Known issues

This page lists known issues for Config Connector.

Many of the issues listed here have been fixed. The Fixed version column indicates the version in which the fix was introduced. To receive this fix, upgrade to the listed version or later.

If you're part of the Google Developer Program, save this page to receive notifications when a release note related to this page is published. To learn more, see Saved Pages.

To filter the known issues by a product version or category, select your filters from the following drop-down menus.

Select your Config Connector version:

Select your problem category:

Or, filter the known issues:

Category Earliest identified version Fixed version Issue and workaround
Resources 1.134 1.134.2

Affected SQLInstance resources might become stuck in an Updating state with an error message regarding the state-into-spec annotation.

invalid value "merge" for "cnrm.cloud.google.com/state-into-spec" annotation

This issue occurs because a stricter validation introduced in version 1.134.0 incorrectly rejected the "merge" strategy for SQLInstance resources, even when applied as a default. This is fixed in later versions.
Direct Reconciler 1.131 1.134

Config Connector version 1.132.0 introduced issues affecting resources that use the direct reconciler.

Duplicate controller name error

In version 1.132.0, resources such as SpannerInstance and SecretManagerSecret might fail to reconcile with an error indicating that a controller with the same name already exists:

error registering controller: error adding direct controller for SpannerInstance to a manager: error creating new controller: controller with name spannerinstance-controller already exists.

SecretManagerSecret "last-applied-configuration" error

If you enable the direct reconciler for a SecretManagerSecret (using the alpha.cnrm.cloud.google.com/reconciler: direct annotation) in versions 1.131.0 through 1.133, reconciliation might fail with an API error:

[kubectl.kubernetes.io/last-applied-configuration] must follow pattern [a-z0-9A-Z]+([_\.\-]*[a-z0-9A-Z]+)*), be less than 64 characters, and must have a UTF encoding of less than 128 bytes

This occurs because Config Connector incorrectly attempts to pass internal Kubernetes annotations as labels to the Secret Manager API.
Installation, Networking Kubernetes 1.19 1.43

If you see an error similar to the following example, you might be experiencing an issue with certificates:

Error from server (InternalError): error when creating "/mnt/set-weaver-dns-record.yml": Internal error occurred: failed calling webhook "annotation-defaulter.cnrm.cloud.google.com": Post "https://cnrm-validating-webhook.cnrm-system.svc:443/annotation-defaulter?timeout=30s": x509: certificate relies on legacy Common Name field, use SANs or temporarily enable Common Name matching with GODEBUG=x509ignoreCN=0

This issue occurs with Kubernetes versions 1.19 and later. Config Connector versions 1.43 and later support the automatic generation of the certificate, which should prevent this issue.


Workaround:

To workaround this issue, you can delete the relevant certificates and the Pods:

kubectl delete -n cnrm-system secrets cnrm-webhook-cert-abandon-on-uninstall
kubectl delete -n cnrm-system secrets cnrm-webhook-cert-cnrm-validating-webhook
kubectl delete -n cnrm-system pods -l "cnrm.cloud.google.com/component=cnrm-webhook-manager"

After you have deleted these resources, the correct certificate should regenerate.

Back to top