Docs

Code Pipeline

Multi-cloud disaster recovery

Fault recovery for dedicated lines in multi-cloud disaster recovery

Code Pipeline has introduced new disaster recovery pipeline templates to help you quickly implement service recovery in disaster recovery scenarios and reduce the risk of business interruption. This article describes how to create and set up the dual-cloud active-active dedicated line failure disaster recovery pipeline.

Introduction to scenarios

This scenario simulates an active-active dual-cloud environment in which, after system's dedicated line is restored, some traffic switches back from the third-party cloud to system.

tip

In dual-cloud active-active scenarios, if the dedicated line of fails, see Dedicated line failover in multi-cloud disaster recovery for the process of switching all traffic to a third-party cloud.

Prerequisites

Workloads for business applications deployed in a dual-cloud, active-active architecture have been created. For actions to create workloads, see Workloads.

Cloud scheduling GTM has been created. GTM controls changes in the direction of data traffic before and after switching. For GTM operating procedures, see GTM Cloud Scheduling Quick Start.

Procedure

Step 1: Create a pipeline

In the left navigation bar, select workspace. On the workspace page, click the target workspace name to enter the current workspace.

Click Create pipeline, and on the Select pipeline templates page, select Disaster recovery > Dual-cloud active-active—Dedicated line failure disaster recovery.

4 Click confirm.

Step 2: Restore global traffic after the dual-cloud split

By enabling the service address of the dedicated line, some of the traffic is redirected to system.

Within the GTM instance, select the GTM instance that controls this traffic switch.

In Address Pool, select the address pool where the dedicated line service is located.

In Address Online Mode, set the addresses for both third-party cloud and dedicated lines to Always Enabled.

Step 3: Verify the dual-cloud process splitting

Set up an approval workflow to verify successful dual-cloud recovery by following these steps. During pipeline execution, after verifying that the business application is running properly, click Pass in the pipeline to proceed to the next step.

tip

This step triggers the approval workflow for validating the dedicated line switch. The practical validation actions must be performed manually by the guiding personnel.

In Approval method, select the approval method.

Select the personnel who need to verify dedicated line switching from Approver.

Step 4: Scale down the application

After the dedicated line service is restored, you need to reduce the number of workload instance replicas on the third-party cloud that were previously expanded back to their initial value. In this scenario, the replica count of the workload for the business application deployed in the third-party cloud is reduced from 4 to 2.

What to do next

Run pipeline

Manage pipelines

Last updated: 2025.09.10 17:37:29