createOrReplace on Secrets throws in case the resource already exists

Describe the bug

Version of the kubernetes-client: 5.11.2

Performing 2 createOrReplace on the same Secret cause the second call to fail with this error:

io.fabric8.kubernetes.client.KubernetesClientException: Failure executing: GET at: https://192.168.64.3:8443/api/v1/secrets/test. Message: the server could not find the requested resource. Received status: Status(apiVersion=v1, code=404, details=StatusDetails(causes=[], group=null, kind=null, name=null, retryAfterSeconds=null, uid=null, additionalProperties={}), kind=Status, message=the server could not find the requested resource, metadata=ListMeta(_continue=null, remainingItemCount=null, resourceVersion=null, selfLink=null, additionalProperties={}), reason=NotFound, status=Failure, additionalProperties={}).

Fabric8 Kubernetes Client version

next (development version)

Steps to reproduce

Running a Main like this would reproduce the issue:

public class Main {

    private static Secret getSecret() {
        return new SecretBuilder()
                .withNewMetadata()
                .withName("test")
                .withNamespace("default")
                .endMetadata()
                .build();
    }
    
    public static void main(String[] args) {
        Config config = new ConfigBuilder().withNamespace(null).build();
        KubernetesClient client = new DefaultKubernetesClient(config);

        client.secrets().createOrReplace(getSecret());
        client.secrets().createOrReplace(getSecret());
    }
    
}

Expected behavior

The resource is seamlessly replaced.

Runtime

minikube

Kubernetes API Server version

1.22.3@latest

Environment

macOS

Fabric8 Kubernetes Client Logs

io.fabric8.kubernetes.client.KubernetesClientException: Failure executing: GET at: https://192.168.64.3:8443/api/v1/secrets/test. Message: the server could not find the requested resource. Received status: Status(apiVersion=v1, code=404, details=StatusDetails(causes=[], group=null, kind=null, name=null, retryAfterSeconds=null, uid=null, additionalProperties={}), kind=Status, message=the server could not find the requested resource, metadata=ListMeta(_continue=null, remainingItemCount=null, resourceVersion=null, selfLink=null, additionalProperties={}), reason=NotFound, status=Failure, additionalProperties={}).
    at io.fabric8.kubernetes.client.dsl.base.OperationSupport.requestFailure (OperationSupport.java:683)
    at io.fabric8.kubernetes.client.dsl.base.OperationSupport.requestFailure (OperationSupport.java:662)
    at io.fabric8.kubernetes.client.dsl.base.OperationSupport.assertResponseCode (OperationSupport.java:613)
    at io.fabric8.kubernetes.client.dsl.base.OperationSupport.handleResponse (OperationSupport.java:556)
    at io.fabric8.kubernetes.client.dsl.base.OperationSupport.handleResponse (OperationSupport.java:519)
    at io.fabric8.kubernetes.client.dsl.base.OperationSupport.handleGet (OperationSupport.java:488)
    at io.fabric8.kubernetes.client.dsl.base.OperationSupport.handleGet (OperationSupport.java:458)
    at io.fabric8.kubernetes.client.dsl.base.BaseOperation.handleGet (BaseOperation.java:696)
    at io.fabric8.kubernetes.client.dsl.base.BaseOperation.getMandatory (BaseOperation.java:182)
    at io.fabric8.kubernetes.client.dsl.base.BaseOperation.require (BaseOperation.java:163)
    at io.fabric8.kubernetes.client.dsl.base.HasMetadataOperation.requireFromServer (HasMetadataOperation.java:120)
    at io.fabric8.kubernetes.client.dsl.base.HasMetadataOperation.replace (HasMetadataOperation.java:187)
    at io.fabric8.kubernetes.client.dsl.base.HasMetadataOperation.replace (HasMetadataOperation.java:141)
    at io.fabric8.kubernetes.client.utils.CreateOrReplaceHelper.replace (CreateOrReplaceHelper.java:68)
    at io.fabric8.kubernetes.client.utils.CreateOrReplaceHelper.createOrReplace (CreateOrReplaceHelper.java:60)
    at io.fabric8.kubernetes.client.dsl.base.BaseOperation.createOrReplace (BaseOperation.java:316)
    at io.fabric8.kubernetes.client.dsl.base.BaseOperation.createOrReplace (BaseOperation.java:83)
    at io.fabric8.kubernetes.client.dsl.base.BaseOperation.createOrReplace (BaseOperation.java:306)
    at io.fabric8.kubernetes.client.dsl.base.BaseOperation.createOrReplace (BaseOperation.java:83)
    at org.keycloak.Main.main (Main.java:27)
    at org.codehaus.mojo.exec.ExecJavaMojo$1.run (ExecJavaMojo.java:254)
    at java.lang.Thread.run (Thread.java:829)

Additional context

I suspect the issue sits in the replace method, but I haven’t narrowed it down.

[shawkins]

We’ll address this in 6.0.0 - if the config has a null namespace, whether because there is not one on the context or the user calls withNamespace(null), the logic will assume an appropriate default - likely “default”. In the event of a checkNamespace call failing, the logic should inform the user to use inNamespace or the original namespace was null, to set a better default on the Config.

[shawkins]

I can add that the error doesn’t occur for e.g. deployments

The dsl logic will be different depending on the resource type - for deployments there is rolling logic that is implemented for a replace.

To restate what is happening under the covers in your test shown above:

Config config = new ConfigBuilder().withNamespace(null).build();
KubernetesClient client = new DefaultKubernetesClient(config);
client.secrets()...

is like calling:

client.secrets().inAnyNamespace()...

Note what operations are available when you call client.secrets().inAnyNamespace() - those method calls will still work as expected using withNamespace(null). However anything else you can call that has access to a passed in item may not be accounting for the possibility of a null namespace.

I can address this with a narrow change to the require logic: https://github.com/fabric8io/kubernetes-client/compare/master...shawkins:master

However all of the other operations would need similarly checked. If it’s the case that we always expect the initial config passed into the client to specify a default namespace, I’d rather add a validation just for that.