Last week, I wrote:
------------- Begin Forwarded Message -------------
I have a Solaris 2.5 system (an x86 Pentium Pro from Dell) that
is regularly going south in a most unusual way.
The system appears to lock up and become *mostly* unresponsive at
the console, though it's still pingable on the net. telnetting to
the system succeeds part-way (i.e., the initial connection message
appears, but no login prompt). Exported NFS filesystems become
unavailable.
Here's the wierd part: The user reports that windows on
the system which are *remote* logins to other systems continue
to work, but attempting to do anything with the window manager itself
or anything in a local window fails.
Methinks this is an IDE disk problem, but Dell's diagnostics
don't show anything.
Any ideas?
------------- End Forwarded Message -------------
Thanks to the following for their reponses:
carlo@hub.eng.wayne.edu
Charles Holbrook <cat@northernway.net>
cda <cda@cosmoslink.net
"K. M. Sherif" <sherif@bigboy74.West.Sun.COM>
Marc Tchamitchian <arm@avignon.inra.fr>
Steve Boronski <spb@stoke.gov.uk>
Brian_Easton@enterprise.inet.net (Brian Easton)
"Rodney C. Marable" <marable@firefly.net>
"John Meagher" <jmeagher@clark.net>
Carlo hit the nail on the head (though this started dawning on me
before I get his message--great minds...).
This was not a hardware problem after all, but rather a problem with
corrupted NIS maps. Slave servers had differing views of the
passwd, hosts, and automounter maps. In addition, the master
server was missing the netid map.
Creating the netid map and recreating all the slave servers
(ypinit -s masterservername) has cleared up the problem.
-- Tim Evans | E.I. du Pont de Nemours & Co. tkevans@eplrx7.es.dupont.com | Experimental Station (302) 695-9353/8638 (FAX) | P.O. Box 80357 EVANSTK AT A1 AT ESVAX | Wilmington, Delaware 19880-0357
This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:11:59 CDT