Ask Your Question

Revision history [back]

click to hide/show revision 1
initial version

How to restart Neutron properly?

We have a control node that also acts as the network node and multiple compute nodes running Icehouse.

Sometimes it's necessary to restart the control node due to maintenance. After the control node is back up, the network connections between virtual machines running on the computes nodes as well as their connections to the outside are often broken. Only a restart of the compute node helps.

This is ridiculous. Do you guys have any advice on how to repair this situation without the need to reboot the whole cluster of nodes? I'd be happy for any workflow you successfully use.