The high temperature threshold cannot be changed, but you can modify the relative temperature percentage. You can adjust the temperature percentage under DFM > Configuration > Polling and Thresholds > Managing Thresholds. Click on the Switches and Hubs group, then adjust the relative temperature threshold. Something like 34% should be more in line with what you're looking for.
As for the other event, this is CSCta08882:
For the VSS switch, components are recognized correctly:
1. The sub-modules on the device are listed under the Card section of Detailed Device View. Some of them are obviously incorrect, such as "Chassis 1 CPU of Switching Processor 5" , which should be listed under CPU section. Instead, as Cards, they are given status of NotPresent, and throw alarms for Card OperationallyDown.
2. Non-temperature sensors (such as voltage sensors) are miscategorized as temperature sensors. This results is false temperature out of range alerts.
Conditions:
DFM 3.1.x
VSS switch (sysObjectID .1.3.6.1.4.1.9.1.896)
Workaround:
1. To avoid the false alerts, a SNMP view can be used to mask out the Entity MIB entries for the submodules that are mistakenly being listed under Card section.
eg.
snmp-server view test entPhysicalEntry.*.1000 include
snmp-server view test entPhysicalEntry.*.2000 include
...
snmp-server view test entPhysicalEntry exclude
snmp-server view test iso include
However, there is no workaround to the fact that those submodules will not be properly managed/monitored.
Further Problem Description:
The processor component is not discovered from Incharge. Thus the sub-modules of processor components are displayed randomly in Card component of DDV, and have a status of NotPresent, and will generate false alerts.
