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

[Thor Lancelot Simon <tls%panix.com@localhost>]

On Wed, Jul 15, 2026 at 12:59:02PM +0200, Timo Buhrmester wrote:
> 
> Wouldn't it be relatively straightforward to guard against changing the
> send/recvspace sysctls past sbmax?

My very dim recollection is that a few of us have looked at this and
not done it, over the years, because it's annoying to precompute the
overhead that will eventually cause the allocation to fail.  But likely
not impossible.

I do think as a hard maximum rather than any kind of per-socket default
the current value is much too low.

-- 
Thor Lancelot Simon	                                     tls%panix.com@localhost
  "Indeed, someone who is a citizen in a democracy is often not one in
   an oligarchy."
					- Aristotle, Politics III.1