Bug#906337: Any clue how to circumvent issues with MPI on autobuilders

On Fri, Sep 14, 2018 at 9:21 AM, Andreas Tille wrote: > Control: tags -1 help > > Hi, > > in the bug log of #906337 the test suite aborts with > > ... > -- > MPI_ABORT was invoked on rank 0 in communicator MPI_COMM_WORLD >

Bug#863829: python2: dolfin 'module' object has no attribute 'cpp'

On Wed, May 31, 2017 at 6:37 PM, Drew Parsons wrote: > Weird, the new python3 module seems to have broken the python2 dolfin > module. That's not good. > > Importing dolfin gives the error: > AttributeError: 'module' object has no attribute 'cpp' The problem is that the

Bug#804465: dolfin: FTBFS: wants ufc 1.5.0

dolfin version 1.6.0 will fix this problem. It is currently sitting in the new queue. Johannes On Sun, Nov 8, 2015 at 8:09 PM Chris West (Faux) < solo-debianb...@goeswhere.com> wrote: > Source: dolfin > Version: 1.5.0-4 > Severity: serious > Justification: fails to build from source > Tags: sid

Bug#787796: fenics fresh installation is broken

On Fri, Jun 5, 2015 at 9:03 AM Massimiliano Leoni leoni.massimilia...@gmail.com wrote: Versions of packages fenics depends on: ii dolfin-bin 1.5.0-2 ii dolfin-doc 1.5.0-2 ii libdolfin-dev 1.5.0-2 ii python-dolfin 1.5.0-2 I can't reproduce this with the latest dolfin

Bug#780359: Options w.r.t. libcgal11 transition

On Wed, May 13, 2015 at 12:19 AM, Jonathan Wiltshire j...@debian.org wrote: Ok, please go ahead now. To upload to unstable from experimental, just add a changelog entry with Upload to unstable and an incremented version. Thank you, I have uploaded dolfin to unstable now. Your only r-dep is

Bug#780359: Options w.r.t. libcgal11 transition

On Mon, May 11, 2015 at 11:35 PM, Joachim Reichel joachim.reic...@gmx.de wrote: Hi, dolfin is now the only reverse dependency holding up the libcgal11 transition. What are the options? (1) Fix dolfin without causing another transition (2) Remove dolfin from testing (3) Upload 1.5.0-1 from

Bug#780359: Options w.r.t. libcgal11 transition

On Tue, May 12, 2015 at 10:07 AM, Johannes Ring joha...@simula.no wrote: On Mon, May 11, 2015 at 11:35 PM, Joachim Reichel joachim.reic...@gmx.de wrote: Hi, dolfin is now the only reverse dependency holding up the libcgal11 transition. What are the options? (1) Fix dolfin without causing

Bug#780359: Options w.r.t. libcgal11 transition

On Tue, May 12, 2015 at 10:29 AM, Jonathan Wiltshire j...@debian.org wrote: On 2015-05-12 09:07, Johannes Ring wrote: On Mon, May 11, 2015 at 11:35 PM, Joachim Reichel joachim.reic...@gmx.de wrote: Hi, dolfin is now the only reverse dependency holding up the libcgal11 transition. What

Bug#739697: src:dolfin: FTBFS on armhf and mips: virtual memory exhausted: Cannot allocate memory

On Mon, Feb 24, 2014 at 12:35 AM, peter green plugw...@p10link.net wrote: peter green wrote: I've been doing some tests on this interestingly it failed for me much earlier than on the buildd, maybe different systems give slightly different ammounts of usable address space or something. Anyway

Bug#679762: Please give back dolfin (hurd-i386 and i386)

Ping? Johannes On Mon, Aug 6, 2012 at 10:28 AM, Johannes Ring joha...@simula.no wrote: Hi, dolfin 1.0.0-7 failed to build on hurd-i386 and i386 due to a bug in gcc-4.7. This bug was fixed in gcc-4.7 4.7.1-6, so please rebuild dolfin on those architectures: gb dolfin_1.0.0-7 . hurd-i386

Bug#679762: Please give back dolfin (hurd-i386 and i386)

Hi, dolfin 1.0.0-7 failed to build on hurd-i386 and i386 due to a bug in gcc-4.7. This bug was fixed in gcc-4.7 4.7.1-6, so please rebuild dolfin on those architectures: gb dolfin_1.0.0-7 . hurd-i386 i386 This would clear the FTBFS bug #679762 (CCed). Thanks, Johannes -- To UNSUBSCRIBE,

Bug#679762: dolfin: FTBFS: Armadillo could not be found. Be sure to set ARMADILLO_DIR. (missing: ARMADILLO_TEST_RUNS)

On Thu, Aug 2, 2012 at 1:30 AM, Kumar Appaiah aku...@debian.org wrote: #680931 has been fixed by the latest gcc download, so I don't believe that the workaround upload for Armadillo is necessary. Thanks Kumar. The latest gcc-4.7 package (version 4.7.1-6) fixed the problem. I guess a binNMU for

Bug#679762: dolfin: FTBFS: Armadillo could not be found. Be sure to set ARMADILLO_DIR. (missing: ARMADILLO_TEST_RUNS)

On Sun, Jul 1, 2012 at 1:09 PM, Christoph Egger christ...@debian.org wrote: Your package failed to build on the buildds: -- Performing Test ARMADILLO_BLAS_TEST_RUNS - Failed -- Performing Test ARMADILLO_GFORTRAN_TEST_RUNS -- Performing Test ARMADILLO_GFORTRAN_TEST_RUNS - Failed CMake Error

Bug#667599: dolfin-dev: fails to upgrade from squeeze - apt does not find an upgrade path

On Sat, Apr 7, 2012 at 3:10 AM, David Kalnischkies kalnischkies+deb...@gmail.com wrote: In bug #667599 we have a report from piuparts that APT is unable to upgrade delfin-dev from squeeze to wheezy. The problem is that APT has to decide in this process if it wants to have libhdf5-1.8 (=

Bug#674014: python-dolfin: Doesn't find python-scientific

Hi Torquil, Thanks for the report. On Tue, May 22, 2012 at 4:02 PM, Torquil Macdonald Sørensen torq...@gmail.com wrote: However, python-scientific is installed, and import Scientific works fine when I run it myself in the python shell. Yes, but does it work if you do import

Bug#667411: Another patch

The same fix should also be applied to Code/Algorithms/itkCurvesLevelSetFunction.h. Johannes --- insighttoolkit-3.20.1.orig/Code/Algorithms/itkCurvesLevelSetFunction.h 2012-05-14 19:55:57.353934793 +0200 +++ insighttoolkit-3.20.1/Code/Algorithms/itkCurvesLevelSetFunction.h 2012-05-14

Bug#672208: python-dolfin: Extension module cpp not found/built

Hi ST, Thanks for the report. On Wed, May 9, 2012 at 8:22 AM, LUK ShunTim shuntim@polyu.edu.hk wrote: It seems that in the current (1.0.0-4+b2) version of python-dolfin, the extension module cpp is somehow not built. Correct. The build log [1] tells us why: CMake Warning at

Bug#651997: python-dolfin: unknown symbol: wrapper_dgesv_

[Adding Kumar (maintainer of Armadillo) in Cc] On Tue, Dec 20, 2011 at 10:16 AM, Julien Cristau julien.cris...@logilab.fr wrote: On Mon, Dec 19, 2011 at 12:46:46 +0100, Johannes Ring wrote: On Mon, Dec 19, 2011 at 10:37 AM, Andreas Kloeckner inf...@tiker.net wrote: This being as it is, could

Bug#651997: python-dolfin: unknown symbol: wrapper_dgesv_

Hi Andreas, Thank you for your report, but I couldn't reproduce your bug. On Tue, Dec 13, 2011 at 11:34 PM, Andreas Kloeckner inf...@tiker.net wrote: When I run any one of the simple fenics demos, I get this error message: ImportError: /usr/lib/libdolfin.so.1.0: undefined symbol:

Bug#651997: python-dolfin: unknown symbol: wrapper_dgesv_

On Mon, Dec 19, 2011 at 10:37 AM, Andreas Kloeckner inf...@tiker.net wrote: Did, and it helped. Thanks very much! Good! This being as it is, could you upload a new package with tightened dependencies? The dependency on libarmadillo2 is added automatically by ${shlibs:Depends}. I guess I

Bug#651997: python-dolfin: unknown symbol: wrapper_dgesv_

On Mon, Dec 19, 2011 at 3:50 PM, Andreas Kloeckner inf...@tiker.net wrote: Hmm, I feel like automatic shared library dependencies should have been able to catch this, i.e. someone is supposed to have done something with the soname at some point--I'm just not sure what... :) Yes, I agree. It is

Bug#642712: dolfin: FTBFS: g++: error: =format-security: No such file or directory

Mònica, Thanks for the report. This issue has been fixed upstream: http://bazaar.launchpad.net/~dolfin-core/dolfin/main/revision/6306 Upstream plans to do a new release soon so I will hold on a few more days before I patch and upload. Johannes -- To UNSUBSCRIBE, email to

Bug#636869: vmtk: FTBFS on all plateforms: No rule to make target `/usr/lib/libpng.so'

Hi! This is no longer a problem with the latest CMake and VTK packages in Debian. What is the correct way to close this bug? Thanks, Johannes -- To UNSUBSCRIBE, email to debian-bugs-rc-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#628243: syfi: FTBFS: syfi/swig/SyFi_wrap.cc:3109:50: error: 'const class GiNaC::basic' has no member named 'tinfo'

Hi Luca, On Sun, Aug 14, 2011 at 10:33 PM, Luca Falavigna dktrkr...@debian.org wrote: I uploaded swig2.0 2.0.4-2.1 which provides the fix in order to let syfi to build again. Could you please look whether it's OK now? Yes, the problem with swig2.0 is resolved with the 2.0.4-2.1 upload. Thank

Bug#631589: python-dolfin: python -c import dolfin fails with MPI error

Hi Luk, thanks for the report and sorry for the late reply. On Sat, Jun 25, 2011 at 7:57 AM, LUK ShunTim shuntim@polyu.edu.hk wrote: $ python -c import dolfin *** The MPI_comm_size() function was called before MPI_INIT was invoked. *** This is disallowed by the MPI standard. *** Your MPI

Bug#631589: python-dolfin: python -c import dolfin fails with MPI error

On Tue, Aug 16, 2011 at 1:00 PM, LUK ShunTim shuntim@polyu.edu.hk wrote: Yes, it does. Good. I will add import PyTrilinos at the top of dolfin/__init__.py to make sure that it will be imported before dolfin. Now an OT bug. :-) I added the import PyTrilinos to demo_poisson.py and got a

Bug#636869: vmtk: FTBFS on all plateforms: No rule to make target `/usr/lib/libpng.so'

Hi Sylvestre, Thanks for your report. On Sat, Aug 6, 2011 at 5:39 PM, Sylvestre Ledru sylves...@debian.org wrote: Your patckage FTNFS on all plateforms: :) [snip] make[3]: *** No rule to make target `/usr/lib/libpng.so', needed by `bin/libvtkvmtkComputationalGeometry.so.0.9.0'.  Stop.

Bug#628243: syfi: FTBFS: syfi/swig/SyFi_wrap.cc:3109:50: error: 'const class GiNaC::basic' has no member named 'tinfo'

Hi, Thanks for the report and sorry for the late reply. This bug has been fixed upstream some time ago [1] and I was going to patch and reupload. However, there is a problem building syfi with swig 2.0.4 [2] which is currently unresolved. The bug could very well be in swig and not in syfi. [1]

Bug#629564: fenics: not installable in sid

Hi Ralf, Thanks for your report. On Tue, Jun 7, 2011 at 8:40 PM, Ralf Treinen trei...@free.fr wrote: fenics is not installable in sid on any architecture. On amd64 (I didn't check the other architectures) this is the case since May 19. Explanation: fenics (= 11.5-1) depends on syfi-doc (=

Bug#621037: scotch: missing inter-library links

Hi Adam, On Wed, Apr 6, 2011 at 1:47 PM, Adam C Powell IV hazel...@debian.org wrote: Just put a fix in alioth, let me know if it gives you trouble.  I plan to upload within 4-5 hours. Sorry, I tried your fix but it gives me trouble because libptscotch is linked against libscotch. This is the

Bug#621037: scotch: missing inter-library links

On Wed, Apr 6, 2011 at 4:29 PM, Adam C Powell IV hazel...@debian.org wrote: Thanks very much Johannes. On Wed, 2011-04-06 at 16:11 +0200, Johannes Ring wrote: On Wed, Apr 6, 2011 at 3:12 PM, Adam C Powell IV hazel...@debian.org wrote: Hi Johannes, On Wed, 2011-04-06 at 14:37 +0200

Bug#621037: scotch: missing inter-library links

On Wed, Apr 6, 2011 at 3:12 PM, Adam C Powell IV hazel...@debian.org wrote: Hi Johannes, On Wed, 2011-04-06 at 14:37 +0200, Johannes Ring wrote: Hi Adam, On Wed, Apr 6, 2011 at 1:47 PM, Adam C Powell IV hazel...@debian.org wrote: Just put a fix in alioth, let me know if it gives you

Bug#619935: Bugs #619935 #612621 : most debian/* files rewritten + patches

Hi Adam and Pierre, On Fri, Apr 1, 2011 at 1:38 PM, Adam C Powell IV hazel...@debian.org wrote: * I noticed that you removed -I/usr/include/lam from the CCS and CCD commands. I don't remember the exact reasons, but those were required for the LAM architectures.

Bug#619935: src:scotch: FTBFS on several arches due to lex issue

Hi Pierre, On Wed, Mar 30, 2011 at 2:24 PM, Pierre Saramito pierre.saram...@imag.fr wrote: Hi Adam and Johannes, From Adam: parser_ll.l:123:31: error: 'yylval' undeclared (first use in this function) From Johannes: I couldn't (for some unknown reason) reproduce this in a pbuilder I

Bug#619935: src:scotch: FTBFS on several arches due to lex issue

Hi Adam, On Mon, Mar 28, 2011 at 3:46 PM, Adam C Powell IV hazel...@debian.org wrote: Package: src:scotch Version: 5.1.11.dfsg-4 Severity: serious Scotch FTBFS on i386, ia64, kfreebsd-i386, mips[el] and sparc: (flex parser_ll.l                    \                                        

Bug#612602: dolfin: FTBFS: error: 'SCOTCH_Dgraph' was not declared in this scope

Jakub, Thanks for your report. This error is because ptscotch.h in libptscotch-dev does not define SCOTCH_PTSCOTCH as it did before. I have filed a bug report for that package (#612606). Thanks again, Johannes On Wed, Feb 9, 2011 at 2:58 PM, Jakub Wilk jw...@debian.org wrote: Source: dolfin

Bug#576865: [Pkg-scicomp-devel] Bug#576865: python-pytrilinos: PyTrilinos modules not linking to correct libraries.

Hi, This bug does not affect this package in Debian since version 9.0.3.dfsg-1 is no longer in Debian. Johannes On Wed, Apr 7, 2010 at 10:05 PM, Nuno Sucena Almeida slug.bugrep...@aeminium.org wrote: Package: python-pytrilinos Version: 9.0.3.dfsg-1 Severity: grave Justification: renders

Bug#571780: [Pkg-scicomp-devel] Bug#571780: python-pydolfin0: Cannot import dolfin from python.

Hi, Thanks for your report. On Sun, Feb 28, 2010 at 12:25 AM, Guillaume Yziquel guillaume.yziq...@citycable.ch wrote: Package: python-pydolfin0 Version: 0.9.7-1 Severity: grave Justification: renders package unusable Recently upgraded to the newest packaging of dolfin. I cannot import

Bug#543097: [Pkg-scicomp-devel] Bug#543097: Bug#543097: dolfin: FTBFS: Please install SWIG version 1.3.36 or recompile UFC with present SWIG.

Hi, UFC and DOLFIN are now built with the same version of SWIG and the bug reported here is no longer an issue. Could you please close this bug? Thanks, Johannes On Mon, Aug 24, 2009 at 1:12 PM, Johannes Ring joha...@simula.no wrote: Hi Lucas, Thanks for the report. The problem is that both

Bug#543097: [Pkg-scicomp-devel] Bug#543097: dolfin: FTBFS: Please install SWIG version 1.3.36 or recompile UFC with present SWIG.

Hi Lucas, Thanks for the report. The problem is that both UFC and DOLFIN has to be built with the same version of SWIG. In this case, UFC has been built with SWIG 1.3.36 while DOLFIN was rebuilt with SWIG 1.3.39. The proper fix in this situation would be to rebuild UFC with the current SWIG and