Skip to content

Handling nesting timeouts #5

@danielcompton

Description

@danielcompton

When I'm making HTTP requests, I have several levels of timeouts I'd like to be able to compose:

  • Granular connection and response timeouts (unfortunately not exposed in Fetch)
  • HTTP Request level timeouts - timeout for a single HTTP request
  • re-frame HTTP request deadline - cumulative time spent attempting to send this logical request, perhaps spread over multiple attempts, and including any exponential backoff time.

I would like to be able to specify an overall deadline for a request after which is it aborted (say 30s), set a per request timeout of 20s, and specify a retry policy. At the moment, it looks like I'd need to build in my own retry logic to handle the overall deadline. Would that be a useful general feature to have?

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions