Restoring Palo Alto form AWS snapshot

[pink-panther]

I've got a process to do this (recover Palo Alto from AWS snapshot) but no way to practice before a disaster. Can anyone confirm I have the important steps covered with respect to networking?

1. Create AMI from snapshot.

2. Launch instance using that AMI.

3. Ensure it goes in the correct management subnet when doing the above (other stuff like security group).

4. Move all ENIs from old/broken PA over to this one with exception of eth0 .

If anyone knows of a tutorial on this (specifically Palo Alto) please let me know. Not sure if there is a free way to rehearse this on AWS ?

[LAYER_8]

You will also want to export the Palo NGFW configuration. Though I know we are just duplicating the same data into a new instance, I've seen less than perfect replication occur before. I'm not sure there is a free way to test this, but, we do have an IronSkillet on our Github that scripts an AWS HA failover. I'm sure some of that could be frankenstein'd into taking down a network for coldstart failover. 

[rmonge]

Do we support taking snapshots in AWS for Palo Alto Network firewalls or Panorama?

Environment:

• Virtual Palo Alto firewall
• Virtual Panorama    
• Deployment: AWS

Answer

Palo Alto VM firewall or Panorama does not support AWS snapshots.

[daz12]

Why ? Is there a technical reason ?

[LAYER_8]

In general, if vendors did not block this, their customers could deploy as many of the boxes as they wanted. Each time a CPUID/UUID changes (restoring from snapshot, redeploying in a different hypervisor, etc) there is a mismatch on the license of the boxes. There isn't a technical reason, this is working as designed. However this is an opinion, there are other vendor products that allow for this and can be dupe'd in lab environments.