VMware NSX: instance launch failed: TooManyExternalNetworks: More than one external network exists

The "TooManyExternalNetworks" exception in neutron looks suspicious. It's thrown here:

https://github.com/openstack/neutron/blob/master/neutron/db/external_net_db.py#L150

It's just a guess, but could there be a configuration failure? Do you maybe have more than one external network?

Looks like a configuration issue. Maybe one of the vmware neutron folks can help out?

hello,

Thank you for your mail back, and I inform you that I have only one external
network, and there is a configuration at neutron file for integration vmware
esxi and OpenStack juno

2015-09-18 8:54 GMT+01:00 Markus Zoeller (markus_z) <email address hidden>:

> The "TooManyExternalNetworks" exception in neutron looks suspicious.
> It's thrown here:
>
>
> https://github.com/openstack/neutron/blob/master/neutron/db/external_net_db.py#L150
>
> It's just a guess, but could there be a configuration failure? Do you
> maybe have more than one external network?
>
> --
> You received this bug notification because you are subscribed to the bug
> report.
> https://bugs.launchpad.net/bugs/1496929
>
> Title:
> instance luanch failed
>
> Status in OpenStack Compute (nova):
> New
>
> Bug description:
> Hello, I followed the documentation " http://docs.openstack.org/kilo
> /config-reference/content/vmware.html " to connect ESXi with OpenStack
> Juno, i put the following configuration on the compute node, nova.conf
> file :
>
> [DEFAULT]
> compute_driver=vmwareapi.VMwareVCDriver
>
> [vmware]
> host_ip=<vCenter host IP>
> host_username=<vCenter username>
> host_password=<vCenter password>
> cluster_name=<vCenter cluster name>
> datastore_regex=<optional datastore regex>
>
> And in the nova-compute.conf :
>
> [DEFAULT]
> compute_driver=vmwareapi.VMwareVCDriver
>
>
> But in vain, on the juno OpenStack Dashboard when i whant to launch
> instance, i have error " Error: Failed to launch instance "Test": Please
> try again later [Error: No valid host was found. ]. ", plz there is an idea
> for launce instance in my ESXi.
>
> attached the logs on the controller and compute node:
>
> ======> nova-conductor
>
> ERROR nova.scheduler.utils [req-618d4ee3-c936-4249-9f8c-7c266d5f9264
> None] [instance: 0c1ee287-edfe-4258-bb43-db23338bbe90] Error from last
> host: ComputeNode (node domain-c65(Compute)): [u'Traceback (most recent
> call last):\n', u' File
> "/usr/lib/python2.7/dist-packages/nova/compute/manager.py", line 2054, in
> _do_build_and_run_instance\n filter_properties)\n', u' File
> "/usr/lib/python2.7/dist-packages/nova/compute/manager.py", line 2185, in
> _build_and_run_instance\n instance_uuid=instance.uuid,
> reason=six.text_type(e))\n', u'RescheduledException: Build of instance
> 0c1ee287-edfe-4258-bb43-db23338bbe90 was re-scheduled: Network could not be
> found for bridge br-int\n']
> 2015-09-17 15:31:34.921 2432 WARNING nova.scheduler.driver
> [req-618d4ee3-c936-4249-9f8c-7c266d5f9264 None] [instance:
> 0c1ee287-edfe-4258-bb43-db23338bbe90] NoValidHost exception with message:
> 'No valid host was found.'
>
>
> =====> neutron
> 2015-09-17 12:36:09.398 1840 ERROR oslo.messaging._drivers.common
> [req-775407a3-d756-4677-bdb9-0ddfe2fac50c ] Returning exception More than
> one external network exists to caller
> 2015-09-17 12:36:09.398 1840 ERROR oslo.messaging._drivers.common
> [req-775407a3-d756-4677-bdb9-0ddfe2fac50c ] ['Traceback (most recent call
> last):\n', ' File
> "/usr/lib/python2.7/dist-packages/oslo/messaging/rpc/dispatcher.py", line
> 134, in _dispatch_and_reply\n incoming.message))\n',...

What network mode you are using? nova-network or neutron ?
If you use nova-network , neutron config is invalid. Or you are using neutron, you should config nsx_plugin as core plugin.

i use neutron, Please, how i can implement nsx_plugin as core plugin, you
can help me on this.

2015-09-18 12:51 GMT+01:00 Dongcan Ye <email address hidden>:

> What network mode you are using? nova-network or neutron ?
> If you use nova-network , neutron config is invalid. Or you are using
> neutron, you should config nsx_plugin as core plugin.
>
> --
> You received this bug notification because you are subscribed to the bug
> report.
> https://bugs.launchpad.net/bugs/1496929
>
> Title:
> instance launch failed: TooManyExternalNetworks: More than one
> external network exists
>
> Status in neutron:
> New
> Status in OpenStack Compute (nova):
> New
>
> Bug description:
> Hello, I followed the documentation " http://docs.openstack.org/kilo
> /config-reference/content/vmware.html " to connect ESXi with OpenStack
> Juno, i put the following configuration on the compute node, nova.conf
> file :
>
> [DEFAULT]
> compute_driver=vmwareapi.VMwareVCDriver
>
> [vmware]
> host_ip=<vCenter host IP>
> host_username=<vCenter username>
> host_password=<vCenter password>
> cluster_name=<vCenter cluster name>
> datastore_regex=<optional datastore regex>
>
> And in the nova-compute.conf :
>
> [DEFAULT]
> compute_driver=vmwareapi.VMwareVCDriver
>
>
> But in vain, on the juno OpenStack Dashboard when i whant to launch
> instance, i have error " Error: Failed to launch instance "Test": Please
> try again later [Error: No valid host was found. ]. ", plz there is an idea
> for launce instance in my ESXi.
>
> attached the logs on the controller and compute node:
>
> ======> nova-conductor
>
> ERROR nova.scheduler.utils [req-618d4ee3-c936-4249-9f8c-7c266d5f9264
> None] [instance: 0c1ee287-edfe-4258-bb43-db23338bbe90] Error from last
> host: ComputeNode (node domain-c65(Compute)): [u'Traceback (most recent
> call last):\n', u' File
> "/usr/lib/python2.7/dist-packages/nova/compute/manager.py", line 2054, in
> _do_build_and_run_instance\n filter_properties)\n', u' File
> "/usr/lib/python2.7/dist-packages/nova/compute/manager.py", line 2185, in
> _build_and_run_instance\n instance_uuid=instance.uuid,
> reason=six.text_type(e))\n', u'RescheduledException: Build of instance
> 0c1ee287-edfe-4258-bb43-db23338bbe90 was re-scheduled: Network could not be
> found for bridge br-int\n']
> 2015-09-17 15:31:34.921 2432 WARNING nova.scheduler.driver
> [req-618d4ee3-c936-4249-9f8c-7c266d5f9264 None] [instance:
> 0c1ee287-edfe-4258-bb43-db23338bbe90] NoValidHost exception with message:
> 'No valid host was found.'
>
>
> =====> neutron
> 2015-09-17 12:36:09.398 1840 ERROR oslo.messaging._drivers.common
> [req-775407a3-d756-4677-bdb9-0ddfe2fac50c ] Returning exception More than
> one external network exists to caller
> 2015-09-17 12:36:09.398 1840 ERROR oslo.messaging._drivers.common
> [req-775407a3-d756-4677-bdb9-0ddfe2fac50c ] ['Traceback (most recent call
> last):\n', ' File
> "/usr/lib/python2.7/dist-packages/oslo/messaging/rpc/dispatcher.py", line
> 134, in _dispatch_and_reply\n incoming.message))\n', ' File
> "/usr/lib/python2.7/dist-packages/oslo/messaging/rpc/dispatcher.py", line
> 177, in _dispatch\n retu...

i have the both network mode in my openstack plateforme, how i can view the
mode used

2015-09-18 14:35 GMT+01:00 Mrhari El Mehdi <email address hidden>:

> i use neutron, Please, how i can implement nsx_plugin as core plugin, you
> can help me on this.
>
> 2015-09-18 12:51 GMT+01:00 Dongcan Ye <email address hidden>:
>
>> What network mode you are using? nova-network or neutron ?
>> If you use nova-network , neutron config is invalid. Or you are using
>> neutron, you should config nsx_plugin as core plugin.
>>
>> --
>> You received this bug notification because you are subscribed to the bug
>> report.
>> https://bugs.launchpad.net/bugs/1496929
>>
>> Title:
>> instance launch failed: TooManyExternalNetworks: More than one
>> external network exists
>>
>> Status in neutron:
>> New
>> Status in OpenStack Compute (nova):
>> New
>>
>> Bug description:
>> Hello, I followed the documentation " http://docs.openstack.org/kilo
>> /config-reference/content/vmware.html " to connect ESXi with OpenStack
>> Juno, i put the following configuration on the compute node, nova.conf
>> file :
>>
>> [DEFAULT]
>> compute_driver=vmwareapi.VMwareVCDriver
>>
>> [vmware]
>> host_ip=<vCenter host IP>
>> host_username=<vCenter username>
>> host_password=<vCenter password>
>> cluster_name=<vCenter cluster name>
>> datastore_regex=<optional datastore regex>
>>
>> And in the nova-compute.conf :
>>
>> [DEFAULT]
>> compute_driver=vmwareapi.VMwareVCDriver
>>
>>
>> But in vain, on the juno OpenStack Dashboard when i whant to launch
>> instance, i have error " Error: Failed to launch instance "Test": Please
>> try again later [Error: No valid host was found. ]. ", plz there is an idea
>> for launce instance in my ESXi.
>>
>> attached the logs on the controller and compute node:
>>
>> ======> nova-conductor
>>
>> ERROR nova.scheduler.utils [req-618d4ee3-c936-4249-9f8c-7c266d5f9264
>> None] [instance: 0c1ee287-edfe-4258-bb43-db23338bbe90] Error from last
>> host: ComputeNode (node domain-c65(Compute)): [u'Traceback (most recent
>> call last):\n', u' File
>> "/usr/lib/python2.7/dist-packages/nova/compute/manager.py", line 2054, in
>> _do_build_and_run_instance\n filter_properties)\n', u' File
>> "/usr/lib/python2.7/dist-packages/nova/compute/manager.py", line 2185, in
>> _build_and_run_instance\n instance_uuid=instance.uuid,
>> reason=six.text_type(e))\n', u'RescheduledException: Build of instance
>> 0c1ee287-edfe-4258-bb43-db23338bbe90 was re-scheduled: Network could not be
>> found for bridge br-int\n']
>> 2015-09-17 15:31:34.921 2432 WARNING nova.scheduler.driver
>> [req-618d4ee3-c936-4249-9f8c-7c266d5f9264 None] [instance:
>> 0c1ee287-edfe-4258-bb43-db23338bbe90] NoValidHost exception with message:
>> 'No valid host was found.'
>>
>>
>> =====> neutron
>> 2015-09-17 12:36:09.398 1840 ERROR oslo.messaging._drivers.common
>> [req-775407a3-d756-4677-bdb9-0ddfe2fac50c ] Returning exception More than
>> one external network exists to caller
>> 2015-09-17 12:36:09.398 1840 ERROR oslo.messaging._drivers.common
>> [req-775407a3-d756-4677-bdb9-0ddfe2fac50c ] ['Traceback (most recent call
>> last):\n', ' File
>> "/usr/...

You can make a confirmation that what component in the VMware side you are using.

If you using NSX, you can use neutron as network backend. It's configuration can refer:
http://docs.openstack.org/admin-guide-cloud/networking_arch.html
http://docs.openstack.org/admin-guide-cloud/networking_config-plugins.html#configure-nsx-mh-plug-in

Another solution is nova-network, this can drive ESXi standard vSwitch and distributed vSwitch.

I removed a tag which doesn't has a subteam which watches for them.

I'd suggest reaching out to Gary Kotton (garyk on freenode irc) or the #openstack-vmware irc channel in freenode about this, I suspect it's a configuration issue.

Also, what version of each component (nova, neutron, vmware) are you running?

Oh I guess the description says Juno, I doubt if a lot of the neutron vmware parts work with nova in Juno, you'd have to ask Gary Kotton.

Also, Juno is basically end of life now so can you recreate this with Liberty?

@El Mehd:
This bug report was opened against Juno (which has reached its EOL) [1].
I'm going to deprecate it with "invalid". If the issue arises again in
master code or a supported stable release, just reopen the report by
setting it to "New".

References:
[1] http://releases.openstack.org/