If a dedicated host has potential risks, it is still available. However, the dedicated host and the Elastic Compute Service (ECS) instances on it may fail at any time. To protect your business against risks that result in possible failures of the dedicated host, we recommend you migrate the dedicated host to another dedicated host.

Prerequisites

All the ECS instances on the dedicated host are stopped.
Warning Service interruptions may occur when you stop instances. We recommend that you stop instances during off-peak hours.

Procedure

Important You cannot perform self-serviced migration on ECS instances that run on local SSD hosts. Submit a ticket for technical assistance.
  1. Log on to the ECS console.
  2. In the left-side navigation pane, choose Instances & Images > Dedicated Hosts.
  3. In the top navigation bar, select a region.
  4. On the Dedicated Hosts page, find the dedicated host. In the Status column, move the pointer over Physical Machine Risk. In the Note message, click Change Host.
    Change the dedicated host
  5. Perform operations based on the dedicated host type.
    • If the dedicated host is not a local SSD host, click OK.
    • If the dedicated host is a local SSD host, click Submit Ticket.
      Warning The data on the local disk of the dedicated host is lost after the migration.
    After all the ECS instances on the dedicated host are migrated to another dedicated host, the host ID and the metadata of the ECS instances remain unchanged. This metadata includes the instance ID, private IP address, and public IP address of each ECS instance. However, the machine ID of the dedicated host is changed.