I encounter this error in a lightly complicated situation:
OS : ubuntu
Branch : Icehouse
When booting a vm with admin role and assigning it with another tenant's private network, it's fine to work.
But the update info cache action will started periodically, at this time, the privous mentioned other tenant's
private network is filtered out from the function "get_available_network()" if do not transfer the provisioned
net_ids parameter. Then bad things happened as below:
The worst affect is the vm can not be deleted normally.
2014-01-20 11:03:19.632 DEBUG nova.compute.manager [-] An error occurred from (pid=5521) _heal_instance_info_cache /opt/stack/nova/nova/compute/manager.py:4488
2014-01-20 11:03:19.632 TRACE nova.compute.manager Traceback (most recent call last):
2014-01-20 11:03:19.632 TRACE nova.compute.manager File "/opt/stack/nova/nova/compute/manager.py", line 4484, in _heal_instance_info_cache
2014-01-20 11:03:19.632 TRACE nova.compute.manager self._get_instance_nw_info(context, instance, use_slave=True)
2014-01-20 11:03:19.632 TRACE nova.compute.manager File "/opt/stack/nova/nova/compute/manager.py", line 891, in _get_instance_nw_info
2014-01-20 11:03:19.632 TRACE nova.compute.manager instance)
2014-01-20 11:03:19.632 TRACE nova.compute.manager File "/opt/stack/nova/nova/network/api.py", line 50, in wrapper
2014-01-20 11:03:19.632 TRACE nova.compute.manager res = f(self, context, *args, **kwargs)
2014-01-20 11:03:19.632 TRACE nova.compute.manager File "/opt/stack/nova/nova/network/neutronv2/api.py", line 439, in get_instance_nw_info
2014-01-20 11:03:19.632 TRACE nova.compute.manager result = self._get_instance_nw_info(context, instance, networks)
2014-01-20 11:03:19.632 TRACE nova.compute.manager File "/opt/stack/nova/nova/network/neutronv2/api.py", line 446, in _get_instance_nw_info
2014-01-20 11:03:19.632 TRACE nova.compute.manager nw_info = self._build_network_info_model(context, instance, networks)
2014-01-20 11:03:19.632 TRACE nova.compute.manager File "/opt/stack/nova/nova/network/neutronv2/api.py", line 1043, in _build_network_info_model
2014-01-20 11:03:19.632 TRACE nova.compute.manager subnets)
2014-01-20 11:03:19.632 TRACE nova.compute.manager File "/opt/stack/nova/nova/network/neutronv2/api.py", line 952, in _nw_info_build_network
2014-01-20 11:03:19.632 TRACE nova.compute.manager for net in networks:
2014-01-20 11:03:19.632 TRACE nova.compute.manager UnboundLocalError: local variable 'network_name' referenced before assignment
Hi team,
I encounter this error in a lightly complicated situation:
OS : ubuntu
Branch : Icehouse
When booting a vm with admin role and assigning it with another tenant's private network, it's fine to work. network( )" if do not transfer the provisioned
But the update info cache action will started periodically, at this time, the privous mentioned other tenant's
private network is filtered out from the function "get_available_
net_ids parameter. Then bad things happened as below:
The worst affect is the vm can not be deleted normally.
2014-01-20 11:03:19.632 DEBUG nova.compute. manager [-] An error occurred from (pid=5521) _heal_instance_ info_cache /opt/stack/ nova/nova/ compute/ manager. py:4488 manager Traceback (most recent call last): manager File "/opt/stack/ nova/nova/ compute/ manager. py", line 4484, in _heal_instance_ info_cache manager self._get_ instance_ nw_info( context, instance, use_slave=True) manager File "/opt/stack/ nova/nova/ compute/ manager. py", line 891, in _get_instance_ nw_info manager instance) manager File "/opt/stack/ nova/nova/ network/ api.py" , line 50, in wrapper manager res = f(self, context, *args, **kwargs) manager File "/opt/stack/ nova/nova/ network/ neutronv2/ api.py" , line 439, in get_instance_ nw_info manager result = self._get_ instance_ nw_info( context, instance, networks) manager File "/opt/stack/ nova/nova/ network/ neutronv2/ api.py" , line 446, in _get_instance_ nw_info manager nw_info = self._build_ network_ info_model( context, instance, networks) manager File "/opt/stack/ nova/nova/ network/ neutronv2/ api.py" , line 1043, in _build_ network_ info_model manager subnets) manager File "/opt/stack/ nova/nova/ network/ neutronv2/ api.py" , line 952, in _nw_info_ build_network manager for net in networks: manager UnboundLocalError: local variable 'network_name' referenced before assignment
2014-01-20 11:03:19.632 TRACE nova.compute.
2014-01-20 11:03:19.632 TRACE nova.compute.
2014-01-20 11:03:19.632 TRACE nova.compute.
2014-01-20 11:03:19.632 TRACE nova.compute.
2014-01-20 11:03:19.632 TRACE nova.compute.
2014-01-20 11:03:19.632 TRACE nova.compute.
2014-01-20 11:03:19.632 TRACE nova.compute.
2014-01-20 11:03:19.632 TRACE nova.compute.
2014-01-20 11:03:19.632 TRACE nova.compute.
2014-01-20 11:03:19.632 TRACE nova.compute.
2014-01-20 11:03:19.632 TRACE nova.compute.
2014-01-20 11:03:19.632 TRACE nova.compute.
2014-01-20 11:03:19.632 TRACE nova.compute.
2014-01-20 11:03:19.632 TRACE nova.compute.
2014-01-20 11:03:19.632 TRACE nova.compute.
2014-01-20 11:03:19.632 TRACE nova.compute.