Hi all,
Thank-you very much to those of you who responded to my help email.
Seems like this is a pretty common error as we can tell from the similar
responses which is all enclosed below. I did find an error in my NIS passwd
file, a missing ":", which caused me all that headache. Thanks again to all
of you who helped me out on this.
*****************************************************************************
From: john@mlb.semi.harris.com (John M. Blasik)
You probably have a malformed password entry. use /usr/etc/pwck to find it.
-- john
From: cil@bilkent.edu.tr (Serkan Cil)
hi,
I had this problem on our Sun 4/490 running SunOS 4.1.3 before. In order to
solve the problem, I found the source of Berkeley finger program and
debuged it. What I found was a syntax error in passwd file. It was such an
error that NIS didn't complain about it while generating the NIS map of
the password file. May be you should search through the password file
to catch a tiny syntax error.
good luck,
______________________________________________________________________
Serkan Cil, UNIX Systems Admin. e-mail: cil@bilkent.edu.tr
Bilkent University phone: 90 (312) 266 4000/2023
Computer Center fax: 90 (312) 266 4126
Bilkent, 06533 Ankara, TURKEY
----------------------------------------------------------------------
From: tony@vegas.es.hac.com (Anthony (Tony) Reeves)
Could be the way the finger client is resolving the address of the person/host
your trying to reach.. you did not say if your using DNS or NIS or NIS+
From: Karl Morgan <exuksm@exu.ericsson.se>
I recently had a problem with finger also. An entry in the password
file had been added (in NIS) which did not have a group ID associated
with it something like:
exuksm:xxxxxxxxxxxxx:101::Karl Morgan, IS, 7-0393:/home/exuksm:/bin/ksh
(Note no group ID)
Doing a ps -ax showed both in.fingerd and syslogd consuming a more
than average amount of CPU.
Another problem I have seen is with inetd not fork/execing in.fingerd
reliably. This can be seen again with ps -ax, while finger is hanging
and before it times out you will see more than one instance of inetd
running. This seems to be a problem with inetd and daemons which start
with a uid other than root. (At least thats how I reported it to Sun).
Hope its one of these for you. Otherwise double check the entry for
in.fingerd in /etc/inetd.conf.
- Karl
From: "Michael (M.A.) Meystel" <MEYSTMA%DUVM.BITNET@pucc.Princeton.EDU>
Hello,
I have had similar problems, and they have always gone away (temporarily,
but for long enough) when I kill -1 the inetd process.
I would be interested to hear if there are any solutions to this.
Thank you,
Mike
-----
Michael A. Meystel
Systems & Network Administrator
College of Engineering
Drexel University
Philadelphia, PA
(215) 895-5807 / meystma@duvm.ocs.drexel.edu
From: bern@penthesilea.Uni-Trier.DE (Jochen Bern)
Well, over *here*, it's a freaked-out Security-before-first Network Admin
who simply blocked the finger Port on his $@&%|+~\! CISCO Box. Without
informing anybody, of course. Do you have a Freak of this Kind who might
have done the same Trick to you?
Regards,
J. Bern
******************************************************************************
Ying-Meei Chew
Systems Administrator
Telebit Corporation
ymc@telebit.com
This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:09:08 CDT