[SMT-LIB] Past SMT-COMP results on starexec

[Aaron Stump]

Hi, Alberto.

On Thu, Mar 30, 2017 at 1:30 AM, Alberto Griggio <griggio at fbk.eu> wrote:

> Hello Aaron,
>
> > I am sorry for this problem, particularly if it is due to a bug in
> > StarExec.  If someone could send me a space id or job id (or id for
> > solver or benchmark) that is supposed to be there but is not -- and of
> > course, if someone accidentally deleted it that would explain the
> > problem -- I will investigate and let you know.
> >
> > Best wishes, and thanks for your patience while we try to see what is
> > going on.  If it is a StarExec bug, I would say it is much more likely
> > that it is an interface issue rather than actual deleted data, just FYI.
>
> Thanks! I think all SMT-COMP jobs are missing (I have sampled randomly,
> didn't try any single one of them...). I thought they were supposed to
> be there indefinitely, was I wrong?
>

StarExec will preserve all jobs indefinitely -- unless someone deletes them!

So the question is if StarExec has a bug and is not displaying or even has
incorrectly deleted a job (the latter seems unlikely to me), or if the
owner of such jobs has accidentally removed them.

This is why it would be very helpful to have the actual StarExec id of a
specific job or solver or something that is supposed to be there but
appears to be missing.  Then we can try to understand if it is a StarExec
issue or user issue.

Another question: do you have any guess about when these jobs seem to have
disappeared?  We do keep backups of StarExec, and so could attempt to
resurrect deleted primitives (solvers/benchmarks) certainly, if we know how
far back to look -- and again, what to look for specifically.

Cheers,
Aaron


> Best,
> Alberto
>