diff --git a/terraform/environments/performance-hub/application_variables.json b/terraform/environments/performance-hub/application_variables.json
index 2f19b37aa52..426032fb547 100644
--- a/terraform/environments/performance-hub/application_variables.json
+++ b/terraform/environments/performance-hub/application_variables.json
@@ -15,7 +15,7 @@
       "container_memory": "3072",
       "task_definition_volume": "upload_volume",
       "network_mode": "none",
-      "db_enabled": false,
+      "db_enabled": true,
       "db_instance_class": "db.m5.large",
       "db_user": "admin",
       "db_allocated_storage": "100",
diff --git a/terraform/environments/performance-hub/database.tf b/terraform/environments/performance-hub/database.tf
index 2efc0e34865..d5a6c44d0fa 100644
--- a/terraform/environments/performance-hub/database.tf
+++ b/terraform/environments/performance-hub/database.tf
@@ -20,7 +20,7 @@ resource "aws_db_instance" "database" {
   storage_encrypted                   = true
   iam_database_authentication_enabled = false
   vpc_security_group_ids              = [aws_security_group.db.id]
-  snapshot_identifier                 = format("arn:aws:rds:eu-west-2:%s:snapshot:%s", data.aws_caller_identity.current.account_id, local.app_data.accounts[local.environment].db_snapshot_identifier)
+  #snapshot_identifier                 = format("arn:aws:rds:eu-west-2:%s:snapshot:%s", data.aws_caller_identity.current.account_id, local.app_data.accounts[local.environment].db_snapshot_identifier)
   backup_retention_period             = 30
   maintenance_window                  = "Mon:00:00-Mon:03:00"
   backup_window                       = "03:00-06:00"
@@ -31,7 +31,10 @@ resource "aws_db_instance" "database" {
   db_subnet_group_name                = aws_db_subnet_group.db.id
   enabled_cloudwatch_logs_exports     = ["error"]
   ca_cert_identifier                  = "rds-ca-rsa2048-g1"
-  apply_immediately = true
+  # BE VERY CAREFUL with apply_immediately = true. Useful if you want to see the results, but can cause a reboot
+  # of RDS meaning the connected app will fail.
+  # When apply_immediately=false, RDS changes are applied during the next maintenance_window
+  apply_immediately                   = false
 
   # timeouts {
   #   create = "40m"