Re: rpc_doio sending evil packet

Christof Petig (christof.petig@wtal.de)
Wed, 11 Mar 1998 16:21:44 +0100

Nicolai Langfeld wrote
>periodically one (not the same each time) of my machines get into this
state:
>
>Mar 10 12:17:47 persephone kernel: RPC: rpc_doio sending evil packet:
>Mar 10 12:17:47 persephone kernel: caa99047 01000000 00000000
00000000 0000
>0000 00000000 00000000 0182780b
>Mar 10 12:17:47 persephone kernel: RPC: rpc_send sending evil packet:
>Mar 10 12:17:47 persephone kernel: caa99047 01000000 00000000
00000000 0000
>
>over and over, with 1 minute intervals. When they have been in this
state
>a while they become impossible to log into by telnet/rsh/ssh and the
console
>is not usable either. Reset button is only way out.
>
>It _seems_ related to NFSds that die and hang, but I cannot offer any
>good observations verifying this, I have never been able to observe
>this decay process first hand, I only discover it when the machine
>stops responding to maintenance routines...

Hi Nicolai,

can you tell us your kernel version (and perhaps more about your setup).

I'm very interested in this thread because our central router is running
Linux 2.0.33 and sometimes starts printing these messages. I thought
them harmless, because the machine continues to work as exspected.
Though, yesterday I found the static routes dropped (no idea why).

Our machine serves as a NFS+NIS+... server for several linux machines
(2.0.0-2.0.32 and 2.1.57+2.1.89). I felt these messages correspond to a
connection to the Internet via isdn4linux (HiSax). It is set up as a
firewall and connects to two out of three ethernet segments.

Christof

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu