ophyd-tango-demo

jupyter env

for easy access we added a jupyter lab to the accompanying docker-compose.yaml this should work without the need create any environment on the host system.

just enter into the jypyter lab with

http://localhost:8888/lab?token=myToken

Getting started / preparation of the environment

Starting point of what is described is an empty mamba/conda environment

• prepare the mamba/conda env

mamba install --file requirements.txt -c conda-forge

• fetch the git submodule ophyd-async https://github.com/desy-fsec/ophyd-async.git
• install ophyd+bluesky

cd ophyd-async/
git submodule init
git submodule update
pip install -e .
cd ..

• run docker-compose to start the tango db and device servers

docker-compose up -d

run the first demo script

based on ophyd-async/docs/user/examples/tango_scan.py. a modified version of this file is in the root of this repo that uses the tango path available in the docker container.

TANGO_HOST=localhost:10000 python tango_scan.py

Note 1: make sure that the environment variable TANGO_HOST is set so that the python process can connect to the tango db inside the docker container Note 2: it may take a few seconds before the scan starts. ... for the first scan it may even take almost a minute before the scan actually runs.

in case of trouble with tango servers not starting

try to flush the mariadb and start servers again after having used

docker-compose down -v

Resources

• the forked ophyd-async repo with tango support: https://github.com/desy-fsec/ophyd-async.git
• docker container + docker compose to run a tango DB + servers (preconfigured like at DESY): https://gitlab.desy.de/marcus.fleck/fsecservers-in-default-tango-cs-docker

to inspect Tango

in order to inspect the tango DB the most common tool to use would be jive. If conda is at hand, it can e.g. installed using the conda env file provided in tango_tool_env/tango_tools.yml with

conda env create -n tango_tools --file tango_tool_env/tango_tools.yml
conda activate tango_tools
TANGO_HOST=localhost:10000 jive