Manage disaster recovery type for local peers

This page describes how to manage disaster recovery type for your local peer databases in Oracle Database@Google Cloud.

When you create an Autonomous AI Database, a local peer is automatically created. By default, its disaster recovery type is backup-based. You can change the disaster recovery type to Autonomous Data Guard.

You can choose the disaster recovery type that meets your Recovery Time Objective (RTO) and Recovery Point Objective (RPO) requirements. The RTO is the maximum amount of time required to restore connectivity to a database after a manual or automatic failover is initiated. The RPO is the maximum duration of potential data loss on the primary database.

  • Backup-based disaster recovery: doesn't incur additional costs beyond automatic backups, but has a higher RTO than Autonomous Data Guard. We recommend that you use backup-based disaster recovery only for non-production workloads.

    Backup-based disaster recovery objectives are as follows:

    • RTO: 1 hour + 1 hour per 5 TB
    • RPO: 10 seconds

  • Autonomous Data Guard: creates a standby database that is continuously updated with the changes from the primary database. We recommend that you use Autonomous Data Guard for production workloads.

    Autonomous Data Guard disaster recovery objectives are as follows:

    • RTO: 2 minutes
    • RPO: 10 seconds

Application Continuity when using Autonomous Data Guard

Autonomous Data Guard combined with Application Continuity can mask outages from end users and applications. Oracle applications should use Application Continuity whenever possible to minimize the impact of failover and switchover events.

You can enable Application Continuity for the predefined services. You can configure client connectivity with the CONNECT_TIMEOUT, RETRY_DELAY, RETRY_COUNT, and TRANSPORT_CONNECT_TIMEOUT parameters in the connection string.

An example connection string is as follows:

localdr1 =
  (description =
    (retry_count=50)
    (connect_timeout=90)
    (retry_delay=3)
    (transport_connect_timeout=3)
    (address =
      (protocol=tcps)
      (port=1521)
      (host=localdr1.adb.uk-london-1.oraclecloud.com)
    )
    (connect_data=
      (service_name=g24da7e94756f60_localdr1_tp.adb.oraclecloud.com)
    )
    (security=
      (ssl_server_dn_match=no)
    )
  )

Change disaster recovery type to Autonomous Data Guard

To change the disaster recovery type of a local peer database from backup-based to Autonomous Data Guard, enable Autonomous Data Guard and set an automatic failover data loss limit.

Autonomous Data Guard monitors the status of the primary database and the local standby assumes the role of primary database in cases where the primary database becomes unavailable, according to the RTO and RPO.

Autonomous Data Guard automatically performs a failover operation according to the data loss limit. The default data loss limit is 0, which means that Autonomous Data Guard only performs an automatic failover where there is no data loss. You can also specify an automatic failover data loss limit between 0 and 3600 seconds (1 hour).

Automatic failover doesn't occur if the primary database becomes unavailable and Autonomous Data Guard determines that the possible data loss is greater than the data loss limit. In such a situation, you can still perform a manual failover with data loss.

To enable Autonomous Data Guard and set an automatic failover data loss limit, do the following:

Console

  1. Go to the Autonomous AI Database Service page in the Google Cloud console.

    Go to Autonomous AI Database Service

  2. Click the name of your database.

  3. On the Autonomous AI Database details page, select the Disaster recovery tab.

  4. For the peer database on which you want to enable Autonomous Data Guard, click View actions, and then click Update disaster recovery.

  5. On the Update disaster recovery page, select the disaster recovery type as Autonomous Data Guard.

  6. Specify Automatic failover with data loss limit in seconds. It should be between 0-3600 seconds. The default value is 0 seconds.

  7. Click Save.

On the Disaster recovery tab, in the Peer database section, the Status field shows the progress of the update operation. After the update is complete, the peer database becomes the standby, and its DR type field shows "Autonomous Data Guard".

gcloud

To enable Autonomous Data Guard and set automatic failover data loss limit, run the gcloud oracle-database autonomous-databases update command:

 gcloud oracle-database autonomous-databases update DATABASE_ID \
 --project=PROJECT_ID \
 --location=REGION \
 --properties-local-data-guard-enabled \
 --properties-local-adg-auto-failover-max-data-loss-limit-duration=MAX_DATA_LOSS_LIMIT

Replace the following:

  • DATABASE_ID: the ID of your primary database
  • PROJECT_ID: the ID of your Google Cloud project
  • REGION: the region where your primary database is located
  • MAX_DATA_LOSS_LIMIT: automatic failover data loss limit between 0 and 3600 seconds (1 hour)

API

To enable Autonomous Data Guard and set automatic failover data loss limit, run the following curl command:

curl -X PATCH \
-H "Authorization: Bearer $(gcloud auth print-access-token)" \
-H "Content-Type: application/json" \
"https://oracledatabase.googleapis.com/v1/projects/PROJECT_ID/locations/REGION/autonomousDatabases/DATABASE_ID"
-d \
'{
  "properties": {
    "localDataGuardEnabled": true,
    "localAdgAutoFailoverMaxDataLossLimitDuration": MAX_DATA_LOSS_LIMIT,
    "encryptionKey": {
         "provider": "ORACLE_MANAGED"
      }
   }
}'

Replace the following:

  • PROJECT_ID: the ID of your Google Cloud project
  • REGION: the region where your primary database is located
  • DATABASE_ID: the ID of your primary database
  • MAX_DATA_LOSS_LIMIT: automatic failover data loss limit between 0 and 3600 seconds (1 hour)

Change disaster recovery type to backup-based

To change the disaster recovery type of a local peer database from Autonomous Data Guard to backup-based, disable the Autonomous Data Guard.

Console

  1. Go to the Autonomous AI Database Service page in the Google Cloud console.

    Go to Autonomous AI Database Service

  2. Click the name of your database.

  3. On the Autonomous AI Database details page, select the Disaster recovery tab.

  4. For the peer database on which you want to disable Autonomous Data Guard, click View actions, and then click Remove Autonomous Data Guard.

  5. To confirm, enter the database ID and click Confirm.

On the Disaster recovery tab, in the Peer database section, the Status field shows the progress of the update operation. After the update is complete, the peer database becomes the backup copy, and its DR type field shows "Backup-based".

gcloud

To disable Autonomous Data Guard, run the gcloud oracle-database autonomous-databases update command:

 gcloud oracle-database autonomous-databases update DATABASE_ID \
 --project=PROJECT_ID \
 --location=REGION \
 --localDataGuardEnabled=false

Replace the following:

  • DATABASE_ID: the ID of your primary database
  • PROJECT_ID: the ID of your Google Cloud project
  • REGION: the region where your primary database is located

API

To disable Autonomous Data Guard, run the following curl command:

curl -X PATCH \
-H "Authorization: Bearer $(gcloud auth print-access-token)" \
-H "Content-Type: application/json" \
"https://oracledatabase.googleapis.com/v1/projects/PROJECT_ID/locations/REGION/autonomousDatabases/DATABASE_ID"
-d \
'{
  "properties": {
    "localDataGuardEnabled": false
   }
}'

Replace the following:

  • PROJECT_ID: the ID of your Google Cloud project
  • REGION: the region where your primary database is located
  • DATABASE_ID: the ID of your primary database

What's next