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

[Annalisa Minelli]

Ok,
with the following command:
v.net.iso input=prova_punto4 output=prova_punto_err type=line,boundary
alayer=1 nlayer=2 ccats=1-100000 costs=1000,5000,10000,20000

I obtain the "same" error.. and the last few lines are:
0026f000-00270000 rw-p 0004d000 08:01 137602
/usr/local/grass-6.5.svn/lib/libgrass_gis.6.5.svn.so
00270000-00271000 rw-p 00000000 00:00 0
00271000-0035a000 r-xp 00000000 08:01 2363060
/usr/lib/libstdc++.so.6.0.13
0035a000-0035b000 ---p 000e9000 08:01 2363060
/usr/lib/libstdc++.so.6.0.13
0035b000-0035f000 r--p 000e9000 08:01 2363060
/usr/lib/libstdc++.so.6.0.13
0035f000-00360000 rw-p 000ed000 08:01 2363060
/usr/lib/libstdc++.so.6.0.13
00360000-00367000 rw-p 00000000 00:00 0
00367000-00394000 r-xp 00000000 08:01 2362767
/usr/lib/libgssapi_krb5.so.2.2
00394000-00395000 r--p 0002d000 08:01 2362767
/usr/lib/libgssapi_krb5.so.2.2Aborted

the lines of debug just above the error are:
D3/5: Line 2 : length = 56.568542
D3/5: Arc centres: 0 0 (nodes: 3 4)
D3/5: Vect_net_get_node_cost(): node = 3
D3/5:   -> cost = 0.000000
D3/5: Vect_net_get_node_cost(): node = 4
D3/5:   -> cost = 0.000000
D5/5: Vect_net_get_line_cost(): line = 2, dir = 1
D5/5: Vect_net_get_line_cost(): line = 2, dir = 2
D5/5: Vect_net_get_line_cost(): edge_bcosts = 56.568542
D3/5:   s1cost = 13900.749000 n1cost = 0.000000 e1cost = 56.568542
D3/5:   s2cost = 13957.317000 n2cost = 0.000000 e2cost = 56.568542
D3/5:   -> arc is reachable
D3/5:   npnts1 = 1
D3/5:   npnts2 = 1
D3/5:   pnt1 = 0 dist1 = 0.000000 iso1 = 2 max iso2 = 2
D3/5:   npnts1 cut = 1
D3/5:   pnt2 = 0 dist2 = 56.568542 iso2 = 2 max iso1 = 2
D3/5:   npnts2 cut = 1
D3/5:   npnts1 2. cut = 1
D3/5:   npnts2 2. cut = 1
*** glibc detected *** v.net.iso: realloc(): invalid next size: 0x08df1450
***

moreover I ran more than one time the command and the last row of the
error-output (first attached) seems to refer to SOME different libraries
(not even the same)..
I'm confused.

Many thanks, Annalisa

2010/6/23 Martin Landa <landa.martin at gmail.com>

> Hi,
>
> 2010/6/23 Annalisa Minelli <annagrass6 at gmail.com>:
> > I'm testing v.net.iso on some river networks to calculate the "distance
> > zones" along the river from a date point (cross section), and *sometimes*
> I
> > found this strange error (in more versions of GRASS: I updated software
> > yesterday, compiled and tested on 64, 65 and 70.. with the same results):
> >
> > Number of centres: 1 (nlayer 2)
> > Calculating costs from centre 1...
> >  100%
> > Generating isolines...
> > *** glibc detected *** v.net.iso: realloc(): invalid next size:
> 0x085ecfd8
> > ***
> > ======= Backtrace: =========
> > /lib/tls/i686/cmov/libc.so.6(+0x6b591)[0x3ea591]
> > /lib/tls/i686/cmov/libc.so.6(+0x70cbd)[0x3efcbd]
> > /lib/tls/i686/cmov/libc.so.6(realloc+0xdd)[0x3eff9d]
> > /usr/local/grass-6.5.svn/lib/libgrass_gis.so(G__realloc+0x35)[0x11eb55]
> > v.net.iso(main+0x15b9)[0x804a889]
> > /lib/tls/i686/cmov/libc.so.6(__libc_start_main+0xe6)[0x395bd6]
> > v.net.iso[0x8049231]
> > ======= Memory map: ========
> > 00110000-0015d000 r-xp 00000000 08:01 137602
> > /usr/local/grass-6.5.svn/lib/libgrass_gis.6.5.svn.so
> > 0015d000-0015e000 r--p 0004c000 08:01 137602
> > /usr/local/grass-6.5.svn/lib/libgrass_gis.6.5.svn.so
> >
> > and so on for some tens of lines.
> > The very strange thing is that in some cases the module produces the
> right
> > output and gives no errors, in other zones I obtain this error evenif the
> > input data are prapared in the same way.
>
> try to set debug level to 5 and run command again.
>
> g.gisenv DEGUG=5
>
> Post last few lines from the log.
>
> Martin
>
> --
> Martin Landa <landa.martin gmail.com> * http://gama.fsv.cvut.cz/~landa<http://gama.fsv.cvut.cz/%7Elanda>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.osgeo.org/pipermail/grass-user/attachments/20100623/b5152dc4/attachment.html