laptop does not recover from suspend

Bug #884017 reported by Rolf Leggewie
14
This bug affects 2 people
Affects Status Importance Assigned to Milestone
gnome-power-manager (Ubuntu)
Confirmed
Undecided
Unassigned

Bug Description

I believe the fix for the rather unimportant bug 620693 introduced a serious regression for me. Recently, quite often I cannot recover a suspended laptop. A hard reboot is the only way out and this has meant lost data and other serious trouble for me at inopportune times. I will use this ticket to collect the necessary information and confirm that bug 620693 is indeed to blame for the regression.

Rolf Leggewie (r0lf)
description: updated
description: updated
Revision history for this message
Rolf Leggewie (r0lf) wrote :

I've noticed that whenever the computer fails to recover from suspend this way on the next reboot there are a bunch of warnings in syslog from ecryptfs like

Nov 13 03:35:01 localhost kernel: [ 183.641252] Valid eCryptfs headers not found in file header region or xattr region
Nov 13 03:35:01 localhost kernel: [ 183.641275] Either the lower file is not in a valid eCryptfs format, or the key could not be retrieved. Plaintext passthrough mode is not enabled; returning -EIO

This is after I cleaned them out from the last time by looking for the 0-sized files in $HOME/.Private. This should not happen since that partition is ext4 with a journal. Could this be yet another ecryptfs bug? Tyler, any comment?

Revision history for this message
Rolf Leggewie (r0lf) wrote :

Now, the funny thing is that even though I got these warnings on the hard reboot that was in order just about 15 minutes ago which produced about 250 lines of ecryptfs-related lines in syslog, there are NO 0-sized files in $HOME/.Private as verified by "sudo find $HOME/.Private/ -xdev -size 0c | tee /tmp/0size.txt".

Revision history for this message
Rolf Leggewie (r0lf) wrote :

although the problem started happening around the time I updated gnome-power-manager, downgrading the package to the previous version does not seem to prevent the problem. Woke up to a computer again this morning that was not coming out of suspend. It seems that if I keep the lid of the netbook open it recovers fine, but not if I close it. The good news is that I believe today at around 10 when I tried to wake the computer up it was able to write a log just before crashing which I attach here. I will remove the non-free slamr kernel module to see if that helps. It's mentioned in the log around the time the crash happened. FWIW, I installed sl-modem packages on Oct. 21 which is also roughly in line with the time the problem started.

I experience this on all installed kernels, I believe.

$ ls -1 /boot/vmlinuz-*
/boot/vmlinuz-2.6.32-23-generic
/boot/vmlinuz-2.6.32-33-generic
/boot/vmlinuz-2.6.32-34-generic
/boot/vmlinuz-2.6.32-35-generic
/boot/vmlinuz-3.0.0-13-generic
/boot/vmlinuz-3.1.0-999-generic

Revision history for this message
Launchpad Janitor (janitor) wrote :

Status changed to 'Confirmed' because the bug affects multiple users.

Changed in gnome-power-manager (Ubuntu):
status: New → Confirmed
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.