We are using one network node, I'll see if I can provoke this issue and check the network queue to see if events are stacking up in there. Likewise, we'll tweak some of the lease timing and see if that helps. I'm puzzled by why it appears to require restarting nova-network to fix this though, presumably eventually the queue would drain of events and things would return to normal.
We are using one network node, I'll see if I can provoke this issue and check the network queue to see if events are stacking up in there. Likewise, we'll tweak some of the lease timing and see if that helps. I'm puzzled by why it appears to require restarting nova-network to fix this though, presumably eventually the queue would drain of events and things would return to normal.