Page 1 of 3 123 LastLast
Results 1 to 10 of 23

Thread: Computer Failed To Suspend Maverick Meerkat

  1. #1
    Join Date
    Apr 2010
    Beans
    53
    Distro
    Ubuntu

    Maverick Fails to Suspend with laptop-mode-tools

    Hello,

    After installing Ubuntu 10.10 x64 today, my laptop will not suspend. The suspend option no longer appears in my indicator applet. If I hit the suspend hotkey on my keyboard, I get the message:

    Computer failed to suspend.

    Failure was reported as: Cannot suspend

    Any ideas? I didn't have a single problem with suspending on 10.04 x64. The ability to sleep is pretty important to me. My laptop is an Acer Aspire 6930.

    [UPDATE] I found in another thread that installing laptop-mode-tools automatically removes the pm-utils package, without which my computer will not suspend. I reinstalled pm-utils, but unfortunately that uninstalls laptop-mode-tools. I suggest a bug be written for this, if it has not already.
    Last edited by powerofpi; October 22nd, 2010 at 04:41 PM. Reason: more precise title needed.

  2. #2
    Join Date
    Oct 2008
    Beans
    105
    Distro
    Ubuntu

    Re: [SOLVED] Computer Failed To Suspend Maverick Meerkat

    Any hope to continue to use laptop-mode-tools while being able to suspend? ...

  3. #3
    Join Date
    Apr 2010
    Beans
    53
    Distro
    Ubuntu

    Re: [SOLVED] Computer Failed To Suspend Maverick Meerkat

    Quote Originally Posted by lores View Post
    Any hope to continue to use laptop-mode-tools while being able to suspend? ...
    Not that I can tell, but I hope someone posts back to the contrary.

  4. #4
    Join Date
    Oct 2008
    Beans
    105
    Distro
    Ubuntu

    Re: [SOLVED] Computer Failed To Suspend Maverick Meerkat

    Not good... I'd suggest u delete the "SOLVED" tag.

  5. #5
    Join Date
    Aug 2008
    Beans
    2

    Re: Computer Failed To Suspend Maverick Meerkat

    On my Toshiba A200-1MB (C2D T5300, Intel 945 with integrated graphics), suspend also broke in Maverick (the 64-bit version). That is, it still works when I boot with 2.6.32 kernel which is a leftover from before the update. On the current 2.6.35-22-generic, when I try to either suspend or hibernate, the computer freezes, and I can only reboot it. Surprisingly (?), SysRq+B does the reboot, which hints that the kernel is still working quite properly.

    Anyway, the current 2.6.35-22-generic kernel seems to be a bit "off" overall. First of all, the "quiet" kernel option is not working, and I still get all the boot messages until the splash shows up. Not that I care, I just treat it as a symptom that something is not right.

    Nothing interesting in the logs, but when I try to use hibernate, I can see the kernel messages, and the last is:

    Code:
    Suspending console(s) (no_console_suspend to debug?)
    When I booted the system with no_console_suspend appended to kernel options and tried to hibernate, it displayed a couple of messages, and then... was still working, at least I could Alt+Fx to the terminals. When I tried to Alt+F7 back to X, it hanged, though.

    I hope for a kernel update in a couple of days... I can live without suspend and hibernation for a while, I don't use them that often, but it would be nice to have them back working.

  6. #6
    Join Date
    Oct 2010
    Beans
    1

    Re: Computer Failed To Suspend Maverick Meerkat

    Yes.I have the same problem. After I reinstall pm-utils, the suspend option appear again. However, the laptop-mode-tools was automatically removed.
    Hope next kernel update would solve it.

  7. #7
    Join Date
    Oct 2008
    Beans
    105
    Distro
    Ubuntu

    Re: [SOLVED] Computer Failed To Suspend Maverick Meerkat

    After removing the laptop-mode-tools package and reinstalling pm-utils and acpi-support, I actually managed to apply laptop-mode-tools directly from http://samwel.tk/laptop_mode/packages/tarball , but there are multiple issues. At the moment, I am trying to make it "refresh" automatically when AC-state changes (if I run "service laptop-mode start", all settings seem to be applied correctly and the battery-life gain is obvious; but otherwise, the laptop-mode won't get activated nor deactivated).

    @rnamatrix: just a few mins ago, there has been a kernel upgrade - nothing changed.

  8. #8
    Join Date
    Oct 2008
    Beans
    105
    Distro
    Ubuntu

    Re: [SOLVED] Computer Failed To Suspend Maverick Meerkat

    OK. I think I got it to work. My quick workaround-howto follows. (Warning: it really is a quick workaround - this includes the possibility of multiple errors/omissions in my instructions, as I was recreating them after getting LMT to work).

    (You need to have pm-utils and acpi-support installed). "Install" (fortunately, it's just copying over a few scripts, I suppose) LMT from http://samwel.tk/laptop_mode/packages/tarball (nb: this is a newer version than the one available in the reps; it has one quite noticeable improvement):

    Code:
    wget http://samwel.tk/laptop_mode/tools/downloads/laptop-mode-tools_1.55.tar.gz
    tar xzf laptop-mode-tools_1.55.tar.gz
    su
    cd laptop-mode-tools_1.55.tar.gz
    ./install.sh
    This allows you to issue "laptop_mode" as root and have LMT automatically detect the AC-state and behave appropriately. If you want to automate this process (which you probably should, as it used to be done automatically in 10.04 with the repo's LMT version and config), you can use my approach. It is redundant, as I find it quite crucial for LMT to always kick in automatically - even if some of the soft fails. You can remvoe redundancy (or shorten the 2nd timeout), thus risking exposure to conflicts/overwrites resulting from pm-utils'/acpi-support's doing equivalent job as LMT.

    Code:
    su
    nano /etc/acpi/power.sh
    add
    Code:
    /root/power.sh.4sec
    /root/power.sh.70sec
    directly after
    Code:
    #!/bin/sh
    save with CRTL O and exit with CTRL X

    Code:
    nano /etc/pm/sleep.d/10_lm
    insert
    Code:
    #!/bin/sh
    sleep 4 && /usr/sbin/laptop_mode auto force > /root/4_pm_log_lm_reload &
    Code:
    chmod +x /etc/pm/sleep.d/10_lm
    nano /etc/pm/sleep.d/10_lm70
    insert
    Code:
    #!/bin/sh
    sleep 70 && /usr/sbin/laptop_mode auto force > /root/70_pm_log_lm_reload &
    Code:
    chmod +x /etc/pm/sleep.d/10_lm70
    ln -s /etc/pm/sleep.d/10_lm70 /etc/pm/power.d/10_lm70
    ln -s /etc/pm/sleep.d/10_lm /etc/pm/power.d/10_lm
    
    nano /root/power.sh.4sec
    insert
    Code:
    #!/bin/sh
    sleep 4 && /usr/sbin/laptop_mode auto force > /root/s4h_power_log_lm_reload &
    Code:
    nano /root/power.sh.70sec
    insert
    Code:
    #!/bin/sh
    sleep 70 && /usr/sbin/laptop_mode auto force > /root/s70h_power_log_lm_reload &
    Code:
    chmod +x /root/power.sh.*
    This will force a LMT reload every time the PC changes it's power state (AC/bat) or goes to sleep/resumes. There are to delays: the first one of 4 s and the second one of 70 s, as I experienced LMT's settings' being overwritten after the initial (4 s) reload. Both of them are being executed by both PM and acpid, so there's 2x2x redundancy. After some testing or restricting PM-utils, it probably could be reduced significantly.

    Go ahead and customise as you please. The humble "log" files in /root/ along with their timestamps are good for debugging. Check status with "laptop_mode status" (or a shortened version, e. g. "laptop_mode status|head -n 28|tail -n 10").


    Let's hope to see a serious solution enter the repos soon...


    EDIT:
    After testing my workaround for some time, I concluded that it is enough to use pm only (no need for power.sh in acpi, so you can comment out the two lines) and that the second timeout can safely be lowered to 16 s. I'm looking into rerunning "laptop_mode" when the battery's state becomes critical in order to disable data-sensitive features - atm, it doesn't do that automatically.
    Last edited by lores; October 19th, 2010 at 12:07 PM.

  9. #9
    Join Date
    Apr 2010
    Beans
    53
    Distro
    Ubuntu

    Re: Computer Failed To Suspend Maverick Meerkat

    Thank you for the workaround! I've been wondering, why the heck is LMT not installed by default with Ubuntu and integrated along with a nice GUI into the power settings? Without LMT my battery life is considerably worse than Windows, but with it (configured) it's comparable or better. Considering the number of laptops Ubuntu finds its way onto, I would expect this to be a standard feature...

  10. #10
    Join Date
    Oct 2008
    Beans
    105
    Distro
    Ubuntu

    Re: Computer Failed To Suspend Maverick Meerkat

    I agree. The argument against integrating LMT into Ubuntu used to be compatibility issues (LMT seemed to sometimes hang some PCs; some options didn't work as expected). However, there are multiple options it provides that are very safe while still providing considerable power savings. And even the "hard-core" options would be way easier to enable and test if the soft were installed by default...

    BTW: After testing my workaround for some time, I concluded that it is enough to use pm only (no need for power.sh in acpi, so you can comment out the two lines) and that the second timeout can safely be lowered to 16 s. I'm looking into rerunning "laptop_mode" when the battery's state becomes critical in order to disable data-sensitive features - atm, it doesn't do that automatically.

Page 1 of 3 123 LastLast

Tags for this Thread

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •