crds-tcx-platform Appilcation Fails to Reconcile or Killed During Deployment

If the crds-tcx-platform application fails to reconcile or killed then you must increate the KAPP controller resource on TKG.

You can perform the following steps to increase KAPP Controller resources if the crds-tcx-platform appilcation fails to reconcile or killed.

With Management Cluster's KUBECONFIG

Set CLUSTER_NAME to the name of the workload cluster export CLUSTER_NAME=<your workload cluster name>.

Run the following command on the management cluster to pause the reconciliation of the workload cluster kapp-controller application.

export KUBECONFIG=<<KUBECONFIG file location of Management Cluster>>

kubectl patch packageinstall ${CLUSTER_NAME}-kapp-controller --type "json" -p '[{"op":"add","path":"/spec/paused","value": true }], {"op":"add","path":"/spec/canceled","value": true }]'

With Workload Cluster's KUBECONFIG

Run the following command to patch the kapp-controller's resource allocation.

export KUBECONFIG=<<KUBECONFIG file location of Workload Cluster>>

kubectl set resources deployment kapp-controller --limits memory=6Gi,cpu=4,ephemeral-storage=2Gi --requests memory=100Mi,cpu=120m,ephemeral-storage=256Mi -n tkg-system