ProLiant Servers (ML,DL,SL)
cancel
Showing results for 
Search instead for 
Did you mean: 

ML350 & DL380: Temperature Warnings etc.

 
SOLVED
Go to solution

ML350 & DL380: Temperature Warnings etc.

Hi;

I have several servers, all proliant.
Some of them often get the warning, event ID 1135:
System Information Agent: Health: A Temperature Sensor Condition has been set to degraded. The system may or may not shutdown depending on the state of the thermal degraded action value '3'.
Chassis: '0'; Location: '6'

The server often reboots afterwards.

Is there a way to configure when to reboot, when to sent warnings, .. In other words, to configure the temperature ?

Or perhaps, is there something wrong when this type of error occurs.
The temperature in the server room was 26°C yesterday when it last occured.

Thanks !

Frank Jacobs
Senior System Administrator
2 REPLIES
Andrew Young_2
Honored Contributor
Solution

Re: ML350 & DL380: Temperature Warnings etc.

Hi Frank

The answer indirectly is NO the temperature thresholds are hardcoded, usually in the circuitry itself.

Just be warned that because the ambient temperature in your computer room is low, that does not mean that there are not hotspots.

Computer cooling is more a function of airflow than actual temperature. Our last datacentre the airconditioning was so badly designed with the vents located in all the wrong places and as result we often had overtemp conditions. The only viable solution was to drop the ambient temperature to as low as 19C.

Also remember that servers internally run naturally warm, usually about 10-15C over the ambient temperature of their environment so its very easy to get to a 40-50C internal temperature and trigger an overtemp condition.

Nearly all servers these days have airflow running front to rear and not the bottom to top of days of old - its still a bit freaky talking like that about another era only 15-20 years ago. The only equipment I still have with that configuration is an old XP array circa 2000.

So just watch for hot spots in your computer room and make sure there are no airflow obstructions.

Regards

Andrew Y

Si hoc legere scis, nimis eruditionis habes
David Claypool
Honored Contributor

Re: ML350 & DL380: Temperature Warnings etc.

You can view the current temperature at each sensor and the thresholds on the Environment page of the System Management Homepage (Insight agent). Warning and Critical thresholds are set based on the spec of the component being protected, the proximity of the sensor to the component and the airflow past the sensor. These are not configurable.

You can configure the system's behavior when the system reaches an overtemp condition to be a graceful shutdown or continue operation.