| jizaymes | 2024.2 Dalmation. | 01:34 |
|---|---|---|
| opendevreview | Merged openstack/neutron master: [eventlet-removal] Replace ``eventlet.spawn_n`` usage https://review.opendev.org/c/openstack/neutron/+/938541 | 02:14 |
| opendevreview | Miro Tomaska proposed openstack/neutron master: Retreive IPs from the external_ids column https://review.opendev.org/c/openstack/neutron/+/938767 | 03:04 |
| opendevreview | Miro Tomaska proposed openstack/neutron master: [Trivial] When logging use port logical_port name instead of uuid https://review.opendev.org/c/openstack/neutron/+/942970 | 03:04 |
| opendevreview | yatin proposed openstack/neutron master: [OVN] Add option to allow configuring dns ovn-owned https://review.opendev.org/c/openstack/neutron/+/942373 | 04:48 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: DNM - Test errors in ``neutron-functional-with-pyroute2-master`` https://review.opendev.org/c/openstack/neutron/+/942859 | 06:51 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: DNM - Test errors in ``neutron-functional-with-pyroute2-master`` https://review.opendev.org/c/openstack/neutron/+/942695 | 06:52 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: DNM == WIP == [eventlet-removal] Remove eventlet in the Metadata Agent https://review.opendev.org/c/openstack/neutron/+/942916 | 07:48 |
| opendevreview | Vasyl Saienko proposed openstack/neutron master: Enable atop by for dsvm jobs https://review.opendev.org/c/openstack/neutron/+/942792 | 07:56 |
| lajoskatona | ralonsoh: Hi, quick question, I started my OVN based devstack env and I have ugly tracebacks in neutron-api.log, like: | 08:24 |
| lajoskatona | Feb 27 12:23:17 myowen devstack@neutron-api.service[158220]: File "/opt/stack/data/venv/lib/python3.10/site-packages/oslo_messaging/_drivers/pool.py", line 70, in put | 08:24 |
| lajoskatona | Feb 27 12:23:17 myowen devstack@neutron-api.service[158220]: self._cond.notify() | 08:24 |
| lajoskatona | Feb 27 12:23:17 myowen devstack@neutron-api.service[158220]: File "/usr/lib/python3.10/threading.py", line 370, in notify | 08:24 |
| lajoskatona | Feb 27 12:23:17 myowen devstack@neutron-api.service[158220]: raise RuntimeError("cannot notify on un-acquired lock") | 08:24 |
| lajoskatona | I beleive I have to pull in master from oslo_messaging, or all oslo libs? | 08:24 |
| ralonsoh | lajoskatona, do you have the latest version? | 08:25 |
| lajoskatona | ralnsoh: from neutron? | 08:25 |
| ralonsoh | yes | 08:25 |
| lajoskatona | yes, I used master | 08:26 |
| ralonsoh | lajoskatona, do you have RPC workers or RPC state report workers? | 08:26 |
| lajoskatona | I have neutron-rpc-server and neutron-periodic-workers & neutron-ovn-maintenance-worker from the process list | 08:29 |
| ralonsoh | lajoskatona, but what is the value of rpc_state_report_workers? | 08:31 |
| ralonsoh | is this a devstack env? | 08:31 |
| ralonsoh | give me one sec because now I'm with a ml2/ovs env (implementing the metadata stuff) | 08:32 |
| lajoskatona | yes, devstack, I see now in neutron.conf, rpc_state_report_workers is 0 | 08:32 |
| lajoskatona | as I see in zuul jobs, rpc_state_report_workers there is also 0, and api_workers is 2 for example in tempest job as in my env | 08:34 |
| ralonsoh | lajoskatona, let me re-stack my env | 08:34 |
| ralonsoh | with the latest code | 08:34 |
| ralonsoh | I remember seeing that but because I was still using eventlet | 08:35 |
| ralonsoh | check if you see this message | 08:35 |
| ralonsoh | one sec | 08:35 |
| ralonsoh | This program is using eventlet and has been monkey_patched | 08:35 |
| ralonsoh | ^ do you see it in the API logs? | 08:35 |
| lajoskatona | ralonsoh: you are right I have eventlet monkey patched | 08:38 |
| ralonsoh | hmmmm is that a fresh env? | 08:39 |
| ralonsoh | if not, try to execute clean.sh | 08:39 |
| lajoskatona | yes, but I rerun stack.sh and force it to pull oslo libs also | 08:39 |
| ralonsoh | and then use RECLONE=True and PIP_UPGRADE=True | 08:40 |
| lajoskatona | ralonsoh: ack, let me rerun it after clean | 08:40 |
| ralonsoh | perfect | 08:40 |
| opendevreview | Lajos Katona proposed openstack/neutron master: If OVS Manager creation failes retry to set values https://review.opendev.org/c/openstack/neutron/+/939117 | 09:35 |
| ralonsoh | lajoskatona, ahhhhhh | 09:58 |
| ralonsoh | I know what is happening | 09:58 |
| ralonsoh | I think you have an "old" (some days) version of devstack | 09:59 |
| ralonsoh | that is including, by default, this: | 09:59 |
| ralonsoh | Loaded mechanism driver names: ['ovn', 'logger'] | 09:59 |
| ralonsoh | remove "logger" from the mech driver list in the config | 09:59 |
| ralonsoh | (I used another OVN system and re-stacked again, having the same problem) | 09:59 |
| ralonsoh | and please, rebease devstack to the latest commit too | 10:00 |
| lajoskatona | ralonsoh: let me check my local.conf | 10:13 |
| lajoskatona | ralonsoh: ack, I fetch devstack master to be sure and removed logger, let's see | 10:15 |
| ralonsoh | slaweq, no rush (these patches are not small) | 11:12 |
| ralonsoh | https://review.opendev.org/c/openstack/neutron/+/941922 and the other 2 on top of this one | 11:12 |
| ralonsoh | whenever you have time | 11:12 |
| ralonsoh | thanks! | 11:12 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: [eventlet-removal] Remove eventlet in the Metadata Agent https://review.opendev.org/c/openstack/neutron/+/942916 | 11:21 |
| opendevreview | Merged openstack/neutron-tempest-plugin master: Redefine ``configure_vlan_transparent`` method again https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/942776 | 11:45 |
| slaweq | ralonsoh sure, I will review them after lunch | 11:56 |
| ralonsoh | cool! | 12:13 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: [eventlet-removal] Remove eventlet in the Metadata Agent https://review.opendev.org/c/openstack/neutron/+/942916 | 12:20 |
| ralonsoh | lajoskatona, I've approved https://review.opendev.org/c/openstack/neutron/+/942393 | 12:21 |
| ralonsoh | can you check the upper patch? | 12:21 |
| ralonsoh | https://review.opendev.org/c/openstack/neutron/+/942530 | 12:21 |
| lajoskatona | ralonsoh: checking | 12:27 |
| lajoskatona | ralonsoh: I pushed a wip or sriov agent: https://review.opendev.org/c/openstack/neutron/+/942884, but I need some more time to test it | 12:28 |
| ralonsoh | lajoskatona, let me check, the problem is that this agent (if I'm not wrong) is usign oslo.service to spawn the agent | 12:30 |
| ralonsoh | let me check | 12:30 |
| lajoskatona | ralonsoh: possible that there's still something I missed, but all the eventlet lines disappeared from the logs at least and the reporting worked as expected, but I have to find a way to test end-to-end | 12:33 |
| ralonsoh | lajoskatona, this is a long shot (I still didn' | 12:36 |
| ralonsoh | didn't have time to test it | 12:37 |
| ralonsoh | it is possible to create a VM, passing a physical NIC to this VM acting as a SRIOV interface | 12:37 |
| ralonsoh | so it could be possible to have a VM running openstack with a SRIOV capable interface in your own laptop | 12:38 |
| lajoskatona | ralonsoh: yeah, I have a vm like that, and recently I tried this one from gibi: https://github.com/gibizer/vagrant-devstack which automates the whole thing | 12:40 |
| ralonsoh | lajoskatona, ok, I think this agent is not using oslo.service | 12:44 |
| ralonsoh | so your patch should be enough | 12:44 |
| opendevreview | Rodolfo Alonso proposed openstack/neutron master: [eventlet-removal] Remove eventlet in the Metadata Agent https://review.opendev.org/c/openstack/neutron/+/942916 | 12:47 |
| opendevreview | Lajos Katona proposed openstack/tap-as-a-service master: Doc: fix devstack plugin location in README https://review.opendev.org/c/openstack/tap-as-a-service/+/943002 | 14:16 |
| opendevreview | Lajos Katona proposed openstack/neutron-tempest-plugin master: Tap Mirror API and scenario tests https://review.opendev.org/c/openstack/neutron-tempest-plugin/+/886004 | 14:17 |
| opendevreview | Takashi Kajinami proposed openstack/ovn-bgp-agent master: Validate ovn/ovs db connection options https://review.opendev.org/c/openstack/ovn-bgp-agent/+/943003 | 14:21 |
| opendevreview | Takashi Kajinami proposed openstack/neutron master: Validate ovn_nb/sb_connection in config parser https://review.opendev.org/c/openstack/neutron/+/942589 | 14:21 |
| opendevreview | Takashi Kajinami proposed openstack/neutron master: Validate ovn_nb/sb_connection in config parser https://review.opendev.org/c/openstack/neutron/+/942589 | 14:56 |
| opendevreview | Takashi Kajinami proposed openstack/neutron master: Use openvswitch interface driver by default https://review.opendev.org/c/openstack/neutron/+/942161 | 15:02 |
| opendevreview | Merged openstack/networking-bagpipe unmaintained/2023.1: [CI] Remove periodic stable jobs from unmaintained branch https://review.opendev.org/c/openstack/networking-bagpipe/+/942584 | 15:28 |
| opendevreview | Merged openstack/neutron master: Update the resource request in custom made member actions https://review.opendev.org/c/openstack/neutron/+/941922 | 15:33 |
| opendevreview | Merged openstack/neutron master: Policy validator "type:list_of_dict*" also contains sub attributes https://review.opendev.org/c/openstack/neutron/+/942079 | 15:34 |
| jamesdenton_ | johnsom are you aware of any performance testing writeups for Octavia using amphora? | 15:40 |
| opendevreview | Elod Illes proposed openstack/networking-bagpipe unmaintained/zed: [CI] Remove periodic stable jobs from unmaintained branch https://review.opendev.org/c/openstack/networking-bagpipe/+/943020 | 15:48 |
| opendevreview | Takashi Kajinami proposed openstack/python-neutronclient master: Remove unused iso8601 https://review.opendev.org/c/openstack/python-neutronclient/+/943027 | 16:11 |
| opendevreview | Takashi Kajinami proposed openstack/python-neutronclient master: Remove unnecessary direct dependency on iso8601 https://review.opendev.org/c/openstack/python-neutronclient/+/943027 | 16:11 |
| johnsom | jamesdenton_ Sadly, no. The last benchmark I did was when I was back at Rackspace years ago. Currently I don't have access to hardware that has enough networking capacity to actually stress it. | 16:37 |
| opendevreview | Merged openstack/neutron master: [eventlet-removal] Remove eventlet from DHCP agent https://review.opendev.org/c/openstack/neutron/+/942393 | 16:48 |
| frickler | I've seen this failure twice today in reqs CI, is this a known unstable issue? https://zuul.opendev.org/t/openstack/build/6fb3e27b3cec415bb4d08cdfab6a1a03 | 17:32 |
| haleyb | frickler: i have not seen that one | 18:38 |
| frickler | haleyb: seems this has recently been updated by the eventlet removal, maybe this has made it racey somehow https://review.opendev.org/c/openstack/neutron/+/939095/4/neutron/tests/unit/notifiers/test_batch_notifier.py | 18:42 |
| haleyb | frickler: it probably is related to eventlet, we'll have to raise it at our monday CI meeting | 19:11 |
| opendevreview | Merged openstack/networking-bagpipe unmaintained/2023.1: Update .gitreview for unmaintained/2023.1 https://review.opendev.org/c/openstack/networking-bagpipe/+/935083 | 19:37 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!