Closed kasperk81 closed 1 month ago
I'm a little concerned about not seeing the curl errors at all - there are some error-handling routines in the curl/wget functions that return specific error messages, but sometimes a status code isn't enough to provide context for investigation. If we swallow all of these errors, how would users get detailed diagnostics? Is there maybe a way to preserve these error logs for later investigation?
when an "attempt" fails, it shows:
the middle line is redundant with
--verbose
and isn't part of the expected output in regular usage. removing it to avoid being flagged as an error by systems like CanonicalError.