In my lab we have two martinos center-configured computers we use to
process MEG data. In my primary computer everything is fine, but in the
other does not like us as much apparently. When we open terminal and set
up the environment (using the exact same commands for the computer that
works) we get this error:
[wittgenstein:dgow] (nmr-stable5.1-env) mne_analyze
Using a 4096-point FFT in filtering of raw data with 2048-sample tapers
X Error of failed request: BadMatch (invalid parameter attributes)
Major opcode of failed request: 143 (GLX)
Minor opcode of failed request: 5 (X_GLXMakeCurrent)
Serial number of failed request: 1015
Current serial number in output stream: 1015
We encounter no such error when we use mne_browse_raw. We only use the
mne_analyze interface to align digitization points. Do you know anything
about this error or know a way to help us?
mne_analyze uses OpenGL for 3D rendering but not mne_browse_raw.
Are you accessing the machine locally of via ssh? Can you tell us more
about the machines? Are they both on Centos administered by help at nmr?
OpenGL could be the problem indeed. I'm accessing the machine locally.
They are both on Centos administered by help at nmr. The Graphics card on the
problem computer is Nvidia Quadro NVS 295, driver version 256.40. The
computer that works has a Nvidia Quadro FX 1800, driver version 185.18.36.
Do you want any other graphics or other information?
I just logged on to "Wittgenstein", the machine that seems to have the
issues, and started mne_analyze without problems. mne_opengl_test runs
fine too. If you are really connected locally, there must be an issue
with the X-server setup. I would suggest to run glxinfo, glxgears, and
check your /etc/xorg.conf
Manfred, thanks for checking. glxinfo comes back with an error. glxgears
does not work and /etc/xorg.conf does not exist. I'm also not sure how to
interpret these messages and what the next step is. It does make sense
that it is an issue with the way x-server is setup
Matti, mne_opengl_test fails (good idea to test the problem)
Sorry, I meant to say /etc/X11/xorg.conf which does exist and contains
the line to load the glx module. But looking at your /var/log/Xorg.0.log
file, it's quite obvious what goes wrong:
(EE) Dec 02 13:03:24 NVIDIA(0): Failed to initialize the GLX module;
please check in your X
(EE) Dec 02 13:03:24 NVIDIA(0): log file that the GLX module has
been loaded in your X
(EE) Dec 02 13:03:24 NVIDIA(0): server, and that the module is the
NVIDIA GLX module. If
(EE) Dec 02 13:03:24 NVIDIA(0): you continue to encounter problems,
Please try
(EE) Dec 02 13:03:24 NVIDIA(0): reinstalling the NVIDIA driver.
I think the problem is that the wrong GLX module - not the NVIDIA one,
but AIGLX - gets loaded. Check with help at nmr