I recently had an issue where I had selected the
optimal strategy for the
instanceTypes on my Compute environment parameters, yet my jobs were appearing stuck in
runnable. This didnt make any sense, the reason I choose
optimal was to allow AWS to grab any instance for me from a range.
After a chat to the very helpful AWS Support I found out that the documentation was misleading, it says “choose optimal to pick instance types from the C, M, and R instance families”.
What it should says is “Optimal chooses the best fit of M4, C4, and R4 instance types available in the region”.
The CPU and Memory requirements I had set for my container, exceeded what any of those instance types could do.
The AWS support engineer said a ticket has been logged to update the documentation.