Azure blob not remounted after node restart

[jakub-xen]

Problem Description

If the batch node is restarted the mounted azure blob using blobfuse is not automatically remounted

Batch Shipyard Version

shipyard.py, version 3.8.2

Steps to Reproduce

Create a pool with
volumes:
shared_data_volumes:
azureblob_vol:

click reboot node from the web portal

Expected Results

azureblob volume should be mounted on start as with azure files

Actual Results

azureblob is not present

Redacted Configuration

INSERT RELEVANT YAML FILES

Additional Logs

INSERT ADDITIONAL LOGS HERE

Additonal Comments

[alfpark]

Please provide more information as requested in the redacted configuration section. Posting a pool.yaml at the minimum would be great (config.yaml would be good too, redacted as necessary).

[jakub-xen]

For reference:

config.yaml:
batch_shipyard:
storage_account_settings: batchstore
global_resources:
docker_images:

• alpine
• jakubjs/azopenfoamv1806-intel
  volumes:
  shared_data_volumes:
  azureblob_vol:
  volume_driver: azureblob
  storage_account_settings: batchstore
  azure_blob_container_name: blob
  container_path: $AZ_BATCH_NODE_SHARED_DIR/azblob
  mount_options:
  - --use-https=true
  bind_options: rw

pool.yaml:
pool_specification:
id: workpool
vm_configuration:
platform_image:
publisher: OpenLogic
offer: CentOS-HPC
sku: '7.4'
vm_count:
dedicated: 0
low_priority: 1
vm_size: STANDARD_H16R
inter_node_communication_enabled: true
per_job_auto_scratch: true
ssh:
username: shipyard
virtual_network:
name: xenof-vnet
resource_group: xenof-group
create_nonexistant: true
address_space: 10.0.0.0/16
subnet:
name: batch-vm-subnet
address_prefix: 10.0.0.0/18

[alfpark]

Thanks for the info, this has been fixed in the develop branch (and develop-cli Docker image).