Bug#1143794: [Debichem-devel] Bug#1143794: Bug#1143794: cp2k: FTBFS: The Open MPI wrapper compiler was unable to find the specified compiler gfortran-15 in your PATH.

On Sun, Aug 23, 2026 at 06:57:19PM +0200, Michael Banck wrote:

> Yeah, that's what I've been seeing all weekend as well trying to build
> the new upstream releases 2026.1 and 2026.2 in experimental.
> 
> I now tried with the old unstable version and hit it there as well
> (after it went through the configure stage fine).
> 
> Interesting that it only started to fail yesterday, I'll try with an
> earlier openmpi version then.

Note: yesterday is only when I noticed that it failed again, not
necessarily when it actually started to fail (so I expressed it
poorly, sorry). This is my build history:

Status: successful  cp2k_2025.2-3_amd64-20260810T155131.511Z
Status: successful  cp2k_2025.2-3_amd64-20260810T155219.633Z
Status: failed      cp2k_2025.2-3_amd64-20260822T111909.499Z
Status: failed      cp2k_2025.2-3_amd64-20260822T145012.811Z
Status: failed      cp2k_2025.2-3_amd64-20260823T154806.629Z
Status: failed      cp2k_2025.2-3_amd64-20260823T154806.528Z
Status: failed      cp2k_2025.2-3_amd64-20260823T154903.834Z
Status: failed      cp2k_2025.2-3_amd64-20260823T154905.454Z

(No history before that because I usually remove old failed logs after
a successful build).

In theory, it would be possible to use debbisect to determine the
exact point where it started to fail, but I did not try because the
package has a long build time.

Thanks.