lkml.org 
[lkml]   [1999]   [Mar]   [15]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: 2.2.x - IPX overbroadcasting (badly)
On Mon, Mar 15, 1999 at 08:21:01AM -0600, Gilbert Ramirez wrote:
> When 2.2.x started, and in the 2.2.0-pre kernels, I also had problems with
> IPX, but not with broadcasts. The storm of packets I saw were directed from
> my Linux box to a NW file server. I too was using the Olicom token-ring
> driver (as I infer from the vendor portion of your MAC address).
>
> I tried to get my box to repeat the symptoms the next day, and it didn't.
> Unfortunately, I no longer have access to my box to test it.

Yeah, I remember that. In fact, we conversed via email on that a while
back (my address from work is <Greg_Norris@mail.dor.state.mo.us>).

> There was an IPX broadcast problem reported by an ethernet user, but a
> change to how IPX handled netbios broadcasts seems to have fixed that
> problem. Perhaps this is an IPX/oltr problem? Is there anyone using oltr,
> IPX, and kernel 2.2.x successfully?

I spoke with him as well, around the time 2.2.[01] came out. After
bouncing a few messages back and forth, I believe that we were
experiencing the same problem... both of us were seeing massive floods
of IPX broadcast packets, which would begin almost immediately. I
suspect it's a problem with IPX itself, rather than with the oltr driver
(though you're quite correct about it being an Olicom card), because he
was using a 3c509 (I think that's right). Last time I heard from him,
it still didn't work without pruning the kernel. Of course, I've been
wrong before. ;)

The fix that was suggested to him was to remove/disable the section of
code referencing IPX_TYPE_PPROP in linux/net/ipx/af_ipx.c. This seems
to work for me as well, though I haven't had a chance to try it yet with
2.2.3 (I keep hoping that "it's ok now" whenever a new version appears
<g>). Unfortunately, this seems more like a temporary workaround than a
proper fix, as I'm sure a block of code that size must be there for a
reason... so I was hoping that providing a tcpdump might point one of
"them smart fellers" to the cause.

Take care!


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-03-22 13:50    [W:0.161 / U:1.300 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site