AWS EC2 availability zones

[Mark Rendle]

Hi,

Please excuse the newbie question, but my Google-fu has failed me.

The Fleet docs have various references to AWS availability zones, but I have not been able to find a straightforward example of scheduling instances across them. That is, where my cluster has N machines, where N is a multiple of 3 and those machines are split equally across three AWS availability zones, how do I ensure that an instance of a process is run in each of the three zones?

Any pointers or links would be *very* much appreciated.

Thanks,
Mark

[Seán C. McCord]

Add the availability zone of each instance into its fleet metadata (most likely by way of your CloudFormation template, but however you define your instance cloudconfigs). Then schedule one unit to run in each availability zone, matched by the fleet metadata matching that zone (the `MachineMetadata` X-Fleet tag).

There is no direct access to such EC2-specific data as availability zones from Fleet.  You simply apply that data to the fleet metadata to make it work.