[Paraview] How to debug Paraview

[Moreland, Kenneth]

Actually, I think the answer is no.  If you want to see the image in the GUI window, then the answer is no.  Get rid of the -tdx and -tdy flags.

-Ken

> -----Original Message-----
> From: Weiguang Guan [mailto:guanw at rhpcs.mcmaster.ca]
> Sent: Wednesday, March 12, 2008 7:54 AM
> To: Moreland, Kenneth
> Cc: paraview at paraview.org
> Subject: RE: [Paraview] How to debug Paraview
>
> I think the answer to your question is YES. I am running Paraview on HP's
> SVA cluster. The paraview client (paraview) is running on one of the
> display nodes of the cluster.
>
> I lauch paraview through a script called sva_paraview.sh, which does job
> scheduling, allocating nodes, starting X servers besides mpirun-ing
> pvserver-s.
>
> Weiguang
>
> On Wed, 12 Mar 2008, Moreland, Kenneth wrote:
>
> > Why do you have the -tdx and -tdy flags?  Are you driving a remote
> display?
> >
> > -Ken
> >
> >> -----Original Message-----
> >> From: Weiguang Guan [mailto:guanw at rhpcs.mcmaster.ca]
> >> Sent: Wednesday, March 12, 2008 7:22 AM
> >> To: Moreland, Kenneth
> >> Cc: paraview at paraview.org
> >> Subject: Re: [Paraview] How to debug Paraview
> >>
> >> Hi Kenneth,
> >>
> >> I appreciate your time very much. I attach two screen shots ---
> generated
> >> by using 1 and 4 pvserver-s respectively.
> >>
> >> What I did are:
> >> (1) On node ic-rb18, "paraview --server=ic-rb1" is issued;
> >> (2) mpirun -f abc, where file abc has the contents shown below
> >> -h ic-rb1 -np 1 -e
> >>
> PATH=.:/home/guanw/visualization/software/install/bin:/opt/sva/chromium/bi
> >> n/Linux:/opt/sharcnet/par
> >>
> aview/bin:/opt/sharcnet/compile/bin:/opt/hptc/bin:/opt/hptc/lsf/top/6.2/li
> >> nux2.6-glibc2.3-x86_64-slurm/etc:/opt/hptc/l
> >> sf/top/6.2/linux2.6-glibc2.3-x86_64-
> >>
> slurm/bin:/usr/kerberos/bin:/usr/bin:/bin:/usr/sbin:/sbin:/usr/X11R6/bin:/
> >> opt/hpmp
> >> i/bin:/opt/sharcnet/pathscale/current/bin:/opt/sharcnet/pgi/pgi-
> >> 6.1/linux86-64/6.1/bin:/opt/sharcnet/sq/bin:/opt/sva/b
> >> in: -e DISPLAY=:0 pvserver -rc --client-host=ic-rb18 -tdx=1 -tdy=1
> >> -h ic-rb10 -np 1 -e
> >>
> PATH=.:/home/guanw/visualization/software/install/bin:/opt/sva/chromium/bi
> >> n/Linux:/opt/sharcnet/pa
> >>
> raview/bin:/opt/sharcnet/compile/bin:/opt/hptc/bin:/opt/hptc/lsf/top/6.2/l
> >> inux2.6-glibc2.3-x86_64-slurm/etc:/opt/hptc/
> >> lsf/top/6.2/linux2.6-glibc2.3-x86_64-
> >>
> slurm/bin:/usr/kerberos/bin:/usr/bin:/bin:/usr/sbin:/sbin:/usr/X11R6/bin:/
> >> opt/hpm
> >> pi/bin:/opt/sharcnet/pathscale/current/bin:/opt/sharcnet/pgi/pgi-
> >> 6.1/linux86-64/6.1/bin:/opt/sharcnet/sq/bin:/opt/sva/
> >> bin: -e DISPLAY=:0 pvserver -rc --client-host=ic-rb18 -tdx=1 -tdy=1
> >> -h ic-rb11 -np 1 -e
> >>
> PATH=.:/home/guanw/visualization/software/install/bin:/opt/sva/chromium/bi
> >> n/Linux:/opt/sharcnet/pa
> >>
> raview/bin:/opt/sharcnet/compile/bin:/opt/hptc/bin:/opt/hptc/lsf/top/6.2/l
> >> inux2.6-glibc2.3-x86_64-slurm/etc:/opt/hptc/
> >> lsf/top/6.2/linux2.6-glibc2.3-x86_64-
> >>
> slurm/bin:/usr/kerberos/bin:/usr/bin:/bin:/usr/sbin:/sbin:/usr/X11R6/bin:/
> >> opt/hpm
> >> pi/bin:/opt/sharcnet/pathscale/current/bin:/opt/sharcnet/pgi/pgi-
> >> 6.1/linux86-64/6.1/bin:/opt/sharcnet/sq/bin:/opt/sva/
> >> bin: -e DISPLAY=:0 pvserver -rc --client-host=ic-rb18 -tdx=1 -tdy=1
> >> -h ic-rb12 -np 1 -e
> >>
> PATH=.:/home/guanw/visualization/software/install/bin:/opt/sva/chromium/bi
> >> n/Linux:/opt/sharcnet/pa
> >>
> raview/bin:/opt/sharcnet/compile/bin:/opt/hptc/bin:/opt/hptc/lsf/top/6.2/l
> >> inux2.6-glibc2.3-x86_64-slurm/etc:/opt/hptc/
> >> lsf/top/6.2/linux2.6-glibc2.3-x86_64-
> >>
> slurm/bin:/usr/kerberos/bin:/usr/bin:/bin:/usr/sbin:/sbin:/usr/X11R6/bin:/
> >> opt/hpm
> >> pi/bin:/opt/sharcnet/pathscale/current/bin:/opt/sharcnet/pgi/pgi-
> >> 6.1/linux86-64/6.1/bin:/opt/sharcnet/sq/bin:/opt/sva/
> >> bin: -e DISPLAY=:0 pvserver -rc --client-host=ic-rb18 -tdx=1 -tdy=1
> >>
> >> I notice, in step (1) paraview --server=ic-rb1, there is a redundance
> of
> >> using the --server as I use --client-host for pvserver. But if I don't
> use
> >> --server for paraview, the connection can not be established.
> >>
> >> Weiguang
> >>
> >> On Tue, 11 Mar 2008, Moreland, Kenneth wrote:
> >>
> >>> Can you post a screen shot?  That might help identify the problem.
> >>>
> >>> -Ken
> >>>
> >>>
> >>> On 3/11/08 3:59 PM, "Weiguang Guan" <guanw at rhpcs.mcmaster.ca> wrote:
> >>>
> >>> I have tried to run paraview not in mpirun. Although it doesn't solve
> >> the
> >>> problem, at least it gets rid of the error message "MPI Application
> rank
> >> 0
> >>> exited before MPI_Init() with status 0".
> >>>
> >>> Weiguang
> >>>
> >>> On Tue, 11 Mar 2008, Moreland, Kenneth wrote:
> >>>
> >>>> Are you running the paraview application in mpirun?  You shouldn't.
> >>>> The server, pvserver, is an MPI application but the client, paraview,
> >> is
> >>> not.
> >>>>
> >>>> -Ken
> >>>>
> >>>>
> >>>> On 3/11/08 1:11 PM, "Weiguang Guan" <guanw at rhpcs.mcmaster.ca> wrote:
> >>>>
> >>>> Hi everyone,
> >>>>
> >>>> I have run into a weird problem with Paraview and can not get a
> >> solution
> >>>> from the mailing list. Does anyone know how to debug Paraview? Does
> >>>> Paraview spit out log file?
> >>>>
> >>>> The problem I have is missing polygon from rendering when paraview
> run
> >> in
> >>>> client/server mode. This problem happens even if I run one pvserver.
> >>>>
> >>>> No abnormal message has been seen until we exit Paraview --- "MPI
> >>>> Application rank 0 exited before MPI_Init() with status 0"
> >>>>
> >>>> Weiguang
> >>>> _______________________________________________
> >>>> ParaView mailing list
> >>>> ParaView at paraview.org
> >>>> http://www.paraview.org/mailman/listinfo/paraview
> >>>>
> >>>>
> >>>>
> >>>>
> >>>
> >>>
> >>>
> >>>
> >