[4301] in testers
Re: sgi [8.3.8]: rcp
daemon@ATHENA.MIT.EDU (Greg Hudson)
Thu Jul 8 10:13:41 1999
Date: Thu, 8 Jul 1999 10:13:24 -0400 (EDT)
Message-Id: <199907081413.KAA16303@rcn-sucks.fsck.com>
From: Greg Hudson <ghudson@MIT.EDU>
To: Jonathon Weiss <jweiss@MIT.EDU>
CC: testers@MIT.EDU, jweiss@MIT.EDU
In-reply-to: "[4257] in testers"
I tested from rcn-sucks (NetBSD/Athena 8.2) to whirlpool (Athena 8.3,
manually updated kshd binary). A krb4 rcp appears to succeed if
encrypted logins aren't required; otherwise I get:
kshd: You must use encryption.
kshd isn't currently smart enough to notice that you are doing an
unencrypted krb4 rsh in order to do an encrypted rcp.
> rcp: Not enough space
> rcp: I/O error
These errors are reminiscent of the kind of errors one gets when one's
dotfiles produce output and screw up the rcp protocol. (rcp layered
on top of rsh was one of the things Kerberos should have thrown out
the window alongside the use of urgent data, but that was then.)
> tla:~: rcp -px pie:/var/athena/sshd.truss /tmp/foo$$
> rcp: Arg list too long
This is probably a different problem (different krb4 rsh conventions
of some kind), although I'm not too motivated to look into it just
now.
