Fsck output does not match tutorials...

[Bo Berglund]

On Tue, 22 Mar 2022 17:45:36 -0500, Keith <keith at caramail.com> wrote:

>On 3/22/22 3:21 PM, Bo Berglund wrote:
>> I have an erroneous disk (it ran out of disk space during a sudo apt
>> full-upgrade operation and Linux crashed).
>> Linux won't boot now.
>>
>> I can put it into a USB carrier and examine it on another system:
>>
>> sudo lsblk -o UUID,NAME,FSTYPE,SIZE,MOUNTPOINT,LABEL,MODEL
>> UUID                                 NAME        FSTYPE  SIZE MOUNTPOINT LABEL
>> MODEL
>>                                       sda                 7.4G
>> UHSII_uSD_Reader
>> 1591-0E10                            +-sda1      vfat    256M
>> c14f2d5f-b499-41df-b7ce-15bdf933cdff +-sda2      ext4    7.2G <== repair?
>>
>> It turned out that free space on sda2 is zero, that is why Linux crashed, I
>> guess.
>> But now I have freed up 40% by deleting a few very big files in the home dir.
>> But there is more to do because Linux will not boot anyway.
>>
>> So I figured I could do as several on-line tutorials suggest:
>>
>> sudo fsck -N /dev/sda2
>> fsck from util-linux 2.33.1
>> [/sbin/fsck.ext4 (1) -- /dev/sda2] fsck.ext4 /dev/sda2
>>
>> This response is not shown in any of the tutorials I found....
>>
>> What am I doing wrong?
>> Can the disk be repaired?
>>
>
>-N is not going to anything and the output is exactly what you would
>expect. If you want a non-destructive check then try:
>
>sudo fsck.ext4 -n -v -f /dev/sda2
>
>This performs a forced, verbose, read-only filesystem check.
>this is what it looks like for my USB drive /dev/sdf1
>
>sudo fsck.ext4 -n -v -f /dev/sdf1
>e2fsck 1.45.5 (07-Jan-2020)
>Warning!  /dev/sdf1 is mounted.
>Warning: skipping journal recovery because doing a read-only filesystem
>check.
>Pass 1: Checking inodes, blocks, and sizes
>Pass 2: Checking directory structure
>Pass 3: Checking directory connectivity
>Pass 4: Checking reference counts
>Pass 5: Checking group summary information
>
>          23532 inodes used (0.19%, out of 12189696)
>            779 non-contiguous files (3.3%)
>             19 non-contiguous directories (0.1%)
>                # of inodes with ind/dind/tind blocks: 0/0/0
>                Extent depth histogram: 23271/184
>       13375930 blocks used (27.44%, out of 48740328)
>              0 bad blocks
>              1 large file
>          20841 regular files
>          2613 directories
>              0 character device files
>              0 block device files
>              0 fifos
>              0 links
>             69 symbolic links (69 fast symbolic links)
>              0 sockets
>------------
>          23523 files
>
>If it finds problems, unmount the drive and rerun without the "-n"
>option and with the "-p" option to autofix any filesystem errors that
>doesn't require human intervention.
>
>sudo fsck.ext4 -v -f -p /dev/sda2

Thanks Keith!
So the tutorial I used did not show the correct options...

I checked the drive again now, first with and then without -n:

$ sudo fsck.ext4 -n -v -f /dev/sdb2
e2fsck 1.46.2 (28-Feb-2021)
Pass 1: Checking inodes, blocks, and sizes
Pass 2: Checking directory structure
Pass 3: Checking directory connectivity
Pass 4: Checking reference counts
Pass 5: Checking group summary information

      163625 inodes used (34.78%, out of 470496)
         191 non-contiguous files (0.1%)
         205 non-contiguous directories (0.1%)
             # of inodes with ind/dind/tind blocks: 0/0/0
             Extent depth histogram: 144828/73
     1159490 blocks used (61.71%, out of 1879040)
           0 bad blocks
           1 large file

      131668 regular files
       13015 directories
           0 character device files
           0 block device files
           0 fifos
        1136 links
       18856 symbolic links (18639 fast symbolic links)
          77 sockets
------------
      164752 files


$ sudo fsck.ext4 -v -f /dev/sdb2
e2fsck 1.46.2 (28-Feb-2021)
Pass 1: Checking inodes, blocks, and sizes
Pass 2: Checking directory structure
Pass 3: Checking directory connectivity
Pass 4: Checking reference counts
Pass 5: Checking group summary information

      163625 inodes used (34.78%, out of 470496)
         191 non-contiguous files (0.1%)
         205 non-contiguous directories (0.1%)
             # of inodes with ind/dind/tind blocks: 0/0/0
             Extent depth histogram: 144828/73
     1159490 blocks used (61.71%, out of 1879040)
           0 bad blocks
           1 large file

      131668 regular files
       13015 directories
           0 character device files
           0 block device files
           0 fifos
        1136 links
       18856 symbolic links (18639 fast symbolic links)
          77 sockets
------------
      164752 files

Did this actually do anything?
It looks like it did not.
So the problem is not errors on the file system but probably the truncation of
some or several files due to disk space running out during an apt full-upgrade
operation.

THis cannot be solved by fsck.

I have bitten the bullet and started over rebuilding the system from scratch on
a new bigger disk. I can check my settings from the damaged disk since it is
readable when mounted on a running system.

I will now update my Linux notes and add these working fsck commands to them.


-- 
Bo Berglund
Developer in Sweden