TKG Management Cluster Creation Fails with Error: "unable to install kapp-controller to bootstrap cluster"
Article ID: 403339
Updated On:
Products
Issue/Introduction
When creating a Tanzu Kubernetes Grid (TKG) management cluster using the UI, the process fails during the bootstrap phase with the following error:
Error:
unable to install kapp-controller to bootstrap cluster: error while waiting for deployment kapp-controller: pods are not yet running for deployment 'kapp-controller' in namespace 'tkg-system'
This error causes the deployment to stall and eventually triggers automatic cleanup of the bootstrap Kind cluster. The issue is consistent across retries and may persist even after reinstalling the kapp CLI tool.
Cause
The issue is caused by insufficient CPU and memory resources allocated to the bootstrap environment (typically a Kind-based jumpbox). When the system does not have enough resources, core components like kapp-controller fail to schedule or start within the Kind cluster, leading to deployment timeouts.
In the reported case, the bootstrap environment was running with only 1 vCPU and 2 GB RAM, which is not sufficient to support a full bootstrap operation.
Resolution
To resolve the issue, increase the compute resources allocated to the bootstrap Kind environment. The following minimums are recommended:
- CPU: 4 vCPU
- Memory: 16 GB RAM
Once the system resources are increased, reattempt the management cluster deployment. The bootstrap cluster should now initialize successfully, and the kapp-controller pods will be able to start as expected.
No changes to the deployment configuration or TKG CLI are necessary beyond adjusting system resources.
Additional Information
Feedback
