Eva R Hocks wrote:
|>
|>
|> Eversince I installed the 2.1.x kernel I noticed a misleading information
at which x did you start ?
|> from the netstat command. There are not that many Transmit-errors (TX-ERR)
|> on the interfaces, neither lo nor ippp0. From the /proc/net/dev
|> information looks like the columns are shifted to the right for 1 col.
|>
|> netstat -in
which version ? (which distribution ?)
|> Kernel Interface table
|> Iface MTU Met RX-OK RX-ERR RX-DRP RX-OVR TX-OK TX-ERR TX-DRP TX-OVR Flags
|> lo 3584 0 42 0 0 0 0 2933 42 0 LRU
|> ippp0 1500 1 8505 0 0 0 0 5377025 11865 0 OPRU
|>
|>
|> /proc/net/dev has the correct packet information :
|>
|> Inter-|Receive |Transmit
|> face |bytes packets errs drop fifo frame compressed multicast|bytes
|> packets errs drop fifo colls carrier compressed
|> lo: 2933 42 0 0 0 0 0 0 2933
|> 42 0 0 0 0 0 0
|> ippp0: 684187 7907 0 0 0 0 0 0 5368904
|> 11032 0 0 0 0 0 0
|>
looks like an additional column introduced into /proc/net/dev at _some_
point in the 2.1.xx development. (who kept track of the changes made here ?)
so the byte count of the transmitter is mis-interpreted as error count...
You're right, the numbers listed under TX-DRP belong to TX-ERR and so on,
so the introduced column must be in the rcv part...
|>
|> Did anybody else notice and know why ?
I'm still at 2.0.33 *blush*
Steffen
--
Steffen Grunewald = steffen(at)gfz-potsdam.de = fax +49-331-288-1266
try to `finger steffen@gfz-potsdam.de` for details
My opinions are mine. You can rent them. Just ask me for prices :-)
"Emacs is a nice operating system, but Unix has the better editor."
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu