How-To Recover a VM booting in Emergency Mode

Last Updated: Nov 05, 2025

Introduction

After restarting a VM, you may sometimes be unable to SSH into the instance. This typically occurs when the VM fails to boot properly and enters emergency mode. Using serial console access, you can investigate and recover the VM.

Prerequisites

• Crusoe cloud CLI access

• Serial console access enabled for the VM

Step-by-Step Instructions

1. Connect to the VM Using Serial Console

   Use the following command to connect to the VM via serial console:

   $ crusoe compute vms serial-console --name $VM_NAME --project-id $PROJECT_ID

   Note: Replace $VM_NAME and $PROJECT_ID with the appropriate VM and project identifiers.

2. Verify the VM’s Boot Mode

   If the VM has booted into emergency mode, the terminal prompt will appear similar to the following:

   # root@hostname:~#

   This indicates the system encountered a critical boot issue preventing normal startup.

3.  Identify and Fix the /etc/fstab Issue

   In most cases, the VM enters emergency mode due to an invalid or missing entry in /etc/fstab.
   A common cause is a reference to a RAID device or mount point that no longer exists.

   • Open /etc/fstab using a text editor such as vi or nano:

     # vi /etc/fstab

   • Locate any invalid entries. For example:

     /dev/md0 /var/ xfs defaults 0 0

   • Comment out the problematic line by adding a # at the beginning:

     # /dev/md0 /var/ xfs defaults 0 0

   • Save and exit the file.

4. Check System Logs (Optional)

   To further diagnose the issue, review the boot logs using:

   # journalctl -xb

   This can help identify other issues preventing the VM from booting normally.

5. Reboot the VM

   Once the /etc/fstab issue has been corrected, reboot the VM:

   # sudo reboot

   After the reboot, try connecting via SSH to verify normal operation.

Additional Resources

• Managing serial console access in Crusoe VMs