OSWinException: Duplicate resource name found [closed]
Hi,
We've recently migrated to a Windows 2016 Failover Cluster with Storage Spaces Direct. Backend is based on an Ocata packstack installation. The cluster has 3 win2016dc node.
So far, everything appears to be working as expected. However, when taking a look at the neutron's log file on one of the node, we're seeing the following errors, literally every 2-3 seconds. The neutron logfile keeps growing:
2017-05-25 11:31:13.686 3732 ERROR hyperv.neutron.hyperv_neutron_agent [req-df69e481-a1b7-46c6-9af0-825de349f020 - - - - -] Error in agent event loop
2017-05-25 11:31:13.686 3732 ERROR hyperv.neutron.hyperv_neutron_agent Traceback (most recent call last):
2017-05-25 11:31:13.686 3732 ERROR hyperv.neutron.hyperv_neutron_agent File "C:\PROGRA~1\CLOUDB~1\OPENST~1\Nova\Python27\lib\site-packages\hyperv\neutron\hyperv_neutron_agent.py", line 474, in daemon_loop
2017-05-25 11:31:13.686 3732 ERROR hyperv.neutron.hyperv_neutron_agent self._port_enable_control_metrics()
2017-05-25 11:31:13.686 3732 ERROR hyperv.neutron.hyperv_neutron_agent File "C:\PROGRA~1\CLOUDB~1\OPENST~1\Nova\Python27\lib\site-packages\hyperv\neutron\hyperv_neutron_agent.py", line 296, in _port_enable_control_metrics
2017-05-25 11:31:13.686 3732 ERROR hyperv.neutron.hyperv_neutron_agent self._metricsutils.enable_port_metrics_collection(port_id)
2017-05-25 11:31:13.686 3732 ERROR hyperv.neutron.hyperv_neutron_agent File "C:\PROGRA~1\CLOUDB~1\OPENST~1\Nova\Python27\lib\site-packages\os_win\utils\metrics\metricsutils.py", line 91, in enable_port_metrics_collection
2017-05-25 11:31:13.686 3732 ERROR hyperv.neutron.hyperv_neutron_agent port = self._get_switch_port(switch_port_name)
2017-05-25 11:31:13.686 3732 ERROR hyperv.neutron.hyperv_neutron_agent File "C:\PROGRA~1\CLOUDB~1\OPENST~1\Nova\Python27\lib\site-packages\os_win\utils\metrics\metricsutils.py", line 269, in _get_switch_port
2017-05-25 11:31:13.686 3732 ERROR hyperv.neutron.hyperv_neutron_agent return self._unique_result(ports, port_name)
2017-05-25 11:31:13.686 3732 ERROR hyperv.neutron.hyperv_neutron_agent File "C:\PROGRA~1\CLOUDB~1\OPENST~1\Nova\Python27\lib\site-packages\os_win\utils\metrics\metricsutils.py", line 283, in _unique_result
2017-05-25 11:31:13.686 3732 ERROR hyperv.neutron.hyperv_neutron_agent _('Duplicate resource name found: %s') % resource_name)
2017-05-25 11:31:13.686 3732 ERROR hyperv.neutron.hyperv_neutron_agent OSWinException: Duplicate resource name found: 3cef1e9b-9925-4926-83ae-e90d3824fc2b
We have tried to migrate the specific instance who has that port binded to a different node, but the error still shows up on the node where the instance has been migrated to.
I have verified and it doesn't appear to be the same "name" / "id" for the specified resource ("3cef1e9b-9925-4926-83ae-e90d3824fc2b"):
# neutron port-list|grep 3cef1e9b-9925-4926-83ae-e90d3824fc2b
| 3cef1e9b-9925-4926-83ae-e90d3824fc2b | | 171b50d4c3a84aa1bae27798788cd25f | fa:16:3e:5e:88:52 | {"subnet_id": "e5e8584d-6a31-4708-ae16-04d17be40fb8", "ip_address": "149.56.181.217"} |
Any ideas what might be wrong ?
Thanks!
Bump! We're still seeing these errors and they are quite filling up the log file. Thanks!
Hi! It's a very bizarre exception, IMO. It basically means that there are at least 2 switch ports with the same ElementName on the same host, which should not be possible. On the host on which you are getting this exception, can you run the following command in powershell:
gwmi -Namespace 'root/virtualization/v2' -Class Msvm_SyntheticEthernetPortSettingData | ? ElementName -eq 3cef1e9b-9925-4926-83ae-e90d3824fc2b There should only be one WMI object returned.
well well... that's very odd; the error hasn't been showing up since the last few days.. I'll close the issue for now and will re-open if needed. Thanks for the follow up ! :)