Bogus packet size???

From: Jeff Beers (jdbeers@SND.SOFTFARM.COM)
Date: 06/13/98


I don't believe this is circle code causing this problem, but
I need some expert opinions.  The sysadmin believes we are
causing it.  I can't argue with him because, I lack knowledge in
this area.

We recently moved to a new site:
Red Hat Linux release 4.1 (Vanderbilt)
Kernel 2.0.27 on an i486

/var/log is getting spammed with 100s of messages a second like
this:

May 22 18:18:07 mudd kernel: loading device 'eth0'...
May 22 18:18:07 mudd kernel: ne.c:v1.10 9/23/94 Donald Becker
(becker@cesdis.gsf
c.nasa.gov)
May 22 18:18:07 mudd kernel: NE*000 ethercard probe at 0x300: 00 60 67
00 6d da
May 22 18:18:07 mudd kernel: eth0: NE2000 found at 0x300, using IRQ 3.
May 22 18:26:02 mudd kernel: eth0: bogus packet size: 32864, status=0x21
nxpg=0x
fc.
May 22 18:26:02 mudd kernel: eth0: next frame inconsistency, 0xfc
May 22 18:26:02 mudd kernel: eth0: bogus packet size: 32918, status=0x21
nxpg=0x
cd.
May 22 18:26:02 mudd kernel: eth0: next frame inconsistency, 0xcd
May 22 18:26:02 mudd kernel: eth0: bogus packet size: 32918, status=0x21
nxpg=0x
cd.
May 22 18:26:02 mudd kernel: eth0: next frame inconsistency, 0xcd
May 22 18:26:02 mudd kernel: eth0: bogus packet size: 32918, status=0x21
nxpg=0x
cd.

I've only seen one place in circle where we write packets.
comm.c uses the write() command, but write decides how many packets
to write at any one time, hence the loop.

Any assistance would be appreciated.

Thanks


     +------------------------------------------------------------+
     | Ensure that you have read the CircleMUD Mailing List FAQ:  |
     | http://democracy.queensu.ca/~fletcher/Circle/list-faq.html |
     +------------------------------------------------------------+



This archive was generated by hypermail 2b30 : 12/15/00 PST