AWS Lambda: Ultimate Guide to Error Handling

Поділитися
Вставка
  • Опубліковано 18 гру 2024

КОМЕНТАРІ • 4

  • @MrDomoArigato
    @MrDomoArigato Рік тому +2

    Hey ! Awesome video, just want to add one caveat for 9:19 - “Caching” Message ID’s / Sequence ID’s in the function handler is not persistent across sandboxes, which is an issue.
    Whenever the poller fleet invokes Lambda, the requests get routed to any of the sandboxes (Concurrent Executions) that are idle. If none are idle, an on-demand gets created. This is a regular synchronous InvokeFunction. There is never a guarantee that same batch will go to the same sandbox.
    DynamoDB is by far the best option for production workloads, but I understand having it in your function code is definitely better than nothing - totally depends on the concurrency

    • @theburningmonk
      @theburningmonk  Рік тому

      That's a good point. It's an issue when you're running with more than one shard (which surprisingly is enough for a lot of workloads because of how mouth throughput you get per shard). Should have mentioned that caveat... shame UA-cam doesn't let you reupload :-/

  • @lorezyra
    @lorezyra 8 місяців тому +1

    Is there blog/medium page that highlights these best practices?

  • @theburningmonk
    @theburningmonk  Рік тому

    This is a preview lesson from my cohort-based workshop, "Production-Ready Serverless". If you liked this, then please check out the full curriculum at productionreadyserverless.com.
    You can get 15% OFF with the code "UA-cam15" during checkout.
    Hope to see you there!