Showing results for 
Search instead for 
Did you mean: 

How should we configure NetBackup CC media server in EC2 instance?

Level 2

We are currently using Netbackup version 3.2.
Backup master server is configured as in on-premises and VM of VMware system is backed up in S3 bucket through CC.
We are planning to recover backed up data of S3 bucket through EC2 instance in AWS system.
In order to recover, CloudCatalyst media server must be prepared within EC2 instance.
How should we configure NetBackup CC media server?
Can you inform us the EC2 recovery process procedures?


Level 6
Partner    VIP    Certified

1) Do you not have an on-prem MSDP server too ?

2) Or is there only one Master on-prem, plus one CC on prem, and no MSDP on-prem ?

3) If I understand, then you want to be able to import and inventory to be able to "see and manage" this "one set of backups" (in the S3 bucket) from two independent NetBackup domains "at the same time".  This is most definitely not advisable, because both Master Servers will manipulate the contents by each trying to manage expiry and so both Master Servers will quickly each effectively corrupt and likely crash each others catalog.

4) You might be able to achieve developing some DR import and inventory procedures to execute upon a luke-warm (pre-built, empty, forever waiting) NetBackup Master server which will effectively sit forever dormant n EC2 - but personally, I would develop these procedures using a demo/test on-prem environment with a demo/test in-cloud environment - and keep practising until you are confident that the same (or at least a very similar) procedure will work for production.

5) Alternatively, re-think your design and "replicate" AIR your backups to a hot live forever instantly DR ready NetBackup domain in EC2.

Level 6

I think what you are looking for guidance on is the new feature in 8.2 with CC where the images sent to the cloud are self-describing, thus allowing recovery without the original infrastructure. If so then...

Read the section in the 8.2 Deduplication Guide regarding "About automated disaster recovery in cloud using
CloudCatalyst". This describes what is required to recover the CC backups from the cloud bucket. It starts from page 280.


1&2) We have an on-prem MSDP server(Netbackup Appliance 8.2). Furthermore, we also have Netbackup CloudCatalyst Appliance on-prem. We have one Master as well as one CC on prem

3) We are currently using just one Netbackup domain. Hence, collusion is irrelevant at this point.

4&5) Our main objective is to perform automated disaster recovery in clouding using S3 bucket.


[NetBackup CloudCatalyst Media Server -> S3 Bucket ->EC2 Instance]



이미지 1.jpg


According to deduplication guide in page 281, there is information about installing NetBackup, creating instance based on RHEL 8.0. However, we are unsure of how to proceed with this installation. Is it appropriate to install with NetBackup CC media server? Where can we refer to installing with NetBackup CC media server? Or is there a virtual appliance?

Level 6
Partner    VIP    Certified

On page 281 it says "NetBackup should be an all-in-one setup. Refer to the NetBackup Installation Guide for more information." - so this implies to me one "NetBackup Master + CC Server", i.e. one single multi-role server.


Also, I think you need to search the NetBackup documentation page for "IAM role" and "AutoDR" and read through.  For example, I searched, and found two articles immediately:

How to configure Amazon Web Services when using I AM Role with NetBackup

Import and restore jobs in AutoDR server fail. Enable AutoDR with manual KMS key transfer.

AWS source account configuration may fail with "Access Denied. IAM role with sufficient privileges must be attached to the CloudPoint EC2 host instance"

Partner    VIP    Accredited Certified

My understanding is that you don't need CC media server in AWS - just a single master/media server.

The 8.2 Dedupe Guide refers to this on page 281 : 'NetBackup should be an all-in-one setup.'

With 8.2, there is the added feature to CloudCatalyst to send NBU Catalog meta data with the backup data.
The AWS master/media server will simply import the dedupe images. 

The manual describes the process under these headings: 
Configure automated DR
Using automated DR

As we went through the Guide, there is mentioning of NetBackup Image Sharing.
According to the Guide, VMware recovery is possible from S3 bucket through NetBackup Image Sharing.
What are the procedures for configuring NetBackup Image Sharing?
In addition, in some documents, it is written: NetBackup Master & CC; in some other documents, it is written: NetBackup Master/ Media & CC; in another documents, it is written: Master Media Server. Among these expressions, which one is the right one?
Furthermore, some document mentions "All-in-One". What does it mean by "All-in-One"?

Partner    VIP    Accredited Certified

Best to use the Dedupe Guide for HOWTO instructions. 

Master server installation contains ALL the binaries that are needed. 
So, you do not need a separate CC media server.

Extracts from page 281: 
NetBackup should be an all-in-one setup.
Refer to the NetBackup Installation Guide for more information.

Using automated DR  (after Configure automated DR steps were executed) :
Run the nbimageshare command to list and import the virtual machine and standard images and then recover the virtual machines.
The path to access the command is: /usr/openv/netbackup/bin/admincmd/
For more information about the nbimageshare command, refer to the NetBackup Commands Reference Guide.