[Pm-utils] Bug? in pm-powersave

[Robby Workman]

On Mon, 3 Jan 2011 00:35:02 -0600
Robby Workman <rw at rlworkman.net> wrote:

> Hi Victor and list,
> 
> I think I've encountered a bug in pm-powersave, or maybe it's
> a bug in how the powersave hooks are processed, or perhaps I'm
> just being braindead ;-) but here's a summary:
> 
> My laptop drive doesn't like the default DRIVE_POWER_MGMT_BAT
> value of "1" - it goes to sleep every 30 seconds and causes
> a delay of 1-2 seconds that makes the system feel like it's
> crawling along.  
> 
> From pm-powersave(8):
>   FILES
>        /etc/pm/power.d/, /usr/lib/pm-utils/power.d/
>            When you run pm-powersave it combines the scripts in 
>            these two directories and executes them in sorted order.
>            If both directories contain a script with the same name,
>            the one in /etc/pm/power.d/ has a higher precedence and 
>            only this one will be executed.
> 
> 
> Therefore, I grabbed $libdir/pm-utils/power.d/harddrive, placed
> a copy at /etc/pm/power.d/harddrive, made sure it retained its
> executable permissions, and edited it down to the following:
> 
>   #!/bin/sh
>   [ -x /sbin/hdparm ] || exit $NA
>   DRIVE_WRITE_CACHE_BAT=0
>   DRIVE_POWER_MGMT_BAT=128
>   # Default devices to operate on
>   DRIVE_LIST="/dev/[hs]d[a-z]"
> 
> However, when I unplug the AC adapter (or run "pm-powersave true"
> manually), the value shown with "hdparm -B /dev/sda" is not 128;
> it stays at whatever it was previously.  /var/log/pm-powersave.log
> shows this:
> 
>   Running hook /etc/pm/power.d/harddrive true:
>   
>   /etc/pm/power.d/harddrive true: success.
> 
> If I remove the /etc/pm/power.d/harddrive file completely, the 
> value shown by hdparm afterward is the default value defined in
> $libdir/pm-utils/power.d/harddrive, and this is what shows in 
> /var/log/pm-powersave.log:
> 
>   Running hook /usr/lib64/pm-utils/power.d/harddrive true:
>   Enabling power management for /dev/sda...Done.
>     
>   /usr/lib64/pm-utils/power.d/harddrive true: success.
> 
> 
> Am I doing something wrong, or is there a bug here?


Well, just damn.  Of course, I was doing something wrong.
It would be *extremely* helpful if I hadn't deleted the
part of the harddrive script that actually *does* something.

Sheesh.  Why can't this sort of mistake ever be noticed 
*before* I mail a public list?  :-)

-RW