Re: [Freeipmi-devel] sensor identifcation

[Albert Chu]

Hmmm, something does not seem right here AB.

Record ID: 52
Sensor type: HSC SCSI BP Temp (Temperature)
Sensor System Software ID: 60h
Sensor owner Lun: 0h
Sensor number: #1

Record ID: 53
Sensor type: Pwr Unit Status (Power Unit)
Sensor System Software ID: 10h
Sensor owner Lun: 0h
Sensor number: #1

Both of the sensors have the same sensor number.  The differing software
IDs mean the sensor number is ok (BTW, I checked that they are Software
IDs and not slave addresses).  However, both sensor readings are read by
passing sensor_number=1 to the "Get Sensor Reading" command.  But a
temperature reading is a threshold based reading and a power unit
reading is a discrete reading.  That indicates to me the sensor data is
being read incorrectly.

Should we be ignoring and not reading one of these sensor records?? 
According to table 5-4, software IDs of value > 47h are reserved and
should probably be ignored.  The reading of the first sensor above is
0.00 degrees.

A different potential issue is that the code does not seem to check the
"Discrete/Threshold Reading Mask".  My understanding, is that this mask
(field 19/20 of a full record SDR) tells us which fields within the 
"Get Sensor Reading" response are useful and which are meaningless.

Al

--
Albert Chu
address@hidden
Lawrence Livermore National Laboratory

----- Original Message -----
From: Anand Babu <address@hidden>
Date: Tuesday, March 23, 2004 8:01 pm
Subject: Re: [Freeipmi-devel] sensor identifcation

> Simpler solution is to use SDR record number as index.
> Every SDR rec# maps uniquely to a sensor on the system.
> Sensors without SDR records has no meaning.
> 
> -ab
> 
> ,----[ Albert Chu <address@hidden> ]
> | Hey AB,
> | 
> | I'm guessing now the only way to uniquely id a sensor is:
> | 
> | sensor owner ID && sensor owner LUN && sensor number
> | 
> | Sound about right?
> | 
> | Al
> `----
> 
> ----- Original Message -----
> From: Albert Chu <address@hidden>
> Date: Tuesday, March 23, 2004 12:09 pm
> Subject: [Freeipmi-devel] sensor identifcation
> 
> > Hey AB,
> > 
> > Is the sensor number and the sensor ID string the only ways to 
> > uniquelyidentify a particular sensor?  It seems to be that way, 
> but 
> > I just want
> > to double check.
> > 
> > Al
> > 
> > --
> > Albert Chu
> > address@hidden
> > Lawrence Livermore National Laboratory
> > 
> > 
> > 
> > _______________________________________________
> > Freeipmi-devel mailing list
> > address@hidden
> > http://mail.nongnu.org/mailman/listinfo/freeipmi-devel
> > 
> 
> 
> 
> _______________________________________________
> Freeipmi-devel mailing list
> address@hidden
> http://mail.nongnu.org/mailman/listinfo/freeipmi-devel
> 
> 
> -- 
> _.|_ 
> (_||_)
> Free as in Freedom <www.gnu.org>
>