my working machine leaves me with the emergency console and won’t mount any partition. It happened after i had to press the reset button because of a hang up.
journal says ext4 is an unknown filesystem type, it also mentions it was unable to load a kernel module… although don’t know wich one.
fdisk -l says a secondary disk has a GPT PMBR size problem but the backup appears to be ok.
I’m typing this from a borrowed notebook, so i’ll try to post fdisk & journal output as soon as i reach my machine.
thanks for your answer, booting from rescue disk leaves me empty, can’t even see the devices. Anyway, i found an old gparted live CD and used it to boot and get the reports. Smartctl
It seems there isn’t an ext4 module, so it can be the problem… but don’t know. Partitions seems to be ok from Gparted live, although havn’t checked the RAID yet.
If is there any other thing i could provide please feel free to request it.
I’m sorry, your’re right i should have mentioned it. I assummed it was a software problem not a physical one.
Did not mention RAID because my main concern were the other partitions, my bad again, i should have been more specific.
lsmod only shows info from the running OS which I assume is the rescue disk???
I’ve chrooted to /dev/sda6 (which was old /) …shouldn’t be lsmod from the machine in that case?. lsmod from live CD has much more modules running.
Can you mount the partition from the rescue??
Yes, i can mount sda6(btrfs), sda7 & sdb1 (both ext4) partitions with the live CD, didn’t tried on RAID (sdb2, sdc)because i could deal with it later, even with some loss.
Partitions content seems to be OK, i used Gparted to verify those 3 and it seems there’s no problem… but don’t know how to make the kernel recognize them again, so the boot process completes.
Sorry i didn’t mention those details before, i really appreciate your help.
I’ve verified them with Gparted live CD … althought i don’t have the logs. Running manually from live cd console
root@debian:~# btrfs check --repair /dev/sda6
enabling repair mode
Checking filesystem on /dev/sda6
UUID: a0eb3d90-ff91-4a22-b848-5e9e25389163
cache and super generation don't match, space cache will be invalidated
found 35044495360 bytes used err is 0
total csum bytes: 29984660
total tree bytes: 1756069888
total fs tree bytes: 1648558080
total extent tree bytes: 66174976
btree space waste bytes: 291871281
file data blocks allocated: 169988112384
referenced 98030682112
root@debian:~# e2fsck -pfv /dev/sda7
201701 inodes used (2.72%, out of 7421952)
1517 non-contiguous files (0.8%)
101 non-contiguous directories (0.1%)
# of inodes with ind/dind/tind blocks: 0/0/0
Extent depth histogram: 201274/258
16670595 blocks used (56.20%, out of 29665536)
0 bad blocks
7 large files
170373 regular files
31135 directories
0 character device files
0 block device files
0 fifos
0 links
180 symbolic links (157 fast symbolic links)
4 sockets
------------
201692 files
root@debian:~# e2fsck -pfv /dev/sdb1
21522 inodes used (0.07%, out of 30531584)
2486 non-contiguous files (11.6%)
35 non-contiguous directories (0.2%)
# of inodes with ind/dind/tind blocks: 0/0/0
Extent depth histogram: 21411/103
53284393 blocks used (43.64%, out of 122094587)
0 bad blocks
13 large files
20261 regular files
1252 directories
0 character device files
0 block device files
0 fifos
0 links
0 symbolic links (0 fast symbolic links)
0 sockets
------------
21513 files
The other partitions belongs to the RAID setup, i could try to reassembly the array but most probably it will fail.
Check /etc/fstab and be sure that looks ok and not corrupted. If one partition/array does not mount the boot fails.
I removed the RAID partition, rebooted and it failed again.
After that i removed all ext4 partitions from fstab and it booted, still got failing services on logs but at least i got root partition working.
Moreover, i manually mount sda7 & sdb1 (both ext4) and it worked, got these lines on logs
BadYuyu kernel: EXT4-fs (sdb1): mounted filesystem with ordered data mode. Opts: (null)
BadYuyu kernel: EXT4-fs (sda7): mounted filesystem with ordered data mode. Opts: (null)
So, the question is … why doesn’t recognize ext4 filesystem at boot?..it never worked on emergency console either, i was expecting the same.
If fstab is the problem, can it be generated via a command line (ie mdadm -D --scan >> mdadm.conf) ?
Otherwise, what should i look after?
There were no errors when i manually mount the partitions…but on previous boot, automounting failed with message “can’t recognize ext4 filesystem” i was expecting something similar on manual mount, to see it working was a surprise. Fsck says everything is fine on every single partition, no errors detected, all clean.
The good news is …this morning i manually edited fstab to add /home and the other partitions, rebooted and it worked like a charm, i’ve the full system working again (even the **** RAID array, although as md0 instead of md127). Just don’t understand WTH happened, i’ll get an external HDD make a data backup and then blow up the GPT, can’t trust this thing. Maybe later i’ll do something similar on SSD too.
I really wanna thank you for your help, you pointed me in the right direction even when my post were not accurate. Chapeau!!
Didn’t have dmesg log from that boot, i’ve uploaded journal to expire box with full boot sequence since yesterday:
