Azure Databox setup

Azure Databox setup.md

Steps required to setup Azure data box - personal observations

Based on: https://docs.microsoft.com/en-us/azure/databox/data-box-deploy-ordered?tabs=portal and observations during setup. Also : https://docs.microsoft.com/en-us/azure/databox/data-box-deploy-picked-up?tabs=in-us-canada-europe

Before you begin:

Requires: i) 1x Azure Databox

ii) 1x Cat 6 cable

iii) 1x Cat 6A cable OR 2x SFP+ links

iv) Laptop computer - ensure that you have permissions to modify the network IP

v) 10GbE (ideal) or 1GbE (optional) for data transfer

Step 0:

Make sure that you have the Azure Databox password - obtain from the Azure portal

Step 1:

Open the data box on both ends - check for power cable. If no power cable, find one (minimum 15AWG)

Step 2:

Connect MGMT port to laptop computer using Cat6 cable

Step 3:

IF using Cat6A ethernet, connect Data3 to ethernet port on Server. If using SFP+, connect 2x to Data 1+2 on Databox, with matching SFP+ ports on the server

Step 4:

Set IP settings on the laptop ethernet card:

Static IP = 192.168.100.5
Subnet Mask = 255.255.255.0

Step 5:

Open browser on laptop, and type in https://192.168.100.10

Step 6:

Go to "set network interfaces" to get IP address for both Databox and Gateway

Step 7:

Hit the arrow next to your storage account on "Connect and Copy" and identify the Block blob path

Step 8:

Click Connect and Copy -> NFS -> and input the IP address of the gateway

Step 9:

Go to your linux server, and create a directory to serve as a mount point - can be anything, as long as it is an empty directory

Step 10:

On the linux server, type:

sudo mount -t nfs <IP address of Databox>:<storage account name>_BlockBlob /path/to/mount/point/in/step/9

OR, as of the latest databox,

sudo mount -t nfs <IP address of Databox>:<storage account name>_BlockBlob_Archive /path/to/mount/point/in/step/9

To specifically move to archive storage if that is intended.

Step 11:

Navigate to the newly created mount point, and create a master folder to serve as the data container.

NOTE: No underscores, please no underscores, please please please no underscores when creating this container folder. The names can only have letters, numbers, and hyphens.

Step 12:

Start transferring data - likely by Rsync. Permissions didn't allow for rsync -a, and symbolic links were rejected by Azure, so rsync -rt seems good to be both recursive and keep modification times.

Return

Step 13:

Recommended to a) cancel any running transfers (eg. pkill screen if data transfers were setup using screen); b) unmount the drive from the linux server (eg. sudo umount -l /path/to/mountpoint); c) disconnect data3 from server to ensure no lingering data transfers running.

Step 14:

Repeat Steps 1-5 with laptop

Step 15:

Go to "Prepare to ship" and click "Start Preparation". This may take days!

Step 16a:

Check for errors - if none, then e-ink screen on front will display return shipping label and the databox will lock. Can download file list and error list to laptop.

Step 16b:

If there are errors to fix, easiest to do by re-mounting the databox. Mounting in Windows (eg. https://docs.microsoft.com/en-us/azure/databox/data-box-deploy-copy-data) was easy way to change corrupting names

Step 17:

Arrange UPS pickup or drop it off at pickup point.