I may have spotted a bug while using the /dev/loop devices.
It seem that when you use filesystem image via. /dev/loop when the
filesystem image is on a NSF mounted volume *something* goes terribly
wrong. Everytime the device is used it give an I/O error during
write. For example (try to make a ext2 file system, then just try to
cat the /dev/loop0, both fail):
- -> # mount -t nfs nfs.system.com:/test /nfsvolume1
- -> # dd if=/dev/zero of=/nfsvolume1/ext2-image.img bs=1024 count=1024
- -> # losetup /dev/loop0 /nfsvolume1/ext2-image.img
- -> # losetup /dev/loop0
/dev/loop0: [0002]:225313 (/nfsvolume1/j3gum/ext2-image.img)
offset 0, no encryption
- -> # mke2fs /dev/loop0
mke2fs 1.06, 7-Oct-96 for EXT2 FS 0.5b, 95/08/09
Linux ext2 filesystem format
Filesystem label=
256 inodes, 1024 blocks
51 blocks (4.98%) reserved for the super user
First data block=1
Block size=1024 (log=0)
Fragment size=1024 (log=0)
1 block group
8192 blocks per group, 8192 fragments per group
256 inodes per group
Writing inode tables: done
Warning: could not erase block 0: Attempt to write block from filesystem resulted in short write
Writing superblocks and filesystem accounting information: done
- -> # cat /dev/loop0
cat: /dev/loop0: I/O error
Sincerely Confused,
- -jeffrey hundstad
-----BEGIN PGP SIGNATURE-----
Version: 2.6.2
iQCVAwUBMv87NqLa8RgOE1M5AQGeRAP/dYUyOW1UG2TVJ2pI/sU4Cizp/qf3kdUa
7s9nZ4hy96LXVTca/ANM5lNPyk+xi/S6mS/fJZvrn3mek3g44R/FqBqasrqyKVKW
rb9tQ0IfwLT653Lh2QwoFoYwhAEqmroggWxV3yBxVwmsShRNFTy7fvxHzqgTFqBa
dlEO7m5GtE0=
=AFDu
-----END PGP SIGNATURE-----