astefanutti commented on a change in pull request #2284:
URL: https://github.com/apache/camel-k/pull/2284#discussion_r649187560



##########
File path: e2e/common/kustomize/operator_test.go
##########
@@ -0,0 +1,89 @@
+// +build integration
+
+// To enable compilation of this file in Goland, go to "Settings -> Go -> 
Vendoring & Build Tags -> Custom Tags" and add "integration"
+
+/*
+Licensed to the Apache Software Foundation (ASF) under one or more
+contributor license agreements.  See the NOTICE file distributed with
+this work for additional information regarding copyright ownership.
+The ASF licenses this file to You under the Apache License, Version 2.0
+(the "License"); you may not use this file except in compliance with
+the License.  You may obtain a copy of the License at
+
+   http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software
+distributed under the License is distributed on an "AS IS" BASIS,
+WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+See the License for the specific language governing permissions and
+limitations under the License.
+*/
+
+package kustomize
+
+import (
+       "fmt"
+       "testing"
+
+       . "github.com/apache/camel-k/e2e/support"
+       . "github.com/onsi/gomega"
+)
+
+func TestBasicOperator(t *testing.T) {
+       defer Uninstall()
+
+       WithNewTestNamespace(t, func(ns string) {
+               ExecMake(t, Make("setup-cluster", fmt.Sprintf("NAMESPACE=%s", 
ns)))

Review comment:
       > That is true, although whenever kamel install is called the 
cluster-setup functions are always carried out. kamel install --cluster-setup 
just aborts the camel-k install once the cluster resources have been applied. 
Thus, the e2e tests are in fact doing the job twice (once in the github action 
and the other in the code). I found this when executing the e2e tests locally 
on my own openshift cluster.
   
   Right. I didn't pay attention to it, indeed, CRDs existence is checked every 
time... Yet, we may have the occasion to simplify things.
   
   > This needs more thought from the perspective of RBAC. The kind cluster has 
no concern for user privilege (the e2e tests use the default admin? user) and 
so executing all commands is not a problem.
   
   Right. This is actually the major reason we maintain the execution of the 
test suite in the OpenShift 3.x workflow.
   
   So my opinion for _Kustomize_ installation is that we should be biased for 
simplicity, rather than trying to cover all the cases, or duplicate existing 
approaches. As the operator model matures, there is an inclination to deploy a 
single operator instance, cluster-wide, (hence) by administrators. I think, the 
time when a _standard_ users, with only admin role on her/his namespace project 
will be gone, if it has ever be a thing. Developers that install an operator 
either use clusters like KinD, with no RBAC, or own the cluster, making them 
de-facto cluster admins. So I'd be inclined to drop support for low-privilege 
install, given the complexity it introduces, and for no real usage 
realistically.




-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


Reply via email to