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'am using a machine as a router, it works sort of fine, however if I do a ping -t whatever.server it will always cause a 3-4% package loss, no matter what server. Looking at ifconfig or netstat -i show no errors what so ever, what are some of the tools I might use to analyse this error?

Using wireshark tells me that it does a request but it gives no reply, so the errors seems to be on the incoming end.

Any idea or suggestions is very welcome, I am kind of lost here.

Thanks in advance,

UPDATE: It seems I get this reply from pinging a VM on the same machine also, I will try and replace the network cable to see if this resolves the issue. I will update this question when I have done the switch. Thanks for the help so far.


share|improve this question
s/packages/packets/ in the title. – mattdm Dec 3 '10 at 14:52
@mattdm, Thank you buddy. – Anders Dec 3 '10 at 14:53

Impossible to say without knowing your setup. Does it happen when you ping the upstream network equipment? What if off of the network and just attached to another computer?

Have you tried replacing the network cable?

share|improve this answer
Good suggestions, I will try these and come back. – Anders Dec 3 '10 at 15:06

To get the beginnings of a starting point, try tracerouting to a remote point, then ping all intermediate hops.

If you find that going from hop N (with no ping loss) to hop N+1 ()with the same ping loss as to the remote end), there's a decent chance that's where the issue is. However, it is possible for intermediate hops to drop (or be slow to reply to) pings directed at themselves, without there being an actual issue.

share|improve this answer
Or use mtr/winmtr, that bundles it all in a handy package. – Joris Dec 4 '10 at 16:14
@joris: As long as you remember that any "loss" or "slowness" seen in the middle of the trace, exceeding that to the end, is down to how intermediate routers respond to ICMP... I've had more than one "fault" reported because people can't properly interpret MTR reports. – Vatine Dec 4 '10 at 19:00

It might be a sign of a funky cable of a NIC going bad. The easiest test would be to plug in another known to be good NIC (USB one would probably be the best if you cant have downtime) and ping that interface. Another test would be to look at routing, if there's a loop somewhere and every now and then a packet ends up going in circles until it runs out of TTL, that'd be another reason for packets to disappear.

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.