Jonathan Lemon writes: > On the surface, I don't see a problem with it, but I wonder why you > believe it is needed. Changing the value of rexmttresh is probably > not a good idea, and I would definitely prefer not to expose it as a > user tunable knob. > > On Sat, Nov 17, 2001 at 06:51:02PM -0600, Alfred Perlstein wrote: > > Any problems with this? > > > > -static int tcprexmtthresh = 3; > > +static unsigned int tcprexmtthresh = 3; > > +SYSCTL_UINT(_net_inet_tcp, OID_AUTO, rexmtthresh, CTLFLAG_RW, > > + &tcprexmtthresh, 0, "Max duplicate acks before fast rexmit"); > > +
How about a kernel option TCP_TWEAKS or something that would expose these 'dangerous' sysctl's for people to play with. -Archie __________________________________________________________________________ Archie Cobbs * Packet Design * http://www.packetdesign.com To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-net" in the body of the message
