NIS: Connection reset by peer

NIS: Connection reset by peer

Post by Fran L » Fri, 19 Nov 1999 04:00:00


Since installing the latest version of NIS, our cluster has been
sending out these messages all during the day:

yp_all: clnt_call: RPC: Unable to receive; errno = Connection reset by peer

These appear as a result of cron jobs "/sbin/rmmod -as"
and "run-parts /etc/cron.hourly".

When I run these commands manually, I do not get this message.

This message does not seem to occur every time the cron is
scheduled, and it happens on all of the machines, including
the master.

The machines nor the yp master were not down when these messages
occurred, and I have restarted the yp services to no avail -- these
messages continue to occur at different times throughout the day.
Even as soon as we get one of these cron messages, I can still
login into the yp domain as a user & rcp files with no problem.

The machines (pentium IIIs & alpha boxes) all have RedHat 6.0 installed.
The versions of yp that are running are the following:

 ypserv - NYS YP Server version 1.3.9 (with tcp wrapper)
 ypbind version 3.3-glibc5

 yp-tools-2.2-1 is installed across the board.

Does anyone have any advice on this situation?  These messages
are getting bothersome.  I really don't know if NIS is hosed
or not.

Fran Lawas-Grodek
(remove "**" to reply)

Frances J. Lawas-Grodek       |    
NASA Glenn Research Center    |     phone: (216) 433-5052
21000 Brookpark Rd, MS 142-2  |     fax  : (216) 433-8000