Kilo: n-cauth fails with "too many connections"

asked 2015-06-24 15:43:36 -0500

opstkusr gravatar image

Have a devstack based Kilo install with compute nodes.

See the following failure when attempting to add a new compute node:

[[01;31m2015-06-24 16:29:03.102 TRACE nova.openstack.common.threadgroup ^[[01;35m^[[00m raise original_exception ^[[01;31m2015-06-24 16:29:03.102 TRACE nova.openstack.common.threadgroup ^[[01;35m^[[00mOperationalError: (OperationalError) (1040, 'Too many connections') None None ^[[01;31m2015-06-24 16:29:03.102 TRACE nova.openstack.common.threadgroup ^[[01;35m^[[00m [stack@grunt108 ovs]$ ^M^[[K[stack@grunt108 ovs]$ /usr/bin/nova-consoleauth --config-file /etc/nova/nova.conf & echo $! >/opt/stack/status/stack/n-cauth.pid; fg || echo "n-cauth fai^Miled to start" | tee "/opt/stack/status/stack/n-cauth.failure"

Is there any recommendation on what needs to be changed to address this?

Updating the max_pool_size in nova.conf seems to address this. Are there any recommended values for these for database and database_api configs in nova.conf?

edit retag flag offensive close merge delete