From: John Stone (
Date: Mon Mar 22 2004 - 11:21:55 CST

  I got your note about the client code. I do have sockets code that
compiles on RH9 etc, I suspect that the problem with your code is just
include files or something. It could be that there's some code in the
MSMS socket interface that isn't 32-bit/64-bit identical and thus it would
require a separate Athlon64 build to work correctly. I guess this is what
we've got to figure out. Do you guys have any Athlon64/Opteron machines there?


On Fri, Mar 19, 2004 at 04:05:30PM -0800, Michel Sanner wrote:
> Hello all,
> Sorry for the delay in replying to this but I was out of my office.
> John would by any chance still have the client C program to connect to MSMS
> when started in server mode? This would help me in trying to reproduce
> the described
> problem.
> -Michel
> John Stone wrote:
> >Jerry,
> > Interesting, if you and Michel find a general cure for this problem
> >you ran into that'd be good to know, since I suspect there will be a lot
> >of other people testing out the 64-bit x86_64 VMD binaries in the next
> >year, and most will likely want to use MSMS if at all possible. As a rule
> >of thumb I _always_ tell people to get MSMS to use with VMD, so do keep us
> >informed.
> >
> >Thanks,
> > John Stone
> >
> >
> >On Mon, Mar 15, 2004 at 12:49:18PM -0600, Jerry Ebalunode wrote:
> >
> >
> >>Hi John,
> >>I finally figured out why I couldn't get the MSMS program to work alone or
> >>
> >>even work with VMD using the socketport option. First, because I was using
> >>my new Gentoo Linux Athlon 64 workstation in a full 64 bit linux
> >>environment, I could not just copy the dynamically linked MSMS I
> >>installed on my older 32 bit linux work station. This verison of MSMS (
> >>about 2 years old) required some 32 bit glibc libraries. To fix this
> >>problem I contacted Dr. Sanner about the availablity of a statically
> >>linked MSMS binary for linux. Well he told me that the current ones
> >>present at there download site are statically linked. After installation,
> >>I have being trying to get them to run using the socketports option,
> >>without any success. This weekend I installed a secondary 32 bit
> >>environment under my original 64 bit environment (i.e., glibc 32 bit and
> >>64 bit in the same box ) and out of curiousity I installed my two year old
> >>MSMS binary which is dependent on the 32 bit glibc libraries. Amazingly,
> >>MSMS worked with the socketport option likea charm. So the problem wasn't
> >>with my platform but with the newer MSMS binaries. However, these new set
> >>of binaries work fine if you are not using sockport option. This
> >>observation might be the reason why we could not use the MSMS along with
> >>vmd on the DEC ALPHA workstation which we reported last year (i.e. ports
> >>could not be opened). I guess the way the new statically linked MSMS
> >>binaries might be the cause of these problems. Just want to let you know
> >>and I am CCing this e-mail to Dr. Sanner
> >>--
> >>Cheers,
> >>
> >>Jerry Ebalunode
> >>Graduate Research Assistant
> >>RM 402F Houston Science Center
> >>Phone: 713-743-8367
> >>Dept. of Biology and Biochemistry
> >>University of Houston
> >>4800 Calhoun Road
> >>Houston, TX 77204
> >>
> >>
> >
> >
> >
> --
> -----------------------------------------------------------------------
> o
> / Michel F. Sanner Ph.D. The Scripps Research Institute
> o Associate Professor Department of Molecular Biology
> \ 10550 North Torrey Pines Road
> o Tel. (858) 784-2341 La Jolla, CA 92037
> / Fax. (858) 784-2860
> o
> -----------------------------------------------------------------------

NIH Resource for Macromolecular Modeling and Bioinformatics
Beckman Institute for Advanced Science and Technology
University of Illinois, 405 N. Mathews Ave, Urbana, IL 61801
Email:                 Phone: 217-244-3349              
  WWW:      Fax: 217-244-6078