I have a Dockerfile
FROM ubuntu:18.04
RUN apt-get update && apt-get install -yq build-essential autoconf libnetcdf-dev libxml2-dev libproj-dev valgrind wget unzip git nano
# pulls ADBM from github and unzips in folder ADMBcode
RUN mkdir /ADMBcode
RUN wget https://github.com/admb-project/admb/releases/download/admb-12.2/admb-12.2-linux.zip
RUN mv admb-12.2-linux.zip /ADMBcode
RUN unzip ADMBcode/admb-12.2-linux.zip -d /ADMBcode
# pulls hydra repo from github into folder HYDRA
RUN mkdir /HYDRA
RUN git clone https://github.com/NOAA-EDAB/hydra_sim.git /HYDRA
WORKDIR /HYDRA
# incase of windows stuff
RUN chmod -R 777 /HYDRA
# compile the model
RUN /ADMBcode/admb-12.2/admb hydra_sim.tpl
CMD ./runModel.sh
It installs software, pulls from Github, and compiles a model, then it should execute a shell script.
If i use the command
podman run imageName
then all seems to run ok. i can enter the container and type ./runModel.sh and i can confirm the model runs.
But if i type
podman run --mount "type=bind,src=/path_to_input_files/,dst=/HYDRA/" imageName
where the path_to_input_files contains two files hydra_sim.dat and hydra_sim.pin i get the error
/bin/sh: 1: ./runModel.sh: not found
This file (runModel.sh) exists in the WORKDIR (it is part of the cloned repo). But if i add it to the path_to_input_files directory that is mounted i still get the same error.
What am i missing here? Thanks
When you run:
podman run --mount "type=bind,src=/path_to_input_files/,dst=/HYDRA/" imageName
You are replacing the contents of /HYDRA with the contents of /path_to_input_files. Anything else that was included in that directory in the image is no longer accessible.
That means in particular that /HYDRA/runModel.sh is not available.
The fix is to mount your input files at another location.