Persistent reconciliation diff when adopting DBCluster and DBInstance with RDS Controller 1.6.2 #2653
Description
Describe the bug
When adopting an RDS Aurora cluster with instances, the reconciliation loop consistently detects differences in the desired state, causing the instances to enter the "Modifying" state every 2–3 minutes.
Steps to reproduce
Adopt an RDS Aurora cluster and instance by using the following manifest:
apiVersion: rds.services.k8s.aws/v1alpha1
kind: DBCluster
metadata:
name: test-rds-cluster
namespace: default
annotations:
services.k8s.aws/deletion-policy: retain
services.k8s.aws/adoption-policy: "adopt-or-create"
spec:
engine: aurora-mysql
engineVersion: 8.0.mysql_aurora.3.08.0
...
---
apiVersion: rds.services.k8s.aws/v1alpha1
kind: DBInstance
metadata:
name: test-rds-instance-0
namespace: default
annotations:
services.k8s.aws/deletion-policy: retain
services.k8s.aws/adoption-policy: "adopt-or-create"
spec:
dbInstanceClass: db.t4g.medium
dbInstanceIdentifier: test-rds-instance-0
...The controller logs show attempts to reconcile due to detected differences, even though no actual changes have been made:
dbcluster.log
{
"level": "info",
"ts": "2025-10-10T16:47:17.676Z",
"logger": "ackrt",
"msg": "desired resource state has changed",
"kind": "DBCluster",
"namespace": "default",
"name": "test-rds-cluster",
"account": "XXXXXXXXXXXX",
"role": "",
"region": "us-east-1",
"is_adopted": true,
"generation": 9,
"diff":
[
{
"Path":
{
"Parts":
[
"Spec",
"NetworkType"
]
},
"A": null,
"B": "IPV4"
}
]
}dbinstance.log
{
"level": "info",
"ts": "2025-10-10T16:45:10.820Z",
"logger": "ackrt",
"msg": "desired resource state has changed",
"kind": "DBInstance",
"namespace": "default",
"name": "test-rds-instance-0",
"account": "XXXXXXXXXXXX",
"role": "",
"region": "us-east-1",
"is_adopted": true,
"generation": 21,
"diff":
[
{
"Path":
{
"Parts":
[
"Spec",
"AllocatedStorage"
]
},
"A": null,
"B": 1
},
{
"Path":
{
"Parts":
[
"Spec",
"AutoMinorVersionUpgrade"
]
},
"A": null,
"B": false
},
{
"Path":
{
"Parts":
[
"Spec",
"BackupRetentionPeriod"
]
},
"A": null,
"B": 7
},
{
"Path":
{
"Parts":
[
"Spec",
"CopyTagsToSnapshot"
]
},
"A": null,
"B": false
},
{
"Path":
{
"Parts":
[
"Spec",
"DBName"
]
},
"A": null,
"B": "testdb"
},
{
"Path":
{
"Parts":
[
"Spec",
"DatabaseInsightsMode"
]
},
"A": null,
"B": "standard"
},
{
"Path":
{
"Parts":
[
"Spec",
"DeletionProtection"
]
},
"A": null,
"B": false
},
{
"Path":
{
"Parts":
[
"Spec",
"KMSKeyID"
]
},
"A": null,
"B": "arn:aws:kms:us-east-1:XXXXXXXXXXXX:key/XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX"
},
{
"Path":
{
"Parts":
[
"Spec",
"LicenseModel"
]
},
"A": null,
"B": "general-public-license"
},
{
"Path":
{
"Parts":
[
"Spec",
"MasterUsername"
]
},
"A": null,
"B": "admin"
},
{
"Path":
{
"Parts":
[
"Spec",
"MonitoringInterval"
]
},
"A": null,
"B": 0
},
{
"Path":
{
"Parts":
[
"Spec",
"MultiAZ"
]
},
"A": null,
"B": false
},
{
"Path":
{
"Parts":
[
"Spec",
"PreferredBackupWindow"
]
},
"A": null,
"B": "08:00-09:00"
},
{
"Path":
{
"Parts":
[
"Spec",
"PreferredMaintenanceWindow"
]
},
"A": null,
"B": "mon:02:00-mon:03:00"
},
{
"Path":
{
"Parts":
[
"Spec",
"PromotionTier"
]
},
"A": null,
"B": 0
},
{
"Path":
{
"Parts":
[
"Spec",
"PubliclyAccessible"
]
},
"A": null,
"B": false
},
{
"Path":
{
"Parts":
[
"Spec",
"StorageEncrypted"
]
},
"A": null,
"B": true
},
{
"Path":
{
"Parts":
[
"Spec",
"StorageThroughput"
]
},
"A": null,
"B": 0
},
{
"Path":
{
"Parts":
[
"Spec",
"StorageType"
]
},
"A": null,
"B": "aurora"
}
]
}Expected outcome
Having the resources adopted without any differences detected, and the instances remained stable without constant modifications during each reconciliation loop.
Environment
- Kubernetes version: 1.32.6
- Using EKS (yes/no), if so version? yes
- AWS service targeted (S3, RDS, etc.): RDS and ACK RDS Controller v.1.6.2