Take the 2-minute tour ×
Server Fault is a question and answer site for professional system and network administrators. It's 100% free, no registration required.

I was working on the mirroring of an EC2 instance. I followed these steps

  1. create an image(AMI) of the instance you would like to mirror
  2. Launch an instance out of the image
  3. ssh into the new instance and launch the processes

The new instance worked like a charm.

I also tried the another method.

  1. Create a snapshot of an instance
  2. Create an image from snapshot
  3. Try and launch the instance from the image.

But somehow the server is not getting set up properly and it fails the status checks built in Amazon console. I need to figure out the reason as it may be critical at some point to be able to launch a server from the snapshot.

share|improve this question

migrated from stackoverflow.com May 2 '12 at 15:02

This question came from our site for professional and enthusiast programmers.

2 Answers 2

up vote 2 down vote accepted

The most common cause of the scenario you describe is a failure to specify the correct kernel (AKI) when registering the snapshot as an AMI.

Compare the AKI of the AMI that works with the AKI of the AMI that does not work.

share|improve this answer
    
Thank you. It was that the kernel ID was not correct –  ranjith May 2 '12 at 4:50
    
Thank you soooooooo much!!! I have wasted hours with this... –  Rafa May 7 '12 at 15:28
    
If the original instance is gone and you only have the snapshot, how do you determine the AKI? Any automated way to store the AKI-id in the snapshot description or as a tag? Can ec2-consistent-snapshot do this? –  Martijn Heemels Aug 13 at 15:07
1  
@martijnheemels I think most PV AMIs have been using the same PV AKI. Yes, snapshots can have tags, and ec2-consistent-snapshot can add tags, but why not register snapshot as an AMI at create time? Shouldn't add any cost over the snapshot itself. –  Eric Hammond Aug 31 at 6:24
    
@EricHammond We create daily snapshots of multiple volumes on a dozen instances. We rarely need to recover from one. I suppose I thought having so many AMI's would be hell to maintain and clean up. Good idea though. Any suggestions on how to automate such a system? –  Martijn Heemels Sep 1 at 14:38

Just to add to the answer:

You can troubleshoot boot issues by viewing the system log. You can do this in the AWS console on the EC2 tab by selecting the instance and clicking 'Instance Actions' > 'Get System Log'.

share|improve this answer
    
My System Log never got generated... :-( –  Rafa May 7 '12 at 15:27
    
FYI: The system log sometimes takes a while to appear as available in the console. I've had it take up to 30. –  Dave Forgac Jun 27 '12 at 3:12

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.