I am a noob trying to do a quick proof-of-concept, trying to deploy a KameletBinding into OpenShift and having a hard time deploying into my Camel K Operator.
All of my attempts are basically getting stuck on the "Creating" stage, then eventually failing with the following message:
Integration "redhat-test" does not have a readiness condition
Below is an example of a basic YAML deployment:
apiVersion: camel.apache.org/v1alpha1
kind: KameletBinding
metadata:
name: redhat-test
spec:
source:
ref:
apiVersion: camel.apache.org/v1alpha1
name: timer-source
kind: Kamelet
properties:
period: '10000'
message: Hello
sink:
ref:
apiVersion: camel.apache.org/v1alpha1
name: log-sink
kind: Kamelet
properties:
showHeaders: 'true'
Any chance I can get a little help from you to identify the issue here?
I've tried the same actions using the Red Hat Integration Camel K operator with no luck.
@Squake is correct. The Issue raised on GitHub was spot on, and was the culprit behind our problem as well.
For further explanation, we quickly configured an OpenShift cluster to explore Camel-K (and some future microservices work). As this was all done in haste, configuration of the OpenShift internal registry was not completed. That proved to be the culprit behind the
Integration "redhat-test" does not have a readiness conditionerror message.There's some merit to the comments on the Github Issue saying that there should be an error message/validation within the Camel-K Operator letting people know if the Internal Registry is not configured since that is a requirement for deploying a KameletBinding. But, perhaps we too are an outlier in that setting up the OpenShift Internal Registry is such a core step in setting up OpenShift, that most people would not miss that step, and we were simply a victim of our own haste.
Whatever the "fix" (if there is one) is, the error message in the Camel-K Operator is pretty misleading as it leads you to think there is an issue in the DSL, where the problem really has to do with OpenShift. That "unknown" sparked a days-long debate between our developers and SysOps guys until we stumbled across the abovementioned GitHub Issue.
My apologies, Stackoverflow community, for not posting an Answer to this question sooner.