Description of problem: The services nova-api and the nova-metadata-api are both using the same port, 8775. Thus, the services are 'competing' for the port and one of them will not work. Version-Release number of selected component (if applicable): Red Hat Enterprise Linux Server release 6.5 Beta (Santiago) python-novaclient-2.15.0-1.el6ost.noarch python-nova-2013.2-0.24.rc1.el6ost.noarch openstack-nova-network-2013.2-0.24.rc1.el6ost.noarch openstack-nova-common-2013.2-0.24.rc1.el6ost.noarch openstack-nova-console-2013.2-0.24.rc1.el6ost.noarch openstack-nova-compute-2013.2-0.24.rc1.el6ost.noarch openstack-nova-conductor-2013.2-0.24.rc1.el6ost.noarch openstack-nova-novncproxy-2013.2-0.24.rc1.el6ost.noarch openstack-nova-scheduler-2013.2-0.24.rc1.el6ost.noarch openstack-nova-api-2013.2-0.24.rc1.el6ost.noarch openstack-nova-cert-2013.2-0.24.rc1.el6ost.noarch How reproducible: everytime Steps to Reproduce: 1. Install Havana on RHEL 6.5 AIO installation. 2. 3. Actual results: Either the openstack-nova-api or the openstack-nova-metadata-api service is down. Expected results: Both services are up and running. Additional info: The error from /var/log/nova/metadata-api.log: 2013-10-09 10:54:21.975 4776 INFO nova.network.driver [-] Loading network driver 'nova.network.linux_net' 2013-10-09 10:54:22.036 4776 DEBUG nova.wsgi [-] Loading app metadata from /etc/nova/api-paste.ini load_app /usr/lib/python2.6/site-packages/nova/wsgi.py:484 2013-10-09 10:54:22.076 4776 INFO nova.openstack.common.periodic_task [-] Skipping periodic task _periodic_update_dns because its interval is negative 2013-10-09 10:54:22.079 4776 CRITICAL nova [-] [Errno 98] Address already in use 2013-10-09 10:54:22.079 4776 TRACE nova Traceback (most recent call last): 2013-10-09 10:54:22.079 4776 TRACE nova File "/usr/bin/nova-api-metadata", line 10, in <module> 2013-10-09 10:54:22.079 4776 TRACE nova sys.exit(main()) 2013-10-09 10:54:22.079 4776 TRACE nova File "/usr/lib/python2.6/site-packages/nova/cmd/api_metadata.py", line 33, in main 2013-10-09 10:54:22.079 4776 TRACE nova server = service.WSGIService('metadata') 2013-10-09 10:54:22.079 4776 TRACE nova File "/usr/lib/python2.6/site-packages/nova/service.py", line 318, in __init__ 2013-10-09 10:54:22.079 4776 TRACE nova max_url_len=max_url_len) 2013-10-09 10:54:22.079 4776 TRACE nova File "/usr/lib/python2.6/site-packages/nova/wsgi.py", line 123, in __init__ 2013-10-09 10:54:22.079 4776 TRACE nova self._socket = eventlet.listen(bind_addr, family, backlog=backlog) 2013-10-09 10:54:22.079 4776 TRACE nova File "/usr/lib/python2.6/site-packages/eventlet/convenience.py", line 38, in listen 2013-10-09 10:54:22.079 4776 TRACE nova sock.bind(addr) 2013-10-09 10:54:22.079 4776 TRACE nova File "<string>", line 1, in bind 2013-10-09 10:54:22.079 4776 TRACE nova error: [Errno 98] Address already in use 2013-10-09 10:54:22.079 4776 TRACE nova The error from the nova-api log: 2013-10-09 11:17:04.520 6048 CRITICAL nova [-] [Errno 98] Address already in use 2013-10-09 11:17:04.520 6048 TRACE nova Traceback (most recent call last): 2013-10-09 11:17:04.520 6048 TRACE nova File "/usr/bin/nova-api", line 10, in <module> 2013-10-09 11:17:04.520 6048 TRACE nova sys.exit(main()) 2013-10-09 11:17:04.520 6048 TRACE nova File "/usr/lib/python2.6/site-packages/nova/cmd/api.py", line 51, in main 2013-10-09 11:17:04.520 6048 TRACE nova server = service.WSGIService(api, use_ssl=should_use_ssl) 2013-10-09 11:17:04.520 6048 TRACE nova File "/usr/lib/python2.6/site-packages/nova/service.py", line 318, in __init__ 2013-10-09 11:17:04.520 6048 TRACE nova max_url_len=max_url_len) 2013-10-09 11:17:04.520 6048 TRACE nova File "/usr/lib/python2.6/site-packages/nova/wsgi.py", line 124, in __init__ 2013-10-09 11:17:04.520 6048 TRACE nova self._socket = eventlet.listen(bind_addr, family, backlog=backlog) 2013-10-09 11:17:04.520 6048 TRACE nova File "/usr/lib/python2.6/site-packages/eventlet/convenience.py", line 38, in listen 2013-10-09 11:17:04.520 6048 TRACE nova sock.bind(addr) 2013-10-09 11:17:04.520 6048 TRACE nova File "<string>", line 1, in bind 2013-10-09 11:17:04.520 6048 TRACE nova error: [Errno 98] Address already in use 2013-10-09 11:17:04.520 6048 TRACE nova
netstat: tcp 0 0 0.0.0.0:8775 0.0.0.0:* LISTEN 32327/python
As stated on upstream bug, the metadata-api should be already spawned by nova-api if specified in nova.conf:enabled_apis (which is the default when installing with packstack) and that's why you can't start it on a separate process if already running nova-api. The question is, did packstack start nova-api-metadata? If so it's a problem with packstack that should be fixed. Having said that, I couldn't reproduce this with the latest puddle (i.e. nova-api-metadata was not started after the deploy).
Closing as it doesn't look like a bug. Feel free to reopen if you thing there is something we should solve.