Resume after pm-suspend with read-only ext4 root file system on USB disk (ext4 journal commit I/O error) )

Bug #901622 reported by Lars Steinke
26
This bug affects 5 people
Affects Status Importance Assigned to Milestone
pm-utils (Ubuntu)
Confirmed
Undecided
Unassigned

Bug Description

After the upgrade to Oneiric resume from pm-suspend no longer works, might be connected to change of kernel from 2.6.38 to 3.0.0... Issue also reported for Arch Linux and kernel 2.6.39 here: https://bbs.archlinux.org/viewtopic.php?pid=1008743
See attached dmesg output for details

This is for kernel 3.0.0-14-server x86_64 (but also occured on 3.0.0-13) and a NAS booting from a USB stick.
---
ApportVersion: 1.23-0ubuntu4
Architecture: amd64
DistroRelease: Ubuntu 11.10
Package: pm-utils 1.4.1-8ubuntu1
PackageArchitecture: all
ProcEnviron:
 PATH=(custom, no user)
 LANG=de_DE.UTF-8
 SHELL=/bin/bash
ProcVersionSignature: Ubuntu 3.0.0-14.23-generic 3.0.9
Tags: oneiric running-unity
Uname: Linux 3.0.0-14-generic x86_64
UpgradeStatus: Upgraded to oneiric on 2011-10-17 (51 days ago)
UserGroups: adm admin audio cdrom dialout dip fax floppy fuse libvirtd lpadmin netdev plugdev powerdev sambashare scanner tape video

Revision history for this message
Lars Steinke (lss) wrote :
Revision history for this message
Lars Steinke (lss) wrote : Dependencies.txt

apport information

tags: added: apport-collected oneiric running-unity
description: updated
tags: added: server
removed: running-unity
Revision history for this message
Lars Steinke (lss) wrote :

Nope, booting with 2.6.38-13-server exhibits the issue also, so something else must have changed between natty and oneiric...

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

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

Changed in pm-utils (Ubuntu):
status: New → Confirmed
Revision history for this message
ShamrockUK (uplift) wrote :

This is affecting me too running XBMCbuntu, which I believe is 11.10 under the hood. No problem with suspend/resume when running from a hard drive, but now I've switched to using an SD card instead I get the same dmesg entries and ext4 errors followed by a read-only filesystem on resume. It's not quite 100% repeatable but certainly happens more often than not.

I'd be happy to provide logs etc, but may need a little guidance on what to provide if anyone is looking at this further.

Revision history for this message
Brian Harkness (maestro-bwh) wrote :

Affects me on a MacBook air 5,2 both with card in reader and USB mounted reader but on both asus computers I have, suspend from USB works and ext4 remains mounted as rw. Using Ubuntu 12.04.

Revision history for this message
Brinstar (brinstar) wrote :

Not sure if this is the same bug but I'm having problems with an ext4 USB drive being recognised in Ubuntu 16.04.3 after waking from suspend

Revision history for this message
Adrian Fonseca (af2623) wrote :

I have the same problem. I have /home in a separated filesystem, and every time a wakeup my lap this ext4 partition is mounted as ro.

Revision history for this message
Donovan (dgxinfinity) wrote :

Same issue as everyone else. I have my system(/home and the likes) on the SD Card and /boot on the internal SSD. I have my OS installed this way is because my Laptop BIOS doesn't allow booting directly from SD Card(perhaps because internally the SD Reader is connected via the PCI bus) This way the laptop boots to the regular destination(Internal SSD) then GRUB loads and then I can select my desired OS, then GRUB will find the Kernel(/boot) and then handoff to the SD Card. This all works as expected with the only problem being that my system will not suspend no matter how I try. I see a flashing cursor on screen while trying to resume from suspend and then I will get read-only errors. The only way to get past this is to do a full reboot. Most of the time requiring a full fsck to fix the orphaned inodes, etc. So what is the advisable solution? Is the problem the fact that my SD Card reader is on the PCI bus and therefore is having trouble with the hotplug driver? Could a plausible solution be to disallow the system from unmounting my SD Card? I mean after all this SD Card while *removable* isn't actually going to be removed during operation of the system, much less while in hibernate. As far as I can tell the system unmounts all sd cards so as to prevent data loss, assuming that it can be removed during suspend, and that is all fine as the majority of people benefit from this decision however how do I disable this feature? Any help would be greatly appreciated.

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.