OK, it appears that a simple way to trigger this bug is to run
sudo losetup -f /var/lib/snapd/snaps/bare_5.snap
Every time this command is run, the error is reported. If you add the
"-P" option, which should be used to force a partition scan, then you
get an additional warning line:
Sep 20 10:07:48 boxxo2-VirtualBox kernel: loop_reread_partitions:
partition scan of loop19 (/var/lib/snapd/snaps/bare_5.snap) failed
(rc=-5)
I'm going to mark this as invalid for snapd, because the squashfs images we use
are perfectly valid, and the bug seems to be in the kernel, and at a first
sight it appears that there are two things going wrong:
1) a partition scan occurs even though it was not explicitly requested
2) the RDB driver should be less verbose, and just assume that if the device is
smaller than 8k, then it's not a RDB partition -- no need to report anything in
the kernel logs.
** Changed in: snapd (Ubuntu)
Status: Confirmed => Invalid
** Also affects: linux (Ubuntu)
Importance: Undecided
Status: New
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubuntu.
https://bugs.launchpad.net/bugs/1979633
Title:
The "bare" snap is the cause of error messages "unable to read
partition table" and "partition table beyond EOD, truncated"
Status in linux package in Ubuntu:
New
Status in snapd package in Ubuntu:
Invalid
Bug description:
Hi,
here is the problem:
# journalctl | grep loop4
Jun 23 09:49:45 vougeot kernel: loop4: detected capacity change from 0 to 8
Jun 23 09:49:45 vougeot kernel: Dev loop4: unable to read RDB block 8
Jun 23 09:49:45 vougeot kernel: loop4: unable to read partition table
Jun 23 09:49:45 vougeot kernel: loop4: partition table beyond EOD, truncated
The loop4 device is used to mount the "bare" snap:
$ df -a | grep /snap/bare/
[...]
/dev/loop4 128 128 0 100%
/snap/bare/5
Note that on another system, the loop device will probably have
another number. The error messages can be found with the following
commands:
# journalctl | grep "unable to read RDB block"
Jun 23 09:49:45 vougeot kernel: Dev loop4: unable to read RDB block 8
# journalctl | grep "unable to read partition table"
Jun 23 09:49:45 vougeot kernel: loop4: unable to read partition table
# journalctl | grep "partition table beyond EOD"
Jun 23 09:49:45 vougeot kernel: loop4: partition table beyond EOD, truncated
ProblemType: Bug
DistroRelease: Ubuntu 22.04
Package: snapd 2.55.5+22.04
Uname: Linux 5.18.6-051806-generic x86_64
ApportVersion: 2.20.11-0ubuntu82.1
Architecture: amd64
CasperMD5CheckResult: unknown
CurrentDesktop: KDE
Date: Thu Jun 23 11:39:16 2022
Snap.Changes: no changes found
SourcePackage: snapd
UpgradeStatus: No upgrade log present (probably fresh install)
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1979633/+subscriptions
--
Mailing list: https://launchpad.net/~kernel-packages
Post to : kernel-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kernel-packages
More help : https://help.launchpad.net/ListHelp
