Re: internal drops with tcp, kernel 2.2.16

[Chris Wedgwood <cw@xxxxxxxx>]

On Thu, Jan 18, 2001 at 12:51:32PM +0100, Andi Kleen wrote:

    The common problem of the dynamic IP changing is already mostly
    but not completelycovered by the ip_dynaddr socket address
    rewrite hack.  The main problem left in there is in the user
    space resolver library btw with its too short timeout.

is there any reason why the ip_dynaddr stuff can't be compiled out
for those like myself who find it an esthetically unpleasant kernel
hack to solve a user-land problem?



  --cw