Related: https://pagure.io/koji/issue/3170
Just a few minor things.
The comment block before this change seems to be explaining why the old code did not retry, and should probably be adjusted.
We might want to log when we retry, for consistency (the normal retry mechanism logs retries at info level).
I'm a little concerned about using is_conn_error here. It might be a little too narrow, but I guess it's a good starting point. This function is intended to identify the cases where we need to reset the connection an retry (once) in _sendCall, so this is overloading a bit. I think it's fine to check it, but if it proves to be insufficient, then we need to be really careful about modifying it.
is_conn_error
rebased onto 68d51b348822bcd5b02056c1a04eeee9d099d091
thanks for the updates! :thumbsup:
Metadata Update from @tkopecek: - Pull-request tagged with: testing-ready
:thumbsup:
Commit 8e3973ad fixes this pull-request
Pull-Request has been merged by tkopecek
Metadata Update from @mfilip: - Pull-request untagged with: testing-ready - Pull-request tagged with: no_qe
Related: https://pagure.io/koji/issue/3170