ndt-users list created

[Jeremy Schafer <>]

comments inline...

Richard Carlson wrote:
> It looks like the server is failing to find the proper web100 connection 
> ID so the limits are not being set/used.  The server process sends data 
> to the stack faster than the stack can sent it out on the network.  Thus 
> a queue builds up in the server (you should see this with a netstat -nat 
> command) so when the 5 sec send timer expires there is still data in the 
> queue that needs to be sent to the client.  The client's watchdog timer 
> is reached and it kills the client, while the server is still in trying 
> to deliver the packets.  Finally the server's watchdog timer goes off, 
> and the server exits.

It didn't dawn on me to add that to the symptoms list. I noticed the 
queue filling up on the server, whether it be ipv6 or ipv4. If there is 
a next time, I will make a note of it. Thanks!

> There are a couple of options we can take to work around this problem, 
> and I'll experiment with them to see what works best.  I'll  push out a 
> new test version soon and let you know when it's ready.

I am willing to experiment if you need a guinea pig. Let me know what 
you want to test. My server is in the next room over so logistically it 
isnt a problem.

> PS thanks for the script patch, I really appreciate patches and updates 
> that help make the tool better for the community.

Anytime.