This topic helps you troubleshoot issues with Tanzu Developer Tools for IntelliJ.
If the namespace field of the debug launch configuration is empty, the workload is re-applied even if it exists on the cluster.
Internally, workloads are gathered in the cluster in the current namespace and compared with the information that you specify. If the namespace field is empty, it is considered null and the internal checks fail.
Do not leave the namespace field blank.
If your debug configuration is created from the launch configuration drop-down menu, it re-applies the workload even if the workload already exists on the cluster.
There is internal logic that is not run when debug configuration is created from the drop-down menu. However, the logic is run when debug configuration is selected from the right-click pop-up menu.
Select debug configuration from the right-click pop-up menu.
When connecting to Google’s GKE clusters, an error appears with the text WARNING: the gcp auth plugin is deprecated in v1.22+, unavailable in v1.25+; use gcloud instead.
GKE authentication was extracted into a separate plug-in and is no longer inside the Kubernetes client or libraries.
Download and configure the GKE authentication plug-in. For instructions, see the Google documentation.
When you run or debug a launch configuration, IntelliJ deactivates the launch controls.
IntelliJ deactivates the launch controls to prevent other launch configurations from being launched at the same time. These controls are reactivated when the launch configuration is started. As such, starting multiple Tanzu debug and live update sessions is a synchronous activity.
Unable to open debugger portYou try to start a Tanzu Debug session and it immediately fails with an error message similar to:
Error running 'Tanzu Debug - fortune-teller-fortune-service': Unable to open debugger port (localhost:5005): java.net.ConnectException "Connection refused"
Old Tanzu Debug launch configurations sometimes appear to be corrupted after installing a later version of the plug-in. You can see whether this is the problem you are experiencing by opening the launch configuration:
workload.yaml.com.vmware.tanzu.tanzuBeforeRunPortForward and com.vmware.tanzu.tanzuBeforeRunWorkloadApply.Because these two tasks are unknown causes, these steps of the debug launch are not run. This in turn means that the target application is not deployed and accessible on the expected port, which causes an error when the debugger tries to connect to it.
It might be that although the launch configuration appears corrupt when seen in the launch config editor, in fact there is no corruption. It’s suspected that this problem only occurs when you install a new version of the plug-in and start using it before first restarting IntelliJ.
There is possibly an issue in the IntelliJ platform that prevents completely or correctly initializing the plug-in when the plug-in is hot-swapped into an active session instead of loaded on startup.
Closing and restarting IntelliJ typically fixes this problem. If that doesn’t work for you, delete the old corrupted launch configuration and recreate it.
When you attempt to Live Update your workload, the following error message appears in the log:
ERROR: Build Failed: apply command timed out after 30s - see }}{{https://docs.tilt.dev/api.html#api.update_settings{{ for how to increase}}
Kubernetes times out on upserts over 30 seconds.
Add update_settings (k8s_upsert_timeout_secs = 300) to the Tiltfile. For more information, see the Tiltfile documentation.
On macOS, the Tanzu Panel doesn’t display workloads or any other resources when using a GKE cluster. Other tools, such as the Tanzu CLI Apps plug-in, display resources correctly.
gke-cloud-auth-plugin is required to properly authenticate to a GKE cluster. However, when starting IntelliJ from Dock or Spotlight, environment variables set by using .profile, .bash_profile, or .zshrc are not available. For more information, see this YouTrack issue.
This might cause gke-cloud-auth-plugin to be missing from PATH when launching IntelliJ and prevent the Tanzu Panel from reaching the cluster.
Open IntelliJ from the CLI. Example command:
open /Applications/IntelliJ\ IDEA.app
The pop-up menu Describe action in the Activity panel fails when used on PodIntent resources. The error message is similar to the following:
Warning: conventions.apps.tanzu.vmware.com/v1alpha1 PodIntent is deprecated; \
use conventions.carto.run/v1alpha1 PodIntent instead
Error from server (NotFound): podintents.conventions.apps.tanzu.vmware.com "my-app" not found
Process finished with exit code 1
When there are multiple resource types with the same kind, attempting to describe a resource of that kind without fully qualifying the API version causes this error.
There is no workaround for this issue at present. A fix is planned for this issue in the next version.
The Tanzu panel shows workloads but doesn’t show Kubernetes resources in the center panel of the activity pane.
When switching the Kubernetes context, the activity pane doesn’t automatically update the namespace, but the workload pane detects the new namespace. Therefore, the Tanzu panel shows workloads but doesn’t show Kubernetes resources in the center panel of the activity pane.
Restart IntelliJ to properly detect the context change.
ClusterBuilderLive Update does not work when using the Jammy ClusterBuilder.
A fix is planned for Tanzu Application Platform v1.5.1.
When an application is applied from IntelliJ it restarts frequently.
An application or environment behavior is triggering the application to restart.
Observed trigger behaviors include:
Prevent the trigger behavior. Example solutions include: