Ok this is a strange one, details as follows:
Servers on IP Range 192.168.1.x
Exchange 2007 Sp2 CAS
Exchange 2007 Mailbox Cluster
Server 2003 Running AD + DNS resolution (internal to network)
Router on ip range 192.168.1.49
Cisco 2800 - ACL and NAT outside IP to internal Exchange 2007 CAS
Switches on Ip range - 5 running RSTP
Configuration was running properly with no issues as of yesterday, no changes were made to the router or sever in the past 24 hrs.
Isssue:
Router looses connection to ONLY the exchange CAS server IP address.
Router is able to ping ALL other switches servers and client machines on network (that i've tried so far anyways).
Router is able to DNS resolve all other hardware on the network.
All hardware on the network can ping the Router Ip address With the exception of the Exchange CAS server.
Exchange CAS server can ping all ip/DNS resolved hardware on the network with the exception of the Router.
So specifically it's a communication problem from router -> CAS and from CAS -> Router, however both devices remain in contact with the rest of the network.
HEre's the wiggy part.
When the communication goes down, IF you ping FROM the router to an unresolved DNS entry forcing it to contact the DNS server and pull down a numeric address.. communication to the CAS server starts to flow again for approx 10-15mins and then fails.
Repeating this process I've manually kept communications up and running with the mail services being delivered and sent and OWA running.
I've been comparing old configs with new on the router and there's no difference in config so far.
at this point any ideas would really be helpfull!!
I'll add more details as I come across but perhaps this description will be enough for a genius out there.
Help me Obiwan..
Post
New Detail:
Doesn't actually even need to be a resolvable/pingable address.
I added a DNS entry to the server to a IP address that doesn't exist, and when I complete the bandaid task of pinging this new DNS entry, the communication returns even tho the ping process from the router fails.
To me this seems to point to some kind of loss of communication with the DNS server on the router side?
When we force the router to resolve a DNS name through the DNS server communication returns.
New Detail:
Added to Router Config: no ip domain-lookup
Connection to CAS fell down (Expected), and obviously local domain resolution was no longer valid on the router.
Was unable to bring the connection to CAS back up until I removed the No IP domain-lookup command and ran a ping namedhost.
