awsbatch: Tags are not getting propagtaed to compute instance volumes for awsbatch scheduler

[manikandan-thangavelu-rl]

Required Info:

• AWS ParallelCluster version: 3.1.4

• Full cluster configuration without any credentials or personal data.

  Region: ap-south-1
  Image:
  Os: alinux2
  HeadNode:
  InstanceType: t3.micro
  Networking:
  SubnetId:
  Ssh:
  KeyName: MountTest_Omicron
  Imds:
  Secured: false
  Scheduling:
  Scheduler: awsbatch
  AwsBatchQueues:
  - Name: queue1
  CapacityType: SPOT
  ComputeResources:
  - Name: optimal
  InstanceTypes:
  - c4.large
  MinvCpus: 0
  DesiredvCpus: 0
  MaxvCpus: 10
  SpotBidPercentage: 100
  Networking:
  SubnetIds:
  -
  Tags:
  - Value:
  Key: cost_resource
  - Value:
  Key: project_name
  - Value:
  Key: researcher_name

• Output of pcluster describe-cluster command.

  {
  "creationTime": "2022-08-12T08:32:13.754Z",
  "headNode": {
  "launchTime": "2022-08-12T08:35:06.000Z",
  "instanceId": "",
  "publicIpAddress": "",
  "instanceType": "t3.micro",
  "state": "running",
  "privateIpAddress": ""
  },
  "version": "3.1.4",
  "clusterConfiguration": {
  "url": ""
  },
  "tags": [
  {
  "value": "3.1.4",
  "key": "parallelcluster:version"
  },
  {
  "value": "",
  "key": "researcher_name"
  },
  {
  "value": "",
  "key": "cost_resource"
  },
  {
  "value": "",
  "key": "project_name"
  }
  ],
  "cloudFormationStackStatus": "CREATE_COMPLETE",
  "clusterName": "RG-Pcluster-083d1f4950c6e4304",
  "computeFleetStatus": "ENABLED",
  "cloudformationStackArn": "",
  "lastUpdatedTime": "2022-08-12T08:32:13.754Z",
  "region": "ap-south-1",
  "clusterStatus": "CREATE_COMPLETE",
  "scheduler": {
  "type": "awsbatch"
  }
  }

Bug description and how to reproduce:

Tags that are predefined in the ParallelCluster configuration file are not getting applied to all EBS volumes associated with the Compute nodes when the sceduler type is awsbatch. The same works fine with slurm as the scheduler.

P.S: The same has been tested with the latest Pcluster verison (3.2.0) also.