Skip to content

Avoid retry storms after permanent clip watcher HTTP errors #10

Description

@Luqueee

Problem

In the Darwin watcher internal/clip/watcher_darwin.go:100-126, lastTS is updated only after a successful PUT.

A permanent 413 or 415 response leaves the same clipboard hash pending, so the watcher retries the same rejected image every tick and can generate sustained traffic and logs.

Acceptance criteria

  • Classify permanent client errors separately from transient transport or server errors.
  • Do not retry an image forever after a permanent rejection.
  • Continue retrying transient failures.
  • Add Darwin-specific tests for 413, 415, 5xx, and successful retry behavior.

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions