Tek-Tips is the largest IT community on the Internet today!
Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!
-
Congratulations derfloh on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!
TCP Resets
- Thread starter jearnold
- Start date
- Thread starter
- #3
Our users are getting error message (no logon domain controllers found) when logging into the domain. We performed a network capture and found the tcp sessions were being reset by the server.<br>
<br>
Once we reboot the server the problem is solved, but eventually the server runs out of resources and starts to reset any new tcp sessions.<br>
<br>
The netstat command shows us most of the connections are in the last_ack state, meaning that the server is in the final step to close the connection, but his never happens even after the (1) hour SessionKeepAlive has expired.<br>
<br>
This is happening only on domain controllers that are connected to the workstations via a Bay Networks 5000 switch. Our NT servers are running NT 4.0 with SP5.<br>
<br>
<br>
Appreciate any help.<br>
<br>
Once we reboot the server the problem is solved, but eventually the server runs out of resources and starts to reset any new tcp sessions.<br>
<br>
The netstat command shows us most of the connections are in the last_ack state, meaning that the server is in the final step to close the connection, but his never happens even after the (1) hour SessionKeepAlive has expired.<br>
<br>
This is happening only on domain controllers that are connected to the workstations via a Bay Networks 5000 switch. Our NT servers are running NT 4.0 with SP5.<br>
<br>
<br>
Appreciate any help.<br>
Some or none of the following may be worth trying:<br>
<br>
Increase the TTL to cut down the requests (won't solve the problem but may get round it).<br>
Turn off spanning tree protocol on the Bay.<br>
Enable port fast on the Bay and check the port settings, some of the timings may be awry making the acknowledgement fail.<br>
I'd say it is a switch issue and not an NT issue for sure.
<br>
Increase the TTL to cut down the requests (won't solve the problem but may get round it).<br>
Turn off spanning tree protocol on the Bay.<br>
Enable port fast on the Bay and check the port settings, some of the timings may be awry making the acknowledgement fail.<br>
I'd say it is a switch issue and not an NT issue for sure.
- Status
Similar threads
- Locked
- Question