I have a Hetzner EX4 server that the sites stopped working, and I asked Hetzner what is wrong as I could not login via SSH or view the websites. They said there is something wrong with the bootloader and that I should put the server in rescue mode and mount the drive again.

This I did, but server restart again didn't load my correct files.

When I put it in rescue mode I get this:

Hardware data:

CPU1: Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz (Cores 8)
Memory: 15948 MB
Disk /dev/sda: 3000 GB (=> 2794 GiB)
Total capacity 2794 GiB with 1 Disk

root@rescue ~ # df -h
Filesystem Size Used Avail Use% Mounted on
rootfs 7.8G 500K 7.8G 1% /
udev 7.8G 0 7.8G 0% /dev
213.133.99.101:/nfs 247G 138G 97G 59% /root/.oldroot/nfs
aufs 7.8G 500K 7.8G 1% /
tmpfs 1.6G 256K 1.6G 1% /run
tmpfs 5.0M 0 5.0M 0% /run/lock
tmpfs 3.2G 0 3.2G 0% /run/shm

I followed this guide on Hetzner Rescue wiki, but whenever I restart it, the server goes back to what is was before rescue mode.


Can anyone drop me some advice please?

Please can someone give me some suggestions what to look for? My server has been down for a week.

This tells us nothing. That wiki is a bunch of suggestions - what did you attempt ?. Merely booting the rescue system then back to your hosted system will do nothing.
Best guess would be to run fsck against your filesystem(s) as the wiki suggests - the /boot if it is a separate filesystem, and the root (/) just to be sure. Then chroot into your system and if it really is a bootloader issue run "grub-install /dev/sda" (presuming grub and /dev/sda) and "update-grub". It's generally pretty safe to run those - especially on a broken system.

Thank you for replying.

I'll give you step by step what I have done:

As per Hetzner Rescue mode wiki:

fsck -C0 /dev/md2 resulted in:
I also did fsck -C0 /dev/md1
I then did:

cd /mnt
mkdir rescue
mount /dev/md2 /mnt/rescue

Then I did the following:
mount /dev/md1 /mnt/rescue/boot

Then:
mount --bind /dev /mnt/rescue/dev/
mount --bind /proc /mnt/rescue/proc/
mount --bind /sys /mnt/rescue/sys/

Then:
chroot /mnt/rescue /bin/bash

then I ran cat /proc/mdstat

cat /proc/mdstat
Personalities : [raid1]
md2 : active raid1 sda3[1]
1073741688 blocks super 1.0 [2/1] [_U]

md1 : active raid1 sda2[1]
524276 blocks super 1.0 [2/1] [_U]

unused devices: <none>

Then I've tried to run grub-install /dev/sda and it gives me:

root@rescue [/]# grub-install /dev/sda
Could not find device for /boot

running df -h:
root@rescue [/]# df -h
Filesystem Size Used Avail Use% Mounted on
/dev/md2 1016G 12G 954G 2% /


I'm lost now. Any further suggestions?

That looks suspiciously like a broken RAID1 - presumably /dev/sdb should be paired with it. As I am not a user of mdadm I can't offer any direct experience - especially why the md1 isn't being recognised. Talk to your provider to see what has happened to sdb.
Might be time to rebuild it completely (after a backup of course) - seems they have that documented.

Thanks for the imput. Unfortunately Hetzner just respond each time, "it's a software" problem, we don't do software"
I'm going to try and do the backup again and just redo the whole installation it looks like