---
title: Manually mounting the filesystem of your Dedibox in rescue mode
description: This page explains how to manually mount the filesystem of your Dedbox in rescue mode to access your data.
dates:
  validation: 2025-11-05
  posted: 2025-11-05
tags: dedibox scaleway
---

This guide explains how to manually mount your server’s filesystem when running in [rescue mode](/dedibox/how-to/use-rescue-mode/).
This procedure is commonly used to diagnose and repair issues on the root disk (e.g., filesystem corruption, kernel upgrade failure, lost credentials, etc.).

<Message type="important">
  Filesystem repair tools such as `fsck` or `xfs_repair` can cause data loss if used incorrectly. Always create a backup when possible.
</Message>

1. Boot the server into rescue mode.
   Refer to [How to use rescue mode](/dedibox/how-to/use-rescue-mode/) for detailed instructions.

2. Log in via SSH. Use the temporary credentials shown on the server’s status page in the Scaleway console.
   ```bash
   ssh root@<DEDIBOX_IP>
   ```

3. Identify your disk and partitions:
   ```bash
   lsblk -f   # shows size, type, and filesystem label
   ```
   Example output (your layout will differ):

   ```
   NAME         MAJ:MIN RM   SIZE RO TYPE MOUNTPOINTS
   sda            8:0    0   447G  0 disk
   ├─sda1         8:1    0   512M  0 part /boot
   │   vfat     FAT32
   └─sda2         8:2    0 446.5G  0 part /
       ext4     1.0
   ```

  <Message type="tip"> 
   Use `blkid` or `fdisk -l` if you need UUIDs or more details.
   Disk names may vary with names like `/dev/nvme0n1`, `/dev/vda`, etc. Replace `/dev/sdaX` in later steps with the correct device.
  </Message>

4. Check the filesystem integrity (recommended).

   <Message type="important">
     Always run repair tools on unmounted filesystems.
   </Message>

   ### ext4
   ```bash
   fsck -f /dev/sda2
   ```

   ### XFS
   ```bash
   xfs_repair /dev/sda2
   ```

   ### Btrfs
   ```bash
   btrfs check --repair /dev/sda2
   ```

5. Create a mount point:
   ```bash
   mkdir -p /mnt/root
   ```

6. Mount the root filesystem.
   Replace `/dev/sda2` with your actual root partition:
   ```bash
   mount /dev/sda2 /mnt/root
   ```
   Verify the mount:
   ```bash
   ls /mnt/root
   ``
   If you have a separate `/boot` (often `sda1`):
   ```bash
   mount /dev/sda1 /mnt/root/boot
   ```
   For UEFI systems with an EFI System Partition (ESP):
   ```bash
   mount /dev/sda1 /mnt/root/boot/efi
   ```

7. (Optional) Mount additional partitions.
   Common examples:
   ```bash
   mount /dev/sda3 /mnt/root/home
   mount /dev/sda4 /mnt/root/var
   ```

8. Bind system directories (required for chroot).
   ```bash
   mount -t proc none /mnt/root/proc
   mount -t sysfs none /mnt/root/sys
   mount --bind /dev /mnt/root/dev
   mount --bind /run /mnt/root/run
   ```

9 Enter the chroot environment to perform your desired rescue actions:
   ```bash
   chroot /mnt/root
   ```

10. Unmount and clean up once you are done.
   Exit chroot first (if used):
   ```bash
   exit
   ```

   Then unmount everything recursively:
   ```bash
   umount -R /mnt/root
   ```

   Remove the temporary directory:
   ```bash
   rmdir /mnt/root
   ```

---

## Troubleshooting

### Mount fails with “wrong fs type”
Run the following command to identify the filesystem type of the partition:
```bash
blkid /dev/sdaX
```
Then specify the filesystem type explicitly:
```bash
mount -t ext4 /dev/sdaX /mnt/root
```

### Encrypted partitions (LUKS)
If your partition is encrypted, you have to open it before mounting it:
```bash
cryptsetup luksOpen /dev/sdaX encrypted_root
mount /dev/mapper/encrypted_root /mnt/root
```

### LVM volume not found
If you are using LVM and cannot find your LVM volume, run the following commands to identify the volume:
```bash
vgscan
vgchange -ay
lvdisplay
```
Mount the logical volume:
```bash
mount /dev/<VG_NAME>/<LV_NAME> /mnt/root
```

### RAID volume not detected
Run the following command to scan for RAID devices:
```bash
mdadm --assemble --scan
cat /proc/mdstat
```
Mount the assembled device:
```bash
mount /dev/md0 /mnt/root
```