[1598] in Kerberos
Re: knfs speed (was Re: How good is kerberos in Transaction processing)
daemon@ATHENA.MIT.EDU (Steve Dyer)
Mon Oct 7 12:06:44 1991
Date: 7 Oct 91 15:05:02 GMT
From: dyer@spdcc.com (Steve Dyer)
To: kerberos@shelby.Stanford.EDU
In article <1991Oct7.133425.1554@ncsu.edu> mauney@adm.csc.ncsu.edu (Jon Mauney) writes:
> I would assume that Kerberized NFS retains the stateless nature of
>standard NFS (implicit question number 1), and thus each request
>for a file read would have to be authenticated; individual sessions last
>only long enough to send one disk block.
In the Kerberized NFS as implemented at Project Athena, a "session" lasts
as long as a student stays logged in to a workstation (or until a locker
[mount point] is detached [unmounted]). State is kept on the server
as regards the mapping of credentials. but this state can easily be
reloaded if a server crashes and is rebooted.
--
Steve Dyer
dyer@ursa-major.spdcc.com aka {ima,harvard,rayssd,linus,m2c}!spdcc!dyer
