Hi Marc Hedgley
Dealing with large file processing really is a frustrating job sometimes and i guess you've taken the right steps already by adjusting the settings, there are a few additional considerations I can recommend.
Instead of processing the entire 25GB file in a single run, consider breaking it down into smaller chunks and processing them in parallel. This way, each smaller chunk has a better chance of completing within the timeout limits.
Rather than processing the entire file synchronously within a single function call, you can design your Azure Function to handle chunks of the file asynchronously. Use Azure Queues, and Azure Service Bus to trigger subsequent processing for each chunk.
and finally, Azure Durable Functions allows you to write stateful functions in a serverless environment. It's designed to handle long-running operations and can be used to implement workflows that span multiple functions.
Hopefully, this helps, kindly accept the answer if you find it useful thanks much.