# blk_update_request .. failed to mount /sysroot

Hi, group I wish someone from this nice community here could help me. Happened that following a routine updating (I run a fedora25 on a 32Gb flash driver) my little tool hasn't booted any more but in emergency mode. I tried all kernels from the pool of them in the grub menu but unsuccessful. From reading the rdsosreport journalclt o log file I got that it's related with a bad block so the root lv part cannot get mounted. I could reinstall everything and start again on another flash-driver device but I got in my home dir there some important files I wanted to rescue.

Would anyone know how-to I could do for, e.g. skip out, flag or something those bad-blocks, mount the home dir and mv mi files, please?? I could manage to cp the /run/initramfs/rdsosreport.txt to a USB flashdriver but it's a so big file, I wasn't sure about attach it here.. (I have not got those 60 points minimum required to upload files anyway)

edit retag close merge delete

1

use a live dvd or usb to boot, and see if you can see the home on the original flash drive, and copy it to another usb flash drive.

( 2017-05-27 17:05:42 -0500 )edit

you should try to mount the partition that contains the /home directory that you want to keep. then copy that directory to the new flash drive.

( 2017-05-27 23:34:00 -0500 )edit

Sort by » oldest newest most voted

I don't know whether I'm putting it properly the case is that because a bad sector found, as it reads on the journalctl or reporting log file automatically created, the system (a fedora25 distro on a USB 32Gb flash drive) stop booting and get into emergency mode, a dracut framework, a tiny kind of command line which just to answer command not found to everything.

I just to put home in a dedicated partition but when I set it up, fedora25 installer gave not option other than make it "auto" and put the home dir in the same partition than root so I got 2 ones: the swap (ext4) /dev/sdb1 partition and root, a LVM partition or /dev/sdb2 which content also home.

I also tried to mount /dev/sdb2 the part in question from a CD live but I always got the same output: "not possible to mount due to bad blocks found...".

I wanted to know whether or not possible to skip those bad sectors even temporally so I may rescue as many files as possible. As when applied # badblocks -v -s /dev/sdb2 >badblocks.log # fsck -t -y -f -c /dev/sdb2 (a recomm found somewhere) I saw that it's not such a big i/o error. The badblock.log file reads there are 3 bb ones found.

3040268
8288736
8288740


Below I enclose the rdsosreport.txt log file the booting process generate automatically if someone could give me a hand. I'd appreciate it so much.

+ cat /lib/dracut/dracut-044-77.fc25
dracut-044-77.fc25
+ cat /proc/cmdline
BOOT_IMAGE=/vmlinuz-0-rescue-6e937743751b4535abbbf5cd1e93fa8b root=/dev/mapper/fedora-root ro rd.lvm.lv=fedora/root rd.lvm.lv=fedora/swap rhgb quiet
+ '[' -f /etc/cmdline ']'
+ for _i in '/etc/cmdline.d/*.conf'
+ '[' -f '/etc/cmdline.d/*.conf' ']'
+ break
+ cat /proc/self/mountinfo
0 0 0:1 / / rw shared:1 - rootfs rootfs rw
17 0 0:17 / /sys rw,nosuid,nodev,noexec,relatime shared:2 - sysfs sysfs rw
18 0 0:4 / /proc rw,nosuid,nodev,noexec,relatime shared:7 - proc proc rw
19 0 0:6 / /dev rw,nosuid shared:8 - devtmpfs devtmpfs rw,size=1519524k,nr_inodes=202064,mode=755
20 17 0:18 / /sys/kernel/security rw,nosuid,nodev,noexec,relatime shared:3 - securityfs securityfs rw
21 19 0:19 / /dev/shm rw,nosuid,nodev shared:9 - tmpfs tmpfs rw
22 19 0:20 / /dev/pts rw,nosuid,noexec,relatime shared:10 - devpts devpts rw,gid=5,mode=620,ptmxmode=000
23 0 0:21 / /run rw,nosuid,nodev shared:11 - tmpfs tmpfs rw,mode=755
24 17 0:22 / /sys/fs/cgroup ro,nosuid,nodev,noexec shared:4 - tmpfs tmpfs ro,mode=755
25 24 0:23 / /sys/fs/cgroup/systemd rw,nosuid,nodev,noexec,relatime shared:5 - cgroup cgroup rw,xattr,release_agent=/usr/lib/systemd/systemd-cgroups-agent,name=systemd
26 17 0:24 / /sys/fs/pstore rw,nosuid,nodev,noexec,relatime shared:6 - pstore pstore rw
27 24 0:25 / /sys/fs/cgroup/cpu,cpuacct rw,nosuid ...
more

when you try to mount /dev/sbd2 add error=continue to see if it forces the mount.

( 2017-05-28 10:38:36 -0500 )edit

According this reading: http://www.system-rescue-cd.org/lvm-g... fedora developers know that mounting from external location when into eventualities like this one following not regular procedures and anyone gives instructions for handling errors. It's probably that even the standard mounting process at booting ignores those LVM particularities. Thats why it applies on flying those fsck handling checkings it reports on the journalctl file and just confusing users when looking up for solutions..

( 2017-05-29 12:56:20 -0500 )edit

some more (probably) info related to my issue.. If someone here could give a hand with how-to & details I' d appreciate it.. http://www.system-rescue-cd.org/lvm-g...

( 2017-05-29 15:20:12 -0500 )edit

excuse if dummy but I need to precise: do you mean "copy" as # cp source target is it that? (well, I did: first created a dir in NuevVol called reskatt and afterward cp as:

$mkdir /run/media/liveuser/NuevoVol/reskatt $ sudo cp /dev/sdb2 /run/media/liveuser/NuevoVol/reskatt/

What I got now is the a copy called sdb2 but how I can access to it? Is it a img? should I mount it? sorry, I'm some confused. Thank you.

more

As declared somewhere in the beginning of this thread I tried first, when at installing fedora-25 (this is a case of installing on an external HD) to place my mounting points at will & above all on dedicated partitions so that making easy to access to my home & files if something got wrong. This time, however, I only had at hand a 32Gb USB flash-driver and due this fact probably, the room available variable, the installer gave me not option other than an auto LVM partitioning, so when at repairing disk commands I applied first as tracked above did not work. - So the first step was to get myself familiar with those LVM commands & tools as documented e.g. here https://linux.die.net/man/8/lvm Even the journalctl.txt file got wrong about recommending commands, it seems to ignore it's about a LVM system. So your first commands kit are # pvdisplay # vgdisplay # lvdisplay so getting your partitions as they were created at installation. Mine, the partition I wanted to rescue, as output from them was "/dev/fedora/root" so, as it will be referenced onward.

• Second, try with a proper rescue CD live e.g. http://www.system-rescue-cd.org/

• As explained here http://www.webupd8.org/2009/12/fixing... so, I tried the following command & got as output a list of blocks # mke2fs -n /dev/fedora/root

• then, also as explained above I applied the following command & starting with the first in the output list as: # e2fsck -y -b 32768 /dev/fedora/root

• And it worked at once with the first in list, so applied the following one to the partition target # e2fsck -y -c /dev/fedora/root

• Following that I tried to (first creating the mounting to dir) & mount it (& this time after hundred tries I got it done w/o issues) # mkdir /media/recovery # mount /dev/fedora/root /media/recovery

• As everything seemed going OK I tried to mount the host disk (where I was going to cp into my rescuing stuff) using the command mount.ntfs-3g (with writing support) as explained here http://www.system-rescue-cd.org/manual/Mounting_ntfs_filesystems/

• First creating the dir where I was going to mount my HD host.

# mkdir /mnt/windows/n-vol

Note here cdrescue has a dedicated mounting area for windows file-systems so I just did this part on there. (where "n-vol" is my win partition host)

# mount.ntfs-3g /dev/sda2 /mnt/windows/n-vol

• Then create to the recovery dir (called here rescue-host) in HD host..

# mkdir mnt/windows/n-vol/rescue-host

Located at home with cd
# cd /media/recovery/home

• then just cp my home stuff to HD host dir (& voilá, DONE) # cp -r user /mnt/windows/n-vol/rescue-host

The complete rescue session if useful, here:

root@sysresccd /root % dumpe2fs /dev/mapper/fedora-root
dumpe2fs 1.43.3 (04-Sep-2016)
dumpe2fs: Bad magic number in super-block while trying to open /dev/mapper/fedora-root
Couldn't find valid file-system superblock.
root@sysresccd /root % blkid
/dev/loop0 ...
more