BFD-2935: Update BFD Server to use default RDS reader endpoint (#1952)

ops/terraform/services/server/modules/bfd_server_asg/main.tf

@@ -2,7 +2,7 @@ locals {
   env = terraform.workspace
 
   # When the CustomEndpoint is empty, fall back to the ReaderEndpoint
-  rds_reader_endpoint = data.external.rds.result["CustomEndpoint"] == "" ? data.external.rds.result["ReaderEndpoint"] : data.external.rds.result["CustomEndpoint"]
+  rds_reader_endpoint = data.external.rds.result["ReaderEndpoint"]
 
   additional_tags = { Layer = var.layer, role = var.role }
 

ops/terraform/services/server/modules/bfd_server_asg/scripts/rds-cluster-config.sh

@@ -2,7 +2,7 @@
 #######################################
 # Results in two calls to the AWS RDS APIs for describe-db-clusters and describe-db-instances.
 # Returns a well-formatted json object including the following keys:
-# "DBClusterIdentifier", "Endpoint", "ReaderEndpoint", "WriterAZ", "WriterNode" and "CustomEndpoint".
+# "DBClusterIdentifier", "Endpoint", "ReaderEndpoint", "WriterAZ", and "WriterNode".
 #
 # This exists to accommodate the desire for placing write-intensive workloads in the same AZ as the
 # writer node. As of May 2022, the data source for DB instances does not expose the `IsClusterWriter`
@@ -12,34 +12,26 @@
 #
 # Globals:
 #   CLUSTER_IDENTIFIER mapped to the "$1" positional argument
-#   BFD_ENV mapped to the "$2" positional argument
 #   CLUSTER a modified json object from the aws rds describe-db-clusters command
 #   WRITER_NODE a string representing the writer node's db-instance-identifier
 #   WRITER_AZ a string representing the writer node's availability zone
-#   WRITER_CONFIG a json object including writier-specific keys
+#   WRITER_CONFIG a json object (as a string) including writer-specific keys
 #
 # Arguments:
 #   $1 maps to the CLUSTER_IDENTIFIER
 #######################################
 set -euo pipefail
 
 CLUSTER_IDENTIFIER="$1"
-BFD_ENV="$2"
-
-ENDPOINT_IDENTIFIER="bfd-${BFD_ENV}-ro"
 
 CLUSTER="$(aws rds describe-db-clusters \
-    --query 'DBClusters[].{DBClusterIdentifier:DBClusterIdentifier,Endpoint:Endpoint,ReaderEndpoint:ReaderEndpoint,Members:DBClusterMembers}[0]' \
-    --db-cluster-identifier "$CLUSTER_IDENTIFIER")"
-
-CUSTOM_ENDPOINT="$(aws rds describe-db-cluster-endpoints --db-cluster-identifier "$CLUSTER_IDENTIFIER" | jq -r --arg endpoint "$ENDPOINT_IDENTIFIER" '.[][] | select(.DBClusterEndpointIdentifier == $endpoint).Endpoint')"
+  --query 'DBClusters[].{DBClusterIdentifier:DBClusterIdentifier,Endpoint:Endpoint,ReaderEndpoint:ReaderEndpoint,Members:DBClusterMembers}[0]' \
+  --db-cluster-identifier "$CLUSTER_IDENTIFIER")"
 
 WRITER_NODE="$(jq -r '.Members[] | select(.IsClusterWriter == true) | .DBInstanceIdentifier' <<<"$CLUSTER")"
 
 WRITER_AZ="$(aws rds describe-db-instances --db-instance-identifier "$WRITER_NODE" --query 'DBInstances[0].AvailabilityZone' --output text)"
 
 WRITER_CONFIG="$(jq --null-input --arg writer_node "$WRITER_NODE" --arg writer_az "$WRITER_AZ" '{ WriterNode: $writer_node, WriterAZ: $writer_az }')"
 
-# Construct the final object keeping just "DBClusterIdentifier", "Endpoint", "ReaderEndpoint", "WriterAZ" from the $CLUSTER
-# Add the objects representing "WriterNode" and "CustomEndpoint"
-jq --argjson obj1 "$WRITER_CONFIG" --arg str1 "$CUSTOM_ENDPOINT"   '. += $obj1 | . += { "CustomEndpoint": $str1 } | del(.Members)' <<<$CLUSTER
+jq --argjson obj "$WRITER_CONFIG" '. += $obj | del(.Members)' <<<$CLUSTER