I have done a lot of search and event opened a case with Redhat but I wanted to check here if anyone has seen a problem with the /proc/net/dev counters being wrong on occasion. Any help would be appreciated.

Details and example:

$ while true; do cat /proc/net/dev|grep eth5; sleep 1; done
eth5:568591100326 813623955 0 0 0 0 0 6365874228 1932782794340 6366662960 0 0 0 0 0 0
eth5:568591234270 813624096 0 0 0 0 0 6365874233 1932782794718 6366662965 0 0 0 0 0 0
eth5:568591376057 813624238 0 0 0 0 0 6365874240 1932782795264 6366662972 0 0 0 0 0 0
eth5:568591509699 813624381 0 0 0 0 0 6365874247 1932782796044 6366662979 0 0 0 0 0 0
eth5:8894958462077763584 813183088 0 0 0 0 0 787166 3140913848 6367102861 1566 0 0 0 1566 0
eth5:568591788846 813624664 0 0 0 0 0 6365874261 1932782797578 6366662993 0 0 0 0 0 0
eth5:568591927178 813624810 0 0 0 0 0 6365874267 1932782798054 6366662999 0 0 0 0 0 0
eth5:568592070081 813624956 0 0 0 0 0 6365874270 1932782798292 6366663002 0 0 0 0 0 0

$ uname -a
Linux hostname 2.6.9-55.0.2.ELsmp #1 SMP Tue Jun 12 17:58:20 EDT 2007 x86_64 x86_64 x86_64 GNU/Linux
$ cat /etc/redhat-release
Red Hat Enterprise Linux AS release 4 (Nahant Update 5)