On 19/10/16 13:38, Lars Hupel wrote:
> Oh, nothing went wrong: Jenkins sent an email immediately after the
> push, and the status page also indicated the failure. Automation working
> as expected.
But that did not help. Florian experienced the broken repository just in
the way it occasionally happens.
The open question is if the massive use of CPU resources by Jenkins is
justified to deliver its service. It is still unclear to me what this
service is in the first place.
So far there were only axioms about Jenkins, but no proofs.
isabelle-dev mailing list