Tuesday, 2020-10-06

« Monday, 2020-10-05 Index Wednesday, 2020-10-07 »
*** armax has quit IRC00:25
*** armax has joined #openstack-neutron00:27
*** sapd1 has quit IRC00:40
*** hamalq has quit IRC00:41
*** zzzeek has quit IRC00:43
*** terdei has joined #openstack-neutron00:44
*** zzzeek has joined #openstack-neutron00:45
*** zzzeek has quit IRC01:49
*** zzzeek has joined #openstack-neutron01:51
*** armax has quit IRC02:06
*** sapd1 has joined #openstack-neutron02:24
*** armax has joined #openstack-neutron03:03
*** zzzeek has quit IRC03:17
*** zzzeek has joined #openstack-neutron03:19
*** armax has quit IRC03:50
*** zzzeek has quit IRC04:01
*** zzzeek has joined #openstack-neutron04:05
*** vipul has joined #openstack-neutron04:30
*** ratailor has joined #openstack-neutron04:43
*** rcernin_ has joined #openstack-neutron04:48
*** rcernin has quit IRC04:50
*** mmethot_ has quit IRC04:54
*** mmethot_ has joined #openstack-neutron04:55
*** abdysn has joined #openstack-neutron05:26
*** viks____ has joined #openstack-neutron05:43
viks____Hi, I'm facing issue in setting ipv6 . My ISP do not have SLAAC enabled. So i created IPv6 subnet in slaac/slaac mode. While creating instance, Ipv6 gets assigned which i can see in the dashboard but not in the instance. But if i associate Ipv6 subnet with the a router, IPv6 gets assigned in the instance. But default route is getting set to some unknown LLA which i think is of the router. If i set default05:44
viks____route via LLA of upstream router manually, ipv6 works on the instance i.e. i can ping to and from instance. How do i make the default route to be set to upstream router LLA automatically? I tried setting `ip_gateway` in `l3_agent.ini` which did not work. Can someone please help me to understand what is going on?05:44
*** tmorin has joined #openstack-neutron05:56
*** gouthamr has quit IRC06:05
*** fyx has quit IRC06:05
*** dviroel has quit IRC06:05
*** mgoddard has quit IRC06:05
*** openstackgerrit has quit IRC06:05
*** StevenK has quit IRC06:05
*** dansmith has quit IRC06:05
*** dviroel has joined #openstack-neutron06:05
*** StevenK has joined #openstack-neutron06:05
*** dansmith has joined #openstack-neutron06:05
*** mgoddard has joined #openstack-neutron06:05
*** gouthamr has joined #openstack-neutron06:05
*** fyx has joined #openstack-neutron06:05
*** jlibosva has joined #openstack-neutron06:06
*** v1k0d3n has quit IRC06:07
*** lajoskatona has joined #openstack-neutron06:08
*** v1k0d3n has joined #openstack-neutron06:08
*** slaweq has joined #openstack-neutron06:10
*** macz_ has joined #openstack-neutron06:13
*** macz_ has quit IRC06:18
*** tmorin has quit IRC06:28
*** ralonsoh has joined #openstack-neutron06:36
*** ratailor has quit IRC06:38
*** ratailor has joined #openstack-neutron06:41
*** ratailor has quit IRC06:41
*** tmorin has joined #openstack-neutron06:43
*** ratailor has joined #openstack-neutron06:48
*** ratailor has quit IRC06:48
*** ratailor has joined #openstack-neutron06:49
hemanth_nHi.. appreciate reviews on https://review.opendev.org/#/c/752248/ .. thanks07:02
*** slaweq has quit IRC07:03
*** spatel has joined #openstack-neutron07:15
*** tesseract has joined #openstack-neutron07:19
*** spatel has quit IRC07:20
*** ralonsoh has quit IRC07:21
*** ralonsoh has joined #openstack-neutron07:23
*** ociuhandu has joined #openstack-neutron07:36
*** rpittau|afk is now known as rpittau07:45
*** mnasiadka has joined #openstack-neutron07:47
*** jpena|off is now known as jpena07:56
*** lucasagomes has joined #openstack-neutron08:06
*** ratailor has quit IRC08:19
*** ratailor has joined #openstack-neutron08:19
*** rcernin_ has quit IRC08:35
*** dtantsur|afk is now known as dtantsur08:41
*** openstackgerrit has joined #openstack-neutron08:43
openstackgerritMerged openstack/neutron master: Use ebtables-legacy also in ovn multinode jobs  https://review.opendev.org/75580908:43
*** Abdallahyas has joined #openstack-neutron08:46
*** slaweq has joined #openstack-neutron08:47
*** tmorin has quit IRC08:48
slaweqralonsoh: lajoskatona hi, can You check https://review.opendev.org/#/c/756073/ ?08:49
lajoskatonaslaweq: yes08:50
ralonsohsure08:50
*** abdysn has quit IRC08:50
openstackgerritMerged openstack/neutron master: Default dnsmasq --conf-file to /dev/null  https://review.opendev.org/75535608:51
openstackgerritMerged openstack/neutron stable/victoria: Fix dhcp_release when client_id is specified  https://review.opendev.org/75581708:53
slaweqlajoskatona: ralonsoh thx08:54
*** priteau has joined #openstack-neutron08:55
openstackgerritSlawek Kaplonski proposed openstack/neutron master: Convert neutron-grenade-ovn job to be zuulv3  https://review.opendev.org/72959108:58
*** Madkiss has joined #openstack-neutron09:39
Madkisshi folks. Quick question: I run OVN with Neutron.I created a VLAN provider network, but it looks like the appropriate interface is not created on the target host if VMs are started.09:39
*** hjensas__ is now known as hjensas09:42
*** priteau has quit IRC09:54
openstackgerritMerged openstack/neutron master: include python36-devel in bindep, for tests  https://review.opendev.org/75592710:22
openstackgerritMerged openstack/neutron master: Open Wallaby DB branch  https://review.opendev.org/75528510:22
*** dviroel has quit IRC10:33
*** thedac has quit IRC10:33
*** tkajinam has quit IRC10:33
*** jmlowe has quit IRC10:33
*** dviroel has joined #openstack-neutron10:34
*** thedac has joined #openstack-neutron10:34
*** tkajinam has joined #openstack-neutron10:34
*** jmlowe has joined #openstack-neutron10:34
slaweqralonsoh: can You check https://review.opendev.org/#/c/725647/ ?11:01
slaweqwhen You will have some time :)11:01
slaweqjlibosva: lucasagomes: hi, if You will have few minutes, please also check https://review.opendev.org/#/c/729591/11:02
slaweqthx in advance11:02
ralonsohslaweq, right now11:03
lucasagomesslaweq, will take a look!11:03
slaweqthank You guys11:03
slaweqwith those 2 patches we will be finally done with migration to zuul v3 :)11:03
jlibosvaslaweq++11:03
*** vipul has quit IRC11:04
jlibosvaslaweq: do you think it's fine to keep the name of the job as is or shall we add -multinode ? I'm not sure how significant that is since we have just a single grenade-ovn job, just thinking out loud11:04
jlibosvaI'm fine with the name as is11:04
slaweqjlibosva: I wanted to keep old name of the job11:06
jlibosvaslaweq: thanks11:06
openstackgerritSlawek Kaplonski proposed openstack/neutron stable/victoria: Use ebtables-legacy also in ovn multinode jobs  https://review.opendev.org/75626011:07
*** sapd1 has quit IRC11:19
*** Abdallahyas has quit IRC11:21
*** abdysn has joined #openstack-neutron11:22
*** jpena is now known as jpena|lunch11:24
*** tmorin has joined #openstack-neutron11:35
openstackgerritMerged openstack/networking-odl master: Migrate legacy grenade job to be native Zuul v3  https://review.opendev.org/72564711:36
slaweqlajoskatona: can You look at gibi's comment in https://review.opendev.org/#/c/755180/ ?11:45
slaweqit seems that we didn't fixed this nova-next job still11:45
slaweqand I'm not sure if we should go with https://review.opendev.org/#/c/755180/ in stable/victoria11:45
slaweqmaybe we should simply keep it for the next release?11:46
*** CeeMac has quit IRC11:48
gibislaweq: hi! yeah, I can totally accept if the feature is kept for W only.11:48
lajoskatonaslaweq: yeah, I just checked the testcase in nova-next11:48
*** raildo has joined #openstack-neutron11:50
lajoskatonathe problem seems to be that in BEFORE_UPDATE event the case when PUT {qos_policy=None} and PUT {binding_profile=None} results similar dict11:51
*** lpetrut has joined #openstack-neutron11:54
lajoskatonayeah, keep it for Wallaby11:57
slaweqlajoskatona: gibi: ok, thx12:00
slaweqso I will not try to merge this in the victoria beforethis thursday :)12:00
gibiack12:01
lajoskatonaok12:09
*** zzzeek has quit IRC12:17
*** haleyb has quit IRC12:18
*** zzzeek has joined #openstack-neutron12:18
*** ratailor has quit IRC12:30
bcafarelslaweq: so do we still need https://review.opendev.org/#/c/756076/ ?12:33
*** jpena|lunch is now known as jpena12:35
slaweqbcafarel: I don't think so really12:36
*** Luzi has joined #openstack-neutron12:36
*** jamesdenton has quit IRC12:46
*** jamesdenton has joined #openstack-neutron12:47
*** haleyb has joined #openstack-neutron12:56
*** elod is now known as elod_afk13:12
*** nweinber has joined #openstack-neutron13:13
openstackgerritLajos Katona proposed openstack/neutron master: fullstack: Use smaller subnets to avoid clashes  https://review.opendev.org/74904113:19
openstackgerritRodolfo Alonso Hernandez proposed openstack/neutron master: Capture timeout exception when deleting NamespaceFixture  https://review.opendev.org/75493813:39
openstackgerritAlex Katz proposed openstack/neutron-tempest-plugin master: Add a test for overlapping SG rules  https://review.opendev.org/52025113:41
*** jmlowe has quit IRC13:41
*** Luzi has quit IRC13:43
*** jmlowe has joined #openstack-neutron13:43
*** mlavalle has joined #openstack-neutron13:59
*** abdysn has quit IRC14:00
openstackgerritLajos Katona proposed openstack/networking-bgpvpn stable/ussuri: Fix pep8 job  https://review.opendev.org/74348714:18
openstackgerritRodolfo Alonso Hernandez proposed openstack/neutron master: Process ingress multicast traffic for 224.0.0.X separately  https://review.opendev.org/74871914:21
openstackgerritLuigi Toscano proposed openstack/networking-odl stable/victoria: Migrate legacy grenade job to be native Zuul v3  https://review.opendev.org/75632414:27
openstackgerritSlawek Kaplonski proposed openstack/neutron-tempest-plugin master: Use Cirros 0.5.1 in scenario jobs for all branches  https://review.opendev.org/75511214:30
*** armax has joined #openstack-neutron14:37
otherwiseguy_slaweq: on that ovs lower-constraints bump to 2.10, the ovsdbapp/neutron code in the victoria branch has the indexing code and it does break with ovs 2.8 already, so I think we pretty much have to bump the requirement or we're just lying about what will work. My fault for not noticing the req--just nothing actually every installed 2.8 in CI. In fact because of the ovs/ovn split, I'm sure devstack wouldn't even work with14:50
otherwiseguy_something <= 2.12 maybe?14:50
otherwiseguy_I have https://review.opendev.org/#/c/756148/ to bump in ovsdbapp/master, but need to backport to victoria if that sounds ok with you.14:51
slaweqotherwiseguy_: hmm, but now if we want to bump requirements for stable/victoria I'm not sure if we will not need blessing from the requirements team for that14:52
otherwiseguy_slaweq: not sure, but it is effectively the requirement whether they acknowledge it or not...14:52
slaweqotherwiseguy_: so is it that we just "forgot" to bump it some time ago? or we merged very recently something what broke it on 2.8?14:53
fricklerslaweq: ralonsoh: the devstack openstacksdk job is broken on stable/train it seems, error is "openstack.tests.functional.cloud.test_qos_policy.TestQosPolicy.test_create_qos_policy_default [] ... inprogress" see https://zuul.opendev.org/t/openstack/build/82a83f08f9064f18bcb9e12e4b0b989714:53
fricklercan you check whether that is an issue on neutron's side?14:54
otherwiseguy_slaweq: just forgot. Once the indexing patch in ovsdbapp went in, it was a requirement. Just didn't notice that ovs req was at 2.8 (the addition of supporting indexing in the ovs tree was like 2 years old at the point it was added to ovsdbapp).14:54
slaweqfrickler: looking14:54
slaweqotherwiseguy_: ok, lets propose patch to bump it and I will ask requirements team to check that too14:55
otherwiseguy_slaweq: thx. sorry for the trouble.14:55
openstackgerritTerry Wilson proposed openstack/ovsdbapp stable/victoria: Bump ovs minimum version for support of indexing  https://review.opendev.org/75632914:55
*** macz_ has joined #openstack-neutron14:56
slaweqfrickler: it seems that it may be neutron issue14:57
slaweqsee14:57
slaweqOct 05 09:13:07.366606 ubuntu-bionic-vexxhost-ca-ymq-1-0020274479 neutron-server[21940]: INFO neutron.wsgi [None req-a54e9a59-69f7-479f-bca1-f9a77063c627 admin admin] 199.204.45.223,199.204.45.223 "GET /v2.0/qos/policies.json HTTP/1.1" status: 200  len: 563 time: 280.495908014:57
slaweqfrickler: but is that failing every time like that or was it just this once?15:00
slaweqbecause I see there that many API calls around same time are very slow there15:01
slaweqmaybe it was just some node issue?15:01
fricklerslaweq: it was the same before the recheck as far as I can see15:10
fricklerfeel free to do another one to check once more15:10
fricklerI originally wanted to verify the depends-on patch, but tempest doesn't get far enough for that.15:11
*** sapd1 has joined #openstack-neutron15:15
*** elod_afk is now known as elod15:15
mlavalleralonsoh: ping15:15
*** hyang has joined #openstack-neutron15:16
ralonsohmlavalle, hola15:16
*** gcheresh has quit IRC15:17
mlavalleralonsoh: hyang are working on adding address groups to security group rules. I've mentioned to him that you are an expert in the ovs firewall15:18
mlavallehyang and I are working on this15:18
hyang@ralonsoh @mlavalle Hi15:18
ralonsohmlavalle, ok, do you have a patch or do you need guidance?15:19
mlavalleralonsoh: hyang is about to start coding changes to the firewall and he might need some assistancein case we get stuck. So what we need for now is guidance to start with the patch15:19
mlavalleralonsoh: I am also pointing him to https://docs.openstack.org/neutron/latest/contributor/internals/openvswitch_firewall.html and the presentation you gave some time ago in one of the summits15:20
openstackgerritFlavio Fernandes proposed openstack/neutron-tempest-plugin master: Add more port_forwarding tests  https://review.opendev.org/75576015:20
ralonsohmlavalle, this presentation is a bit old15:21
ralonsohand I was working in other OVS firewall15:21
mlavalleralonsoh: yeah, but is a starting point15:21
mlavalleoh, ok, so let's forget about that one15:21
ralonsohthe current one has a feature called OF conjuntion15:21
mlavalleralonsoh: so hyang should start from the doc in the neutron docs15:22
ralonsohany new rule with an IP and a port should be handled there15:22
ralonsohlet me check one thing15:22
ralonsohfirst of all15:22
ralonsohdo you have some document or patch related?15:22
hyangralonsoh: this is the WIP neutron patch https://review.opendev.org/#/c/751110/ and we're adding the remote-address-group to the security group rules15:24
mlavalleralonsoh: and this is the spec: https://specs.openstack.org/openstack/neutron-specs/specs/victoria/address-groups-support-in-security-group-rule.html15:24
slaweqfrickler: it seems that it's failing in same way on every devstack patch :/15:25
slaweqso there is some problem in neutron probably15:25
fricklerslaweq: it may also be an issue in openstacksdk, because it is tested only against master, not in stable branches15:26
mlavalleralonsoh: what we are thinking is to create at least two additional patches to follow the one above: one with whatever changes might be rquired in the rpc channel and another one with changes to the firewall code. This latter patch might actually be split in several15:29
openstackgerritLucas Alvares Gomes proposed openstack/neutron master: [OVN][Doc] DevStack guide: Add command to get the  port's UUID  https://review.opendev.org/75582915:29
ralonsohmlavalle, ok so you have the first one https://review.opendev.org/#/c/75111015:30
ralonsohmodifying the DB15:30
mlavalleand the API15:30
ralonsohyes and the API15:30
ralonsohthe next one thr RPC15:30
ralonsohand the next one the FW15:30
mlavalleyes, I'll work on that one15:30
mlavalleI meant I'll work on the RPC changes15:31
mlavalleand hyang will start working on the FW15:31
hyangyup, that's our plan15:31
ralonsohwhat you need, in the FW, is to handle (somehow) the list of IP addresses in class ConjIdMap15:31
ralonsohnow we can have15:32
ralonsoh1) as SG pointing to a network address15:32
ralonsoh2) as SG allowing a remote SG15:33
ralonsohyou are introducing the 3rd one: a rule that will allow a set of IPs15:33
ralonsohcorrect?15:33
hyangyes15:33
*** jangutter_ has joined #openstack-neutron15:34
*** jangutter has quit IRC15:35
ralonsohhyang, check "def add_flows_from_rules"15:35
*** jangutter has joined #openstack-neutron15:35
ralonsohthis is the method, in the FW, that updates the rule creation15:35
ralonsohthe OF creation, from a rule15:35
ralonsohyou have two parts15:35
ralonsoh        self._add_non_ip_conj_flows(port)15:36
ralonsoh        self.conj_ip_manager.update_flows_for_vlan(port.vlan_tag)15:36
ralonsohyou'll need to handle this 3rd condition15:36
slaweqfrickler: I opened bug https://bugs.launchpad.net/neutron/+bug/189874815:39
openstackLaunchpad bug 1898748 in neutron "[stable/train] Creation of the QoS policy takes ages" [Critical,New]15:39
slaweqI will try to check that tomorrow15:39
mlavalleralonsoh: thanks. I think this is a good starting point. We'll start from here and will get back to you with questions if we need. Is that ok with you?15:40
ralonsohmlavalle, for sure15:41
openstackgerritLajos Katona proposed openstack/networking-odl stable/train: This change is a squash of 2 different patches to pass the gate  https://review.opendev.org/74580615:41
ralonsohmlavalle, and also via mail if needed15:41
mlavalleralonsoh: muchas gracias caballero15:41
ralonsohmlavalle, un placer15:41
hyangralonsoh mlavalle: appreciate for you guidance, I will do my research and sync with guys again!15:41
*** lpetrut has quit IRC15:43
*** tmorin has quit IRC15:47
fricklerslaweq: thx, I'll watch that bug15:47
*** rpittau is now known as rpittau|afk15:48
*** jlibosva has quit IRC15:58
*** lucasagomes has quit IRC16:03
*** tesseract has quit IRC16:06
*** hyang has quit IRC16:08
*** hamalq has joined #openstack-neutron16:12
*** dtantsur is now known as dtantsur|afk16:18
*** ociuhandu has quit IRC16:48
njohnstonfrickler: Hi, are you a good person to help triage a bug that looks to be against neutron-dynamic-routing?  https://bugs.launchpad.net/neutron/+bug/189863416:49
openstackLaunchpad bug 1898634 in neutron "BGP peer is not working" [Undecided,New]16:49
*** jpena is now known as jpena|off17:16
openstackgerritTom Stappaerts proposed openstack/neutron-tempest-plugin master: QOS DSCP marking traffic test  https://review.opendev.org/75636917:19
openstackgerritTom Stappaerts proposed openstack/neutron-tempest-plugin master: QOS DSCP marking traffic test  https://review.opendev.org/75636917:20
*** hyang has joined #openstack-neutron17:35
*** hyang has quit IRC17:44
*** ralonsoh has quit IRC17:52
openstackgerritLajos Katona proposed openstack/networking-odl stable/train: This change is a squash of 2 different patches to pass the gate  https://review.opendev.org/74580617:58
openstackgerritLajos Katona proposed openstack/neutron master: Remove left over code for admin_state_down_before_update  https://review.opendev.org/74490118:02
*** lajoskatona has quit IRC18:02
sean-k-mooneynjohnston: can you add https://review.opendev.org/#/c/753314/ to your review list18:16
njohnstonsean-k-mooney: will do!18:25
openstackgerritMerged openstack/neutron-tempest-plugin master: Wait longer time for FIP to be updated by L3 agent  https://review.opendev.org/75531418:26
openstackgerritBrian Haley proposed openstack/neutron master: Update ovn-octavia-provider contact  https://review.opendev.org/75637618:29
*** viks____ has quit IRC18:35
*** ircuser-1 has quit IRC18:37
openstackgerritBrian Haley proposed openstack/ovn-octavia-provider master: DNM: debug q-qos failure  https://review.opendev.org/75637818:43
*** gcheresh has joined #openstack-neutron19:00
fricklernjohnston: commented, but I assume that that will rather need some DB expert to take a look at19:29
njohnstonthanks frickler19:45
*** ociuhandu has joined #openstack-neutron19:56
*** gcheresh has quit IRC20:04
*** ociuhandu has quit IRC20:08
*** zzzeek has quit IRC20:15
*** zzzeek has joined #openstack-neutron20:16
*** nweinber has quit IRC20:20
*** nweinber_ has joined #openstack-neutron20:21
*** nweinber_ has quit IRC20:26
openstackgerritBrian Haley proposed openstack/ovn-octavia-provider master: DNM: debug qos-fip failure  https://review.opendev.org/75637820:55
*** raildo has quit IRC21:00
*** ircuser-1 has joined #openstack-neutron21:04
*** slaweq has quit IRC21:07
*** slaweq has joined #openstack-neutron21:12
*** slaweq has quit IRC21:44
openstackgerritMoshe Levi proposed openstack/os-vif stable/victoria: Refactor code of linux_net to more cleaner and increase performace  https://review.opendev.org/75640621:58
*** macz_ has quit IRC22:33
*** rcernin_ has joined #openstack-neutron22:56
*** rcernin_ has quit IRC22:58
*** rcernin has joined #openstack-neutron22:58
openstackgerritMerged openstack/neutron master: Filter out port with invalid ofport in OVS firewall  https://review.opendev.org/75331423:01
*** mlavalle has quit IRC23:03
openstackgerritMerged openstack/neutron-specs master: Port NUMA affinity policy, revisit  https://review.opendev.org/74411623:10
« Monday, 2020-10-05 Index Wednesday, 2020-10-07 »

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