The procedure look like this,
2) Copy installation file form host to docker container, and extract files,
cd /mofem_build
cp $HOSTHOME/Downloads/ParaView-5.4.0-RC3-Qt5-OpenGL2-MPI-Linux-64bit.tar.gz .
tar -xvzf ParaView-5.4.0-RC3-Qt5-OpenGL2-MPI-Linux-64bit.tar.gz
3) The run pvserver in docker container. You can then use docker port to find out what host port the container’s port 22 is mapped to:
docker run -P -d --rm=true --name pvserver --volumes-from mofem_build mofem_build /mofem_build/ParaView-5.4.0-RC3-Qt5-OpenGL2-MPI-Linux-64bit/bin/pvserver --server-port=22
docker port pvserver 22
as result you get,
where numbers before double dot is IP address of host (
0.0.0.0) and after port number (
32773).
4) Open preview in your host machine, go to File->Connect, in connect add server, and as a host and port use 0.0.0.0 and 32773 respectively. Next click Configure and save. Select your server and connect to it.
At that point you have access to mofem_build directory in docker container. Note you can run analysis is another container.
This in will work, but is not the best way. We need to compile preview server with mpi and add this to mofem build environment.