openSUSE Forums > Archives > SF Archives > ARCHIVES - 64-bit » "fsck.reiserfs --rebuild-tree" Crashes

Go Back   openSUSE Forums > Archives > SF Archives > ARCHIVES - 64-bit
Forums FAQ Members List Search Today's Posts Mark Forums Read

ARCHIVES - 64-bit Questions specific to 64-bit systems running SUSE Linux
(Questions that apply to both 32-bit and 64-bit systems should be posted in the appropriate mixed architecture forums)

 
 
LinkBack Thread Tools Display Modes
  #1 (permalink)  
Old 21-Dec-2007, 09:58
Mazilo
Guest
 
Posts: n/a
Default

When I tried to download a 10MB file to my account, the download stopped with an error message saying no more space left. So, I deleted the partially downloaded file. Then, I executed telinit 1 to bring down the system to init level 1. I dismounted the home partition and executed fsck.reiserfs --rebuild-tree which ended up with the following error messages:
Code:
Do you want to run this program?[N/Yes] (note need to type Yes if you do):Yes
Replaying journal: Done.
Reiserfs journal '/dev/sdb8' in blocks [18..8211]: 0 transactions replayed
###########
reiserfsck --rebuild-tree started at Fri Dec 21 10:25:42 2007
###########

Pass 0:
####### Pass 0 #######
Loading on-disk bitmap .. ok, 10747456 blocks marked used
Skipping 8538 blocks (super block, journal, bitmaps) 10738918 blocks will be read
0%....20%....40%....60%....80%....100%************************left 0, 7948 /sec
389244 directory entries were hashed with "r5" hash.
********"r5" hash is selected
Flushing..finished
********Read blocks (but not data blocks) 10738918
****************Leaves among those 47522
****************Objectids found 389304

Pass 1 (will try to insert 47522 leaves):
####### Pass 1 #######
Looking for allocable blocks .. finished
0%....20%....40%....60%....80%....100%************************ left 0, 489 /sec
Flushing..finished
********47522 leaves read
****************47306 inserted
****************216 not inserted
####### Pass 2 #######

Pass 2:
0%....20%Not enough allocable blocks, checking bitmap...there are 0 allocable blocks, btw

out of disk space
At this moment, my only option seems to be completely re-formating the home partition to lose all its data in order to gain its space back. I wondered if anyone out here has some solution for me to fix this partition before I proceed to format this home partition. I have already made an image file from this partition right after the above error messages and this image file as well as the home partition no longer want to mount. Before proceeding to re-format the home partition, I thought of resizing this partition and try the above attempt again to see if I will be able to regain its contents. Anyone?
  #2 (permalink)  
Old 21-Dec-2007, 19:21
ken_yap
Guest
 
Posts: n/a
Default

Well, no that's not a crash. That's the program not continuing because it doesn't have the resources it needs.

What possessed you to run --rebuild-tree when the problem was not a filesystem error but simply a lack of free space?

At the moment the only way out that I can see is to take an image of the partition (which you apparently already did), reformat the partition, and then recover your files from the image using loopback mount, something like this:

mount -o loop,ro -t reiser /data/part.img /mnt/oldhome

Hopefully you can get your files off it.
  #3 (permalink)  
Old 21-Dec-2007, 20:50
Mazilo
Guest
 
Posts: n/a
Default

Quote:
What possessed you to run --rebuild-tree when the problem was not a filesystem error but simply a lack of free space?[/b]
I had encountered this before; however, it was because the partition was dirty due to system crashes as well as hibernations (where the system hangs when trying to wake up). df had shown the partition had about 1.6GB space; however, trying to write a 1MB gave the insufficient disk space messages. After fsck --rebuild-tree, I gained back the 1.6GB space and was able to use it. So, when this happened recently, I just did the fsck --rebuild-tree again (not realizing the HD was really full).

Quote:
At the moment the only way out that I can see is to take an image of the partition (which you apparently already did), reformat the partition, and then recover your files from the image using loopback mount, something like this:

mount -o loop,ro -t reiser /data/part.img /mnt/oldhome[/b]
I tried this option before; however, the mount process complains with the following error messages:
Code:
[root@Mi:/opt/tmp 21%] # mount -o loop=/dev/loop0,ro -t reiserfs /opt/tmp/sdb8.img /mnt
mount: cannot mount /dev/loop0 read-only
The other option I have thought of is to put in a new HD. Create a partition of the same size. Use dd to image the partition on to this newly create partition. Then, resize the partition before running the fsck again to see if that will do the trick.
  #4 (permalink)  
Old 24-Dec-2007, 09:30
Mazilo
Guest
 
Posts: n/a
Default

Here is my update on how I resolved this problem:
  1. Using dd to make an image partition to an external USB drive.
  2. Run fsck.reiserfs --rebuild-sb on the USB partition. In this process, I had to select y to create journal to recover this partition. Once recovered, I made sure to manually check the integrities of the files.
  3. Use mkereiserfs to format the unmounted damaged partition. Once formatted, re-mounted the partition and copied all the recovered files from the USB partition to this newly formatted partition to restore the data.
  4. Checked the integrities of copied files before deleting the files on the USB partition.
Now, my system is back to normal with no data lost.
 

Bookmarks

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On




 

Search Engine Friendly URLs by vBSEO 3.3.0 RC2