AWS Standard Stack Install Fails

Support
,
#1

Situation
Hello, new to the forum. Long time sysadmin but new to openemr and aws.

After using and testing AWS express version I decided to upgrade to standard.

I attempted to create using the standard cloud formation stack.
However, during the stack creation, when it gets to WebserverInstance I receive a CREATE_FAILED message under events.

Any idea what may be causing this or how to fix? I have searched around but have not found any information pertaining to this.

Thanks for any help!

EDIT: Forgot to add that this only happens with t3 and not t2 instance.

OpenEMR Version
I’m using OpenEMR version AWS Standard

#2

Are you using the t3a instances? or is this a stock t3? I’m pretty sure we have issues with t3a but I haven’t heard of anyone having problems on the t3. I think we added support for t3 last year as an option to be selected in the template but I haven’t heard of anyone having issues with them. @jesdynf any thoughts here?

#3

I believe the specific one I was trying to install at the time was the t3.medium. I am running fine on t2.medium currently but was curious about it and heard that the t3 were better performance.

#4

No reason jumps out at me as to why this would be so. Immediate options:

  • Relaunch the stack, disabling rollback in CloudFormation, and you’ll be able to log into your instance and look around. There should be logging artifacts from CF’s cloud init that give you a hint about what’s gone wrong. Delete the stack once you’ve had your look.
  • Relaunch the stack as a t2 and then stop and change the instance type – I think this should work? Not 100% sure you can change generations, actually.

This weekend I’ll try to investigate what’s going wrong – that or bite the bullet and upgrade the launch scripts to Ubuntu 20, one of the two.

#5

@jesdynf You can stop and change generations as long as they are the same architecture that the OS on the installed hard drive supports. I’ve done it bunches of times, even switching from t2s/t3s to c5s etc.

I haven’t used the cloudformation though to launch with the t3 generation though, everything I did was my won ec2 launches, but I can’t imagine that being a difference, disabling rollback though is a good idea and you can find out what’s going on.

  • Stephen
#6

Thanks for all the help! Because of another issue I have ran into using the cloud formation(patient portal not functioning correctly) I have decided to try the install using my own setup instead of the stack. Changing the t2 to t3 after creation did work however!