Gazelle Value 3 does not resume from suspend

Bug #108253 reported by AusIV
6
Affects Status Importance Assigned to Milestone
System76
Fix Released
High
Carl Richell
System76 Driver
Fix Released
High
Carl Richell

Bug Description

As of feisty, my Gazelle value fails to return from suspend. It does, however, resume from hibernation.

Tags: gazv3
Revision history for this message
AusIV (linux-ausiv) wrote :

/etc/default/acpi-support is attached

Revision history for this message
Carl Richell (carlrichell) wrote :

to fix:

gksu /etc/default/acpi-support

change the lines that read:

# Change the following to "standby" to use ACPI S1 sleep, rather than S3.
# This will save less power, but may work on more machines
ACPI_SLEEP_MODE=mem

To:

# Change the following to "standby" to use ACPI S1 sleep, rather than S3.
# This will save less power, but may work on more machines
ACPI_SLEEP_MODE=standby

As the note says it doesn't conserve as much power but it is reliable. This was the setting we used with Edgy. Looks like for the time being we'll have to continue with S1 sleep.

Changed in system76:
status: Unconfirmed → Confirmed
Revision history for this message
Neale Pickett (neale) wrote :

I have a Gazelle Value 3 (gazv3 in the driver installer), with the same S3 problem. I changed ACPI_SLEEP_MODE from "mem" to "standby" back in edgy and now it comes back but the wireless card moved from eth2 to eth1 on resume. Today's upgrade to feisty changes that behavior so the card disappears completely on resume. There's nothing in dmesg. Going to "mem" has the same problem.

I have installed system76-drivers. Kernel version is 2.6.20-15-generic #2 SMP

When I issue "echo -n standby > /sys/power/state" as root, oddly, wireless comes back just fine. The LED stays on during suspend when I do this. Could the problem be that acpi-support is trying to remove the networking modules? (lsmod shows the ipw3945 module before and after suspend, but it is in a different position in the output.)

Hibernate doesn't do anything aside from break wireless--it never goes to sleep. I haven't repartitioned the hard drive since I got the machine.

Would a reinstall of the entire OS from scratch maybe help? I'm getting pretty frustrated with this broken machine, the Gazelle Value I bought a year earlier (gazv2) works so well, this newer one is disappointing.

Revision history for this message
AusIV (linux-ausiv) wrote :

I've gone back to edgy for the time being, but I suspect the reason your machine won't hibernate is that the swap partition isn't being loaded. I had the same thing happen when I did a dist-upgrade. Find your swap by typing "sudo fdisk -l", then type "sudo swapon <device>".

This won't solve the problems with wireless, but it should put your machine into hibernate.

Revision history for this message
Scott Henson (scotth) wrote :

I'm having this same problem. I would really like S3 sleep and Ive been poking at my laptop to see if I could figure it out. One thing that I noticed while I was doing that is that the start of resume from suspend looks a lot like a boot that never gets through to post. Are we sure that Ubuntu even gets the chance to resume. It looks to my amatuer eye like this is a bug in the bios.

Changed in system76:
importance: Undecided → High
Changed in system76-driver:
importance: Undecided → High
status: Unconfirmed → Confirmed
Changed in system76:
assignee: nobody → carlrichell
Changed in system76-driver:
assignee: nobody → carlrichell
Revision history for this message
Carl Richell (carlrichell) wrote :

fix committed to system76-driver release 2.0.5 - release expected June 8th or early the week of June 11th.

Changed in system76-driver:
status: Confirmed → Fix Committed
Changed in system76:
status: Confirmed → Fix Committed
Changed in system76-driver:
status: Fix Committed → Fix Released
Changed in system76:
status: Fix Committed → Fix Released
Revision history for this message
Neale Pickett (neale) wrote :

This didn't appear to do anything other than modify /etc/default/acpi-support (it says "standby" right now, I tried "mem" too and the hard drive doesn't even turn off). In "standby" mode the fan never turns off, and 1 time out of 8 or so the machine freezes trying to resume, requiring a cold boot (hold down power for 5 seconds). In "mem" it looks like the screensaver just turns on.

How can I verify that I'm running the correct version of the driver, and what sleep mode does this version of the driver support? What are the chances this ACPI thing will be addressed in the upstream kernel at some point?

I'll probably try to move to gutsy, and if that doesn't work, downgrade to edgy. And if that doesn't work, my wife (it's her computer) will probably lose patience and try to return it on warranty.

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.