search

GoogleCloudPlatform / flink-on-k8s-operator

[DEPRECATED] Kubernetes operator for managing the lifecycle of Apache Flink and Beam applications.
Apache License 2.0
658 stars 266 forks source link

ConfigMap and Secret fails to mount while deploying Flink Job Cluster #373

Closed sugam72 closed 3 years ago

sugam72 commented 3 years ago

I am trying to run Flink job cluster with savepoints using gcs connector. I am following this document.

When I create the Flink cluster, I get the following error in job manage's event. MountVolume.SetUp failed for volume "hadoop-config-volume" : couldn't propagate object cache: timed out waiting for the condition MountVolume.SetUp failed for volume "gcp-service-account-volume" : couldn't propagate object cache: timed out waiting for the condition

Is there any other consideration I need to take care of apart from the ones mentioned in document mentioned above?

Job cluster runs perfectly fine without hadoopConfig and gcpConfig in yaml, but I want to use the failover mechanism.

functicons commented 3 years ago

Can you print the information about hadoop-config-volume and gcp-service-account-volume, to check if they are in good state?

sugam72 commented 3 years ago

I am a little new to this. Can you help me understand how can I describe these volumes when the container has not created? Or you are asking something else?

sugam72 commented 3 years ago

Figured this issue comes if secret and configmap are not in the same namespace as the job cluster. It would be helpful if this consideration can be added to the documentation