I have encountered a problem where Wingate works fine for between 5 and 15 minutes depending on load, and then completely fails to respond to client requests. Although there is a DNS problem that I will describe below, I don't know if this is the cause so I would be very interested in your ideas of what in general could cause this.
The DNS problem is this. We have an active directory Windows 2000 domain server running DHCP and DNS. The DHCP service identifies the primary controller (itself) as the DNS server and the Wingate computer as the default router (gateway) to the network clients. In the DHCP scope properties DNS page I have it set to always automatically update DHCP client information in DNS. In the DNS properties for this server I have a enabled forwarders with a forwarding ip address pointing to the Wingate computer. What appears at the Wingate computer after 5 or 10 minutes of apparently normal activity (users can access Web pages and send/receive email) is a large number of DNS requests, many of which are SOA and A lookups for names on the local domain, and repeated many times over for the same client on the Activity page (as well as in History). At the same time the client PC's get no further response to their requests. It could be that Wingate is simply swamped by dns requests that it can't resolve and further requests just time out. If so, the question is why these requests end up at Wingate when they should not have been forwarded by my DNS server. Am I missing something here? Is there some other possible reason for Wingate to fail?