This topic describes why an instance created from an image generated by a server migration using Server Migration Center (SMC) might fail to start, and provides solutions.
Problem description
After migrating a server using Server Migration Center (SMC), an instance created from the generated image fails to start.
Causes
The instance is I/O optimized, but the source server is missing the required virtio driver.
The boot configuration on the source server is incorrect.
The source server runs an early Linux distribution, such as CentOS 5 or Debian 7, and its built-in GRUB version is earlier than 1.99.
The SHENLONG architecture is incompatible with some older operating systems.
Solution
Work through each cause below until the instance starts successfully.
-
Check the driver.
When you create an I/O optimized instance, ensure that the virtio driver is installed on the source server before migration. If the driver is missing, install it, re-run the migration, and start the instance again. For more information, see Install a virtio driver.
Verify the boot configuration of the source server and correct any errors before retrying the migration.
-
If your source server runs an early Linux distribution (such as CentOS 5 or Debian 7) with a GRUB version earlier than 1.99, you may see the abnormal boot output below when remotely connecting to the instance. If so, upgrade GRUB to version 1.99 or later and retry the migration. For instructions, see How to install GRUB on a Linux server?.
SeaBIOS (version rel-1.7.5-0-ge51488c-20140602_164612-nilsson.home.kraxel.org) Machine UUID 86 50a iPXE (http://ipxe.org) 00:03.0 C980 PCI2.10 PnP PMM+7[...])+7[...] 0 C980 Booting from DVD/CD... Boot failed: Could not read from CDROM (code 0003) Booting from Hard Disk... Boot failed: not a bootable disk No bootable device. The SHENLONG architecture is incompatible with some older operating systems. For details on incompatible operating systems and their solutions, see How to resolve post-migration instance downtime?.