On 7/8/19 2:25 PM, Jonathan Dieter wrote:
On Tue, 2019-07-02 at 16:27 -0700, Kevin Fenzi wrote:
> At the point all these are failing, it would be running the rawhide
> curl/dnf stack, not the native f29 one on the builders, so if it's a
> curl problem I would expect it to happen more often.
>
> So, how can we debug or mitigate this? any ideas?
I think the easiest way to mitigate this would be to fix DNF to retry
when failures happen on a single mirror. I know I said that I'll look
into this, but I've started a new job and haven't really had the time
yet.
Yeah, that might help...sadly we don't even have any logs to look at
from the appliance cases at least. Those are in a loop mounted image and
when it fails, it umounts and removes the failed image. ;(
kevin