
*** yamamoto has joined #openstack-neutron22:38
*** dmitriis has quit IRC22:44
opendevreviewRodolfo Alonso proposed openstack/neutron-tempest-plugin master: Switch neutron-tempest-plugin-api job to be ML2/OVN  https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/79125522:52
*** dmitriis has joined #openstack-neutron22:55
*** yamamoto has quit IRC22:59
*** yamamoto has joined #openstack-neutron23:05
« Monday, 2021-06-07 Index Wednesday, 2021-06-09 »

Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!