Subject: | |
From: | |
Reply To: | |
Date: | Sat, 9 Mar 2002 18:08:43 -0800 |
Content-Type: | text/plain |
Parts/Attachments: |
|
|
Ralph Bagen asks:
==================================
Dear group:
Can anyone share any information on HPMC (high
priority machine check) errors on a 99x server?
The system FLTs with a CxF3 error code (lower
left on the console)... the x has varied from
CAF3, CBF3... CEF3, CFF3... nothing has changed
hardware configuration wise since the box booted
ok this morning.
======================================
Others on the list are much better qualified
than I am to speculate on what all can cause
HPMC errors... I'll just throw out one thing
that I know will cause them: We started
getting HPMC errors on our 959-400 a couple
years ago; after we added two CPU cards to
what was originally a 959-200. Turned out two
CPU boards had intermittant cache problems.
HP eventually replaced the two CPU boards, and
AFAIR we have never had any more HPMC system
halts since then.
SIDEBAR: Don't recall exactly what halts they
caused, but we had another series of system
crashes associated with adding those additional
CPU boards; and with adding more memory to our
959: It was originally powered up new as a
959-200 in December of 1995. Ran very well for
like 4+ years before we made it a 959-400.
After a series of somewhat painful system halts,
I found out that even in a relatively "clean"
environment you do get some surface corrosion
("coating" is perhaps a better word) on
motherboard and memory carrier contacts. It
ended up that all that was required to get a
"solid" connection to the new CPU boards and
SIMMs was to "work them" in and out of the
contacts 10 or 12 times; to scrape through the
surface corrosion... No problem since then..
Ken Sletten
* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *
|
|
|