Originally posted by: Wouter Liefting
If the Domain Controller is also your DNS server (which is very common), then it's a perfectly valid explanation.
Upon establishing a connection, the sshd daemon does a reverse DNS lookup to determine the host name, and possibly compare this hostname to a list in /etc/hosts.equiv or similar. It may optionally also do a forward lookup to see if the same IP address pops up, as an additional security check. So if the DNS server is unavailable, this will lead to significant timeouts.
You may want to modify the /etc/ssh/sshd_config settings, including UseDNS and HostbasedAuthentication, to prevent this in the future. But better yet, your admin team has to understand that a Domain Controller is one of the most crucial bits of infrastructure on your network, and should provide suitable backup/failover solutions in case the PDC is down. And your /etc/resolv.conf file should list all slave DNS servers that are available. Alternatively, you can setup your own slave DNS server on your own machines, and use that.