[cobalt-users] RaQ XTR:fan failure & highly loaded CPU

["Etienne Antoniutti Di Muro" <etienne@xxxxxxxxxxxxx>]

I (admin) received the attached mail messages from the RaQ XTR.
The "C" Logo Badge started blinking and all services (web,mail,ftp,telnet)
were suspended. I needed to force the RaQ to power down as it was not responding.
The server is a new machine, up since a month,
can't believe that ALL 6 fans failed simultaneously.

The machine is receiving about 100 web hits per day.

Actually room temperature is quite hot (30+ deg C = 86+ deg F), maybe the
cause is that the operating anvirorment conditions were close to the limit??
Anyway, why was the CPU  loaded with all services suspended?

any idea?


tnks in advance
etienne

--------------
Original Msg no1

From:admin
To: sys
Subject:One or more fans have failed

One or more fans have failed.  These fans should be replaced immediately to
avoid overheating.  See below for a listing of the problematic fans.

There are six fans which are located in the middle of the unit in two groups
of three, directly behind the drive bays.  Fans 1-3 are located between the
drive bays and the CPU(s) on the left side of the box, when facing the front
panel.  Fan 1 is near the edge, fan 2 is in the middle, and fan 3 is toward
the center of the box.  Fans 4-6 are located on the right side of the box,
between the drive bays and the power supply.  Fan 4 is toward the center of
the box, fan 5 is in the middle, and fan 6 is nearest the edge of the box. 
Consult your documentation for instructions on how to replace the fans.

Below is a detailed listing of each fan's speed in revolutions per minute and
an indication of the acceptibility of that speed.

Fan #      RPM      Status
__________________________
  
Fan 1:     5699     FAILED
Fan 2:     5699     FAILED
Fan 3:     5399     FAILED
Fan 4:     5399     FAILED
Fan 5:     5099     FAILED
Fan 6:     5399     FAILED

---------------
Original Msg no2

From: admin
To:sys
Subject:CPU has been heavily loaded.

Over the past fifteen minutes, the CPU has been heavily loaded.

This will result in noticible performace loss.  Consider moving some of the
services to other Cobalt servers, or reduce the complexity of the CGI
scripts running on the Cobalt server itself.

1 minute load average:	14.99
5 minute load average:	14.88
15 minute load average:	14.15