Karsten Bräckelmann wrote:
Like a successful update from the second mirror, or any other non-stock channel?
I'm thinking this is the case (second mirror) but the log file does not clearly state that the bad mirror is ignored and it continues with a good mirror for an update file.
To my eyes: sa-update failed. I get a message from cron showing me the mirror had a read timeout. Nothing about it rolling on to a second mirror.
Is the cron job an SA problem or is it something the Fedora maintainer made?