[GRASS-user] v.net.iso problem or bug?

[Maris Nartiss]

Hello,
first - GRASS version, OS, Arch?
second - have tested with GRASS 7?
third - if dataset is small, TGZ it and drop somewhere for testing
(opening an issue in trac would also be fine).

I have dataset that fails to work with salesman in 6.4.1. I had no
time to make tests with 6.5 and 7. Issue is different - it reports
some nodes to be unreachable after extensive v.clean and v.digit
cleanup. Migh be interesting to collect sample data that causes
v.net.* to fail to try isolate root cause(s) of such failures.

Maris.


2011/4/8, Pierluigi De Rosa <pierluigi.derosa at gfosservices.it>:
> Dear grass users
>
> I have a strange problem using v.net.iso
> I'm doing an analysis on a river network in a very easy system ( I have
> one main channel and two tributaries) in order to calculate distances
> from outlet.
> i'm running:
> v.net.iso --overwrite input=streams_net output=myriver type=line
> nlayer=2 ccats=1 costs=100,500,5000
> but i have this error:
> *** glibc detected *** v.net.iso: realloc(): invalid next size:
> 0x09c147b8 ***
> ======= Backtrace: =========
> /lib/libc.so.6(+0x6c501)[0x17d501]
> /lib/libc.so.6(+0x71c6d)[0x182c6d]
>
> and so on
>
> few months ago an other user have the same problem but it seems he
> didn't go to a solution.
> If I change the outlet point sometimes could happens v.net.iso work!
> So sometimes it works sometimes not...
> Is this a bug?
> Thanks
> Pierluigi
>
>
> _______________________________________________
> grass-user mailing list
> grass-user at lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/grass-user
>