freeipmi-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Freeipmi-devel] sensors status - current strategy


From: Anand Babu
Subject: [Freeipmi-devel] sensors status - current strategy
Date: Fri, 20 Aug 2004 18:33:32 -0700
User-agent: Gnus/5.1006 (Gnus v5.10.6) Emacs/21.3 (gnu/linux)

Threshold Sensors Mask:
As per proposal, sensors currently uses threshold mask to filter out
non readable fields. But there are some incorrect threshold masks in
the SDR which enables displaying of few out-of-range fields (only in
the verbose mode). This doesn't affect the interpretation of sensor
status in any ways in the current model. 

Status Determination:
Sensor status is determined differently this time. BMC itself has a
feature to find out the correct status, in fact BMC's report is more
accurate and descriptive. Previous versions attempted to calculate
status flag on its own when the BMC threshold-status-flag indicated
OK-status but SDR fields did not make any sense. By looking at the
verbose display of sensors, we incorrectly concluded that the BMC's
status determination logic was broken :( 

OLD MODEL: (overly smart and buggy)
  if ((BMC STATUS REPORT == OK) and (OUT-OF-RANGE SDR FIELDS))
  then
     calculate manually with the normal min/max range.
  else
     return (BMC STATUS REPORT)

NEW MODEL: (much simpler and in accordance to the spec)
  return (BMC STATUS REPORT)

Discrete sensors doesn't suffer from this problem. Their status is
always determined correctly.

-- 
Anand Babu
Free as in Freedom <www.gnu.org>




reply via email to

[Prev in Thread] Current Thread [Next in Thread]