Hi,
Ho hum, versioneer should document its requirements and validate them;
it is after all very easy to check a version string and produce a
well formed error message - even us scientific programmers can do that ;[)
In the bigger picture this is just par for public clusters using redhat.
The ruby cluster didn't come online until April 2015.
Of course, we have other gits available as non defaults:
with the next oldest one, git/1.9.4, parmed is built.
scott
On Wed, Dec 05, 2018 at 10:27:25PM -0500, Jason Swails wrote:
> > On Dec 5, 2018, at 4:22 PM, Scott Brozell <sbrozell.rci.rutgers.edu> wrote:
> > git version 1.7.1
>
> 2010 called and they want their git back. :)
>
> In seriousness it looks like the git version is too old. This doesn???t matter for users since versioneer only tries to pull git information to build the version string when inside a git repository.
>
> As long as the Jenkins CI and the GitHub CI that is run including ParmEd regularly continues to work reliably I???m not overly concerned about this.
>
> It is probably worth an issue so I can get to it at some point, but it won???t be a priority.
>
> Thanks,
> Jason
>
> >
> > Note that make errors from parmed are being ignored:
> >>>> make[2]: *** [parmed] Error 1
> >
> > This is just a build of master on a local machine.
> > "The continuous integration is failing!" was intended to be a joke
> > on chicken little with allusions to current events, but humor is
> > harder than MD apparently:
> > What came first the chicken or the egg - on my face
> > :)
> >
> > scott
> >
> >> On Wed, Dec 05, 2018 at 03:29:52PM -0500, Jason Swails wrote:
> >> The continuous integration at https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fjenkins.jasonswails.com&data=02%7C01%7Csbrozell%40iqb.rutgers.edu%7C821b8f97f9634bbd793408d65b2ac686%7Cb92d2b234d35447093ff69aca6632ffe%7C1%7C0%7C636796636623801868&sdata=473goMUVPs1X%2BtyJGWDO6IQ6qPAzlP7ZuCPYrGdo25k%3D&reserved=0 is working
> >> correctly. Where does this come from?
> >>
> >> It looks like versioneer (a 3rd party module for semantic versioning) is
> >> not using git rev-list the correct way. But this should be failing
> >> everywhere... unless git changed how rev-list works.
> >>
> >> What version of git is being used for this build?
> >>
> >> Thanks,
> >> Jason
> >>
> >> On Wed, Dec 5, 2018 at 3:12 PM Scott Brozell <sbrozell.rci.rutgers.edu>
> >> wrote:
> >>
> >>> Hi,
> >>>
> >>> Maybe no one got the joke since it seems that it was a stretch to expect
> >>> Ræv Skrævto be entered into a
> >>> search bar and get to
> >>> https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fen.wikipedia.org%2Fwiki%2FChicken_Little&data=02%7C01%7Csbrozell%40iqb.rutgers.edu%7C821b8f97f9634bbd793408d65b2ac686%7Cb92d2b234d35447093ff69aca6632ffe%7C1%7C0%7C636796636623801868&sdata=Bs5R7tm0gX7lJk3aI%2FRQ4amZIVQn5xUNI6uhgowU%2B0Q%3D&reserved=0
> >>> :-O
> >>>
> >>> But this bug report was intended to show that parmed was not being built.
> >>> One detail i omitted from the report was
> >>> Red Hat Enterprise Linux Server release 6.10 (Santiago)
> >>>
> >>> This happens before the Traceback and might also be useful:
> >>> ...
> >>> make[2]: Entering directory `/tmp/master/AmberTools/src'
> >>> (cd parmed && PYTHONPATH=/tmp/master/lib/python2.7/site-packages
> >>> /tmp/master/bin/amber.python setup.py install -f --prefix=/tmp/master)
> >>> usage: git rev-list [OPTION] <commit-id>... [ -- paths... ]
> >>> ...
> >>>
> >>> Everything else seems to build.
> >>>
> >>> scott
> >>>
> >>>> On Wed, Dec 05, 2018 at 12:32:19PM -0500, Prof. Ræv Skræv wrote:
> >>>> Hi,
> >>>>
> >>>> The continuous integration is failing!
> >>>>
> >>>> Traceback (most recent call last):
> >>>> File "setup.py", line 160, in <module>
> >>>> version=versioneer.get_version(),
> >>>> File "/tmp/master/AmberTools/src/parmed/versioneer.py", line 1427, in
> >>> get_version
> >>>> return get_versions()["version"]
> >>>> File "/tmp/master/AmberTools/src/parmed/versioneer.py", line 1401, in
> >>> get_versions
> >>>> pieces = from_vcs_f(cfg.tag_prefix, root, verbose)
> >>>> File "/tmp/master/AmberTools/src/parmed/versioneer.py", line 1068, in
> >>> git_pieces_from_vcs
> >>>> pieces["distance"] = int(count_out) # total number of commits
> >>>> TypeError: int() argument must be a string or a number, not 'NoneType'
> >>>> make[2]: *** [parmed] Error 1
> >>>> make[2]: Leaving directory `/tmp/master/AmberTools/src'
> >>>> make[1]: *** [serial] Error 2
> >>>>
> >>>> The continuous integration is failing!
> >>>>
> >>>> Linux ruby 2.6.32-754.3.5.el6.x86_64 #1 SMP Thu Aug 9 11:56:22 EDT 2018
> >>> x86_64 x86_64 x86_64 GNU/Linux
> >>>> ...
> >>>> # On branch master
> >>>> commit 7b2f6baae587931afd4a4f0ba39d90c5f1e4ed45
> >>>> ...
> >>>> ./configure -mkl intel
> >>>> ...
> >>>> Should I download and install Miniconda for you? [y/n] y
> >>>> Downloading an Amber-compatible Python version 2.7 via Miniconda
> >>>> ...
> >>>> icc version 18.0.3 (gcc version 7.3.0 compatibility)
> >>>>
> >>>> The continuous integration is failing!
> >>>>
> >>>>
> >>>> In other news, the sky is falling!
> >>>>
> >>> https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.skyandtelescope.com%2Fastronomy-news%2Fcomet-46p-wirtanen-approaches-earth%2F&data=02%7C01%7Csbrozell%40iqb.rutgers.edu%7C821b8f97f9634bbd793408d65b2ac686%7Cb92d2b234d35447093ff69aca6632ffe%7C1%7C0%7C636796636623801868&sdata=TsBLU9JdxApaE3i6fJpqG59uJ2j37PpLxQkngTgb6VE%3D&reserved=0
> >>>>
_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers
Received on Thu Dec 06 2018 - 08:00:02 PST
