From: jrui_at_ci.uc.pt
Date: Fri Feb 22 2008 - 05:13:16 CST

Hi,

The error arises even after a fresh OS restart. The files are small:
~2KB each.
I can load all my mol2 files as separates molecules: the error arises when I
try to load the files as a trajectory (in a new vmd session), right after the
second frame.
I have 512MB swap and 512MB RAM. When I start VMD, it tells me that there is
375MB of free system memory (74%).
The main memory eaters are Xorg (5.5%) and vmd (5.0%).
I could not find anything strange in the dmesg output, but i'm not sure what I
should be looking for... :)

Here is the output of the "free" command before starting VMD:

             total used free shared buffers cached
Mem: 513932 376448 137484 0 10272 219236
-/+ buffers/cache: 146940 366992
Swap: 506008 0 506008

Here is the output of "free" after launching VMD:
Mem: 513932 376508 137424 0 10328 219260
-/+ buffers/cache: 146920 367012
Swap: 506008 0 506008

And here is the output of "free" after I load two files as a trajectory:
Mem: 513932 376656 137276 0 10436 219252
-/+ buffers/cache: 146968 366964
Swap: 506008 0 506008

For now, I converted my mol2 files to pdb files and load them as a trajectory
and it works fine.

Thanks,
Rui Rodrigues

Quoting Axel Kohlmeyer <akohlmey_at_cmm.chem.upenn.edu>:

> On Thu, 21 Feb 2008, jrui_at_ci.uc.pt wrote:
>
> RR> Hi,
> RR>
> RR> When I type the following commands in vmd console:
> RR> mol new ADP01.mol2
> RR> ls
> RR> mol new ADP02.mol2
> RR> ls
> RR> ... and so on, everything goes fine. However, if try to read the
> different
> RR> conformations as a trajectory:
> RR> mol new ADP01.mol2
> RR> ls
> RR> mol addfile ADP02.mol2
> RR> ls
> RR>
>
> RR> then I get the error "couldn't fork child process: not enough
> RR> memory", although I still can load more frames to the molecule. The
> RR> molecule has only 40 atoms (ADP). Other than that, vmd seems to work
> RR> normally, except for calling linux commands and external programs
> RR> (such as msms or tachyon). What could that be?
>
> how long has your machine been running? how large are the files?
> what else is running on that machine? how much swap do you have
> configured?
>
> it could quite likely be that you are simply running out of (virtual)
> memory and that VMD has nothing to do with it. modern GUIs tend to have
> a lot of gimmicks with sometimes serious memory leaks. also large
> applications like webbrowsers, word processors, graphics tools etc.
> tend to consume large quantities of address space and leak memory
> like a sieve.
>
> cheers,
> axel.
>
> ps.: please check the status of your machine with running 'dmesg'
> in a terminal and check if there is anything odd happening or run
> 'top' and then press Shift-M to see the top memory eaters.
>
> RR>
> RR> I'm using VMD 1.8.6 and Ubuntu 7.04, with 512MB RAM.
> RR>
> RR>
> RR> Thanks,
> RR> Rui Rodrigues
> RR>
>
> --
> =======================================================================
> Axel Kohlmeyer akohlmey_at_cmm.chem.upenn.edu http://www.cmm.upenn.edu
> Center for Molecular Modeling -- University of Pennsylvania
> Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
> tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425
> =======================================================================
> If you make something idiot-proof, the universe creates a better idiot.
>
>