Re: Abnormal behavior after bumping net.inet6.udp6.sendspace

> Is it this?
> 
> 	% /sbin/sysctl -a | grep sbmax
> 	kern.sbmax = 262144
Yes, you are correct!  That was it.

> Setting the individual socket buffer sizes larger than sbmax (or too
> close to it, because there is per-protocol overhead) has always caused
> unobvious behavior, in my experience.
Makes sense (though I wasn't aware of sbmax)

> It may be that sbmax should default to a larger value.
Wouldn't it be relatively straightforward to guard against changing the
send/recvspace sysctls past sbmax?

Thanks!