VMD-L Mailing List

From: John Stone (johns_at_ks.uiuc.edu)
Date: Tue Apr 11 2017 - 03:00:45 CDT

Thomas,
  Thanks for the solution. Yes, if indirect rendering is disabled
in XQuartz, that would completely explain the problem. I assume that
Apple closed off indirect rendering as some sort of approach to
improving the "default security" of the system.

Cheers,
  John Stone
  vmd_at_ks.uiuc.edu

On Tue, Apr 11, 2017 at 05:07:46PM +1000, Thomas Griffiths wrote:
> Hi John, Ryan;
>
> I may have a solution to this Xquartz and VMD issue. It is indeed an issue
> with XQuartz. I believe it has something to do with allowing XQuartz to
> allow indirect GLX contexts. (What that specifically means I don't know. My
> expertise here is lacking, but that's what I found being discussed online.)
> The following fixes my implementation.
>
> The fix is this: Quit XQuartz, then in the macOS Terminal application (on
> your own machine, *not* logged into the reomte cluster) run:
>
> defaults write org.macosforge.xquartz.X11 enable_iglx -bool true
>
> Reload your terminal and you should then be able to log into a remote
> machine and launch remote XQuartz sessions for VMD
>
> ---
>
> More indepth:
> Apparently, version 2.7.8 was the last release of XQuartz where indirect
> glx rendering was enabled by default. Which is why rolling back to this
> version works. There are a couple of posts out there (
> https://www.visitusers.org/index.php?title=Re-enabling_
> INdirect_glx_on_your_X_server) regarding enabling this indirect rendering.
>
> But the best advice I found was here (Actually for XQuartz and Gaussview 5,
> the viewer for Gaussian 09, but enabled remote VMD view again.) See the
> section 'XQuartz and macOS'
>
> https://opus.nci.org.au/display/Help/GaussView
>
> My current system is OS 10.11.6 and XQuartz version 2.7.11 (the latest
> XQuartz at the time of writing). Applying the above defaults allows me to
> run remote VMD sessions again (also fixed Gaussview for any wandering
> quantum chemists who might land on this page).
>
> Before setting the above default, or by setting it to false, I get the same
> errors as Ryan for glxinfo (without the libGL errors though)
>
> name of display: raijin3:14.0
> X Error of failed request: BadValue (integer parameter out of range
> for operation)
> Major opcode of failed request: 149 (GLX)
> Minor opcode of failed request: 3 (X_GLXCreateContext)
> Value in failed request: 0x0
> Serial number of failed request: 19
> Current serial number in output stream: 20
>
> This is fixed with the default set to true. glxinfo then prints a very
> verbose list, too big to paste here.
>
> Hope this helps.
>
> Tom.
>
> On 18 March 2017 at 06:25, John Stone <johns_at_ks.uiuc.edu> wrote:
>
> > Hi Ryan,
> > I bounced a copy of your reply to VMD-L so that others using VMD on
> > MacOS X and and/or with XQuartz can see it as well. The errors you get
> > when you run 'glxinfo' and 'glxgears' on the remote linux machine indicate
> > that the XQuartz window server on your Mac is (for some reason) failing to
> > respond to GLX protocol requests being performed on the remote Linux
> > system.
> > GLX is that part of OpenGL responsible for managing graphics resources
> > and negotiating how the OpenGL framebuffer is mapped onto the
> > windows and shown on the physical display. I'm most concerned with the
> > fact
> > that 'glxinfo' doesn't run, since that is the simplest test of whether the
> > X server has an operable OpenGL interface, and that is failing too.
> > I note that your libGL implementation on the Linux server is emitting
> > some errors about "failed to load driver: swrast", which makes me wonder
> > if the OpenGL system software on your Linux machine is complete or if there
> > are some software components missing. It is possible that the source of
> > trouble is actually the Linux machine rather than the XQuartz server,
> > but you may be able to determine that better based on what configurations
> > had worked for you in the past vs. the specific details of the software
> > configurations you're trying to use here. Did a previous version of
> > XQuartz work properly with the same Linux system? Does a different Linux
> > system work with this XQuartz version? Is XQuartz the only thing that
> > changed vs. the setup that had worked previously?
> >
> > Cheers,
> > John Stone
> > vmd_at_ks.uiuc.edu
> >
> > On Thu, Mar 16, 2017 at 02:28:19PM -0600, Ryan P. Steele wrote:
> > > Thanks for the response, John.
> > >
> > > When I run those two commands, I receive the following:
> > >
> > > kingspeak1 > glxinfo
> > > name of display: 10.242.128.31:20.0
> > > libGL error: No matching fbConfigs or visuals found
> > > libGL error: failed to load driver: swrast
> > > X Error of failed request: GLXBadContext
> > > Major opcode of failed request: 149 (GLX)
> > > Minor opcode of failed request: 6 (X_GLXIsDirect)
> > > Serial number of failed request: 23
> > > Current serial number in output stream: 22
> > >
> > > kingspeak1 > glxgears
> > > libGL error: No matching fbConfigs or visuals found
> > > libGL error: failed to load driver: swrast
> > > X Error of failed request: BadValue (integer parameter out of range
> > > for operation)
> > > Major opcode of failed request: 149 (GLX)
> > > Minor opcode of failed request: 3 (X_GLXCreateContext)
> > > Value in failed request: 0x0
> > > Serial number of failed request: 28
> > > Current serial number in output stream: 30
> > >
> > >
> > > It certainly sounds like something isn't working here, but I'm not
> > > familiar with these sorts of display configurations.
> > >
> > > Ryan
> > >
> > >
> > >
> > >
> > > On 3/16/17 1:56 PM, John Stone wrote:
> > > >Ryan,
> > > > Are you able to run either 'glxinfo' or 'glxgears' on this system in
> > > >the same way that you tried running VMD? If so, do they work, or do
> > they
> > > >fail?
> > > >
> > > >Cheers,
> > > > John Stone
> > > > vmd_at_ks.uiuc.edu
> > > >
> > > >On Thu, Mar 16, 2017 at 01:50:53PM -0600, Ryan P. Steele wrote:
> > > >> Hi, folks -
> > > >>
> > > >> We seem to be having an issue with Mac users when running a
> > server-side
> > > >> VMD window.\001 This version of VMD is a 'common' installation for
> > all
> > > >> users on a cluster.\001 With Windows or all tested Linux
> > terminals, things
> > > >> work fine when running it through tunneled x connections.\001
> > However, the
> > > >> recent version of XQuartz on OS X does not work.\001
> > > >>
> > > >> Upon startup, VMD gives the error copied below (the errors shown
> > here are
> > > >> for a user-installed version, but the common version behaves the
> > same
> > > >> way).\001 A locally installed version on the same Mac machine
> > works fine,
> > > >> and the server version, when run in text-only mode, also works
> > fine.\001
> > > >> It seems as though the x-tunneled graphics are not playing nicely
> > with
> > > >> XQuartz.
> > > >>
> > > >> Any suggestions?
> > > >>
> > > >> Thanks -
> > > >> Ryan
> > > >>
> > > >> > vmd
> > > >> /uufs/chpc.utah.edu/common/home/u0764443/mybin/vmd2/lib/
> > vmd_LINUXAMD64:
> > > >> /usr/lib64/libGL.so.1: no version information available (required
> > by
> > > >> /uufs/chpc.utah.edu/common/home/u0764443/mybin/vmd2/lib/
> > vmd_LINUXAMD64)
> > > >> Info) VMD for LINUXAMD64, version 1.9.3 (November 30, 2016)
> > > >> Info) [1]http://www.ks.uiuc.edu/Research/vmd/
> > > >> Info) Email questions and bug reports to [2]vmd_at_ks.uiuc.edu
> > > >> Info) Please include this reference in published work using VMD:
> > > >> Info)\001\001\001 Humphrey, W., Dalke, A. and Schulten, K., `VMD -
> > Visual
> > > >> Info)\001\001\001 Molecular Dynamics', J. Molec. Graphics 1996,
> > 14.1,
> > > >> 33-38.
> > > >> Info) ------------------------------------------------------------
> > -
> > > >> Info) Multithreading available, 32 CPUs detected.
> > > >> Info)\001\001 CPU features: SSE2 AVX
> > > >> Info) Free system memory: 43GB (67%)
> > > >> NVIDIA: no NVIDIA devices found
> > > >> Info) No CUDA accelerator devices available.
> > > >> libGL error: No matching fbConfigs or visuals found
> > > >> libGL error: failed to load driver: swrast
> > > >> XRequest.149: BadValue (integer parameter out of range for
> > operation) 0x0
> > > >> XRequest.149: GLXBadContext 0xa00001
> > > >> XRequest.149: 0 0x0
> > > >> Info) OpenGL renderer: ErrorUnknown
> > > >> Info)\001\001 Features: STENCIL
> > > >> Info)\001\001 GLSL rendering mode is NOT available.
> > > >> XRequest.149: BadMatch (invalid parameter attributes) 0xa00009
> > > >> NVIDIA: no NVIDIA devices found
> > > >> NVIDIA: no NVIDIA devices found
> > > >> Info) Dynamically loaded 2 plugins in directory:
> > > >> Info)
> > > >> /uufs/chpc.utah.edu/common/home/u0764443/mybin/vmd2/lib/
> > plugins/LINUXAMD64/molfile
> > > >> XRequest.149: GLXBadContext 0xa00001
> > > >> XRequest.149: 0 0x0
> > > >> XRequest.149: BadMatch (invalid parameter attributes) 0xa00009
> > > >> XRequest.149: GLXBadContext 0xa00001
> > > >> XRequest.149: 0 0x0
> > > >> XRequest.149: BadMatch (invalid parameter attributes) 0xa00009
> > > >> XRequest.149: GLXBadContext 0xa00001
> > > >> XRequest.149: 0 0x0
> > > >> XRequest.149: BadMatch (invalid parameter attributes) 0xa00009
> > > >> XRequest.149: GLXBadContext 0xa00001
> > > >>
> > > >> (repeats thousands of times...)
> > > >>
> > > >> --
> > > >> ==========================
> > > >> Ryan P. Steele
> > > >> Assistant Professor
> > > >> University of Utah
> > > >> Dept. of Chemistry
> > > >> 315 South 1400 East
> > > >> Salt Lake City, UT 84112
> > > >> Phone : 801-587-3800
> > > >> Office: Thatcher 4611
> > > >> ==========================
> > > >>
> > > >>References
> > > >>
> > > >> Visible links
> > > >> 1. http://www.ks.uiuc.edu/Research/vmd/
> > > >> 2. mailto:vmd_at_ks.uiuc.edu
> > >
> > > --
> > > ==========================
> > > Ryan P. Steele
> > > Assistant Professor
> > > University of Utah
> > > Dept. of Chemistry
> > > 315 South 1400 East
> > > Salt Lake City, UT 84112
> > > Phone : 801-587-3800
> > > Office: Thatcher 4611
> > > ==========================
> >
> > --
> > NIH Center for Macromolecular Modeling and Bioinformatics
> > Beckman Institute for Advanced Science and Technology
> > University of Illinois, 405 N. Mathews Ave, Urbana, IL 61801
> > http://www.ks.uiuc.edu/~johns/ Phone: 217-244-3349
> > http://www.ks.uiuc.edu/Research/vmd/
> >
> >
>
>
> --
>
>
>
>
>
>
>
> *Thomas GriffithsPhD CandidateSchool of Chemistry | Faculty of Science
> Medicine & HealthUniversity of Wollongong NSW 2522 AustraliaAcademic Senate
> ??? Postgraduate MemberScience Medicine & Health ??? Postgraduate
> RepresentativeT +612 4221 3559 | M +614 1527 1006uow.edu.au
> <http://www.uow.edu.au/science/chem/>*
>
> University of Wollongong CRICOS: 00102E
>
> NOTICE: This email is intended for the addressee named and may contain
> confidential information. If you are not the intended recipient, please
> delete it and notify the sender. Please consider the environment before
> printing this email. 

-- 
NIH Center for Macromolecular Modeling and Bioinformatics
Beckman Institute for Advanced Science and Technology
University of Illinois, 405 N. Mathews Ave, Urbana, IL 61801
http://www.ks.uiuc.edu/~johns/           Phone: 217-244-3349
http://www.ks.uiuc.edu/Research/vmd/