Mastering Oracle ADG: Fast Disaster Recovery Strategies for Critical Systems
This article explains what Oracle Active Data Guard (ADG) is, outlines its basic deployment architecture, and provides detailed step‑by‑step emergency response procedures for four common network and server configurations to ensure rapid failover and business continuity.
1. What is Oracle ADG
ADG is Oracle's disaster‑recovery architecture that ensures services continue from a standby ADG site when the primary site—often using RAC—encounters an issue.
The standby site can be in the same rack, data center, city, or a remote location, and ADG typically relies on manual failover rather than automatic switching, making the speed of manual operations critical.
2. Oracle ADG Basic Deployment Architecture
An example three‑tier structure includes load balancer, application servers, and Oracle RAC (ADG).
When the primary RAC fails, a failover command switches to the ADG database, but the service address changes from the primary IP1 to the ADG IP2, causing the front‑end application servers to lose connectivity.
3. Architecture Planning Changes and Scenarios
Scenario 1: RAC and ADG IPs in the same subnet
Failover to ADG
Change ADG IP2 to the primary IP1
Validate that business services have recovered
If the application server supports reconnection, no further action is required; otherwise, a restart of the application server is needed.
Scenario 2: RAC and ADG in different subnets, sharing the same application servers
Failover to ADG
Modify the application server data source to point to ADG IP2
Restart the application server cluster to connect to ADG
Validate that business services have recovered
When the number of application servers is small, they can be restarted quickly, but with many servers during peak traffic, restarting each one can overload the already‑running servers and prevent recovery.
Scenario 3: RAC and ADG in different subnets, each with its own set of application servers
Failover to ADG
Switch business access to address 2
Validate that business services have recovered
If many business users need to be notified of the address change, the response efficiency may suffer.
Scenario 4: RAC and ADG in different subnets, each with its own application servers, using a load balancer for a unified address
Failover to ADG
Redirect the load balancer to the ADG application‑server cluster
Validate that business services have recovered
Summary
Mind‑map illustration of the emergency response flow.
Signed-in readers can open the original source through BestHub's protected redirect.
This article has been distilled and summarized from source material, then republished for learning and reference. If you believe it infringes your rights, please contactand we will review it promptly.
Efficient Ops
This public account is maintained by Xiaotianguo and friends, regularly publishing widely-read original technical articles. We focus on operations transformation and accompany you throughout your operations career, growing together happily.
How this landed with the community
Was this worth your time?
0 Comments
Thoughtful readers leave field notes, pushback, and hard-won operational detail here.