I'm trying to connect my network to a points layer with a threshold of 4500 meters. I have about 11000 or so entries in my polyline and about 24 in my points layer.

I have already used v.distance, hence the threshold. However, I receive an error (image attached) any time I try the following:

v.net input=network points=cdw1c output=network_v1 operation=connect threshold=4500

Here is the error: enter image description here

Error content is:

Assertion failed: !p->taken[i], file split_q.e, line 69

Can anyone help me to understand what it means?

After a long consulation with a generous user from GRASS community, I still haven't came to a solution. I've tried on two computers with Win 7 as operating system, the following GRASS versions: 6.4.2 (through QGIS and without the plugin), 6.4.3 RCs, 7.0.0

For the first two, I have received the same error; for version 7, I didn't receive the error but the process have never finished. On the other hand, my helper managed to complete it in seconds using my files.

Can the problem have something to do with Windows 7, some background or system process, or so? Should I install any extension or plugin in order to run network analysis programs on Win 7?


2 Answers 2


Well I don't have a verified answer for this. However after few days of hard work, consultation with other GRASS-users via the community mailing lists, and help from the members of this forum, I came to a conclusion that there is a problem running v.net on windows.

I solved it by installing Linux-mint 14 on my computer. I run GRASS-gis using the new OS and it worked like a charm. I attach a link to another question, in which some users here kindly explained to me how to install GRASS on Linux.


For the record: the problem cannot be solved in GRASS 6 for Windows7 (see https://trac.osgeo.org/grass/ticket/1855#comment:3).

But it is working fine in GRASS 7 which will be released this year. To work with it right away, nightly snapshots for Windows are available here: http://wingrass.fsv.cvut.cz/grass70/

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.