4

I was happily pinging in the net when suddenly:

Pinging X with 32 bytes of data:

Reply from .235: bytes=32 time=1444ms TTL=41
Reply from .235: bytes=32 time=1767ms TTL=41
Reply from .235: bytes=32 time=1531ms TTL=41
Reply from .235: bytes=32 time=-1198187ms TTL=41

Ping statistics for .235:
    Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
    Minimum = 1444ms, Maximum = -1198187ms, Average = 1073443462ms

A negative time for the reply. I checked wireshark and it had the same values:

Time delta from previous displayed frame: -1198.187867000 seconds
Time since reference or first frame: -1179.935038000 seconds

I didn't change the time of the machine while pinging. This was made in my local network at home, from a XP VM in windows 7. So I blame the VM (virtualbox). But I was wondering if this strange behaviour (to me) could have a reason, or if any of you have seen this before. Thank you bye.

1
  • 1198187 ms is almost 20 minutes... that seems very long for clock drift caused by virtualization.
    – Keith
    Nov 9, 2012 at 21:56

1 Answer 1

1

This looks like a combination of virtualization driver issue and a signed/unsigned integer overflow. The average time of 1073443462 x 4 is too close to 2^32 for it not to be an integer overflow issue --> 1073443462 * 4 - 1444-1767-1531 - 2^32 = -1198190 ... which is pretty close to -1198187. Maybe my math is off, or there might be a rounding error.

1
  • Timing (of any kind) is notoriously bad in virtual machines. It's possible your VM's clock was corrected by the hypervisor, which could cause readings like this. Don't rely on any time measurements in a VM for anything important.
    – voretaq7
    Nov 10, 2012 at 0:41

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .