hdparm sets power management value to 128 in AC mode

[Nicolò Chieffo]

Binary package hint: hdparm

Currently in lucid, when you boot the laptop you have a value of 128 (instead of 254) in hdparm -B, even if the laptop is plugged in the AC cable.

I tried to manually set apm value in /etc/hdparm.conf but it does not work

ProblemType: Bug
Architecture: amd64
Date: Thu Jan 28 11:09:46 2010
DistroRelease: Ubuntu 10.04
Package: hdparm 9.15-1ubuntu7
ProcVersionSignature: Ubuntu 2.6.32-11.15-generic
SourcePackage: hdparm
Uname: Linux 2.6.32-11-generic x86_64

[Kevin Funk (kfunk)]

Same for me. Here it the -B parameter is even set to 1. This happens after some recent updates (since one week I guess).
Very annoying as -B 1 spins down the disk after a short period. I have to reset the parameter to a higher value each startup to avoid disk damage.
Modifying /etc/hdparm.conf doesn't work aswell.

[thomas-student (thomas-student)]

THE BUG IS:

The file /usr/lib/pm-utils/power.d/95hdparm-apm owned by package "hdparm" version 9.15-1ubuntu9 in lucid checks for presence of laptop-mode-tools at the wrong location on the disk, then proceeds to manage the APM setting for the hard drive itself when laptop-mode-tools should be handling it.

Specifically, the files /usr/sbin/laptop_mode and /var/run/laptop-mode-tools/enabled have AFAICT been deprecated/moved. This:

    if [ -e /usr/sbin/laptop_mode ] ; then
            LMT_CONTROL_HD_POWERMGMT=$(. /etc/laptop-mode/laptop-mode.conf && ec
            if [ "$LMT_CONTROL_HD_POWERMGMT" != 0 ] \
               && [ -e /var/run/laptop-mode-tools/enabled ]
            then
                    # Laptop mode controls hdparm -B settings, we don't.
                    return
            fi
    fi

(which comes from the described file) includes the uses of these paths that urgently need correcting.

[thomas-student (thomas-student)]

Actually, that only fixed AC plugs/unplugs. There is STILL a problem with suspend/resume on AC.

[thomas-student (thomas-student)]

If you purge package laptop-mode-tools and then edit /etc/hdparm.conf to add:

apm = 254
apm_battery = 254

after this line (line 41):

#apm = 255

Then all is well. It doesn't like laptop-mode-tools at all, though.

[thomas-student (thomas-student)]

Responding to an email from someone, purging laptop-mode-tools if it's already uninstalled would be 'sudo dpkg --purge laptop-mode-tools'.

I have no idea why laptop-mode-tools's config files are still causing SOMETHING on the system to call hdparm -B128 / 254 when laptop-mode-tools is uninstalled. It would not be hard to write a wrapper for hdparm and get a stack trace of whoever's calling, but I don't have time for that.

[thomas-student (thomas-student)]

OK, so somehow I had a different, newer version of laptop-mode-tools than lucid. I probably got it accidentally from a PPA. My original ("THE BUG IS") bug report is sort-of bunk as far as lucid goes, though this may appear in the future if (when) hdparm is updated.

The normal "fix" (see above) of purging laptop-mode-tools and adding regular old /etc/hdparm.conf lines for apm=254 and apm_battery=254 worked once I downgraded to the lucid's laptop-mode-tools. For most people having frequent laptop-drive-kerchunks and high load cycle numbers from SMART, this will probably be all that you need to do.

I apologize for the noise.

As for why others are still having this problem, I would suspect that it's either a bug/difference in laptop-mode-tools or (if laptop-mode-tools is removed but not purged) that some other package is utilizing the setting from laptop-mode-tools.conf improperly. I think that some universe gnome-applet or udev script might be doing this, but since I'm not having the problem myself, I can't really help there.

I'm going to leave this set Incomplete so that the other person who is having the problem can still use this bug report to seek a solution.

[Launchpad Janitor (janitor)]

[Expired for hdparm (Ubuntu) because there has been no activity for 60 days.]