[sage-devel] Re: make ptestlong on Big Sur
For the record, I was able to get make ptestlong it to finish by doing: ulimit -n 4096 On Friday, January 8, 2021 at 3:30:19 PM UTC-5 Zachary Scherr wrote: > Interesting. I've noticed the segfaults in the past, but never the "too > many open files" error. I'll try to play around with my ulimit and see if > I can get the testing to finish. > > On Friday, January 8, 2021 at 1:21:20 PM UTC-5 Matthias Koeppe wrote: > >> I think this has started to show up, at least sporadically, with the >> multiprocessing changes made in/for python 3.8, see >> https://trac.sagemath.org/ticket/27754 >> >> Something definitely needs fixing. >> >> On Friday, January 8, 2021 at 9:55:06 AM UTC-8 zsc...@gmail.com wrote: >> >>> Has anyone tried to run "make -j8 ptestlong" on Big Sur? Every time I >>> run it, doctesting breaks with the following traceback: >>> >>> Traceback (most recent call last): >>> File "/Users/zscherr/sage/develop/src/bin/sage-runtests", line 182, in >>> >>> err = DC.run() >>> File >>> "/Users/zscherr/sage/develop/local/lib/python3.9/site-packages/sage/doctest/control.py", >>> >>> line 1236, in run >>> self.run_doctests() >>> File >>> "/Users/zscherr/sage/develop/local/lib/python3.9/site-packages/sage/doctest/control.py", >>> >>> line 937, in run_doctests >>> self.dispatcher.dispatch() >>> File >>> "/Users/zscherr/sage/develop/local/lib/python3.9/site-packages/sage/doctest/forker.py", >>> >>> line 2040, in dispatch >>> self.parallel_dispatch() >>> File >>> "/Users/zscherr/sage/develop/local/lib/python3.9/site-packages/sage/doctest/forker.py", >>> >>> line 1935, in parallel_dispatch >>> w.start() # This might take some time >>> File >>> "/Users/zscherr/sage/develop/local/lib/python3.9/site-packages/sage/doctest/forker.py", >>> >>> line 2207, in start >>> super(DocTestWorker, self).start() >>> File >>> "/usr/local/Cellar/python@3.9/3.9.1_4/Frameworks/Python.framework/Versions/3.9/lib/python3.9/multiprocessing/process.py", >>> >>> line 121, in start >>> self._popen = self._Popen(self) >>> File >>> "/usr/local/Cellar/python@3.9/3.9.1_4/Frameworks/Python.framework/Versions/3.9/lib/python3.9/multiprocessing/context.py", >>> >>> line 224, in _Popen >>> return _default_context.get_context().Process._Popen(process_obj) >>> File >>> "/usr/local/Cellar/python@3.9/3.9.1_4/Frameworks/Python.framework/Versions/3.9/lib/python3.9/multiprocessing/context.py", >>> >>> line 277, in _Popen >>> return Popen(process_obj) >>> File >>> "/usr/local/Cellar/python@3.9/3.9.1_4/Frameworks/Python.framework/Versions/3.9/lib/python3.9/multiprocessing/popen_fork.py", >>> >>> line 19, in __init__ >>> self._launch(process_obj) >>> File >>> "/usr/local/Cellar/python@3.9/3.9.1_4/Frameworks/Python.framework/Versions/3.9/lib/python3.9/multiprocessing/popen_fork.py", >>> >>> line 65, in _launch >>> child_r, parent_w = os.pipe() >>> OSError: [Errno 24] Too many open files >>> make: *** [ptestlong] Error 1 >>> >>> Since I was trying to test in parallel I don't know exactly what's going >>> on, but the last test that was attempted before this error was >>> >>> sage -t --long --warn-long 53.1 --random-seed=0 src/sage/doctest/test.py >>> >>> by itself this test seems to pass but causes a lot of random segfaults, >>> although this is with homebrew's python 3.9.1. >>> >> -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/sage-devel/5d282261-a374-4c21-8101-98873866da0dn%40googlegroups.com.
[sage-devel] Re: make ptestlong on Big Sur
Interesting. I've noticed the segfaults in the past, but never the "too many open files" error. I'll try to play around with my ulimit and see if I can get the testing to finish. On Friday, January 8, 2021 at 1:21:20 PM UTC-5 Matthias Koeppe wrote: > I think this has started to show up, at least sporadically, with the > multiprocessing changes made in/for python 3.8, see > https://trac.sagemath.org/ticket/27754 > > Something definitely needs fixing. > > On Friday, January 8, 2021 at 9:55:06 AM UTC-8 zsc...@gmail.com wrote: > >> Has anyone tried to run "make -j8 ptestlong" on Big Sur? Every time I run >> it, doctesting breaks with the following traceback: >> >> Traceback (most recent call last): >> File "/Users/zscherr/sage/develop/src/bin/sage-runtests", line 182, in >> >> err = DC.run() >> File >> "/Users/zscherr/sage/develop/local/lib/python3.9/site-packages/sage/doctest/control.py", >> >> line 1236, in run >> self.run_doctests() >> File >> "/Users/zscherr/sage/develop/local/lib/python3.9/site-packages/sage/doctest/control.py", >> >> line 937, in run_doctests >> self.dispatcher.dispatch() >> File >> "/Users/zscherr/sage/develop/local/lib/python3.9/site-packages/sage/doctest/forker.py", >> >> line 2040, in dispatch >> self.parallel_dispatch() >> File >> "/Users/zscherr/sage/develop/local/lib/python3.9/site-packages/sage/doctest/forker.py", >> >> line 1935, in parallel_dispatch >> w.start() # This might take some time >> File >> "/Users/zscherr/sage/develop/local/lib/python3.9/site-packages/sage/doctest/forker.py", >> >> line 2207, in start >> super(DocTestWorker, self).start() >> File >> "/usr/local/Cellar/python@3.9/3.9.1_4/Frameworks/Python.framework/Versions/3.9/lib/python3.9/multiprocessing/process.py", >> >> line 121, in start >> self._popen = self._Popen(self) >> File >> "/usr/local/Cellar/python@3.9/3.9.1_4/Frameworks/Python.framework/Versions/3.9/lib/python3.9/multiprocessing/context.py", >> >> line 224, in _Popen >> return _default_context.get_context().Process._Popen(process_obj) >> File >> "/usr/local/Cellar/python@3.9/3.9.1_4/Frameworks/Python.framework/Versions/3.9/lib/python3.9/multiprocessing/context.py", >> >> line 277, in _Popen >> return Popen(process_obj) >> File >> "/usr/local/Cellar/python@3.9/3.9.1_4/Frameworks/Python.framework/Versions/3.9/lib/python3.9/multiprocessing/popen_fork.py", >> >> line 19, in __init__ >> self._launch(process_obj) >> File >> "/usr/local/Cellar/python@3.9/3.9.1_4/Frameworks/Python.framework/Versions/3.9/lib/python3.9/multiprocessing/popen_fork.py", >> >> line 65, in _launch >> child_r, parent_w = os.pipe() >> OSError: [Errno 24] Too many open files >> make: *** [ptestlong] Error 1 >> >> Since I was trying to test in parallel I don't know exactly what's going >> on, but the last test that was attempted before this error was >> >> sage -t --long --warn-long 53.1 --random-seed=0 src/sage/doctest/test.py >> >> by itself this test seems to pass but causes a lot of random segfaults, >> although this is with homebrew's python 3.9.1. >> > -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/sage-devel/048d4535-f0aa-4b72-96b4-f12166a0f364n%40googlegroups.com.
[sage-devel] Re: make ptestlong on Big Sur
I think this has started to show up, at least sporadically, with the multiprocessing changes made in/for python 3.8, see https://trac.sagemath.org/ticket/27754 Something definitely needs fixing. On Friday, January 8, 2021 at 9:55:06 AM UTC-8 zsc...@gmail.com wrote: > Has anyone tried to run "make -j8 ptestlong" on Big Sur? Every time I run > it, doctesting breaks with the following traceback: > > Traceback (most recent call last): > File "/Users/zscherr/sage/develop/src/bin/sage-runtests", line 182, in > > err = DC.run() > File > "/Users/zscherr/sage/develop/local/lib/python3.9/site-packages/sage/doctest/control.py", > > line 1236, in run > self.run_doctests() > File > "/Users/zscherr/sage/develop/local/lib/python3.9/site-packages/sage/doctest/control.py", > > line 937, in run_doctests > self.dispatcher.dispatch() > File > "/Users/zscherr/sage/develop/local/lib/python3.9/site-packages/sage/doctest/forker.py", > > line 2040, in dispatch > self.parallel_dispatch() > File > "/Users/zscherr/sage/develop/local/lib/python3.9/site-packages/sage/doctest/forker.py", > > line 1935, in parallel_dispatch > w.start() # This might take some time > File > "/Users/zscherr/sage/develop/local/lib/python3.9/site-packages/sage/doctest/forker.py", > > line 2207, in start > super(DocTestWorker, self).start() > File > "/usr/local/Cellar/python@3.9/3.9.1_4/Frameworks/Python.framework/Versions/3.9/lib/python3.9/multiprocessing/process.py", > > line 121, in start > self._popen = self._Popen(self) > File > "/usr/local/Cellar/python@3.9/3.9.1_4/Frameworks/Python.framework/Versions/3.9/lib/python3.9/multiprocessing/context.py", > > line 224, in _Popen > return _default_context.get_context().Process._Popen(process_obj) > File > "/usr/local/Cellar/python@3.9/3.9.1_4/Frameworks/Python.framework/Versions/3.9/lib/python3.9/multiprocessing/context.py", > > line 277, in _Popen > return Popen(process_obj) > File > "/usr/local/Cellar/python@3.9/3.9.1_4/Frameworks/Python.framework/Versions/3.9/lib/python3.9/multiprocessing/popen_fork.py", > > line 19, in __init__ > self._launch(process_obj) > File > "/usr/local/Cellar/python@3.9/3.9.1_4/Frameworks/Python.framework/Versions/3.9/lib/python3.9/multiprocessing/popen_fork.py", > > line 65, in _launch > child_r, parent_w = os.pipe() > OSError: [Errno 24] Too many open files > make: *** [ptestlong] Error 1 > > Since I was trying to test in parallel I don't know exactly what's going > on, but the last test that was attempted before this error was > > sage -t --long --warn-long 53.1 --random-seed=0 src/sage/doctest/test.py > > by itself this test seems to pass but causes a lot of random segfaults, > although this is with homebrew's python 3.9.1. > -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/sage-devel/57125c66-070f-4170-bee9-5316ee6c1325n%40googlegroups.com.
[sage-devel] Re: make ptestlong
On Thursday, April 7, 2016 at 5:03:44 PM UTC+2, kcrisman wrote: > > Yes, though again this is a loaner so very little was preinstalled. But > where is Sage invoking Java (as a command) during non-optional tests in the > first place? Running doctests shouldn't be using jmol - or do some of them? > Java is used to pre-render a jmol scene preview, if it is available. There are doctests for that, and they will trigger the java popup on OSX. If we had a reliable way to distinguish the Apple popup from the real "java" then we could prevent it from happening... -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+unsubscr...@googlegroups.com. To post to this group, send email to sage-devel@googlegroups.com. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
[sage-devel] Re: make ptestlong
> Its not that "make ptestlong" does anything special to build Sage, it just > launches our own (parallel) doctester after building the dependencies which > happen to include Sage. Makefile parallelism is some version of > > MAKE='make -j8' make > > You can't really enable makefile parallelism from within the makefile, it > really needs to be specified when invoking make. > > I have an alias in my own computer but this was a loaner. I didn't know it was impossible to invoke any other way - I suppose we could read it off the machine and put that in the script, but maybe that's not a good idea to mess with the Makefile. > > > > The Java thing is really annoying on OSX; Apple, in their infinite wisdom, > included a "java" binary that just pops up the nag window. So whenever a > script tries to launch java you get this popup. Sage is fine with java not > working (its only used for optional things), but we can't un-popup that nag > screen. The easiest fix is to just install java, of course... > > >> Yes, though again this is a loaner so very little was preinstalled. But where is Sage invoking Java (as a command) during non-optional tests in the first place? Running doctests shouldn't be using jmol - or do some of them? I could imagine a new user being confused, anyway. -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+unsubscr...@googlegroups.com. To post to this group, send email to sage-devel@googlegroups.com. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
[sage-devel] Re: make ptestlong
Its not that "make ptestlong" does anything special to build Sage, it just launches our own (parallel) doctester after building the dependencies which happen to include Sage. Makefile parallelism is some version of MAKE='make -j8' make You can't really enable makefile parallelism from within the makefile, it really needs to be specified when invoking make. The Java thing is really annoying on OSX; Apple, in their infinite wisdom, included a "java" binary that just pops up the nag window. So whenever a script tries to launch java you get this popup. Sage is fine with java not working (its only used for optional things), but we can't un-popup that nag screen. The easiest fix is to just install java, of course... On Thursday, April 7, 2016 at 3:28:08 PM UTC+2, kcrisman wrote: > > 1) I was under the (mis)conception that make ptest would also > (automatically) build in parallel. Apparently this is not the case (only > *tests* in parallel), and when you are building gcc too and not in > parallel, the build gives new meaning to 'long'. Should this be > considered a bug? The developer guide/installation guide, I think, don't > promise this, but it would seem to be useful to have this be the default > nowadays, with the same logic as the testing code. > 2) I get some message about "to use the Java command, install JDK" on a > loaner Macbook Air with El Capitan. I didn't think we required java as a > strong dependency (as opposed to an optional dependency) for testing > (notwithstanding that for 3d graphing from command line it's still needed, > I guess?). It claims to only be using optional=mpir,python2,sage. > Thanks! > - kcrisman > -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+unsubscr...@googlegroups.com. To post to this group, send email to sage-devel@googlegroups.com. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
Re: [sage-devel] Re: make ptestlong and binary-pkg
On 2016-03-07 10:44, Dima Pasechnik wrote: > does > $ sage -gp > work on this binary? ? 1+1 %1 = 2 Yes, the basic command above works. But maybe I would have to call something more advanced. -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+unsubscr...@googlegroups.com. To post to this group, send email to sage-devel@googlegroups.com. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
[sage-devel] Re: make ptestlong and binary-pkg
does $ sage -gp work on this binary? (googling " *** Warning: can't expand ~." will show an old sage-devel post and a closed PARI/GP related ticket...) On Monday, March 7, 2016 at 8:36:08 AM UTC, Daniel Krenn wrote: > > I've used https://github.com/sagemath/binary-pkg to create a package > sage-7.1.beta6-Ubuntu_14.04-x86_64.tar.bz2 > Then I've extracted it to some location; relocating (once) seem to have > worked, but with the following issues (with make ptestlong): > > 1a) Before running the tests, all the cythonized files are compiled again. > > 1b) Similarly, the full documentation seems to be rebuilt. > > 2) Some of the doctests fail, for example: > > File "src/sage/lfunctions/lcalc.py", line 376, in > sage.lfunctions.lcalc.LCalc.analytic_rank > Failed example: > lcalc.analytic_rank(E) > Expected: > 1 > Got: > *** Warning: can't expand ~. > *** Warning: can't expand ~. > 1 > > See below for all failures. > > 3) At the end of the building of the package, I saw > CRITICAL:root:todo: strip > (No idea if this should be there or not). > > Because of 1 (takes a long time) and 2 (failing doctests) the created > package is not useable for development. > > Best wishes > > Daniel > > > Failing files: > > sage -t --long src/sage/schemes/elliptic_curves/ell_rational_field.py # > 2 doctests failed > sage -t --long src/sage/plot/plot.py # 1 doctest failed > sage -t --long src/sage/doctest/test.py # 17 doctests failed > sage -t --long src/sage/dev/sagedev.py # 2 doctests failed > sage -t --long src/sage/tests/cmdline.py # 16 doctests failed > sage -t --long src/sage/libs/ppl.pyx # 8 doctests failed > sage -t --long src/sage/plot/line.py # 1 doctest failed > sage -t --long src/sage/lfunctions/zero_sums.pyx # 3 doctests failed > sage -t --long > src/doc/en/thematic_tutorials/explicit_methods_in_number_theory/elliptic_curves.rst > > > # 1 doctest failed > sage -t --long src/sage/schemes/elliptic_curves/lseries_ell.py # 6 > doctests failed > sage -t --long src/sage/repl/interpreter.py # 1 doctest failed > sage -t --long src/sage/misc/temporary_file.py # 1 doctest failed > sage -t --long src/sage/lfunctions/lcalc.py # 3 doctests failed > -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+unsubscr...@googlegroups.com. To post to this group, send email to sage-devel@googlegroups.com. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
