Steve,<br><br>I'm a little confused, well not confused, unclear perhaps. :-)<br>I'm using the dijkstra_sp_delta() function. This doesn't explicitly say whether it uses directions and reverse costs. My results so far suggests it ignores directed graphs completely.<br>
<br>If it is using cost and reverse_cost, presumably I could just set these to the length values.<br>So I tried this, but I'm still getting the crash.<br>I do have some very small lengths - the smallest is being reported as 1.0E-7.<br>
The data was imported by osm2po from planet.osm.<br>I'm assuming this (or these) tiny distances are due to what are essentially duplicated nodes which are sufficiently far apart that osm2po does not recognize them as duplicates.<br>
<br>I've just done a search for links with the same source and target node identifiers - no matches.<br>(for my application, I could actually delete the cul-de-sac loops that you describe)<br><br>The new hard disk is on order for an attempt to build everything in Ubuntu.<br>
<br><br>Richard<br><br><br><div class="gmail_quote">On Mon, Jan 31, 2011 at 1:11 PM, Stephen Woodbridge <span dir="ltr"><<a href="mailto:woodbri@swoodbridge.com" target="_blank">woodbri@swoodbridge.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
<div>On 1/31/2011 10:36 AM, Richard Marsden wrote:<br>
<blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
Anton & Daniel,<br>
<br>
Thanks.<br>
<br>
I saw some references to problems with negative reverse_cost. I've<br>
checked the data and I don't have any negative cost or reverse_cost<br>
values. Plenty of zeroes though.<br>
</blockquote>
<br></div>
Ahhh, I'm guessing the zero cost is just as bad as negative costs, but I might be wrong on this. I would try setting all the zero costs to some really small positive cost and see if that helps.<br>
<br>
Or if you really have zero length segments, then why not reduce the size of your graph by removing all of them.<br>
<br>
You should also check the the source and target ids are not the same. as this is problematic also. If you have a cul d'sac there the start and end are the same but it does a loop, then you should split the loop into two segments.<br>
<br>
-Steve<br>
<br>
<blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;"><div><div></div><div>
Highest node value is 49,608,314 - so I don't think it is the integer<br>
overflow problem I'm seeing discussed.<br>
<br>
I also tried Iran (northern hemisphere, not too big, imperfect road data<br>
in osm) and it also crashes. The common thread seems to be that<br>
countries where I can't match all (geonames-derived) cities with osm<br>
nodes, crash using valid nodes. This isn't a simple "route cannot be<br>
found" as my UK Belfast example demonstrates (no routes possible from<br>
Belfast to the England - no crash occurs)<br>
<br>
One problem is that all I'm getting is a big crash - nothing indicates<br>
what the problem actually is.<br>
<br>
My laptop can boot into Ubuntu but the partition will be too small and<br>
the cpu too slow.<br>
<br>
The machine I'm using is my main batch processing /testing machine. It<br>
currently doesn't have space for a Unix/Linux partition, but has enough<br>
spare disk bays. I was already thinking of buying a large (1TB+) hard<br>
disk for this project. Although I've managed to squeak by with what I<br>
have, and PostGres's requirements aren't too big, manipulating<br>
planet.osm files does take some disk space!<br>
<br>
Perhaps this is the way forward then: New disk and try building an<br>
Ubuntu partition with postgres/etc. Assuming I don't have to rerun<br>
osm2po, it should take me a couple of days to rebuild everything.<br>
Most of my 'stack' should run fine on Ubuntu - Postgres, PostGis,<br>
Python. Windows is being used partly for historical reasons (earlier<br>
versions of my Python scripts used Windows for a good reason which no<br>
longer applies), but also I'm using Excel for my output. Perhaps there's<br>
a cross-platform Python library I can use. If not, I could create CSV<br>
files and "tie them together" with a Windows-based script (they're<br>
simple data sheets but I am relying on some formatting to make the data<br>
easier to use).<br>
<br>
re. building: I actually do most of my dev in Windows but using Visual<br>
Studio. I've seen the pgRouting Windows build instructions and they<br>
don't look very Windows friendly! Believe it or not I would be happier<br>
doing such builds in a Unix/Linux type of environment... (I've built<br>
my own UMN Mapserver with FreeBSD in the past).<br>
<br>
The project can't go forward if these crashes continue. So a little bit<br>
of a gamble that a new disk (or two if I decide on RAID) and Ubuntu will<br>
fix it...<br>
<br>
<br>
Richard<br>
<br>
<br>
<br>
<br>
On Sun, Jan 30, 2011 at 9:51 PM, Daniel Kastl <<a href="mailto:daniel@georepublic.de" target="_blank">daniel@georepublic.de</a><br></div></div><div>
<mailto:<a href="mailto:daniel@georepublic.de" target="_blank">daniel@georepublic.de</a>>> wrote:<br>
<br>
Hi Richard,<br>
<br>
You could try OSGeo Live DVD with already installed pgRouting in<br>
Virtualbox for example just to see if a newer version of would solve<br>
your issue: <a href="http://live.osgeo.org/" target="_blank">http://live.osgeo.org/</a><br>
<<a href="http://live.osgeo.org/" target="_blank">http://live.osgeo.org/</a>>As Anton mentioned, there seem to be not<br>
many developers out there, who are willing (or have the knowledge)<br>
to contribute Windows binaries.<br>
My feeling is that most pgRouting installations are running on Linux<br>
servers, while most users, who just want to try out pgRouting,<br>
prefer to do this on Windows.<br>
<br>
Daniel<br>
<br>
2011/1/31 Anton Patrushev <<a href="mailto:anton.patrushev@georepublic.de" target="_blank">anton.patrushev@georepublic.de</a><br></div>
<mailto:<a href="mailto:anton.patrushev@georepublic.de" target="_blank">anton.patrushev@georepublic.de</a>>><div><br>
<br>
Hi Richard,<br>
<br>
Unfortunately Win binaries are quite old because nobody is<br>
willing to<br>
make new ones.<br>
There can be few different reasons why it crushes (data problem for<br>
example). Do you have any Linux box around to test your data with<br>
newer version of pgRouting?<br>
<br>
Anton.<br>
_______________________________________________<br>
<br>
<br>
<br>
<br></div><div>
_______________________________________________<br>
Pgrouting-users mailing list<br>
<a href="mailto:Pgrouting-users@lists.osgeo.org" target="_blank">Pgrouting-users@lists.osgeo.org</a><br>
<a href="http://lists.osgeo.org/mailman/listinfo/pgrouting-users" target="_blank">http://lists.osgeo.org/mailman/listinfo/pgrouting-users</a><br>
</div></blockquote><div><div></div><div>
<br>
_______________________________________________<br>
Pgrouting-users mailing list<br>
<a href="mailto:Pgrouting-users@lists.osgeo.org" target="_blank">Pgrouting-users@lists.osgeo.org</a><br>
<a href="http://lists.osgeo.org/mailman/listinfo/pgrouting-users" target="_blank">http://lists.osgeo.org/mailman/listinfo/pgrouting-users</a><br>
</div></div></blockquote></div><br>