Hi,
I was trying to check the fixes that Ernesto pushed yesterday, that should fix some of the Bugzillas I had written. But to my surprise, I could not see any difference. I then checked Hudson and it looks like there are several builds that have failed during the night:
and also
They seem to both have failed due to failed downloads from other repos and similar.
Since the former build is only run once every 24 hours, I would appreciate if someone could trigger a manual build to see if it works better. It would be good if the master build could be triggered as well. But hopefully there will be a push to master which will trigger it anyway.
I would also like to take the opportunity to ask if we can run these jobs a bit more frequent. I find it a bit frustrating sometimes to have to wait for up to an hour to be able to test fixes. I do know that the Papyrus builds are run each 20 minutes (if something has been pushed). So when I and Camille worked with improving the import tool, we had a very quick turn-around loop with those builds every 20 minutes.
So I wonder what are the reasons for *not* running these builds a bit more frequent to avoid unnecessary waiting time in the feedback loop? If we have capacity in the build server, it should be able to build 20 minutes after something has been pushed to master. Why is the nightly for the code-generator/run-time only run every 24 hours? Yes, it is traditionally called "nightly builds", but for me it was long time ago that "nightly builds" actually were run only once every night. :)
/Peter Cigéhn
