[plug] ping yields "wrong data byte #0 should be 0x9a etc" -- why?

[Colin Muller]

On Fri, Aug 03, 2001 at 08:35:08AM +0800, Leon Brooks wrote:
> Anthony J. Breeds-Taurima wrote:
> > On Fri, 13 Jul 2001 daniel at iinet.net.au wrote:
> >>wrong data byte #0 should be 0x9a but was 0x9999 f9 4e 3b 6c d a 0
> >>         8 9 a b c d e f 10 11 12 13 14 15 16 17 18 19 1a 1b 1c 1d 1e 1f 20
> >>21 22 23 24 25 26 27
> >>         28 29 2a 2b 2c 2d 2e 2f
> >>64 octets from 203.0.178.191: icmp_seq=4 ttl=252 time=1510.2 ms
> > The Only time I've seen this is on a RedHat box were libc had been upgraded
> > but the ping suit hadn't.
> 
> > If that fits with your situation, then apply all the latest RPMS (You will
> > want to install the iputils package if it isn't already installed)
> 
> The only time I've seen this is on an ISDN link with dud hardware (it 
> drops DMA transfers into the bucket occasionally) and I still see it 
> occasionally because the silly buggers won't spend a grand to replace 
> their hardware, the gateway through which their 
> eight-significant-figures business runs.

I've seen this when pinging a machine which had patchwork-quilt
upgrades along the lines Anthony mentions; it seemed to be more likely
to happen when the in-between route (not the machine itself) was
overloaded than when it wasn't, but I could never quite pin down what
was causing it.

Colin