Preprocessing for Instance Redeployment

Scenarios

When you authorize the redeployment for an ECS with local disks attached in local disk replacement and instance redeployment events, you need to preprocess local disks before the instance redeployment to prevent risks caused by ECS exceptions.

This section describes how to preprocess local disks for disk-intensive (D series) and ultra-high I/O (I series) ECSs running Linux.

Instance redeployment will cause all data loss on the ECS local disks. If you do not need to retain data on the local disks, you can perform the operations below.

If you need to retain data on the local disks, do not authorize the redeployment. Instead, contact technical support.

Procedure

Perform the following operations to add nofail to all data disks in the /etc/fstab configuration file:

1. Remotely log in to the instance to be preprocessed.
2. Run the following command to open and edit the /etc/fstab configuration file:

   vim /etc/fstab

3. Press i to enter editing mode and add parameter nofail to all data disks in the /etc/fstab configuration file.
   Example:

   UUID-8232fee7-f20a-416c-a2e0-cbc8c85a01a2   /mnt/nvme0n1      ext4   defaults,nofail	0 2

   • UUID-8232fee7-f20a-416c-a2e0-cbc8c85a01a2: UUID of a data disk
   • /mnt/nvme0n1: data disk mount point, which can be queried using mount | grep drive-letter.
   • ext4: file system type of the data disk, which can be queried using blkid drive-letter.

   • nofail: allows the system to ignore mounting of a missing drive during boot. When local disks contained in the file system are missing, the system continues booting.

4. Press :wq to exit editing mode.
5. Run the following command for the configuration to take effect:

   systemctl daemon-reload