Offline restore fails with `Create OperandRequest Timeout` warnings

When you restore an offline backup of a Cloud Pak for Data deployment, the restore operation runs for a long time and then fails.

Symptoms

In the CPD-CLI*.log file, you see multiple warnings like in the following example:

Time: <timestamp> level=warning - Create OperandRequest Timeout Warning
Time: <timestamp> level=warning - Create OperandRequest Timeout Warning
Time: <timestamp> level=warning - Create OperandRequest Timeout Warning
Time: <timestamp> level=warning - Create OperandRequest Timeout Warning
Time: <timestamp> level=warning - Create OperandRequest Timeout Warning
Time: <timestamp> level=warning - Create OperandRequest Timeout Warning
Time: <timestamp> level=warning - Create OperandRequest Timeout Warning
Time: <timestamp> level=warning - Create OperandRequest Timeout Warning
Time: <timestamp> level=warning - Create OperandRequest Timeout Warning
Time: <timestamp> level=warning - Create OperandRequest Timeout Warning

Causes

The most likely cause is that the Operand Deployment Lifecycle Manager (ODLM) failed to install role or rolebindings in the Cloud Pak for Data instance project (${PROJECT_CPD_INST_OPERANDS}).

Diagnosing the problem

Run the following commands to show the OperandRequests in the Cloud Pak for Data operators project and in the Cloud Pak for Data instance project:

oc get opreq -n ${PROJECT_CPD_INST_OPERATORS}

oc get opreq -n ${PROJECT_CPD_INST_OPERANDS}

In the output for each command, the PHASE column is empty for some OperandRequests. For example:

oc get opreq -n ${PROJECT_CPD_INST_OPERATORS}
NAME                             AGE     PHASE     CREATED AT
im-service                                         <timestamp>
zen-ca-operand-request                             <timestamp>
zen-service                                        <timestamp>

Resolving the problem

Clean up the Cloud Pak for Data projects and retry the restore.