lkml.org 
[lkml]   [1997]   [Jun]   [16]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
SubjectRe: kerneld/multicast bug (tickled by gated)
Date

Olaf Titz <olaf@bigred.inka.de> writes:
>There is more that can be done with request-route than just starting a
>pppd. My request-route script starts IPIP tunnels as needed, and "as
>needed" includes "if the configuration allows it now for this route".
>Otherwise it just returns and lets the IP packet bounce.

Can you please give a bit more detail on exactly what you are doing
with this? Then maybe I can comment on how it can be done without
the dangers inherent in request-route.

>Even if this could be done with diald at all (I have not checked), it
>is much more simple and efficient this way. This is enough reason for
>me to keep request-route on my system.

As Alexy, Alan and myself keep pointing out, the presense
of request-route breaks the routing code in the kernel. People are
lucky when it doesn't crash, but if you put it under stress it will.
If the request route message never gets generated, you have
nothing to worry about, but if it does, it puts the routing code
to sleep at a place where it is not suppose to sleep, madness
will follow if a routing change is made during this sleep.
This is quite likely on a box running gated or routed, and there
is even quite a bit of danger in the routing changes that will be made
by pppd as it comes up.

--
Eric Schenk www: http://www.dna.lth.se/~erics
Dept. of Comp. Sci., Lund University email: Eric.Schenk@dna.lth.se
Box 118, S-221 00 LUND, Sweden fax: +46-46 13 10 21 ph: +46-46 222 96 38



\
 
 \ /
  Last update: 2005-03-22 13:39    [W:0.039 / U:0.172 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site