package linux-image-3.2.0-23-generic 3.2.0-23.36 failed to install/upgrade: ErrorMessage: subprocess installed post-installation script returned error exit status 2

This bug has been reported on the Ubuntu ISO testing tracker.

A list of all reports related to this bug can be found here:
http://iso.qa.ubuntu.com/qatracker/reports/bugs/986797

Please also see bug 986791, bug 986792, and bug 986798. These are all related to the same Oneiric -> Precise upgrade.

I took a screenshot of the error during upgrade so I'm attaching it, the full text message was, "subprocess installed post-installation script returned error exit status 2".

Examining the contents of VarLogDistupgradeApttermlog.gz I see the following:

update-initramfs: Generating /boot/initrd.img-3.2.0-23-generic
run-parts: executing /etc/kernel/postinst.d/pm-utils 3.2.0-23-generic /boot/vmlinuz-3.2.0-23-generic
run-parts: executing /etc/kernel/postinst.d/update-notifier 3.2.0-23-generic /boot/vmlinuz-3.2.0-23-generic
run-parts: executing /etc/kernel/postinst.d/zz-update-grub 3.2.0-23-generic /boot/vmlinuz-3.2.0-23-generic
/usr/sbin/grub-probe: error: cannot find a device for / (is /dev mounted?).
run-parts: /etc/kernel/postinst.d/zz-update-grub exited with return code 1
Failed to process /etc/kernel/postinst.d at /var/lib/dpkg/info/linux-image-3.2.0-23-generic.postinst line 1010.
dpkg: error processing linux-image-3.2.0-23-generic (--configure):
 subprocess installed post-installation script returned error exit status 2

Which seems to indicate this is actually an error with grub. I'm going to mark this as a duplicate of bug 916077 for now and will follow up with the release team regarding this issue. Please continue to follow 916077 for this issue. Thanks.

Not a duplicate - grub-installer runs during the installer, this clearly isn't that.

Erick, please post the output of 'grub-probe -vv --target=device /' and the contents of /proc/self/mountinfo after you encountered this bug.

No surprise here, I'd installed grub to the root partition (sda16) while installing this test OS:

lance@lance-desktop:~$ grub-probe -vv --target=device /
grub-probe: info: cannot open `/boot/grub/device.map'.
/dev/sda16

Is this the rest of what you wanted:

lance@lance-desktop:~$ cat /proc/self/mountinfo
15 20 0:14 / /sys rw,nosuid,nodev,noexec,relatime - sysfs sysfs rw
16 20 0:3 / /proc rw,nosuid,nodev,noexec,relatime - proc proc rw
17 20 0:5 / /dev rw,relatime - devtmpfs udev rw,size=1018416k,nr_inodes=214663,mode=755
18 17 0:11 / /dev/pts rw,nosuid,noexec,relatime - devpts devpts rw,gid=5,mode=620,ptmxmode=000
19 20 0:15 / /run rw,nosuid,relatime - tmpfs tmpfs rw,size=410268k,mode=755
20 1 259:0 / / rw,relatime - ext4 /dev/disk/by-uuid/1966e720-b7df-41ee-a3d5-15d68cc7784e rw,errors=remount-ro,user_xattr,acl,barrier=1,data=ordered
21 15 0:16 / /sys/fs/fuse/connections rw,relatime - fusectl none rw
22 15 0:6 / /sys/kernel/debug rw,relatime - debugfs none rw
23 15 0:10 / /sys/kernel/security rw,relatime - securityfs none rw
24 19 0:17 / /run/lock rw,nosuid,nodev,noexec,relatime - tmpfs none rw,size=5120k
25 19 0:18 / /run/shm rw,nosuid,nodev,relatime - tmpfs none rw
26 16 0:19 / /proc/sys/fs/binfmt_misc rw,nosuid,nodev,noexec,relatime - binfmt_misc binfmt_misc rw
28 20 0:20 / /home/lance/.gvfs rw,nosuid,nodev,relatime - fuse.gvfs-fuse-daemon gvfs-fuse-daemon rw,user_id=1000,group_id=1000

Or do I need to poke about in the history logs?

I'm super tired ATM so excuse me if I'm being dense. One thing I think I should mention is that of these four errors the first to appear during the upgrade process was bug 986798 if that's helpful.

I fear that the information that would be useful is no longer available
since you rebooted. The only way to figure this out would be to attempt
the upgrade and not reboot before debugging it.

Could you give as exact as possible directions on how somebody might go
about reproducing your upgrade scenario, including a complete summary of
your partitioning setup and exactly how you performed the upgrade?

Oh, and the order doesn't matter - they're all basically the same error.

I could just repeat it myself. But this was the partitioning layout:

lance@lance-desktop:~$ sudo parted -l
[sudo] password for lance:
Model: ATA WDC WD5000AAKS-0 (scsi)
Disk /dev/sda: 500GB
Sector size (logical/physical): 512B/512B
Partition Table: msdos

Number Start End Size Type File system Flags
 1 32.3kB 43.3GB 43.3GB primary ext4 boot
 2 43.3GB 86.2GB 42.8GB primary ext4
 3 86.2GB 129GB 43.0GB primary ext4
 4 129GB 500GB 371GB extended
18 129GB 151GB 21.8GB logical ext4
14 151GB 172GB 21.1GB logical ext4
15 172GB 193GB 21.1GB logical ext4
16 193GB 215GB 21.4GB logical ext4
17 215GB 236GB 21.4GB logical ext4
13 236GB 258GB 22.0GB logical ext4
12 258GB 280GB 21.6GB logical ext4
11 280GB 301GB 21.7GB logical ext4
10 301GB 323GB 21.8GB logical ext4
 5 323GB 374GB 50.8GB logical ext3
 6 378GB 432GB 53.6GB logical ext3
 7 432GB 487GB 54.9GB logical ext3
 8 487GB 498GB 10.7GB logical ext3
 9 498GB 500GB 2517MB logical linux-swap(v1)

Model: ATA WDC WD800JB-00JJ (scsi)
Disk /dev/sdb: 80.0GB
Sector size (logical/physical): 512B/512B
Partition Table: msdos

Number Start End Size Type File system Flags

Yes, the 80GB drive was blank.

The Oneiric media:

Ubuntu 11.10 "Oneiric Ocelot" - Release i386 (20111012)

And the apt history log shows it was installed 2012-04-19, will attach.

Do you want me to repeat? If so should I use the same partition?

I currently have sda14, 17 ,and 18 unused.

Of course installing Oneiric + the upgrade will take approximately 2 1/2 to 3 hours.

Just happened to think that fstab might be helpful (nothing remarkable):

# /etc/fstab: static file system information.
#
# Use 'blkid' to print the universally unique identifier for a
# device; this may be used with UUID= as a more robust way to name devices
# that works even if disks are added and removed. See fstab(5).
#
# <file system> <mount point> <type> <options> <dump> <pass>
proc /proc proc nodev,noexec,nosuid 0 0
# / was on /dev/sda16 during installation
UUID=1966e720-b7df-41ee-a3d5-15d68cc7784e / ext4 errors=remount-ro 0 1
# swap was on /dev/sda9 during installation
UUID=80627269-1ccd-4774-b4ea-a5ef8824ffaa none swap sw 0 0

Just reviewing this and I guess I was somewhat indirect in answering Colin's question in comment #9.

I installed and updated Oneiric on 4/19 in preparation for iso/upgrade testing. Other than being on a multi-boot drive the install was typical - only / + swap - nothing fancy. On 4/22 I booted into that Oneiric again, updated it, and then ran 'update-manager -d -c'. Downloading and installing the updates seemed to go great until I encountered these four errors.

I took screenshots and made notes while the upgrade continued wondering how I'd go about reporting these but when prompted to reboot I did so and apport kicked in doing most of the work for me, but I reported at least two duplicates before I figured out what was going on ;^)

To be honest I don't see a problem. I installed synaptic after all reporting was done just to use it for examining the situation and those packages are installed and updated to the proper version. Could this have just been an apport fluke?

The only other "oddity" I can think to mention is that I use a KVM switch to share two PC's with the same monitor, keyboard and mouse. But that's never been an issue in the past. Still worth mentioning, but I really hadn't done any switching near the time these errors occurred.

I did save this install just in case we might be able to pull any info, but anticipating that we may need to retest I have a fresh, updated Oneiric on sda14 now.

I'd also previously done an Ubuntu Lucid -> Precise upgrade and an Lubuntu Oneiric -> Precise upgrade w/o any errors. But I went ahead and prepared for retesting just in case we have a rebuild.

I changed the status to Invalid because we shouldn't really be faced with Oneiric -> Precise upgrades anymore at this point.