New Juno install - Unable to boot instance, neutron client auth issue

asked 2016-10-28 17:49:50 -0500

czykz gravatar image

updated 2016-11-02 17:05:10 -0500

Hoping I can get some help, been stuck here for a while. After a new install of Juno, all services seem to be working, except for the process of booting an instance. Neutron commands themselves work, and I able to create a gateway, ext-net and internal net without issues.

However, booting an instance ends with: Error: Failed to launch instance "test": Please try again later [Error: No valid host was found. ]. Which is generally a compute node resource issue but I seem to have plenty.

It looks like the scheduler is finding the node and not filtering it out, I don't see any errors on there:

2016-11-02 20:32:06.804 11962 DEBUG nova.filters [req-6ac35dde-af5f-4c8f-8e24-13f180ce79a7 None] Filter RetryFilter returned 1 host(s) get_filtered_objects /usr/lib/python2.7/dist-packages/nova/filters.py:88
2016-11-02 20:32:06.812 11962 DEBUG nova.filters [req-6ac35dde-af5f-4c8f-8e24-13f180ce79a7 None] Filter AvailabilityZoneFilter returned 1 host(s) get_filtered_objects /usr/lib/python2.7/dist-packages/nova/filters.py:88
2016-11-02 20:32:06.813 11962 DEBUG nova.filters [req-6ac35dde-af5f-4c8f-8e24-13f180ce79a7 None] Filter RamFilter returned 1 host(s) get_filtered_objects /usr/lib/python2.7/dist-packages/nova/filters.py:88
2016-11-02 20:32:06.814 11962 DEBUG nova.filters [req-6ac35dde-af5f-4c8f-8e24-13f180ce79a7 None] Filter ComputeFilter returned 1 host(s) get_filtered_objects /usr/lib/python2.7/dist-packages/nova/filters.py:88
2016-11-02 20:32:06.814 11962 DEBUG nova.filters [req-6ac35dde-af5f-4c8f-8e24-13f180ce79a7 None] Filter ComputeCapabilitiesFilter returned 1 host(s) get_filtered_objects /usr/lib/python2.7/dist-packages/nova/filters.py:88
2016-11-02 20:32:06.814 11962 DEBUG nova.filters [req-6ac35dde-af5f-4c8f-8e24-13f180ce79a7 None] Filter ImagePropertiesFilter returned 1 host(s) get_filtered_objects /usr/lib/python2.7/dist-packages/nova/filters.py:88
2016-11-02 20:32:06.864 11962 INFO nova.openstack.common.periodic_task [req-6ac35dde-af5f-4c8f-8e24-13f180ce79a7 None] Skipping periodic task _periodic_update_dns because its interval is negative
2016-11-02 20:32:06.879 11962 DEBUG nova.filters [req-6ac35dde-af5f-4c8f-8e24-13f180ce79a7 None] Filter ServerGroupAntiAffinityFilter returned 1 host(s) get_filtered_objects /usr/lib/python2.7/dist-packages/nova/filters.py:88
2016-11-02 20:32:06.880 11962 DEBUG nova.filters [req-6ac35dde-af5f-4c8f-8e24-13f180ce79a7 None] Filter ServerGroupAffinityFilter returned 1 host(s) get_filtered_objects /usr/lib/python2.7/dist-packages/nova/filters.py:88
2016-11-02 20:32:06.881 11962 DEBUG nova.scheduler.filter_scheduler [req-6ac35dde-af5f-4c8f-8e24-13f180ce79a7 None] Filtered [(node4, node4) ram:29530 disk:427008 io_ops:0 instances:1] _schedule /usr/lib/python2.7/dist-packages/nova/scheduler/filter_scheduler.py:281
2016-11-02 20:32:06.882 11962 DEBUG nova.scheduler.filter_scheduler [req-6ac35dde-af5f-4c8f-8e24-13f180ce79a7 None] Weighed [WeighedHost [host: (node4, node4) ram:29530 disk:427008 io_ops:0 instances:1, weight: 1.0]] _schedule /usr/lib/python2.7/dist-packages/nova/scheduler/filter_scheduler.py:286

This are the error details for the instance:

File "/usr/lib/python2.7/dist-packages/nova/conductor/manager.py", line 616, in build_instances request_spec, filter_properties) File "/usr/lib/python2.7/dist-packages/nova/scheduler/client/__init__.py", line 49, in select_destinations context, request_spec, filter_properties) File "/usr/lib/python2.7/dist-packages/nova/scheduler/client/__init__.py", line 35, in __run_method return getattr(self.instance, __name)(*args, **kwargs) File "/usr/lib/python2.7/dist-packages/nova/scheduler/client/query.py", line 34, in select_destinations context, request_spec, filter_properties ...
(more)
edit retag flag offensive close merge delete

Comments

Juno is 2 years old and not supported anymore afaik. Better use a more recent version. As to your problem, I doubt it has anything to do with Neutron. When a VM is launched, the network is wired after scheduling. You should check nova-scheduler.log for filtering messages (DEBUG log level needed).

Bernd Bausch gravatar imageBernd Bausch ( 2016-11-01 07:11:39 -0500 )edit

Thanks, unfortunately i'm stuck with Juno for now as this is a dev environment for a currently live Juno one.

It looks like the scheduler is finding the node and not filtering it out, I don't see any errors on there. I'll add it to my question as well

czykz gravatar imageczykz ( 2016-11-02 15:59:10 -0500 )edit

I have Mitaka installed and in some point I received Error: No valid host was found. Personelly I didn't figure it out what was the core issue but did manage to get a way around it by simply using other user credentials than admin to launch an instance and it worked.

Ali84 gravatar imageAli84 ( 2016-11-02 23:35:47 -0500 )edit

Thanks, I tried a different user and tried creating a new user/tenant and am still getting the same results. Is there any way to drill down to see why nova scheduler would filter out compute nodes? It logs don't seem to provide much in the way of details.

czykz gravatar imageczykz ( 2016-11-03 11:25:09 -0500 )edit

Neutron authentication issue? Nova accesses Neutron via API, so like anybody else it needs to provide credentials. Normally, user name neutron, password whateveryouconfigured. These credentials are in nova.conf in the [neutron] section (at least in Mitaka). Might be a typo there.

Bernd Bausch gravatar imageBernd Bausch ( 2016-11-05 02:28:51 -0500 )edit