Aspera On Demand: Attaching an EBS volume for User Docroots

Summary

This article provides a basic set of instructions on how to modify your Aspera on Demand (AoD) image to utilize EBS storage.  The default AoD image uses ephemeral storage for all transfer data.  This storage is not preserved in the event of a system failure or if you shut down your image. By switching to EBS storage the data transferred into the image can be preserved even if an instance is shutdown.

Instructions

Step 1 - Login into the AWS management console using your Amazon credentials

Step 2 - Add an EBS volume

First, determine which Availability Zone your Aspera On Demand Instance is located in. You can determine the Zone, by viewing the instance details.  In the image below, you can see the Region and Zone of the currently selected Aspera on Demand instance.

Next, create an EBS Volume within the Same Availability Zone by selecting Volumes under ELASTIC BLOCK STORE from the menu on the left and then clicking the Create Volume button near the top of the page and specifying the desired size of your volume.

Step 3 - Attach the EBS Volume to the Instance

From the EBS Volumes listing right click on the volume and choose Attach Volume from the context menu. Select the instance in the Instances drop-down box and then click the Attach button.

NOTE: It is suggested that you mount the EBS volume starting at /dev/sdf, (sdb/c/d/e are generally used for ephemeral drives.) Device will automatically be populated with /dev/sdf since this is the recommended value.

Step 4 - Create a file system on the newly mounted EBS volume

SSH into your OnDemand instance, and run the following commands to create and mount the disk.

NOTE: The command to execute in each step is specified after the # prompt.

a. Verify the newly attached EBS drive (sdf in this example).

# cat /proc/partitions
major minor #blocks name

8 1 2097152 sda1
8 16 440366080 sdb
8 32 440366080 sdc
9 0 880732032 md0
8 80 1048576 xvdf

b. Make xfs the default file system (Note: Kernel type requires -l option)

# mkfs.xfs -f -l lazy-count=0 /dev/xvdf

c. Create a mount point for /dev/sdf by editing /etc/fstab and adding the entry as highlighted below

# vim /etc/fstab
/dev/sda1 / ext3 defaults 1 1
none /dev/pts devpts gid=5,mode=620 0 0
none /dev/shm tmpfs defaults 0 0
none /proc proc defaults 0 0
none /sys sysfs defaults 0 0
/dev/md0 /mnt/ephemeral xfs defaults 0 0
/dev/xvdf /mnt/ebs xfs defaults 0 0

d. Mount the new EBS volume

# mkdir /mnt/ebs
# mount -a
# df -h /mnt/ebs
Filesystem Size Used Avail Use% Mounted on
/dev/xvdf 1014M 160K 1014M 1% /mnt/ebs

Step 4: Configure each Aspera transfer user's docroot to point to the EBS storage

To move existing users to the new mount use the following command. (Skip if no users have yet been created.)

# mv /mnt/ephemeral/* /mnt/ebs

Through the Aspera OnDemand Console GUI, set the docroot of the aspusers group to /mnt/ebs/$(name) and save changes.

 Once complete Aspera transfers will be to EBS storage.

Have more questions? Submit a request

0 Comments

Article is closed for comments.
Powered by Zendesk