Skip to content

AWS AURORA Secondary clusters (global db) are stuck at "Creating" in ap-southeast-1

0

Hello AWS Support,

We are experiencing an issue where an Aurora PostgreSQL secondary cluster is stuck in the "Creating" state for several hours. This cluster is being added to an existing Global Database.

Resource Details:

Region: ap-southeast-1 (Singapore)

Engine Version: Aurora PostgreSQL 17.4

Instance Class: db.r8g.large

Troubleshooting performed:

Instance Availability: We have verified via describe-orderable-db-instance-options that db.r8g.large is available in all three Availability Zones (1a, 1b, 1c) for this engine version.

KMS Permissions: We identified a missing kms:CreateGrant permission in the secondary region's KMS key policy. This has since been updated to allow rds.amazonaws.com full access (CreateGrant, Decrypt, GenerateDataKey, etc.).

1 Answer
0

Based on the information provided, there are a few potential issues that could be causing your secondary cluster to remain stuck in the "Creating" state:

KMS Key Access Issue: While you've updated the KMS permissions, Aurora Global Database has a specific limitation regarding encryption credentials. If Aurora loses access to the AWS KMS key for the DB cluster, the encrypted DB cluster can go into an inaccessible-encryption-credentials state. Since you mentioned the KMS permissions were initially missing, this could have caused the creation process to fail. Even though you've since corrected the permissions, the cluster creation may have already encountered a terminal state.

Engine Version Compatibility: Ensure that your primary DB cluster is also running Aurora PostgreSQL 17.4. It's recommended that secondary clusters use the same DB engine version as the primary cluster to avoid compatibility issues during the replication setup.

Primary Cluster Source: If your primary DB cluster was created as a replica of an Amazon RDS PostgreSQL instance (not an Aurora cluster), you cannot create secondary clusters from it. Attempts to do so will time out and the secondary cluster won't be created successfully.

Recommended Actions:

  1. Check the primary cluster's engine version matches the secondary (17.4)
  2. Verify the primary cluster wasn't originally created from an RDS PostgreSQL replica
  3. Consider deleting the stuck secondary cluster and recreating it now that KMS permissions are properly configured
  4. Monitor the cluster status and check CloudWatch logs for any specific error messages

If the issue persists after verifying these points, you may need to engage AWS Support directly for deeper investigation into the cluster creation logs.
Sources
Using Amazon Aurora Global Database - Amazon Aurora
Adding an AWS Region to an Amazon Aurora global database - Amazon Aurora

answered 8 months ago

EXPERT

reviewed 8 months ago

You are not logged in. Log in to post an answer.

A good answer clearly answers the question and provides constructive feedback and encourages professional growth in the question asker.