Hi Kevin.
Your RequestInformation looks a little bare. Unless things have changed
recently, you might need to set WHOLE_EXTENT, ORIGIN, SPACING,
CAN_PRODUCE_SUB_EXTENT keys there. and if you support parallel IO honor
the requested extent in RequestData.
Burlen
On 05/04/2015 09:52 AM, Dean,
btw, here is an batch script
https://github.com/burlen/pvserver-configs/blob/master/servers/nautilus/3.10.0-x/start_pvserver-x.qsub
starting ParaView with pvservers assigned to GPU's round robin, from the
SGI UV 1000 with 8 GPU. The other tricky aspect of running PV on
multiple GPU is
Forwarded Message
Subject:Re: [Paraview] Fwd: Image Doesn't load correctly
Date: Tue, 05 May 2015 09:14:06 -0700
From: Burlen Loring blor...@lbl.gov
To: Dean, Kevin kevin.d...@decisionsciencescorp.com
ParaView should set it after it determines if the reader
to the same problem we're
trying to address :). What are your thoughts?
Utkarsh
On Mon, Apr 13, 2015 at 11:48 AM, Burlen Loring burlen.lor...@gmail.com wrote:
OK, I added some notes based on the above discussion.
On 04/13/2015 08:33 AM, Utkarsh Ayachit wrote:
Great, let's start here:
https
going back to your original question. when exec'ing new processes there
are various flavors of exec, including one that uses your environment.
Which made me wonder which flavor is PV using? I wrote a little pvsc to
test this out(attached). Apparently PV does use the exec that includes
your
Hi Utkarsh,
I think environment variables are kind of a band aid around a larger
issue. In addition to leveraging the user's local environment, there's
also no way to get information from the server side environment. for
example list of account numbers a user can charge jobs to, or a list of
: Burlen Loring burlen.lor...@gmail.com
Date: Sunday, April 12, 2015 at 10:34 PM
To: Utkarsh Ayachit utkarsh.ayac...@kitware.com, Rick Angelini
richard.c.angelini@mail.mil
Cc: paraview@paraview.org paraview@paraview.org
Subject: Re: [Paraview] default_servers.pvsc question
Hi Utkarsh,
I think
).read()
default_shell=default_shell.rstrip(\r\n)
except:
default_shell=/usr/krb5/bin/
Rick Angelini
USArmy Research Laboratory
CISD/HPC Architectures Team
Phone: 410-278-6266
-Original Message-
From: Burlen Loring burlen.lor
*From:* Burlen Loring [burlen.lor...@gmail.com]
*Sent:* Thursday, June 04, 2015 5:45 PM
*To:* Göbbert, Jens Henrik; paraview@paraview.org
*Subject:* Re: [Paraview] save 'Surface LIC' as texture
I think it could be done if you're willing to hack into PV. There is
already some mechanism
I think it could be done if you're willing to hack into PV. There is
already some mechanism in there for sucking textures back from the gpu
and writing them to disk for debugging. It's not exactly what you need
but at least will illustrate pulling data from GPU.
a potential issue with this
On Wed, Jun 10, 2015 at 2:49 PM, Jeff Becker
jeffrey.c.bec...@nasa.gov mailto:jeffrey.c.bec...@nasa.gov wrote:
On 06/10/2015 11:46 AM, Burlen Loring wrote:
Hmm. If Bunyk ray cast doesn't work some thing strange is
going on as that is a software(non gpu) algorithm
2 things to try: set the remote render threshold to zero, and if that
doesn't get it to show up, change the mapper to Bunyk ray cast. One of
those usually works depending on what the cause of the issue is.
On 06/10/2015 11:00 AM, Jeff Becker wrote:
Hi. I have an unstructured grid file that
to unstructured data. If that works maybe something with your
input data, or size of it.
On 06/10/2015 11:27 AM, Jeff Becker wrote:
On 06/10/2015 11:06 AM, Burlen Loring wrote:
2 things to try: set the remote render threshold to zero, and if that
doesn't get it to show up, change the mapper to Bunyk ray cast
PM, Burlen Loring wrote:
you ran out of memory.
You may need more than 1 node, and a parallel file format.
You mentioned that you had a grid of 500 x 500 points. You could save
a lot of memory by using vtkImageData, vtkRectilinearMesh or
vtkStructuredGrid if your data is amenable to one
Polytechnique de Montréal/
/LADYF/
//Lab. de dynamique des fluides
(514) 340 4711 ext 2489
Local: C-318.21.3*
*
On Wed, Jun 17, 2015 at 10:19 AM, Burlen Loring
burlen.lor...@gmail.com mailto:burlen.lor...@gmail.com wrote:
You also have to prepend the paths
You also have to prepend the paths with the .so's to LD_LIBRARY_PATH
before starting python.
Burlen
On 06/16/2015 07:19 PM, Felix Salazar wrote:
Hello community,
I normally run my Paraview scripts in pvpython, so far without issues.
But I got some problems using specific Python functions
Hi Jeff,
It really sounds like you are unintentionally using X forwarding. It
could be baked into your ssh config or a bash alias.
I see the window with the gears
if you are really doing remote rendering, you won't see the window.
you may be able to confirm accidental X forward using ssh
A fresh clone as of midday yesterday PV client fails to launch with the
below error. I enabled mpi and python, everything else is default value.
libGL error: No matching fbConfigs or visuals found
libGL error: failed to load driver: swrast
X Error: BadValue (integer parameter out of range for
VTK_DEBUG_LEAKS, although will catch some serious errors, will not catch
all kinds of leaks. For example you can have leaks where data
erroneously accumulates with each time step, but is deleted at program
termination. VTK_DEBUG_LEAKS will not catch that kind of error. It's
probably better to
be willing to try it out if you have more details
(commit hashtag, ticket, etc.), even if I have to figure out how to
backport it from a newer version.
Thanks,
Tim
*From:* Burlen Loring <burlen.lor...@gmail.com>
*From:* ParaView <paraview-boun...@paraview.org> on behalf of
Gallagher, Timothy P <tim.gallag...@gatech.edu>
*Sent:* Friday, May 20, 2016 4:17 PM
*To:* Burlen Loring; Andy Bauer
*Cc:* paraview@paraview.org
*Subject:* Re: [Para
fyi, a common way to encounter
Fatal Python error: PyThreadState_Get: no current thread
is when cmake's find Python modules mix two version of Python. this
happen if you have two installs, one in a system location and one else
where with interpreter in the path. You can identify it in the
certainly seems like a bug worth fixing. let me know if you want me to
start a bug report.
On 07/08/2016 01:19 PM, Burlen Loring wrote:
Thanks Utakarsh. It sounds like it may be possible to work around the
issue by not importing the lookuptable module from simple. However,
it's "import
Hi All,
first of all, is 5.1 production ready? or is 4.4 still the way to go in
a production environment? this would be the only install available for
users so it's important that it function reliably/correctly.
re: 5.1, I'm wondering what the recommended mesa config for use with pv
5.1 on
Hi All,
I'm encountering some issues when trying to use pvbatch from a frozen
build(PARAVIEW_FREEZE_PYTHON=ON). Seems that certain key python modules
are missing.ie math. However, I can use Python shell feature in the GUI
from the same build. Do we have any idea about what may be wrong with
got it thanks!
On 07/05/2016 02:16 PM, Utkarsh Ayachit wrote:
first of all, is 5.1 production ready? or is 4.4 still the way to go in a
production environment? this would be the only install available for users so
it's important that it function reliably/correctly.
See this thread:
Research Laboratory
CISD/HPC Architectures Team
Phone: 410-278-6266
-Original Message-
From: ParaView <paraview-boun...@paraview.org> on behalf of Burlen Loring
<blor...@lbl.gov>
Date: Tuesday, July 5, 2016 at 2:51 PM
To: ParaView <paraview@paraview.org>
Subject: [Non-DoD
hey Cynthia,
if your data is in, or can be converted to, CSV format just open it up,
and apply the table to points filter, followed by glyph filter. table to
points you'll need to specify the coordinate arrays and check keep all
arrays option. in the glyph filter properties you can select a
g the MPI-I/O path anyway.
I would follow Berk’s suggesting of looking into the filesystem before
diving into the guts of MPI-I/O.
-Ken
*From:*ParaView [mailto:paraview-boun...@paraview.org] *On Behalf Of
*Burlen Loring
*Sent:* Tuesday, November 22, 2016 12:55 PM
*To:* Geveci, Berk (Exter
I've hit that before. the RAW reader uses MPI-I/O, and MPI uses signed
int everywhere in its API, so there are limits that you end up hitting.
For instance to partition the data among processors one must give
MPI-I/O the starting offset of the data in the file as a signed int, and
that is
!
Hmm, that's odd. Can you share the plugin code with me? Let's see if I
can reproduce the issue.
Utkarsh
On Tue, Jan 3, 2017 at 2:13 PM, Burlen Loring <blor...@lbl.gov> wrote:
Hi Utkarsh, Happy new year!
I have a couple of questions about the new way.
Shouldn't this be automatically
After upgrading to 5.2 my plugin is not compiling. When I configure the
plugin I see a few pages of the following:
CMake Warning (dev) at io/CMakeLists.txt:54 (add_library):
Policy CMP0028 is not set: Double colon in target name means ALIAS or
IMPORTED target. Run "cmake
OK, good to know. would any one be willing to give a quick review?
https://gitlab.kitware.com/paraview/paraview/merge_requests/1476
On 03/20/2017 12:16 PM, Ben Boeckel wrote:
On Mon, Mar 20, 2017 at 11:31:28 -0700, Burlen Loring wrote:
I had to apply to clang-format manually. it's up
fyi, https://gitlab.kitware.com/paraview/paraview/merge_requests/1476
it complains about clang-format, but it seems that the source file was
not formatted in that way to begin with.
On 03/17/2017 12:11 PM, Burlen Loring wrote:
Thank you guys. I poked around this morning and found
Thank you guys. I poked around this morning and found the following:
The "BlueObeliskElements" is the only cmap in the default presets that
triggers the issue. It has more values than can be displayed on a single
line, and this is what triggers the FPE in the logic.
Enforcing a minimum
install clang-format, run
`git clang-format` in your ParaView source directory, amend your
commit, then force push it with `git gitlab-push -f`.
- Cory
On Fri, Mar 17, 2017 at 3:36 PM, Burlen Loring <burlen.lor...@gmail.com> wrote:
fyi, https://gitlab.kitware.com/paraview/paraview/merge_r
I tried out ParaView 5.3.0 today, and I encountered the above crash.
steps to reproduce: load data (sphere source will work), set color by
var, open color map dialog, change to categorical, open choose presets
dialog, scroll down, when you get near the bottom you will get the crash.
Program
not sure if it matters, but I had been trying to load a color map that
was saved from ParaView 5.1.0. attached.
On 03/16/2017 04:50 PM, Burlen Loring wrote:
I tried out ParaView 5.3.0 today, and I encountered the above crash.
steps to reproduce: load data (sphere source will work), set color
On 04/21/2017 10:19 AM, Ben Boeckel wrote:
On Thu, Apr 20, 2017 at 14:44:47 -0700, Burlen Loring wrote:
[100%] Freezing Python and ParaView Python packages and modules
Removing
/usr/common/software/ParaView/builds/ParaView-5.3.0/Utilities/PythonInitializer/FrozenPython
Error: could not find
Error: No module named math
On 04/24/2017 06:52 AM, Ben Boeckel wrote:
On Fri, Apr 21, 2017 at 12:10:36 -0700, Burlen Loring wrote:
That file lives in: lib/python2.7/config/config.c.in
OK, it looks like the freeze_paraview.py file can't find it. Could you
debug around line 250 to see what it is
Hi Patrick,
Your output shows you enabled some gpu specific drivers, and GLX. I
think that is going to screw things up for you. Best to disable all of
them but the ones you specifically need and to explicitly disable glx.
Here is how I configured OSMesa for a Cray
uld I post on the Mesa forum ?
Patrick
Burlen Loring wrote:
that's the point. this allows you to run without the windowing system
or GPU on the cluster. Most cluster have neither. If you wanted to
provide the GUI then I would suggest you have two installs of both
ParaView and Mesa. One based
.
Burlen
On 05/16/2017 08:38 AM, Patrick Begou wrote:
Burlen Loring wrote:
../mesa-17.0.2/configure --enable-texture-float --disable-glx
--disable-dri --disable-egl --disable-gles1 --disable-gles2
--disable-gbm --disable-driglx-direct --disable-xvmc
--enable-gallium-osmesa --with-gallium-drivers
Hi Patrick,
OSMesa (and friends such as OpenSWR) do not need to make use of GLX. In
fact I think it should be disabled completely for that case in both Mesa
and ParaView builds. For eg, when you configure ParaView you'd specify
-DVTK_USE_X=OFF. At least this is how it was in the past. I
_AVCODECID=ON \
...
Burlen Loring wrote:
- one build for the front end with GUI: it has no GPU but a
windowing system (this is the blocking point at this time)
OK. to accomplish this, first do a Mesa build configured as
follows(glx+software rendering)
../mesa-17.0.6/configure --enable-texture-floa
Hi Fabrizio,
in the shell where you launch your pvserver you could set the following
environment variables:
export LIBGL_ALWAYS_SOFTWARE=1 GALLIUM_DRIVER=swr
this is bash syntax, your shell may be different. When I do that I see
"SWR detected AVX" print in that shell when the connection
301 - 346 of 346 matches
Mail list logo
