Original query:
---------------
> I have been having some problems with an Enterprise 2 server with dual
> 200Mhz CPU modules. It is running Solaris 2.5.1 and is an AFS file and
> database server. Under heavy loads it reboots and gives me the following
> type of errors. It sounds like a hardware error, but is it?
>
> Mar 25 15:16:02 hepnrc unix: panic[cpu1]/thread=0x507b0720: CPU1 Ecache
> SRAM Dat Ecache SRAM Data Parity Error: AFSR 0x00000000 80408000 AFAR
> 0x00000000 60000000 80408000 AFAR 0x00000000 60000000
> Mar 25 15:16:02 hepnrc unix: syncing file systems... [16] 18 [16] 18 [16]
> 18 [16] 18 [16] 18 [16] 18 [16] 18 [
.....
The Solution:
--------------
Thanks for the help. It turns out that the cache on my CPU modules
was bad. The UltraSparc 200mhz rev -04 has a problem with cache
memory. If you experience errors with these contact sun. Thanks
to the following individuals. My appologies to the one person who
specifically pointed out the problem with the -04, his email got
garbled with another message by pine. I lost all trace of him
(well, except for syslog).
akyol@lightning.Stanford.EDU
James Hsieh <jhsieh@soe.ucsd.edu>
Jay Lessert <jayl@latticesemi.com>
Glenn Satchell <Glenn.Satchell@Uniq.com.au>
Thanks to everyone who responded.
Jeff Dingbaum HEP Network Resource Center @ Fermilab
dingbaum@hep.net PO Box 500, MS368
system admin, webmaster, Batavia, IL 60510-0500
postmaster, coffeemaker (630)840-8472 (630)840-8463 fax
This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:11:50 CDT