Configuring warm DR with WarehousePG Disaster Recovery

Use warm DR when you need a short Recovery Time Objective (RTO) and can keep a pre-provisioned DR cluster running continuously. The DR cluster stays close to current by applying WAL at regular intervals. When a failure occurs, promote the DR cluster to take over as the new primary.

Note

You must complete most of the steps on this page before a failure occurs. The DR cluster runs continuously in recovery mode, applying WAL at regular intervals, and doesn't accept client connections until promoted. When a failure occurs, proceed to Failing over.

Prerequisites

Configure the DR cluster for restore before proceeding. See Setting up the DR cluster.

Configuring scheduled restores

Complete these before a failure occurs. Run a one-time full restore to establish the DR cluster's baseline state, then schedule delta restores on the DR cluster to keep it current.

Running the initial restore

Establish a baseline state on the DR cluster before scheduling regular delta restores. Run all commands as the WarehousePG cluster owner, typically gpadmin on the DR coordinator.

Check what restore points are available:
```
whpg-dr list-backup my_cluster
```
Run a full restore to the latest restore point:
```
whpg-dr restore my_cluster --target-name latest
```

Note

A full restore requires empty target data directories. If the directories already contain data from a previous restore, clear them first.

Scheduling delta restores on the DR cluster

Configure a cron job on the DR coordinator that applies new restore points at regular intervals, keeping the cluster current. The restore-point interval determines your RPO. For example, a 15-minute interval gives at most 15 minutes of data loss. RTO is determined by the time for the final delta restore plus promotion, which is low when the WAL gap is small.

As the WarehousePG cluster owner, typically gpadmin, add the following entry to the crontab (crontab -e), offset by a few minutes to allow archiving to complete:

5,20,35,50 * * * * source /usr/edb/whpg7/greenplum_path.sh && /usr/edb/whpg-dr/bin/whpg-dr restore my_cluster --target-name latest --delta >> $HOME/gpAdminLogs/whpg-dr-restore.log 2>&1

This example runs a delta restore every 15 minutes (at :05, :20, :35, and :50), applying the latest available restore point and logging output to $HOME/gpAdminLogs/whpg-dr-restore.log. Adjust the schedule to match your RPO target.

Note

Offset the delta restore cron job from the restore point creation schedule on the primary to allow archiving to complete before the DR job runs. If create-restore-point takes longer than the offset on your cluster, increase the offset accordingly.

Verifying warm DR

To check what restore points the primary cluster has created, run on the primary coordinator:

whpg-dr list-backup my_cluster

Output

Full backup: 20260624T195641_base_backup
└── Restore Points:
    ├── 20260624-195718R_whpgdr_full_backup: 2026-06-24 19:57:18
    ├── 20260624-200623R_delta_test: 2026-06-24 20:06:23
    └── 20260625-141506R_scheduled: 2026-06-25 14:15:06

To check which restore point the DR cluster has applied and how far each segment has replayed WAL, run on the DR coordinator:

whpg-dr list-restore my_cluster

Output

Latest Completed Restore
------------------------
Restore point: 20260625-141506R_scheduled
Backup name: 20260624T195641_base_backup
Restore time: 2026-06-25 14:20:14

Recovery Cluster Segment Status
-------------------------------
Content ID   Status                 Replay End LSN  Host          Path
----------   ------                 --------------  ----          ----
coordinator  shut down in recovery  0/98091348      cdw           /data/coordinator/gpseg-1
segment 0    shut down in recovery  0/80000120      sdw1          /data1/primary/gpseg0
segment 1    shut down in recovery  0/80000120      sdw1          /data1/primary/gpseg1
segment 2    shut down in recovery  0/78000120      sdw2          /data1/primary/gpseg2
segment 3    shut down in recovery  0/78000120      sdw2          /data1/primary/gpseg3

When the DR cluster is current, the restore-point name in list-restore matches the latest entry in list-backup. The Replay End LSN values advance with each delta restore cycle.

Failing over

When the primary cluster fails or becomes unavailable, or when testing your DR setup, promote the DR cluster. The DR cluster is at most one restore-point interval behind, so the RTO is low.

Stop the delta restore cron job on the DR coordinator (crontab -r, or comment out the whpg-dr line) to prevent a partial restore from interfering with promotion. If the primary is still accessible, stop its restore-point cron job too.
Optionally, apply one final delta restore to catch up to the latest available restore point:
```
whpg-dr restore my_cluster --target-name latest --delta
```
Promote the DR cluster:
```
whpg-dr promote my_cluster
```
Start the cluster. Source the WarehousePG environment and set COORDINATOR_DATA_DIRECTORY before running gpstart:
```
source /usr/edb/whpg7/greenplum_path.sh
export COORDINATOR_DATA_DIRECTORY=<coordinator_data_directory>
gpstart -a
```
Rebuild system indexes and collect statistics on each database. Without this step, queries generate warnings about missing table statistics. Repeat for each database in the cluster, replacing <database_name> accordingly:
```
reindexdb --system -d <database_name> -e
analyzedb -as pg_catalog -d <database_name> -p 10 -v
analyzedb -d <database_name> -p 10 -v
```

After these steps, the cluster accepts connections. Extension data is restored as part of the cluster state.

Adding mirrors and standby coordinator

Optionally, add mirrors and a standby coordinator after promotion. whpg-dr doesn't restore these components, so the promoted cluster has only primary segments. See the WarehousePG documentation for enabling segment mirroring and enabling coordinator mirroring.