Migrating data from on-premise to cloud storage

1. Create the policy file policy.xml.

1. Create a volume set with existing volumes.
   Note: Unmount the file system before creating the volume set. After creating the volume set, mount it at the same mount point.
   \# umount mount_path_of_data_volume
\# vxvset -g dg_name make vset_name local_data_volume
\# mount -t vxfs /dev/vx/dsk/dg_name/vset_name \
mount_path_of_data_volume
   1. Create buckets/containers in the cloud storage namespace. See the related cloud vendor documentation for instructions.
   2. Create the cloud volume.
   For block-level data migration:
   \# vxassist -g dg_name make cloudvol_name size vxcloud=on

\# vxassist -g dg_name make cloudvol_name size fscloud=on

1. Configure the Cloud target.

   Connector	Command
   S3	lang=bash
   	\# vxcloud -g diskgroup_name \
   	addcloud cloudvol_name host=host_address \
   	bucket=bucket_name access_key=access_key \
   	type=S3 secret_key=secret_key \
   	where, secret_key and access_key are credentials to access the vendor cloud services.
   	By default, https is set to true and sig_version is set to v4 .
   Glacier	lang=bash
   	\# vxcloud -g diskgroup_name \
   	addcloud vol_name host=host_address \
   	bucket=vault_name access_key=access_key \
   	secret_key=secret_key type=GLACIER
   	where, secret_key and access_key are credentials to access the vendor cloud services.
   	Note: Only file-level tiering is supported with Amazon Glacier.
   BLOB	lang=bash
   	\# vxcloud -g diskgroup_name \
   	addcloud cloudvol_name \
   	host=host_address bucket=bucket_name \
   	access_key=access_key type=BLOB \
   	endpoint=account_name
   	where, access_key is the credential to access the vendor cloud services and endpoint is the storage account name of the user.
   	By default, https is set to true .
   Google Cloud	lang=bash
   	\# vxcloud -g diskgroup_name \
   	addcloud cloudvol_name host=host_address \
   	bucket=bucket_name type=GOOGLE \
   	google_config=config.json_file_path
   	where, config.json is a file that contains the private key, project_id , and client_email values for the Google service account. Download this file in the JSON format from the Service Accounts tab of the GCP Console.
   	By default, https is set to true .

   Note: Veritas recommends that you associate each cloud volume with a separate bucket.

2. Add the cloud volume to the volume set:
   \# vxvset -g dg_name addvol vset_name cloudvol_name
   1. Verify that the cloud volume is appropriately tagged.
   \# fsvoladm queryflags dataonly mount_path_of_data_volume cloudvol_name
3. Assign placement classes to the local and cloud volumes.
   \# vxassist -g dg_name settag local_datavol_name \
vxfs.placement_class.LOCAL
vxassist -g dg_name settag cloudvol_name \
vxfs.placement_class.CLOUD
   1. Assign the policy to the file systems.
   \# fsppadm assign mount_path_of_data_volume policy.xml
4. View an analysis report of the data transfer.
   \# fsppadm analyze mount_path_of_data_volume
   1. Enforce the policy to move the data between the local volume and the cloud volume.
      Note: You can create a cron job to schedule the migration of old data onto the cloud volume.

   \# fsppadm enforce mount_path_of_data_volume
5. Verify the location of files on the local and cloud volumes:
   \# fsmap -a list_file
\# fsmap -a /data1/*
Volume    Extent Type     File Offset     Extent Size     File
localvol         Data               0         1048576     /data1/reports-2016-03
cloudvol         Data               0         1048576     /data1/reports-2016-04
   1. Check the free space and used space across volumes in the volume set using
   \# fsvoladm list mount_path_of_data_volume
\# fsvoladm list /data1
devid   size                 used                 avail                name
0       2097152              356360               1740792              localvol
1       10737418240          40                   10737418200          cloudvol

Related information