Csi.sock error when trying to deploy GCP Persistent Disk on Nomad 1.8.1

Nomad
1

I keep getting the following error whenever I try to deploy GCP-PD-controller or node jobs on my nomad cluster.

Error: CSI plugin failed probe: timeout while connecting to gRPC socket: failed to stat socket: stat /opt/nomad/data/client/csi/plugins/35d9ab7c-9111-ce54-da1f-8f3cabceedcd/csi.sock: no such file or directory

job "plugin-gcp-pd-controller" {
  datacenters = ["dc1"]

  group "controller" {
    task "plugin" {
      driver = "docker"

      env {
        GOOGLE_APPLICATION_CREDENTIALS = "/secrets/service-account.json"
      }
      template {
        data        = <<EOF
    {{ key "service-account" }}
    EOF
        destination = "/secrets/service-account.json"
      }

      config {
        image = "registry.k8s.io/cloud-provider-gcp/gcp-compute-persistent-disk-csi-driver:v1.15.0"

        args = [
          "controller",
          "--endpoint=unix://csi/csi.sock",
          "--logtostderr",
          "--v=5",
        ]

        privileged = true
      }

      csi_plugin {
        id             = "gcp-pd2"
        type           = "controller"
        mount_dir      = "/csi"
        #stage_publish_base_dir = "/local/csi"
        health_timeout = "45s"
      }

      resources {
        cpu    = 500
        memory = 256
      }
    }
  }
}

Debug info:
Nomad v1.8.1
BuildDate 2024-06-19T06:43:57Z
Revision 5022543e4b7b8dcec9df123f86630ae3fdcffbe6

Nomad is running as root and has full permissions on /opt/nomad/data

Not sure how to fix, thanks for any help.

Frank