I got a strange issue on one of our brand new Windows 7 Professional machine which we rolled out as Post Master Enterprise email server. The basic stuffs that I carried/checked out to identify the depth of issue are as below.
- Unable to open any webpage external as well as webmail running on the same machine.
- Emails were not being sent/received from Google Apps accounts.
- Post Master Enterprise email server was responding fine to client connections on both POP3 and SMTP port, due to this emails were stuck into the spool.
- ping response to internal as well as external hosts was fine, e.g. ping 8.8.8.8
- Tracert command was also showing proper connectivity to internal and external hosts, e.g. tracert 8.8.8.8
- DNS resolution was working fine. Tested this by pinging a host which I never opened on the machine, e.g. ping discovery.com
- Got below error while changing Computer/Host name
The following error occurred attempting to rename the computer to "example".
The name limit for the local computer network adapter card was exceeded.
- Got this error while accessing network shares
The name limit for the local computer network adapter card was exceeded.
How I solved “The name limit for the local computer network adapter card was exceeded” issue?
I was required troubleshoot this issue immediately due to Post Master Enterprise email server running on the same machine and all email communication was halted. After rebooting the machine every thing was working fine for a few minutes and then suddenly this kind of behavior re-appeared.
I started googling for “The name limit for the local computer network adapter card was exceeded” and found a Microsoft KB article. I started taking action according to the article but suddenly realized that it was not applicable to Windows 7 Operating System while making some registry changes. But I got a few hints regarding the issue from this Microsoft KB article.
I ran netstat -a command to view all network connections in Established/Listening/Time_Wait/Close_Wait mode as mentioned in Microsoft KB article. I found a lot of CLOSE_WAIT connections, a small portion is pasted below. The command netstat -a was taking too long to generate entire list actually it didn’t complete until I resolved the issue using tcpview.exe. I thought this was trying to fill up the entire TCP port range which was not allowing more genuine TCP connections to open.
TCP 192.168.1.6:49158 174.35.44.58:http CLOSE_WAIT TCP 192.168.1.6:49159 174.35.44.23:http CLOSE_WAIT TCP 192.168.1.6:49160 174.35.44.23:http CLOSE_WAIT TCP 192.168.1.6:49168 174.35.44.23:http CLOSE_WAIT TCP 192.168.1.6:49169 174.35.44.23:http CLOSE_WAIT TCP 192.168.1.6:49170 174.35.44.58:http CLOSE_WAIT TCP 192.168.1.6:49173 174.35.44.23:http CLOSE_WAIT TCP 192.168.1.6:49174 174.35.44.58:http CLOSE_WAIT
To view which application is opening so many connections execute netstat -ab command, however I didn’t do that. I downloaded TCPView utility and executed tcpview.exe utility from the zip file to view real time TCP connections.
tcpview.exe helped me to identify the culprit application which was opening too many connections and was not closing them. I found orbitdm.exe application file which is a part of Orbit Download.
I killed this process orbitdm.exe and then everything came back to normal. I don’t know why orbitdm.exe was opening too many TCP connections. I was unable to identify the actual reason due to time constraint because the machine is our email server and I was required to bring it to normal as soon as possible. That was why I even forgot to capture the snapshot of tcpview.exe utility showing too many connections opened and unable to close (CLOSE_WAIT).