From: Lawrence Paulson <lp15@cam.ac.uk>
I was testing some last-minute, minor changes. Testboard run 917 aborted due to a timeout. You’d think that some of the jobs had timed out, but when I checked the log, it looked like an entirely different process had been run. In particular, the log file is more than double the length of its predecessors:
job 615: 10,793 lines
job 616: 10,775 lines
job 917: 22,841 lines
What gives? I’d like to get this test done.
Larry
From: Fabian Huch <huch@in.tum.de>
Builds depend very much on the implicit state of their predecessor -
sessions that were previously built and are still consistent with the
current sources are skipped.
If you go a few builds back, the output is 42,927 lines.
Fabian
From: Lawrence Paulson <lp15@cam.ac.uk>
So how does one trigger a new run without pushing new material?
Larry
isabelle-dev mailing list
isabelle-dev@in.tum.de
https://mailman46.in.tum.de/mailman/listinfo/isabelle-dev
From: Fabian Huch <huch@in.tum.de>
Log in to jenkins with your isabelle TUM credentials (the same ones as
for repository access) and click on "Build with Parameters" > "Build".
Fabian
isabelle-dev mailing list
isabelle-dev@in.tum.de
https://mailman46.in.tum.de/mailman/listinfo/isabelle-dev
From: Lawrence Paulson <lp15@cam.ac.uk>
… is out of action again, the website at least
Larry
isabelle-dev mailing list
isabelle-dev@in.tum.de
https://mailman46.in.tum.de/mailman/listinfo/isabelle-dev
From: Fabian Huch <huch@in.tum.de>
Thanks, the jenkins process crashed -- this technology is getting less
and less reliable.
Fabian
isabelle-dev mailing list
isabelle-dev@in.tum.de
https://mailman46.in.tum.de/mailman/listinfo/isabelle-dev
Last updated: Dec 21 2024 at 16:20 UTC