UDP Failure

[Cei]

Hi all,

Our Mumble server has been up and running for several months now, and had generally been perfectly stable. However, last night it went weird for no reason. No server reboots or anything of the kind, but people started being unable to hear and transmit.

2015-04-11 03:36:33.224 1 => <31:Cole(36)> Requesting crypt-nonce resync

2015-04-11 03:36:38.692 1 => <19:Junkoe(-1)> Requesting crypt-nonce resync

2015-04-11 03:36:41.365 1 => <14:Roll Fizzlebeef(38)> Requesting crypt-nonce resync

2015-04-11 03:36:41.771 1 => <21:cardinaldirection(-1)> Requesting crypt-nonce resync

This is appearing in the log now and then since the issues started. Some of the users are reporting "UDP packets cannot be sent to or received from the server. Switching to TCP mode." error message on connecting, but not consistently. Relogging sometimes fixed this.

I've done some Googling, but I'm at a loss. Murmur.ini has always been bound to an IP address rather than all interfaces. The firewall allow Murmur through, and no changes were made to suddenly produce this issue. Google indicates that it's a UDP problem, but why would this suddenly start happening with no warning?

Thanks :)

[kissaki]

Are you aware of TCP and UDP technology?

It’s probably nothing under your direct influence.

Either a temporary routing issue, or network provider throttling/failing on UDP packets.

[Cei]

That was my suspicion, namely a routing issue with the host. It seems to have stablised over the last few hours, but I'm going to keep an eye on it. Posted mainly due to the fact that some people on Google indicated that it might be other issues.