[16929] in Athena Bugs
Re: telnetd in 8.3
daemon@ATHENA.MIT.EDU (Greg Hudson)
Wed Jun 30 11:03:53 1999
Message-Id: <199906301503.LAA26014@r93aag001866.sbo-smr.ma.cable.rcn.com>
To: "Thomas H. Grayson" <thg@MIT.EDU>
cc: bugs@MIT.EDU, jf@MIT.EDU, karen@MIT.EDU, phils@MIT.EDU
In-reply-to: Your message of "Wed, 30 Jun 1999 10:51:13 EDT."
<199906301451.KAA00650@depot.mit.edu>
Date: Wed, 30 Jun 1999 11:03:37 EDT
From: Greg Hudson <ghudson@MIT.EDU>
Hi. The first and second problems you mention have to do with a
Hesiod problem we've been having with the BIND upgrade in the Athena
8.3 release. (This is related to an operational change made to the
Hesiod servers. I'm not entirely sure why the operational change
hasn't been backed out yet, and I'll complain to ops about that.)
I've included some mail about the problem below.
If you notice a machine losing, I believe you can make to stop losing,
at least temporarily, with the command:
dig @machinename ns.athena.mit.edu ns
with "machinename" replaced with the hostname of the machine. You can
run that command remotely (which is good, since you might possibly
have trouble logging into the problem machine).
We hadn't heard about the third problem yet (cygnus ktelnet with
forced encrypted logins). Hopefully we'll look into it soon.
cfyi [9617]: Hesiod problem on 8.3 machines
Date: Sun, 27 Jun 1999 10:47:27 -0400 (EDT)
Message-Id: <199906271447.KAA126873@antharia.mit.edu>
From: Dan Winship <danw@MIT.EDU>
To: cfyi@MIT.EDU, sipb-office@MIT.EDU, testers@MIT.EDU
Cc: ops@MIT.EDU
Some 8.3 machines have been having intermittent Hesiod problems since
ops changed the IP address of suomi, one of the Hesiod servers, as
part of a machine room reorganization Thursday night. (There have been
no reports of 8.2 or earlier machines having problems.)
Symptoms might include:
- Not being able to log in to a machine ("username unknown")
- Being able to log in to a private workstation, but getting a
temporary homedir
- Having "inc" report that you have no mail when you know that you
do.
Current evidence seems to suggest that forcing the machine to resolve
the address of one of the Hesiod servers (eg, "hostinfo apollo") will
fix the problem at least for a while. We're trying to figure out how
to fix it completely.
-- Dan
*** End of Transaction ***
