Package: squashfs-tools-ng
Version: 0.8-1
Severity: minor
I was doing "mmdebstrap artful delete-me.squashfs", which internally runs tar
and tar2sqfs.
Something broke after tar2sqfs started but before tar did anything useful.
This left a stub delete-me.squashfs lying around, which confused me.
A minimal recipe to demonstrate this problem is as follows:
bash5$ tar2sqfs test.sq </dev/null
reading tar header: short read
bash5$ file test.sq
test.sq: Squashfs filesystem, little endian, version 1024.0, compressed,
6917529027641081856 bytes, 0 inodes, blocksize: 512 bytes, created: Thu Jan 1
00:00:00 1970
bash5$ unsquashfs -ll test.sq
read_filesystem_tables: Bad id count in super block
File system corruption detected
FATAL ERROR:failed to read file system tables
bash5$ rdsquashfs --list=/ test.sq
test.sq: reading super block: data corrupted.
Is it reasonable to delete the output file when something goes wrong?
If so, please do.
(I'm thinking of "git clone" which works like this.)
If it's unreasonable (e.g. because output isn't guaranteed to be a regular
file?),
can rdsquashfs give a clearer error?
-- System Information:
Debian Release: 10.3
APT prefers stable
APT policy: (990, 'stable'), (500, 'stable-updates'), (500,
'proposed-updates'), (500, 'unstable'), (1, 'experimental')
Architecture: amd64 (x86_64)
Kernel: Linux 5.4.0-0.bpo.2-amd64 (SMP w/2 CPU cores)
Kernel taint flags: TAINT_OOT_MODULE, TAINT_UNSIGNED_MODULE
Locale: LANG=en_AU.UTF-8, LC_CTYPE=en_AU.UTF-8 (charmap=UTF-8),
LANGUAGE=en_AU.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)
LSM: AppArmor: enabled
Versions of packages squashfs-tools-ng depends on:
ii libc6 2.29-9
ii libselinux1 2.8-1+b1
ii libsquashfs0 0.8-1
squashfs-tools-ng recommends no packages.
squashfs-tools-ng suggests no packages.
-- no debconf information
