[GRASS5] Re: [bug #3877] (grass) r.to.vect: severe memory leaks, I'm helpless

Daniel Calvelo dca.gis at gmail.com
Tue Dec 6 01:03:27 EST 2005


I really sympathize with your current predicament, Maciek. Just to
clear things a little, I think that what Radim meant is that there are
no memory *leaks*. It's just that the memory *requirements* of your
problem are well beyond grass's architecture current capabilities. You
just don't have enough memory to perform the operation, the machine
goes into swapping out everything, including your X server, and never
goes out of it before you kill it.

The only workaround I can think of is

1) LOTS of swap and
2) LOTS of patience.

This is linux, right? You can try to add swapfiles (to more than the
8GB you already tried) and then, knowing that disk access is in the
order of 10^3 slower than RAM, maybe some timings on small datasets
might allow you to calculate the required time for it to complete.
Then launch it at night. Or week-end. I've been through this kind of
hassle for other applications, and that approach worked for me before.
Especially the "patience" part :)

But anyway, we should indeed put these memory-full-related problems on
the TODO list for 7.0.

Best wishes.

Daniel.

On 12/5/05, Maciek Sieczka <werchowyna at epf.pl> wrote:
> On pon, 2005-12-05 at 13:38 +0100, Radim Blazek via RT wrote:
> > please read old mails on this problem. I dont have time to explain it
> > again and again. AFAIK there are no big memory leaks.
>
> Is it aknowledged by Grass developers that a machine freeze at 5 mln
> vector points file is a BUG (no matter what the reason is)?
>
> If it is aknowledged, can we expect it to be fixed? When - soon, month
> time, year time? Or is it going to be a "feature" and left as is?
>
> Maciek
>
>
> --------------------
> W polskim Internecie s± setki milionów stron. My przekazujemy Tobie tylko najlepsze z nich!
> http://katalog.epf.pl/
>
> _______________________________________________
> grass5 mailing list
> grass5 at grass.itc.it
> http://grass.itc.it/mailman/listinfo/grass5
>


--
-- Daniel Calvelo Aros




More information about the grass-dev mailing list