Fr�d�ric L. W. Meunier
2001-Oct-17 02:21 UTC
"ext2fs_check_if_mount: No such file or directory while determining whether" messages
Hi. I was using 2.4.10 with ext3 0.9.10 and thought it was
time to use -ac for the first time because 2.4.12-ac3
includes 0.9.12.
I don't know what I did to get the following messages, but in
my last boot I removed /etc/mtab (at runtime) and made it a
symlink to /proc/mounts. Not sure if a bad idea, but the only
problem until I rebooted was the need of losetup -d.
When I rebooted, all partitions were unmounted, but I got
device is busy for / and before 'none' showed up as being
mounted read-only. I never received a messages about 'none'.
'none' is devfs and always was only in /proc/mounts, not
/etc/mtab (at least here).
Then I boot with 2.4.12-ac3. My setup is as follows:
devfs is mounted at boot time.
My initialization script uses:
[ -e /dev/.devfsd -a -x /sbin/devfsd ] && /sbin/devfsd /dev
# Check the integrity of the root file system
fsck -C -V -a /
# If there was a failure, drop into single-user mode.
if [ $? -gt 1 ] ; then
echo "fsck failed. Run fsck -v <partition>"
sulogin
reboot -f
fi
# Remount the root file system in read-write mode
mount -v -o remount,rw /
# Check the integrity of all file systems except root
fsck -C -V -A -R -a
# Mount file systems in fstab
mount -v -n -a
My shutdown script uses:
umount -v -n -r -a
mount -v -n -o remount,ro /
I received the following messages:
INIT: version 2.83 booting
Started device management daemon for /dev
fsck 1.25 (20-Sep-2001)
[/sbin/fsck.ext3 -- /] fsck.ext3 -a -C0 /dev/ide/host0/bus0/target0/lun0/part1
ext2fs_check_if_mount: No such file or directory while determining whether
/dev/ide/host0/bus0/target0/lun0/part1 is mounted.
/dev/ide/host0/bus0/target0/lun0/part1: clean, 22842/2560864 files,
178366/2560351 blocks
EXT3 FS 2.4-0.9.12, 10 Oct 2001 on ide0(3,1), internal journal
/dev/ide/host0/bus0/target0/lun0/part1 on / type ext3 (rw)
fsck 1.25 (20-Sep-2001)
Checking all file systems.
[/sbin/fsck.ext3 -- /usr/local/src1] fsck.ext3 -a -C0
/dev/ide/host0/bus0/target0/lun0/part3
ext2fs_check_if_mount: No such file or directory while determining whether
/dev/ide/host0/bus0/target0/lun0/part3 is mounted.
/dev/ide/host0/bus0/target0/lun0/part3 has been mounted 20 times without being
checked, check forced.
/dev/ide/host0/bus0/target0/lun0/part3: 30323/512512 files (3.7%
non-contiguous), 113484/512071 blocks
spurious 8259A interrupt: IRQ7.
[/sbin/fsck.ext3 -- /home/ftp/pub] fsck.ext3 -a -C0
/dev/ide/host0/bus0/target0/lun0/part4
ext2fs_check_if_mount: No such file or directory while determining whether
/dev/ide/host0/bus0/target0/lun0/part4 is mounted.
/dev/ide/host0/bus0/target0/lun0/part4 has gone 191 days without being checked,
check forced.
Deleted inode 354821 has zero dtime. FIXED.
/dev/ide/host0/bus0/target0/lun0/part4: 9408/645120 files (5.2% non-contiguous),
179272/644608 blocks
Oct 16 23:28:18 pervalidus kernel: ext3_orphan_cleanup: deleting unreferenced
inode 518811 is in /var/log/debug
I'd like to know why it did check the other file systems. I
used tune2fs -c 0 -i 0 with them mounted some time ago (but
not with 2.4.10). And what caused the "ext2fs_check_if_mount:
No such file or directory while determining whether" messages
? ext2 is compiled as a module, but isn't in /proc/modules.
All partitions use ext3 in /etc/fstab.
--
0@pervalidus.{net, {dyndns.}org} Tel: 55-21-2717-2399 (Niteri-RJ BR)
Theodore Tso
2001-Oct-17 12:02 UTC
Re: "ext2fs_check_if_mount: No such file or directory while determining whether" messages
On Wed, Oct 17, 2001 at 12:21:29AM -0200, Frederic L. W. Meunier wrote:> > I don't know what I did to get the following messages, but in > my last boot I removed /etc/mtab (at runtime) and made it a > symlink to /proc/mounts. Not sure if a bad idea, but the only > problem until I rebooted was the need of losetup -d.This is fine, but you have to make sure that /proc is mounted after you remount the root read/write. E2fsck needs to be able to check whether or not filesystems are mounted, and so that's why it's complaining about these errors:> ext2fs_check_if_mount: No such file or directory while determining whether /dev/ide/host0/bus0/target0/lun0/part4 is mounted.> /dev/ide/host0/bus0/target0/lun0/part3 has been mounted 20 times without being checked, check forced. > /dev/ide/host0/bus0/target0/lun0/part4 has gone 191 days without being checked, check forced. > > I'd like to know why it did check the other file systems. I > used tune2fs -c 0 -i 0 with them mounted some time ago (but > not with 2.4.10).Can you run dumpe2fs -h on those partitions? It looks like the time between checks and number of mounts between check has gotten reset. How or why, that's not clear, but you can use tune2fs -c 0 -i 0 again to avoid the filesystem checks from happening in the future. (It is a good idea to run fsck from time to time, though, just in case kernel bugs or hardware hiccups have introduced corruption in the filesystem.) - Ted
Fr�d�ric L. W. Meunier
2001-Oct-18 02:41 UTC
Re: "ext2fs_check_if_mount: No such file or directory while determining whether" messages
Theodore Tso wrote:> This is fine, but you have to make sure that /proc is mounted > after you remount the root read/write. E2fsck needs to be > able to check whether or not filesystems are mounted, and so > that's why it's complaining about these errors:>> ext2fs_check_if_mount: No such file or directory while determining whether /dev/ide/host0/bus0/target0/lun0/part4 is mounted.Yes, I noticed some hours ago that /proc/mounts isn't mounted before. Would it work if I modify my /etc/fstab and move proc to the first line ? Actually it's: /dev/ide/host0/bus0/target0/lun0/part1 / ext3 defaults 1 1 /dev/ide/host0/bus0/target0/lun0/part3 /usr/local/src1 ext3 defaults 1 2 /dev/ide/host0/bus0/target0/lun0/part4 /home/ftp/pub ext3 defaults 1 2 proc /proc proc defaults 0 0 tmpfs /dev/shm tmpfs defaults,size=256m 0 0 tmpfs /tmp tmpfs defaults,size=256m 0 0 tmpfs /var/lock tmpfs defaults,size=50k 0 0 tmpfs /var/run tmpfs defaults,size=100k 0 0 tmpfs /home/pervalidus/tmp tmpfs defaults,size=256m 0 0> Can you run dumpe2fs -h on those partitions? It looks > like the time between checks and number of mounts between > check has gotten reset. How or why, that's not clear, but > you can use tune2fs -c 0 -i 0 again to avoid the > filesystem checks from happening in the future. (It is a > good idea to run fsck from time to time, though, just in > case kernel bugs or hardware hiccups have introduced > corruption in the filesystem.)# dumpe2fs -h /dev/ide/host0/bus0/target0/lun0/part3 dumpe2fs 1.25 (20-Sep-2001) Filesystem volume name: <none> Last mounted on: <not available> Filesystem UUID: 8c092176-9418-11d4-920a-b0cc6fece225 Filesystem magic number: 0xEF53 Filesystem revision #: 1 (dynamic) Filesystem features: has_journal needs_recovery sparse_super Filesystem state: clean Errors behavior: Continue Filesystem OS type: Linux Inode count: 512512 Block count: 512071 Reserved block count: 0 Free blocks: 381590 Free inodes: 478207 First block: 0 Block size: 4096 Fragment size: 4096 Blocks per group: 32768 Fragments per group: 32768 Inodes per group: 32032 Inode blocks per group: 1001 Last mount time: Tue Oct 16 21:28:11 2001 Last write time: Tue Oct 16 21:28:11 2001 Mount count: 1 Maximum mount count: 20 Last checked: Tue Oct 16 21:27:30 2001 Check interval: 15552000 (6 months) Next check after: Sun Apr 14 20:27:30 2002 Reserved blocks uid: 0 (user root) Reserved blocks gid: 0 (group root) First inode: 11 Inode size: 128 Journal UUID: <none> Journal inode: 8 Journal device: 0x0000 First orphan inode: 0 # dumpe2fs -h /dev/ide/host0/bus0/target0/lun0/part4 dumpe2fs 1.25 (20-Sep-2001) Filesystem volume name: <none> Last mounted on: <not available> Filesystem UUID: a855fb38-9418-11d4-9191-a4ef693c376b Filesystem magic number: 0xEF53 Filesystem revision #: 1 (dynamic) Filesystem features: has_journal needs_recovery sparse_super Filesystem state: clean Errors behavior: Continue Filesystem OS type: Linux Inode count: 645120 Block count: 644608 Reserved block count: 32230 Free blocks: 465479 Free inodes: 635715 First block: 0 Block size: 4096 Fragment size: 4096 Blocks per group: 32768 Fragments per group: 32768 Inodes per group: 32256 Inode blocks per group: 1008 Last mount time: Tue Oct 16 21:28:11 2001 Last write time: Tue Oct 16 21:28:11 2001 Mount count: 1 Maximum mount count: 20 Last checked: Tue Oct 16 21:27:55 2001 Check interval: 15552000 (6 months) Next check after: Sun Apr 14 20:27:55 2002 Reserved blocks uid: 0 (user root) Reserved blocks gid: 0 (group root) First inode: 11 Inode size: 128 Journal UUID: <none> Journal inode: 8 Journal device: 0x0000 First orphan inode: 0 I'll use tune2fs again. BTW, I used it with the partitions mounted (but not with the broken 2.4.10). Now I'm not sure if only for -m 0 :-) I see it's good to run fsck from time to time. For some reason there was "Deleted inode 354821 has zero dtime. FIXED." in part4. The partition was always cleanly unmounted. PS: Don't reply directly to me, as I just moved to a new service (where I forward my messages) now. The old closed without any notice... -- 0@pervalidus.{net, {dyndns.}org} Tel: 55-21-2717-2399 (Niteri-RJ BR)
Frederic L. W. Meunier
2001-Oct-18 04:16 UTC
Re: "ext2fs_check_if_mount: No such file or directory while determining whether" messages
>> This is fine, but you have to make sure that /proc is >> mounted after you remount the root read/write. E2fsck >> needs to be able to check whether or not filesystems are >> mounted, and so that's why it's complaining about these >> errors:>> ext2fs_check_if_mount: No such file or directory while >> determining whether >> /dev/ide/host0/bus0/target0/lun0/part4 is mounted.> Yes, I noticed some hours ago that /proc/mounts isn't > mounted before. Would it work if I modify my /etc/fstab > and move proc to the first line ?Sorry for that. Then I'll mount it after devfsd. -- 0@pervalidus.{net, {dyndns.}org} Tel: 55-21-2717-2399 (Niteri-RJ BR)