freeipmi-devel
[Top][All Lists]
Advanced

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

[Freeipmi-devel] RFC: ipmi-sel, ipmi-sensors, ipmimonitoring output


From: Al Chu
Subject: [Freeipmi-devel] RFC: ipmi-sel, ipmi-sensors, ipmimonitoring output
Date: Wed, 14 Jan 2009 09:35:24 -0800

Hey everyone,

Much of the current output from ipmi-sel, ipmi-sensors, and
ipmimonitoring is very legacy.  It was based on old demands and was
semi-arbitrarily (and not consistently) put together.

I've decided I need to clean up the output, make it easier to read, and
make it relatively consistent between the three tools.  Don't worry,
there are options for legacy output :-)  I'll also add multiple output
formats, more verbose vs. non-verbose options, add comma separated
output if it's easier for script parsing, etc.

The new default output will be very similar to what ipmimonitoring is
now, with fields separated by pipes w/ spacing in between for easier
reading.

Record_ID | Sensor Name | Sensor Group | Monitoring Status| Sensor Units
| Sensor Reading
3 | CPU1 Temp | Temperature | Nominal | C | 18.000000 
4 | CPU2 Temp | Temperature | Nominal | C | 14.000000 

I'd to ask people if they have an opinion on how sensor readings should
be output in ipmi-sensors and ipmimonitoring.  Should it be like:

A)

| C | 18.00

B)

| 18.00 | C

or

C)

18.00 C

'C' is the more natural way to read it, but 'A' and 'B' may be easier
for scripting.  In ipmimonitoring, I put the actual sensor reading last
(after the sensor units) b/c it 1) made it easier to truncate output for
the --quiet-readings option, and 2) IMO makes it easier for parsing if
you don't care about the units [1].

I'm inclined for 'A', but I figure I'll give folks a chance to chime in
if they dislike it for some reason or if they think there is a better
output pattern.

Al

[1] - Obviously everyone cares about units.  I mean don't care w/n the
confines of your script b/c perhaps you already know everything is in
Celsius.

-- 
Albert Chu
address@hidden
Computer Scientist
High Performance Systems Division
Lawrence Livermore National Laboratory





reply via email to

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