| *** zhouhenglc is now known as Guest870 | 00:47 | |
| *** zhouhenglc_ is now known as zhouhenglc | 00:47 | |
| opendevreview | Brian Haley proposed openstack/neutron master: Remove import of 'imp' module https://review.opendev.org/c/openstack/neutron/+/842450 | 01:38 |
|---|---|---|
| *** zhouhenglc__ is now known as zhouhenglc | 01:52 | |
| *** zhouhenglc__ is now known as zhouhenglc | 03:22 | |
| lajoskatona | ykarel: Hi, regarding https://review.opendev.org/c/openstack/tap-as-a-service/+/842134 , as I see we can merge it as tempest passed, and if it is in the other patch https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/842113 can pass also | 04:36 |
| lajoskatona | ykarel: what do you think? | 04:37 |
| ykarel | lajoskatona, as taas patch has depends-on neutron-tempest-plugin patch we can't merge taas patch unless and until neutron-tempest-plugin is green | 04:39 |
| ykarel | if you mean merge with removing depends-on and making job non-voting in taas patch, then yes that can be done | 04:39 |
| ykarel | and revert non-voting change after merging neutron-tempest-plugin fix | 04:41 |
| lajoskatona | ykarel: ohh, I see now, sorry it's too early here.... | 04:43 |
| * ykarel_ network issue | 04:48 | |
| opendevreview | Lajos Katona proposed openstack/tap-as-a-service master: Add context for all SQL transactions https://review.opendev.org/c/openstack/tap-as-a-service/+/842134 | 04:49 |
| *** ykarel_ is now known as ykarel | 04:51 | |
| ykarel | lajoskatona, can drop job from gate pipeline too as it's non-voting now ^ | 04:51 |
| ykarel | and remove depends-on | 04:52 |
| opendevreview | yatin proposed openstack/tap-as-a-service master: Add context for all SQL transactions https://review.opendev.org/c/openstack/tap-as-a-service/+/842134 | 05:30 |
| opendevreview | yatin proposed openstack/neutron-tempest-plugin master: taas: use ssh_key_type from config for taas scenario https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/842113 | 05:36 |
| opendevreview | yatin proposed openstack/tap-as-a-service master: Add context for all SQL transactions https://review.opendev.org/c/openstack/tap-as-a-service/+/842134 | 05:38 |
| opendevreview | yatin proposed openstack/neutron-tempest-plugin master: Add irrelevant files for fwaas job https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/842460 | 05:42 |
| ykarel | lajoskatona, i updated the patches by switching depends-on | 05:43 |
| *** zhouhenglc is now known as Guest896 | 06:05 | |
| *** zhouhenglc_ is now known as zhouhenglc | 06:05 | |
| opendevreview | yatin proposed openstack/neutron-tempest-plugin master: Add irrelevant files for fwaas job https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/842460 | 06:12 |
| lajoskatona | ykarel: thanks, I am not that effective this morning :-) | 06:31 |
| ykarel | not a big deal, happen sometimes | 06:34 |
| ykarel | ralonsoh, hi, i was looking at osc job failures for segment tests shared by frickler and seems those are related to sqlalchemy 2.0 changes | 07:17 |
| ykarel | can you check and confirm | 07:17 |
| ykarel | https://storage.gra.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_d00/841389/3/check/osc-functional-devstack-tips/d0091ef/controller/logs/screen-q-svc.txt | 07:17 |
| ykarel | https://f45708043030637061d5-675a54f2682a71ce0a87525368caccbf.ssl.cf2.rackcdn.com/841389/3/check/osc-functional-devstack/14facfe/controller/logs/screen-q-svc.txt | 07:17 |
| ralonsoh | ykarel, so you have LP bug? | 07:18 |
| ralonsoh | just to know, if not, I'll open one | 07:18 |
| ykarel | ralonsoh, i don't think it was opened | 07:18 |
| ralonsoh | ok, I'll investigate those errors now | 07:19 |
| ykarel | Thanks | 07:19 |
| slaweq | lajoskatona: hi | 07:20 |
| ykarel | what i see from logs were for standardattributes table and it retries after Lock wait timeout exceeded; try restarting transaction | 07:20 |
| slaweq | lajoskatona: I just wanted to let You know that I will not be available on the tomorrow's drivers meeting | 07:20 |
| lajoskatona | slaweq: Hi, thanks | 07:21 |
| slaweq | we have recharge day in Red Hat so I will be mostly off for all day | 07:21 |
| ykarel | and ORM session: SQL execution without transaction in progress, traceback, which made me suspect those sql changes | 07:21 |
| slaweq | and during the drivers meeting I also have to go to the swimming pool with my kids | 07:21 |
| ralonsoh | ykarel, this is because I introduced this message to warn about this possible problem | 07:21 |
| lajoskatona | slaweq: that sounds good:-) Swimming pool day is today for me :-) | 07:22 |
| slaweq | :) | 07:22 |
| ralonsoh | dammit, I prefer swimming pool day rather than recharge day! | 07:22 |
| ykarel | ralonsoh, okk that's message is helpful for this | 07:23 |
| ralonsoh | ykarel, indeed there is an error, I'll push a patch right now | 07:25 |
| ykarel | ack | 07:25 |
| ralonsoh | in _notify_subnet_gateway_ip_update | 07:25 |
| opendevreview | Lajos Katona proposed openstack/networking-bagpipe master: py310: Add required-projects list to py310 job https://review.opendev.org/c/openstack/networking-bagpipe/+/841917 | 07:44 |
| lajoskatona | slaweq, ralonsoh: Hi, could you please check these bagpipe patches when you have some free time: https://review.opendev.org/q/project:openstack/networking-bagpipe+status:open+-age:17d+label:Verified%252B1 ? | 07:49 |
| ralonsoh | sure | 07:50 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: [sqlalchemy-20] Add missing DB contexts in L3 methods https://review.opendev.org/c/openstack/neutron/+/842468 | 07:50 |
| *** zhouhenglc_ is now known as zhouhenglc | 07:51 | |
| ykarel | frickler, fyi ^ fix for osc functional random failures | 07:58 |
| lajoskatona | slaweq, ralonsoh: regarding milestone-1 n-lib release: https://review.opendev.org/c/openstack/releases/+/842362 , do we need these patches or as we already have 2.21.0 for zed just wait with it and release next time when enough content is ready? | 08:09 |
| ralonsoh | lajoskatona, I don't think those patches are strictly needed | 08:10 |
| lajoskatona | ralonsoh: ok, thanks, that was my feeling also | 08:11 |
| ralonsoh | documentation and testing, no features or bugs | 08:11 |
| frickler | ykarel: ralonsoh: thanks | 08:12 |
| frickler | who wo be best to ask about RPC queue things? I have no clue how to review https://review.opendev.org/c/openstack/neutron-dynamic-routing/+/842383 | 08:13 |
| frickler | s/wo/would/, me needs coffee | 08:15 |
| ralonsoh | I'll try to review it | 08:15 |
| frickler | ralonsoh: that would be great, thanks again | 08:15 |
| opendevreview | Merged openstack/neutron stable/xena: [OVN] Handle missing acls during log removal https://review.opendev.org/c/openstack/neutron/+/841944 | 08:16 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: [DNM - testing FT] Skip "PortBindingChassisEvent" if revision number changes https://review.opendev.org/c/openstack/neutron/+/842147 | 08:23 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: Check the Chassis_Private nb_cfg_timestamp with current value https://review.opendev.org/c/openstack/neutron/+/842471 | 08:37 |
| slaweq | lajoskatona: sorry, I was in the meeting | 09:03 |
| slaweq | those neutron-lib patches are docs and CI related things, none of them needs to be released now | 09:03 |
| lajoskatona | slaweq: thanks, I abandon then the release patch for n-lib, as we don't really have much thing to release now and we already have one for milestone-1 | 09:09 |
| frickler | slaweq: btw., where can I find your gpg key? I'm always seeing "key not found" when reading your mails | 09:14 |
| opendevreview | Merged openstack/networking-bagpipe master: Update python testing as per zed cycle teting runtime https://review.opendev.org/c/openstack/networking-bagpipe/+/841536 | 09:26 |
| slaweq | frickler: should be here https://keyserver.ubuntu.com/pks/lookup?search=skaplons%40redhat.com&fingerprint=on&op=index | 09:36 |
| frickler | slaweq: ah, thx, somehow I had only imported the newer key, not the older one. works fine now | 09:42 |
| slaweq | frickler: great | 09:44 |
| frickler | lajoskatona: at the PTG if I understood correctly it was communicated that the BGP extension project has been canceled at redhat, can these patches all be abandoned, then? https://review.opendev.org/q/topic:bug%252F1921461 | 12:43 |
| slaweq | frickler: it wasn't abandoned at Red Hat but Ericsson | 12:48 |
| frickler | slaweq: ah, sorry. then I'll ask on the patches directly | 12:53 |
| slaweq | frickler: no worries, probably rubasov will be able to help You with this | 12:53 |
| rubasov | slaweq, frickler: yes, those patches will not be continued | 12:55 |
| rubasov | we reverted the related n-lib patches | 12:55 |
| rubasov | I'm not the owner, so cannot technically abandon the changes, but I guess slaweq's abandon script will do the job next time he or lajoskatona runs it | 12:57 |
| slaweq | rubasov: I usually run it | 12:57 |
| slaweq | and I can abandon those patches if that's needed | 12:57 |
| slaweq | but frickler probably should also be able to do so | 12:57 |
| rubasov | if they bother anybody, feel free to abandon them | 12:58 |
| frickler | rubasov: open patches that are not being worked upon anymore make it more difficult to focus the attention on active patches. I'll abandon and add a note | 13:04 |
| rubasov | thanks | 13:05 |
| *** dasm|off is now known as dasm | 13:21 | |
| frickler | slaweq: rubasov: I think https://review.opendev.org/c/openstack/neutron-specs/+/785349 could then be adandoned, too? also, should the specs that had already been merged be dealt with somehow? | 13:22 |
| rubasov | I believe the open spec can be abandoned too | 13:23 |
| rubasov | I would leave the merged spec around, that may be informative in historical contexts even if it was not implemented | 13:25 |
| hjensas | need some help with a failing ironic job, DHCP namespace and dnsmasq not started for network: 6e733d66-2d78-45cf-a649-77092c7e323f | 13:47 |
| hjensas | logs: https://b310562373a188973e4d-6d67d1055c684f7187f360a048e3d766.ssl.cf2.rackcdn.com/835157/5/check/networking-baremetal-multitenant-vlans/5eaf280/controller/logs/screen-q-dhcp.txt | 13:48 |
| hjensas | I can see _network_create, then _subnet_update called, but never see "Calling driver for network:" for this specific network. The other nets are ok. | 13:49 |
| hjensas | slaweq: ralonsoh: mind taking a look at ^? | 13:52 |
| slaweq | hjensas: sorry but I will be leaving in few minutes so I don't have time to check it now | 13:52 |
| ralonsoh | hjensas, let me check in a few mins | 13:53 |
| hjensas | slaweq: ok, enjoy your time off. :) | 13:53 |
| slaweq | hjensas: thx, I will be around in about 1h but I will have meeting | 13:54 |
| ralonsoh | hjensas, this is what you have for all ports in this network | 13:57 |
| ralonsoh | https://paste.opendev.org/show/bwjvuDmP4bs5Q2opN2yO/ | 13:57 |
| ralonsoh | but I don't see any networking_generic_switch logs | 13:57 |
| ralonsoh | well yes, those ports are bound | 13:58 |
| hjensas | ralonsoh: There is no port with device-owner network:dhcp for network 6e733d66-2d78-45cf-a649-77092c7e323f, but the subnet on that network is dhcp enabled. | 14:06 |
| ralonsoh | this port, if I recall correctly, is created by the DHCP agent | 14:07 |
| mnaser | lajoskatona: i am not able to +2 changes for neutron-vpnaas to stable branches, is that policy or maybe just missed acl? :) | 14:12 |
| ykarel | ralonsoh, slaweq can you check https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/842113 and it depends-on | 14:18 |
| ykarel | thx in advance | 14:19 |
| ykarel | and also https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/842460 | 14:19 |
| ralonsoh | ok | 14:20 |
| frickler | mnaser: stable branches usually have dedicated maintainer groups historically | 14:40 |
| frickler | yes, currently it contains only neutron-stable-maint. we created neutron-dynamic-routing-stable-maint in my case which is likely similar to yours. lajoskatona what do you think? | 14:42 |
| hjensas | ralonsoh: I added some additional debug logging in dhcp/agent.py - https://paste.opendev.org/show/bQbhsKQ24YxTxZlJ03lI/ | 15:27 |
| hjensas | ralonsoh: any(s for s in network.subnets if s.enable_dhcp) - condition is false. But the subnet have enable_dhcp: True. | 15:28 |
| ralonsoh | hjensas, are you deploying this in a routed provider networks environment? | 15:29 |
| hjensas | ralonsoh: yes | 15:29 |
| ralonsoh | hjensas, ok so this network has a subnet in other segment | 15:29 |
| ralonsoh | not in the one assigned to this host running the DHCP | 15:29 |
| hjensas | ralonsoh: https://paste.opendev.org/show/bUEjz2Au4aAqhT0EQY91/ <- segment is on "mynetwork", and bridge_mapping have that. | 15:33 |
| hjensas | ralonsoh: https://paste.opendev.org/show/bQZMmklC9PHlh40PdKX8/ | 15:35 |
| hjensas | agent is on host "devstack" and openvswitch agent configuration have bridge mapping to "mynetwork". | 15:36 |
| ralonsoh | hjensas, what version are you using? | 15:37 |
| ralonsoh | hjensas, more precisely, did you use devstack? What is the git hash of neutron? | 15:37 |
| hjensas | ralonsoh: commit 5c6eff7e15d7b5bf7f0f175293adcec5d953fd6b (HEAD -> master | 15:38 |
| hjensas | ralonsoh: yes, it is devstack. | 15:38 |
| ralonsoh | hjensas, that should work since https://github.com/openstack/neutron/commit/8fd88fd2239bd19794f2a1cf9480e5aa7d4f5a9b | 15:40 |
| ralonsoh | the problem seems to be in the server | 15:40 |
| ralonsoh | that is returning a network without any subnet | 15:40 |
| ralonsoh | when DHCP makes the RPC call to https://github.com/openstack/neutron/blob/master/neutron/api/rpc/handlers/dhcp_rpc.py#L175-L240 | 15:41 |
| ralonsoh | hjensas, can you log, in the neutron server, what L232 is returning? | 15:41 |
| hjensas | ralonsoh: sure | 15:42 |
| lajoskatona | mnaser, frickler: yes, I think we can create similar group for vpnaas | 15:43 |
| opendevreview | Merged openstack/neutron stable/wallaby: [OVN] Handle missing acls during log removal https://review.opendev.org/c/openstack/neutron/+/841945 | 15:49 |
| opendevreview | Merged openstack/neutron-tempest-plugin master: Add irrelevant files for fwaas job https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/842460 | 15:49 |
| hjensas | ralonsoh: https://paste.opendev.org/show/bJ5zE7npvA89HnC5qLnN/ | 15:50 |
| hjensas | 'subnets': [], 'non_local_subnets': [{'id': '0ae02029-0993-4bb3-b84b-952e67853d49' | 15:50 |
| ralonsoh | hjensas, this is why | 15:51 |
| ralonsoh | if not any(s for s in network.subnets if s.enable_dhcp): | 15:51 |
| ralonsoh | self.disable_dhcp_helper(network.id) | 15:51 |
| ralonsoh | return | 15:51 |
| ralonsoh | hjensas, one last time, please check the network segments | 15:53 |
| ralonsoh | because https://github.com/openstack/neutron/blob/master/neutron/api/rpc/handlers/dhcp_rpc.py#L223-L226 is populated in the other way around | 15:53 |
| hjensas | ralonsoh: but the subnet is supposed to be in subnets not in non_local_subnets since the host have the bridge_mapping to 'mynetwork'. | 15:53 |
| ralonsoh | hjensas, the subnet will be in "subnets" if the network segments have this host | 15:54 |
| ralonsoh | hjensas, the agent, when is created, triggers the execution of "_update_segment_host_mapping_for_agent" | 15:59 |
| ralonsoh | check for "Checking segment: " in the Neutron logs (in debug mode) | 16:00 |
| ralonsoh | that will report what agents where created and the mappings | 16:00 |
| hjensas | ralonsoh: https://paste.opendev.org/show/bBYdB3rjO0eTqsnU8puS/ and Checking segment log entries with filter on the segment_id - https://paste.opendev.org/show/bJidK2p4X8UoZSNHJd46/ | 16:08 |
| ralonsoh | hjensas, I don't know why "devstack" node is defined with the name and not the host ID | 16:11 |
| ralonsoh | this could be a problem | 16:11 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: Log when a segment is mapped/unmapped to a host https://review.opendev.org/c/openstack/neutron/+/842576 | 17:12 |
| ralonsoh | hjensas, ^^^ | 17:12 |
| ralonsoh | I'm leaving now, it's a bit late for me | 17:12 |
| mlavalle | lajoskatona: tomorrow all Red Hat employees have a day off. I don't think there will be quorum for the drivers meeting | 17:13 |
| hjensas | ralonsoh: ack, for me too. I'll add a change with depends-on and see how it goes, hopefully more logging may atleast give a clue about what is going on. | 17:14 |
| hjensas | ralonsoh: thank you! | 17:14 |
| ralonsoh | yw | 17:14 |
| lajoskatona | mlavalle: ok, thanks | 17:17 |
| lajoskatona | mlavalle: I should have counted on it, slaweq mentioned this morning that you have recharging day | 17:18 |
| mlavalle | lajoskatona: and that counts mtomaska, the person who submitted the RFE | 17:18 |
| opendevreview | Elvira García Ruiz proposed openstack/neutron master: [OVN] Log drop events per security group https://review.opendev.org/c/openstack/neutron/+/835014 | 17:18 |
| lajoskatona | mlavalle: I answered to your mail and postponed the meeting to next week | 17:19 |
| mlavalle | lajoskatona: thanks... hve a great weekend! | 17:19 |
| lajoskatona | mlavalle: thanks, the same to you :-) | 17:20 |
| mtomaska | thanks mlavalle. Indeed I wont be available tomorrow :) | 17:22 |
| mlavalle | mtomaska: btw, you don't need to attend the meeting to have the RFE discussed. We will discuss it whether you are there or not. It will be discussed next week | 17:22 |
| mlavalle | of course, you are welcome to attend | 17:23 |
| mtomaska | yep I know, but I would like to attend | 17:25 |
| opendevreview | Brian Haley proposed openstack/neutron master: Start using security-groups-shared-filtering from neutron-lib https://review.opendev.org/c/openstack/neutron/+/842581 | 17:34 |
| opendevreview | Merged openstack/neutron master: Remove "distutils" library https://review.opendev.org/c/openstack/neutron/+/842133 | 19:17 |
| opendevreview | Brian Haley proposed openstack/neutron master: Remove import of 'imp' module https://review.opendev.org/c/openstack/neutron/+/842450 | 19:18 |
| opendevreview | Damian Dąbrowski proposed openstack/neutron master: [L3-HA] Always keep qg- port up, but with a few tweaks to avoid issues https://review.opendev.org/c/openstack/neutron/+/839671 | 21:17 |
| *** dasm is now known as dasm|off | 21:26 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!