In message <397E0659AA2DD411843500508B64F1CE1E0723@mail.guardent.com>, Tim.News
ham@guardent.com writes:


> 
>The parameter used for RFC1948 need not be randomly generated
>at boot time.  I was actually wondering if this violates the intent.
>Shouldnt
>ISN values exist in the same space across reboots?

In all seriousness, that depends on now long a reboot takes.  For 
correctness in the TCP sense, if a reboot takes longer than twice the 
maximum segment lifetime the ISN value for a connection can be 
different.  When I implemented RFC 1948, I initialized the secret by a 
hash of a per-machine secret and whatever random values were lying 
around shortly after boot time -- the reboot time was easily long 
enough.

		--Steve Bellovin, http://www.research.att.com/~smb