[Dock-fans] Mysterious crash

Darrell Hurt darrellh at niddk.nih.gov
Fri Oct 28 05:57:11 PDT 2005


The funny thing is that there is no core dump and no error output. But I 
know that the ligand database has more entries than are processed. I am 
unfamiliar with valgrind, I check into it.

Darrell



Kaitlyn Hwang wrote:
> Can you debug with valgrind or any debugging tools to see what happens
> when the app crash? Do you get a core dump? What is the error output?
> 
> I have been working Dock5 with parallel, and see the segfault when I put
> in the bug patches 6&8. So I took those two patches out. It would run
> correctly at first (with the patches in), but after awhile it would crash.
> My time processed for the Dock5 (without the patches) is off, but at least
> it has been able to process 100k ligands.
> 
> 
> kaitlyn
> 
> On Thu, 27 Oct 2005, Darrell Hurt wrote:
> 
> 
>>Hi everyone,
>>
>>I'm using DOCK 5.2 on a Linux box P4-3GHz machine, 1 GB RAM. I'm working
>>with the PubMed, the Sigma-Aldrich, and the Developmental Therapeutics
>>Program/NCI databases in mol2 format. When I dock ATP (the natural
>>substrate) into the defined docking site (grid spacing 0.3, 1100
>>spheres), I get a good orientation and a reasonable score. When I dock
>>any of the databases, DOCK will run with them for about 800 molecules
>>and then crash. "Unlimiting" the memory does not help. I then ran
>>DBfilter 2.2.8 on the pubmed database, which increased the size of the
>>database from ~135 MB to ~150 MB. When I dock with this database, I can
>>process ~6400 molecules before the crash.
>>
>>Any hints?
>>
>>Thanks,
>>Darrell
>>_______________________________________________
>>Dock-fans mailing list
>>Dock-fans at docking.org
>>http://blur.compbio.ucsf.edu/mailman/listinfo/dock-fans
>>
>>



More information about the Dock-fans mailing list