Server Fault is a question and answer site for system and network administrators. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I have two Dell boxes with two ethernet ports a piece. I have a cable directly connecting two of these ports, creating a tiny LAN with 10.3.3.x addresses. The other port on each box is hooked up to a switch and has a DHCP-supplied address to talk to the outside world.

I've noticed that when scp'ing large files from one box to the other over the private LAN, the transfers sometimes stall. It appears that any other network activity on either box will cause the transfer to resume.

Wake-on-LAN is disabled on all interfaces according to ethtool. What else could be causing these stalled transfers?

share|improve this question

Gosh, this sounds pretty strange. Even though I think this should work with no problem whatsoever, are TCP keepalives on in your sshd config?

TCPKeepAlive yes

To try further troubleshooting you can try turning off apm and acpi services to try to see if that is somehow using powermanagment to turn off the nic.

If this happens consistently after a preset time, that time might give more clues about what is happening.

share|improve this answer
Thanks, Kyle. I checked and TCPKeepAlive is turned on (by default) on both machines. acpid is running, but no events have been logged since startup time. I'll try stopping it and see what happens during a big transfer. – sbleon Dec 30 '09 at 0:29
up vote 0 down vote accepted

This problem seems to have gone away after yum updating both servers, including the kernels.

share|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.