Can't bind to UDP

I filed a ticket but just in case anyone knows what's up.

CentOS 7 host, London

All I use it for is an authoritative DNS server - slave - running NSD.

It got migrated automatically few days ago. Today noticed it wasn't responding to UDP requests.

When I ssh into the box -

[root@li354-217 ~]# systemctl status nsd.service
● nsd.service - NSD DNS Server
   Loaded: loaded (/usr/lib/systemd/system/nsd.service; disabled; vendor preset: disabled)
   Active: failed (Result: exit-code) since Mon 2017-02-13 09:28:01 UTC; 32s ago
  Process: 21630 ExecStart=/usr/sbin/nsd -d -c /etc/nsd/nsd.conf $NSD_EXTRA_OPTS (code=exited, status=1/FAILURE)
 Main PID: 21630 (code=exited, status=1/FAILURE)

Feb 13 09:28:01 li354-217 systemd[1]: Starting NSD DNS Server...
Feb 13 09:28:01 li354-217 nsd[21630]: nsd starting (NSD 4.1.14)
Feb 13 09:28:01 li354-217 nsd[21630]: [2017-02-13 09:28:01.759] nsd[21630]: notice: nsd starting (NSD 4.1.14)
Feb 13 09:28:01 li354-217 nsd[21630]: can't bind udp socket: Cannot assign requested address
Feb 13 09:28:01 li354-217 nsd[21630]: [2017-02-13 09:28:01.761] nsd[21630]: error: can't bind udp socket: Cannot assign requested address
Feb 13 09:28:01 li354-217 nsd[21630]: server initialization failed, nsd could not be started
Feb 13 09:28:01 li354-217 nsd[21630]: [2017-02-13 09:28:01.761] nsd[21630]: error: server initialization failed, nsd could not be started
Feb 13 09:28:01 li354-217 systemd[1]: nsd.service: main process exited, code=exited, status=1/FAILURE
Feb 13 09:28:01 li354-217 systemd[1]: Unit nsd.service entered failed state.
Feb 13 09:28:01 li354-217 systemd[1]: nsd.service failed.

Rebooting doesn't help, trying to restart the service doesn't help, it can't bind to UDP port 53.

Fortunately my other two nameservers (both also linode - one in Texas and the other I think is either New Jersey or Califormia, I'd have to look) work just fine, but I need this one to work as well. Redundant redundancy is why I have it, and it isn't very redundant if it isn't running.

3 Replies

The actual problem, when they migrated this linode, they changed the IPv6 address.

The IPv6 address given by ifconfig is different from what it has historically been, what it is listed as in my admin panel for that linode.

And that's why NSD wouldn't start.

So if linode migrates you and suddenly a service won't start, well, check your IP addresses.

We don't just change IPs because you moved hosts. Something else was at play.

-Chris

Well I don't know what else was at play, but the IPv4 address stayed the same.

The IPv6 address, which is what the reverse DNS is set to, use to be 2a01:7e00::f03c:91ff:fe18:8a7e and now /sbin/ifconfig is showing it to be 2a01:7e00::825f:e564:ad53:72fc

I certainly didn't change it, nor do I believe I have the ability to change it beyond the local subnet pool assigned to the VM. So whatever was at play wasn't my end.

Reply

Please enter an answer
Tips:

You can mention users to notify them: @username

You can use Markdown to format your question. For more examples see the Markdown Cheatsheet.

> I’m a blockquote.

I’m a blockquote.

[I'm a link] (https://www.google.com)

I'm a link

**I am bold** I am bold

*I am italicized* I am italicized

Community Code of Conduct