 |
FECO: | 3000 # 64 |
| MODIFICATION No: | N/A | |
| PRODUCT: | NCR 4300 | |
| SUB-ASSEMBLY | N/A | |
| ESTIMATED MAN HOURS | N/A | |
| CLASS OF CHANGE: | INFORMATIONAL | |
| REASON FOR CHANGE: | OVERTEMPERATURE CONDITION |
FECO 3000 # 64
|
FECO: | 3000 # 64 |
| MODIFICATION No: | N/A | |
| PRODUCT: | NCR 4300 | |
| SUB-ASSEMBLY | N/A | |
| ESTIMATED MAN HOURS | N/A | |
| CLASS OF CHANGE: | INFORMATIONAL | |
| REASON FOR CHANGE: | OVERTEMPERATURE CONDITION |
1.Reason for change.
4300 systems may potentially power themselves off due to a
"perceived" over
temperature condition.
2.units affected.
Units manufactured from early december 1996 to late january 1997, running mp-ras.
3.cause>
Noise generated on vcc by the simms, uncovered a software bug that
caused the
problem. Failures occur at room temperature and are configuration
dependent. Units
exhibiting the problem were lightly loaded with disks; with four
processors and four
64mb or 128mb simms.
(tech serv ref 97/20)
(cssb - 97-00031)
4.symptoms.
4.1the problem manifests itself during heavy processor and memory
Activity with "over-temperature" warning messages followed by a system power off.
4.2the problem was found by ncr during a manufacturing stres test
But customer applications could generate the same level of activity
and problem.
Warning messages will appear on the console, stating that the system
temperature is
approaching the upper limits. This may continue for some time
alternating with
messages saying that the temperature status is okay. If the activity is
heavy enough,
the warning messages will state that the system temperature limit has
been exceeded
and that the system will shutdown in 10 seconds.
However, the system may or may not shut down if the software
determines the
temperature is okay before the shutdown starts.
5.resolution
Patch psmiba301 version 97.01.29.13 or later needs to be applied to
the unit. This patch
is available via technical services.
6.if the unit continues to power off due to an over temperature
Condition after the patch is applied, it is likely a real over
temperature condition exists
which should be investigated to avoid damage to the unit.
(tech serv ref 97/20)
(cssb - 97-0031)