Recover

Recover turns a panic in any later middleware or handler into a 500 Internal Server Error response, instead of a dropped connection. Register it near the start of the chain, after logging, so the logger records the 500.

app.Use(middleware.Recover())

The recovered panic goes through your error handler like any other error.

Use RecoverWithConfig to customize the response or log the stack.

app.Use(middleware.RecoverWithConfig(middleware.RecoverConfig{
  Handler: func(c *zinc.Context, err *middleware.RecoverError) error {
    log.Printf("panic: %v\n%s", err.Value, err.Stack)
    return c.Status(zinc.StatusInternalServerError).JSON(zinc.Map{
      "error": "internal server error",
    })
  },
}))

Set DisableStack when stack capture is not needed.