IoT Core OTA Job FAILED retry techniques

Hi there IoT people,

I am looking into the best practice approach to retrying an OTA update to failed devices.

Scenario:

"Let's say we ran an OTA update for 1000 devices, and 42 of them FAILED for some reason. We corrected the "reason" and now need to rerun the job for 42 devices only.

So, I looked thru the API's and I got as far as

Attach a lambda to a IoT Rule that filters for Completed events, as per

• https://docs.aws.amazon.com/iot/latest/developerguide/events-jobs.html
• https://docs.aws.amazon.com/iot/latest/developerguide/iot-lambda-rule.html

The next logical step is call ListJobExecutionsForJob, passing in the job Id, filtering for FAILED jobs only.

But I can't see a way to marry up the Execution to a Thing, because you need the thing name to call DescribeJobExecution.

Do you need to use Fleet Indexing, or is there a better way to programmatically get the list of failed Things for an OTA Job, so we can alert / report / build retry tooling?

Thanks!