Unfortunately no one can be told what fun_plug is - you have to see it for yourself.
You are not logged in.
Hello,
I put in two ST3500630AS (500GB SATA 3.0Gb/s 16MB) disks in a DNS-323 that came with 1.02 firmware and formatted it as RAID1. I then upgraded to 1.04 did e2fsck immediately thereafter, I got errors as per transcript below.
I am not conversant with ext2 file system and would appreciate comments from anyone as to how serious are these errors. How can I format clean? What kind of problems would have arisen had I not done e2fsck?
Kind regards,
Jaya
e2fsck 1.32 (09-Nov-2002) sh: badblocks: not found Pass 1: Checking inodes, blocks, and sizes Inode 15220749, i_blocks is 64, should be 24. Fix<y>? yes Pass 2: Checking directory structure Pass 3: Checking directory connectivity /lost+found not found. Pass 4: Checking reference counts Pass 5: Checking group summary information Block bitmap differences: -(30459916--30459920) Fix<y>? yes Free blocks count wrong for group #929 (32234, counted=32239). Fix<y>? yes /dev/md0: ***** FILE SYSTEM WAS MODIFIED ***** ERROR~ #
Offline
Did you execute this from a telnet shell? Did you unmount the filesystem first? Never run e2fsck on a mounted filesystem!!
Offline
Hi Frodo,
frodo wrote:
Did you execute this from a telnet shell? Did you unmount the filesystem first?
Yes and yes.
I am more interested in what the specific errors mean and possible consequences if I had not fixed them.
Jaya
Offline
jayas wrote:
I am more interested in what the specific errors mean and possible consequences if I had not fixed them.
These might not even be errors - because you've run fsck on a live filesystem, you may just see errors due to the filesystem changing while being fsck'ed. As Gandalf once said - Listen to Frodo...
Offline
HaydnH wrote:
jayas wrote:
I am more interested in what the specific errors mean and possible consequences if I had not fixed them.
These might not even be errors - because you've run fsck on a live filesystem, you may just see errors due to the filesystem changing while being fsck'ed. As Gandalf once said - Listen to Frodo...
Hi HaydnH,
If you read my response to Frodo, I answered his question "Yes and yes". This means I am using telnet and yes it was done with unmounted file system.
My question really is if anyone here knows enough to tell me specifically what the error is in relation to ext file system. Saves me having to negotiate ext2 file system documentation.
Kind regards,
Jaya
Offline
Me too ...
What's your disk configuration? Did you do this on both disks?
I formatted two standard disks after upgrading to 1.04, but got a "94%" error at the end of the format. After a reboot everything seemed fine, but I thought I'd run e2fsck to see. /mnt/HD_b2 unmounted fine and the e2fsck ran and told me it had fixed something (I didn't have the sense to save the output, but it was a similar number of lines as yours).
I couldn't unmount HD_a2 and e2fsck warned me that it would be unwise to run, so I didn't. Did you get the same errors on both disks?
Is it OK to unmount HD_a2? Did you have to do anything special? I noticed that it seemed to have some hidden system folders (with names starting with .)
Offline
jayas wrote:
HaydnH wrote:
jayas wrote:
I am more interested in what the specific errors mean and possible consequences if I had not fixed them.
These might not even be errors - because you've run fsck on a live filesystem, you may just see errors due to the filesystem changing while being fsck'ed. As Gandalf once said - Listen to Frodo...
Hi HaydnH,
If you read my response to Frodo, I answered his question "Yes and yes". This means I am using telnet and yes it was done with unmounted file system.
My question really is if anyone here knows enough to tell me specifically what the error is in relation to ext file system. Saves me having to negotiate ext2 file system documentation.
Kind regards,
Jaya
How did you manage to unmount /dev/md0? Isn't your fun_plug installed on it?
Offline
I would try this http://www.inreto.de/dns323/fsck/
Offline
Hi Frodo,
frodo wrote:
How did you manage to unmount /dev/md0? Isn't your fun_plug installed on it?
My telnet is installed on ramdisk to allow unmounting. I telnet, stop smb and unmount /dev/md0 before doing e2fsck.
Jaya
Edit: added fun_plug installation script
#!/bin/sh # manifests root=/home/root disk=/mnt/HD_a2/fun_path logs=/mnt/HD_a2/fun_logs export PATH=.:$PATH # switch to disk path cd $disk { # boot timestamp busybox date # execution trace set -x # install files busybox cp busybox $root busybox cp telnetd $root busybox cp profile $root/.profile # replace shell busybox mv -f /bin/sh /bin/sh.old busybox ln -s $root/busybox /bin/sh # establish console busybox mknod /dev/ptyp0 c 2 0 busybox chmod 0666 /dev/ptyp0 busybox mknod /dev/ttyp0 c 3 0 busybox chmod 0666 /dev/ttyp0 } >$logs 2>&1 # switch to root path cd $root # restrict telnet users busybox cp /etc/passwd passwd busybox sed -e '2,$s|:/home/.*|:/:/bin/sync|' passwd >/etc/passwd busybox rm -f passwd busybox echo "/etc/passwd updated" >>$logs # clone admin password busybox sed -n -e 's|^admin:|root:|p' /etc/shadow >>/etc/shadow busybox echo "/etc/shadow updated" >>$logs # start telnet daemon telnetd -l /bin/login -d busybox echo "telnet daemon started" >>$logs
Last edited by jayas (2008-02-06 01:54:25)
Offline