We have had an event today that was relatively straightforward in nature. Our primary DNS server failed. 

We have a secondary DNS server configured in your software, and yet the software failed to recognise the failure of the primary DNS server and messages started to queue as DNS lookups associated with anti spam settings failed to connect.

The end result was that on the anti spam administration page, the "ms" time connection being monitored to DNS lookups etc were all either timing out or showing huge times.

The solution was easy. We just put the IP address of the secondary DNS server into the primary position on your software, and things started to run well again.

So I'm requesting that you improve the DNS failover of your software to better support the primary DNS server going offline. Perhaps a setting to allow the time in "ms" before attempting to connect to the secondary server?

At the very least, the function that is there should work. Currently it appears to us that the failover does not work with a failure.