SUMMARY: Using scat to identify CPU hog

I found a helpful command to at least get some of the information:

tlist -s pctcpu

-----Original Message-----
From: sunmanagers-bounces@sunmanagers.org
[mailto:sunmanagers-bounces@sunmanagers.org] On Behalf Of Brian Lucas
Sent: Wednesday, February 15, 2006 8:35 AM
To: sunmanagers@sunmanagers.org
Subject: Using scat to identify CPU hog

Mighty gurus,

Yesterday a Sun E3500 hung and had to be STOP-A'd.  In attempting to
diagnose why this critical system froze, I installed scat to analyze the
crash file hoping for the magic output to appear.  Since I am new to
scat and the documentation isn't too helpful other than giving me
various commands I can run (which I have poured through), I was hoping
someone could give some pointers on crash analysis.  I believe the
system was hung at the CPU level and this line from the analysis output
tends to support that

==== there are runnable threads, may have a cpu hog ====

==== reporting stopped threads ====

  thread        pri pctcpu           idle   pid         wchan command
      0x140a000  96  0.841          0.65s     0             0 sched
  0x2a100017d40 109  0.000 157d16h54m5.30s     0             0 sched
  0x2a1003dfd40 109  0.000 157d16h54m5.30s     0             0 sched


How can I find what specific process froze the CPU?  The only output
from the panic information is that the system was STOP-A'd which of
course makes sense.  I would really like to know what was on the CPU at
the time it froze.

Google and the archives did not reveal this information to me so I turn
to you.  I have no sunsolve contract so contacting them is out of the
question.

Much thanks,
Brian
_______________________________________________
sunmanagers mailing list
sunmanagers@sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers
_______________________________________________
sunmanagers mailing list
sunmanagers@sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers