Robert Schiele wrote:
On Wed, Feb 14, 2007 at 02:10:04PM +0100, Martin Mohring wrote:
@Michael: Why did the server come into the state where the normal "trigger rebuild" does not help?
Also I am not Michael and I don't know what was the exact reason for _this_ case I can explain you a case that can lead to situations like this:
All build rely on a certain set of base backages like the glibc for instance. Without a working glibc version you cannot build any package. If you have an empty project you don't have a glibc in your project but you are still happy because you get this from your "parent" project. Now if you build a glibc package yourself this will be used for future builds. If the glibc you built is broken (but does still build) then you are out of luck because you no longer get the glibc from your parrent project but your own broken version. That way you cannot build _any_ package any more. You even can't rebuild the package that created the broken glibc to fix the issue.
Robert
You explaind exactly the error case I had. My initial _link produced a "glibc replacement", because I was not yet completed by checking in the other files (in this case: the "real" specfile). The only reason for me to have a "clobber & rebuild" was the fact that you can bring the system to a deadlock. There are other way how you can produce such a situation. @Michael: Is is OK for you to add such a "clobber & rebuild" function in addition to "trigger rebuild" or would you change "trigger rebuild" to do more things? Martin --------------------------------------------------------------------- To unsubscribe, e-mail: opensuse-buildservice+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-buildservice+help@opensuse.org