Terraform Enterprise deployment on Podman fails to start with "HostPath, ConfigMap, and PersistentVolumeClaim are currently the only supported VolumeSource"

Avatar
Dmitry Uchuvatov
  • Updated

Problem

Terraform Enterprise deployment on Podman fails to start with
Error: /fdo/kube.yaml: failed to create volume "log": HostPath, ConfigMap, and PersistentVolumeClaim are currently the only supported VolumeSource

Prerequisites

  • Terraform Enterprise on Podman

Cause

Podman version is lower that 4.3.0, which is required as per the System Requirements.

This can be validated by executing podman version command on virtual machine(s) where the installation of TFE failed:

$ podman version 
Version: 4.2.2

The reason why we require to run TFE on Podman v4.3.0 (or newer) is the support of emptyDir block that was unavailable in older versions.

For the reference, we have the following part in the Example Configurations section:

volumes:
...
  - emptyDir:
      medium: "Memory"
    name: "log"
  - emptyDir:
      medium: "Memory"
    name: "run"
  - emptyDir:
      medium: "Memory"
    name: "tmp"

Solution

Make sure to install Podman v4.3.0 or newer.

The steps are specific for different Operating Systems and available in Podman documentation.

Outcome

podman kube play <path_to_YAML_file> didn't return the error message.

Additional Information

Articles in this section

See more

Related articles