It builds! Does it run?

Sunday, December 21, 2008 12:02 - GNOME

It builds!

An important step for the build brigade happened last week, the build master finally got some firewall issues resolved and was hooked to svn-commits-list to get change notifications, and those steps made it good enough to get its official name, it is now up on build.gnome.org, thanks to everyone involved!

It is not over yet and there is still plenty to be done, join the fun:

  • Instructions on live.gnome.org to configure slaves (and master) should be updated, probably on the basis of this mail
  • Current setup requires an attentive eye (or more) to catch regressions; it would be great to have per-module and per-slave feeds, listing build regressions (per-slave is easy, that is just a module who failed, while building the previous time; per-module may be more difficult, as we could want the regression to happen in more than half the slaves failed before assuming it is module regression, especially if we want to hook module regressions to IRC nagging).
  • Way less important but nice would be to have the build master reload its configuration (list of slaves, list of modules) without having to be restarted (I am sure it is almost there but I didn't have the time to finish this).
  • Another nice thing would be to have "administrative tasks" that could be sent by the master, things like updating modulesets or removing a module directory.
  • Particularly useful for the release team would be the handling of more than a single moduleset by a single master; so the release team could tell a slave to build a tarball-moduleset to smoketest it.
  • It would also be interesting to be able to attach extra informations to some elements; distribution, architecture, sponsor… to build slaves; bug numbers to failed builds or modules.

But does it run?

In this week commit digest (to be published later today) I note bug 547885 in ORBit2 got fixed (ORBit2 should not call g_thread_init() itself as it is usually too late).

It has consequences for lots of GNOME modules as they will now have to call g_threads_init() explicitely or suffer errors or even a miserable death. I noticed problems on applets (and I of course blamed Vincent at first) and Python programs (even in their configure scripts, as they would import module that would then abort() because of g_threads_init() missing, see for example bugs 564740 and 564726.

2.25.3 took extra time because of removed libgnomeui from libpanelapplet affecting a few modules, please test your modules with ORBit2 from Subversion to avoid problems in 2.25.4. Thanks!

Last Modification: Sunday, December 21, 2008 13:32