crash after compiling sagemanifolds sage 7.3 (ppa aims)

104 views
Skip to first unread message

aishen

unread,
Sep 23, 2016, 5:40:01 AM9/23/16
to sage-devel
New ubuntu 16.04 fresh install. I got it already on last week on an old xenial.
What do I do ? Forgetting about sagemanifolds ?
Or get the sagemath binary which works ?
I would like to keep the deb because it updating automatically
Thanks for your help
Best
Henri
real    41m33.940s
user    40m36.776s
sys    1m0.760s

Installation of SageManifolds 0.9.1 completed!

pi@pc:/usr/lib/sagemath$ sage
┌────────────────────────────────────────────────────────────────────┐
│ SageMath version 7.3, Release Date: 2016-08-04                     │
│ Type "notebook()" for the browser-based notebook interface.        │
│ Type "help()" for help.                                            │
└────────────────────────────────────────────────────────────────────┘

**********************************************************************

Oops, Sage crashed. We do our best to make it stable, but...

A crash report was automatically generated with the following information:
  - A verbatim copy of the crash traceback.
  - A copy of your input history during this session.
  - Data on your current Sage configuration.

It was left in the file named:
    '/home/pi/.sage/ipython_genutils-0.1.0/Sage_crash_report.txt'
If you can email this file to the developers, the information in it will help
them in understanding and correcting the problem.

You can mail it to: sage-support at sage-s...@googlegroups.com
with the subject 'Sage Crash Report'.

If you want to do it now, the following command will work (under Unix):
mail -s 'Sage Crash Report' sage-s...@googlegroups.com < /home/pi/.sage/ipython_genutils-0.1.0/Sage_crash_report.txt

To ensure accurate tracking of this issue, please file a report about it at:
http://trac.sagemath.org

Hit <Enter> to quit (your terminal may close):
pi@pc:/usr/lib/sagemath$

Eric Gourgoulhon

unread,
Sep 23, 2016, 7:17:16 AM9/23/16
to sage-devel
Hi,

Le vendredi 23 septembre 2016 11:40:01 UTC+2, aishen a écrit :

It was left in the file named:
    '/home/pi/.sage/ipython_genutils-0.1.0/Sage_crash_report.txt'
If you can email this file to the developers, the information in it will help
them in understanding and correcting the problem.


Could you please send the above Sage_crash_report.txt file?
Besides, can you tell which version of SageMath you are using? Is it the one from the Ubuntu PPA package?
Thanks.
Eric.

Eric Gourgoulhon

unread,
Sep 23, 2016, 8:41:43 AM9/23/16
to sage-devel
OK I could reproduce the error, with the following set-up:

- Ubuntu 16.04 x86_64
- SageMath 7.3 from AIMS PPA installed as follows:
    sudo -E apt-add-repository -y ppa:aims/sagemath
    sudo -E apt-get update
    sudo -E apt-get install sagemath-upstream-binary-full
- SageManifold 0.9.1 installed as follows
    cd /usr/lib/sagemath
    sudo wget -N http://sagemanifolds.obspm.fr/spkg/sm-install.sh
    sudo bash sm-install.sh

The error is (cf. the attached crash report file)

ImportError: /usr/lib/sagemath/local/lib/python2.7/site-packages/sage/symbolic/pynac.so: undefined symbol: _ZN5GiNaC8constantC1ENSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEEPFNS_2exEjP7_objectERKS6_j

Actually, this error does not seem to be connected to SageManifolds (which does not touch pynac.pyx) but to the inability of this version of SageMath to rebuild itself: it has already been reported in a context different from SageManifolds:
cf. https://groups.google.com/d/msg/sage-support/Qg5rpO43daE/zDH_Y3gvCQAJ
There, it was suggested this might be due to the pretty old version of gcc (4.8.4) used to buid the binary. Do the AIMS PPA maintainers have an opinion on that ?

Sage_crash_report.txt

Eric Gourgoulhon

unread,
Sep 23, 2016, 9:14:05 AM9/23/16
to sage-devel
PS (to aishen): while this problem is fixed, you may consider installing in your home directory SageMath 7.3 from the file sage-7.3-Ubuntu_16.04-x86_64.tar.bz2 downloaded from http://www.sagemath.org/download-linux.html.
Installing SageManifolds in such a version has been tested successfully.


henri....@gmail.com

unread,
Sep 23, 2016, 9:31:07 AM9/23/16
to sage-...@googlegroups.com

Thank you very much for your help. I already use the binary sage7.3, in between, and it works fine with sagemanifolds. Sorry for the disturbance, but I like very much sagemanifolds and not being able to use it is disturbing for me. But now everything is fine and if they find a workaround I will retry it latter.

Best regards

Henri (aishen)


Le 23/09/2016 à 15:14, Eric Gourgoulhon a écrit :
PS (to aishen): while this problem is fixed, you may consider installing in your home directory SageMath 7.3 from the file sage-7.3-Ubuntu_16.04-x86_64.tar.bz2 downloaded from http://www.sagemath.org/download-linux.html.
Installing SageManifolds in such a version has been tested successfully.


--
You received this message because you are subscribed to a topic in the Google Groups "sage-devel" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/sage-devel/z1XQDRt-OFs/unsubscribe.
To unsubscribe from this group and all its topics, send an email to sage-devel+...@googlegroups.com.
To post to this group, send email to sage-...@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

Jan Groenewald

unread,
Sep 23, 2016, 9:46:45 AM9/23/16
to sage-devel
Hi

Thanks, Eric. The PPA for 16.04 is indeed the 14.04 binary in the PPA copied forward. Let me look at what (time) it takes to upload the proper 16.04 binary.

In principle, I can continue to maintain and increase automation of this, but time might become an issue. We have a customized Ubuntu called aims-desktop that we redistribute, and first created the PPA for that. I know it is used much more widely. For redistribution we contacted Canonical's legal department after their IP policy was in the news, and were told to redistribute modified copies we must 1) remove all Ubuntu branding and rebuild *all*  So in future our incentive to spend time on this might reduce.

Of course it is still not recommended to run sage -b as root and even less so in the PPA. The PPA was always viewed as a temporary solution while people are working on properly debianizing sage, a hard and multi-year project, which will no doubt leave Debian/Ubuntu with a outdated sage.

Still we want to support knowledgable people doing this out of necessity while sage-in-debian is not ready.

Regards,
Jan

On 23 September 2016 at 15:14, Eric Gourgoulhon <egourg...@gmail.com> wrote:
PS (to aishen): while this problem is fixed, you may consider installing in your home directory SageMath 7.3 from the file sage-7.3-Ubuntu_16.04-x86_64.tar.bz2 downloaded from http://www.sagemath.org/download-linux.html.
Installing SageManifolds in such a version has been tested successfully.


--
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+unsubscribe@googlegroups.com.

To post to this group, send email to sage-...@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.



--
  .~.
  /V\     Jan Groenewald
 /( )\    www.aims.ac.za
 ^^-^^ 

Jan Groenewald

unread,
Sep 23, 2016, 9:49:07 AM9/23/16
to sage-devel
Hi

On 23 September 2016 at 15:46, Jan Groenewald <j...@aims.ac.za> wrote:
were told to redistribute modified copies we must 1) remove all Ubuntu branding and rebuild *all*


.. binaries, or 2) sign a license to redistribute to African universities only (how gracious) and they would then not be able to redistribute freely.

(Sorry, accidentally cut some text).

Regards,
Jan
 

Eric Gourgoulhon

unread,
Sep 23, 2016, 10:32:58 AM9/23/16
to sage-devel
Hi Jan,

Thanks for your prompt and detailed answer, and for your efforts in offering this PPA to the community.

Best regards,

Eric.

Jan Groenewald

unread,
Sep 23, 2016, 11:11:34 AM9/23/16
to sage-devel
Hi

If everything debian-builds OK, there will be a package here in a few hours in the dev PPA:
https://launchpad.net/~aims/+archive/ubuntu/sagemath-dev/+packages

which take s about an hour. If successful, publishing that in the PPA could take anything
from 5 minute to 5 hours, depending on how busy launchpad is.

Please send me feedback before I copy to the stable PPA. I can't test that this even opens the notebook now.

Regards,
Jan

--
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+unsubscribe@googlegroups.com.
To post to this group, send email to sage-...@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

henri....@gmail.com

unread,
Sep 23, 2016, 12:42:23 PM9/23/16
to sage-...@googlegroups.com

I just had a look apparently it failed ?

You received this message because you are subscribed to a topic in the Google Groups "sage-devel" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/sage-devel/z1XQDRt-OFs/unsubscribe.
To unsubscribe from this group and all its topics, send an email to sage-devel+...@googlegroups.com.

Jan Groenewald

unread,
Sep 23, 2016, 2:19:54 PM9/23/16
to sage-devel
Thanks, that was the second time. Third time also failed at 10 minutes without buildlog.
launchpad does that sometimes but ususally 1 (or 2)  retries are enough. 4th retry is now showing a buildlog, and busy building, and is at 33 minutes, give it 30 more or so.

Regards,
Jan

To unsubscribe from this group and all its topics, send an email to sage-devel+unsubscribe@googlegroups.com.

To post to this group, send email to sage-...@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

--
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+unsubscribe@googlegroups.com.
To post to this group, send email to sage-...@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

henri....@gmail.com

unread,
Sep 23, 2016, 2:33:54 PM9/23/16
to sage-...@googlegroups.com

Ok I wait a little and try again

To unsubscribe from this group and all its topics, send an email to sage-devel+...@googlegroups.com.

Jan Groenewald

unread,
Sep 23, 2016, 3:08:11 PM9/23/16
to sage-devel

henri....@gmail.com

unread,
Sep 23, 2016, 7:26:37 PM9/23/16
to sage-...@googlegroups.com

OK  works fine :

Writing /usr/lib/sagemath/local/lib/python2.7/site-packages/sage-7.3-py2.7.egg-info

real    0m7.385s
user    0m4.820s
sys    0m1.528s



Installation of SageManifolds 0.9.1 completed!

pi@pc:/usr/lib/sagemath$ ls -al
total 932
drwxr-xr-x   8 root root   4096 sept. 24 01:21 .


Le 23/09/2016 à 20:19, Jan Groenewald a écrit :
To unsubscribe from this group and all its topics, send an email to sage-devel+...@googlegroups.com.

Eric Gourgoulhon

unread,
Sep 24, 2016, 5:06:24 AM9/24/16
to sage-devel
Hi,

I confirm that the new package sagemath-upstream-binary-full, taken from the PPA aims/sagemath-dev works well: SageManifolds could be installed in it without any trouble and its test suite is passed.

Thanks a lot Jan !

Best regards,

Eric.

Jan Groenewald

unread,
Sep 24, 2016, 5:17:06 AM9/24/16
to sage-devel
Hi

OK, requesting copy to stable PPA on launchpad.

Regards,
Jan

--
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+unsubscribe@googlegroups.com.
To post to this group, send email to sage-...@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

henri....@gmail.com

unread,
Sep 24, 2016, 5:27:20 AM9/24/16
to sage-...@googlegroups.com

Wonderfull :)

Thanks a lot Jan and Eric

Best regards

Henri

You received this message because you are subscribed to a topic in the Google Groups "sage-devel" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/sage-devel/z1XQDRt-OFs/unsubscribe.
To unsubscribe from this group and all its topics, send an email to sage-devel+...@googlegroups.com.

Henri Girard

unread,
Apr 28, 2017, 4:14:32 AM4/28/17
to sage-...@googlegroups.com
Hi
sage-8 beta 4 compiled well on ubuntu xenial 16G RAM x8 core AMD
Thanks
Henri
Reply all
Reply to author
Forward
0 new messages