Hey James, (I am adding the sparc porters mailing list in Cc:, hopping they can give us some advice). On Tue, Nov 27, 2012 at 12:12:41AM +0100, James Healy wrote: > On 27 November 2012 00:02, Antonio Terceiro <terceiro@debian.org> wrote: > > Note that -4 is not in testing yet due to a problem on sparc, so -5 will > > hit the archive pretty soon. > > regarding the sparc failures, has anyone else noticed this pattern to > the sparc build logs for the ruby1.9.1 source package? > > stadler - 32 cores - always fails > spontini - 2 cores - sometimes passes > schroeder - 2 cores - sometimes passes > lebrun - 2 cores - sometimes passes > > Is it a co-incidence that the failures happen more often on a many-core machine? I was puzzled with the fact that the sparc builds seemed to fail randomly, but didn't noticed that yet. Thanks for sharing! Dear sparc porters: can you hint us on what could be the cause of this? Were there any similar cases in the past that you could refer us to? For the record, right now I am inclined to just disable running the test suite during the build on sparc. -- Antonio Terceiro <terceiro@debian.org>
Attachment:
signature.asc
Description: Digital signature