Elastic Compute Service:CentOS EOL guidance

Background information

EOL options

To receive software updates and security patches for the operating system, we recommend that you migrate your workloads from CentOS to a different operating system. If you cannot complete the migration in a short period of time, you can continue to use CentOS. For more information, see the Continue using CentOS section of this topic.

Before migration, assess the following items:

1. The operating systems to which you can migrate CentOS workloads.

   When you select an alternative operating system, consider factors such as security, stability, compatibility, budget, and long-term strategy for the operating system. For information about Linux alternatives to CentOS, see the CentOS alternatives section of this topic.

2. The methods that you can use to migrate from CentOS to other operating systems.

   Suggestions for the migration:

   • If your business runs on an Alibaba Cloud Container Service for Kubernetes (ACK) or self-managed Kubernetes cluster, we recommend that you perform a rolling update to migrate nodes (ECS instances) in the cluster from CentOS to other operating systems.

   • In other cases, we recommend that you use the new deployment method to migrate from CentOS to other operating systems by replacing CentOS instances with new non-CentOS instances or replacing CentOS operating systems with other operating systems.

   • If you have difficulty in redeploying business-related runtime environments, we recommend that you perform an in-place migration to migrate from CentOS to other operating systems.

   Migration method	Rolling update (only for scenarios in which your business runs on ACK or self-managed Kubernetes clusters)	New deployment (replacing CentOS instances with new non-CentOS instances)	New deployment (replacing CentOS operating systems with other operating systems)	In-place migration
   Method description	Perform a rolling update to migrate nodes in an ACK or self-managed Kubernetes cluster from CentOS to other operating systems. Important ACK provides the node pool management feature that allows you to perform a rolling update to upgrade the operating systems of nodes in an ACK cluster by replacing the system disks of the nodes. After the system disks are replaced, the original system disks are released and data stored on the original system disks is deleted. Before you perform a rolling update by replacing system disks, make sure that no critical data is stored on the system disks and Kubernetes configurations do not depend on the system disks. Before you replace the system disks, we recommend that you create snapshots for the disks to back up disk data. For more information, see Create a snapshot for a disk.	Create a new instance that runs a different operating system to replace an existing CentOS instance. Then, deploy the applications that are hosted on the existing CentOS instance on the new instance.	Replace the CentOS operating system of an existing ECS instance with a different operating system by replacing the system disk of the instance. Then, redeploy applications on the replacement operating system. Important After you replace the operating system of an ECS instance, the original system disk of the instance is released and all data stored on the disk is deleted. Before you replace the operating system of an ECS instance, create snapshots for the system disk of the instance to back up disk data. For more information, see Create a snapshot for a disk.	An in-place migration allows you to go from an existing operating system to a different operating system by using specific tools and ensures that your applications, user data, and system configurations remain intact.
   Supported scenario	This method is suitable for scenarios in which your business runs on ACK or self-managed Kubernetes clusters.	No limit is imposed.	No limit is imposed.	This method is suitable for scenarios in which operating systems are slightly modified, such as modification to specific default configurations, and contain large amounts of data that are difficult to export.
   Supported alternative operating system	If you use an ACK cluster, see OS images for the supported alternative operating systems. If you use a self-managed Kubernetes cluster, select a supported alternative operating system based on your business requirements.	No limit is imposed. Note You can select Alibaba Cloud public images or use images provided by certified software vendors and operating system replacement services provided by service partners in Alibaba Cloud Marketplace. For more information, see CentOS replacement solutions.	No limit is imposed. Note You can select Alibaba Cloud public images or use images provided by certified software vendors and operating system replacement services provided by service partners in Alibaba Cloud Marketplace. For more information, see CentOS replacement solutions.	Operating systems that are binary compatible with CentOS.
   Business interruption	Business is not interrupted.	Whether business is interrupted during the instance replacement depends on whether primary/secondary backup operations are implemented in the system architecture.	The ECS instance restarts and cannot provide external services during the operating system replacement.	The ECS instance restarts and cannot provide external services during the migration.
   Adaptation of business applications	Business code runs in containers and is not affected by the migration. This eliminates the need to adapt business applications to the new operating system and recompile business applications.	To prevent dynamic linking failures caused by version upgrades of dependency libraries in the new operating system, you must recompile business applications that are written in compiled languages, such as C and C++, for the new operating system.	To prevent dynamic linking failures caused by version upgrades of dependency libraries in the new operating system, you must recompile business applications that are written in compiled languages, such as C and C++, for the new operating system.	Before you perform an in-place migration from CentOS to a different operating system, you must deploy and verify business applications on the new operating system to ensure that the applications can run as expected on the new operating system after the migration.
   Migration risk	Extremely low risk. Compatibility issues are unlikely to occur.	Low risk. The migration process is controllable. If compatibility issues occur, you can roll back to CentOS and troubleshoot the issues with ease.	Low risk. The migration process is controllable. If compatibility issues occur, you can roll back to CentOS and troubleshoot the issues with ease.	High risk. If you made deep modifications and customizations to the original CentOS operating system, such as replacing native .so files or related tools with self-compiled .so files or related tools, the probability of migration failure is high. Large amounts of data and configuration of the original CentOS operating system persist. In this case, compatibility issues that occur during the migration are difficult to troubleshoot.
   Migration cost	Snapshots incur additional costs.	New instances incur additional but controllable costs.	Snapshots incur additional costs.	Snapshots are automatically created during the migration and incur additional costs.
   System maintainability	High system maintainability. No legacy issues exist.	Relatively high system maintainability. You can eliminate all legacy issues by redeploying business applications on the new operating system to ensure long-term system health and maintainability. If data disks are attached to the CentOS instance and you migrate the data disks to a new instance after you replace the CentOS instance with the new instance, the data disks may contain specific data or configurations from the CentOS instance, which may affect system maintainability.	Relatively high system maintainability. You can eliminate all legacy issues by redeploying business applications on the new operating system to ensure long-term system health and maintainability. If data disks are attached to the instance whose CentOS operating system is replaced, the data disks may contain specific data or configurations from the CentOS operating system, which may affect system maintainability.	Low system maintainability. The system disk may contain legacy data and configurations that are incompatible with the new operating system.
   Operation	See Update a node pool.	See the Method 1: New deployment section of this topic.	See the Method 1: New deployment section of this topic.	See the Method 2: In-place migration section of this topic.

Migrate operating systems and workloads

You can use the following methods to migrate operating systems and workloads based on the CentOS alternatives that you select and your business requirements. For information about the advantages, disadvantages, and supported scenarios of the migration methods, see the EOL options section of this topic.

• Method 1: New deployment

  1. Create an ECS instance that runs a different operating system to replace an existing CentOS instance, or replace the operating system of an existing CentOS instance with a different operating system. For more information, see Create an instance on the Custom Launch tab and Replace the operating system (system disk) of an instance.

  2. After you create a replacement instance or replace the operating system of an existing CentOS instance, you must redeploy business-related runtime environments on the replacement instance or operating system. For example, you must re-install software and repositories, such as Advanced Package Tool (APT) and Yellowdog Updater Modified (YUM) repositories, configure environment variables, and redeploy applications. You can perform the following steps. If you replace the operating system of an existing CentOS instance with a different operating system, the replacement instance in the following steps refers to the instance after the CentOS operating system is replaced.

     1. Collect the application and software information of the CentOS instance and export critical application configurations and data files.

     2. Prepare the replacement instance. Install the required applications and software on the replacement instance based on the collected information. Then, import the exported application configurations and data files to the replacement instance.

     3. If a business application is written in a compiled language such as C or C++, recompile the application for the new operating system.

     4. Test and verify that the application services run as expected on the replacement instance.

     5. Stop the application services on the CentOS instance, synchronize application data to the replacement instance, verify that the application services run as expected on the replacement instance, and then start the applications services on the replacement instance.

• Method 2: In-place migration

  The following table describes the migration paths that you can take when you perform an in-place migration.

  Note

  Before you migrate from CentOS 7 to Alibaba Cloud Linux 3, make sure that you are familiar with the differences between CentOS 7 and Alibaba Cloud Linux 3 before you perform the migration. For information about the differences, see Differences between Alibaba Cloud Linux 3 and CentOS 7.

  Source operating system	Destination operating system	References
  CentOS 7/8	Alibaba Cloud Linux	Migrate and upgrade the operating system of an ECS instance
  	Anolis OS	Anolis Migration SIG Migrate and upgrade the operating system of an ECS instance
  	Red Hat Enterprise Linux	Convert2RHEL: Move to Red Hat Enterprise Linux from other Linux distros Migrate and upgrade the operating system of an ECS instance
  CentOS 7	AlmaLinux 8/9	Migrate CentOS 7 to AlmaLinux 8 ELevating CentOS 7 to AlmaLinux 9
  CentOS 8	CentOS Stream 8	Start using CentOS Stream
  CentOS	Rocky Linux 8/9	How to migrate to Rocky Linux from CentOS

Continue using CentOS

References

• For information about the lifecycle plans of other operating systems, which are released on the official websites, see Maintenance support for operating systems.

• For information about image EOL, see FAQ about image EOL.

• Operating system features, performance optimizations, basic software package updates, and security updates may vary based on the operating system version. For more information, see Release notes for public images and Release notes for Alibaba Cloud Linux.