Page History

...

Valerio writes: It's not different from the psana2 approach at all.  First you need to package the source in a tar.gz like we always did, as shown in the "Checklist for building psana1 environments from feedstocks" section above.  The py2 and py3 psana recipes have now the pinning file (conda_build_config.yaml) in them. You just need to update the meta.yaml file, run "conda build recipe" (did this as user "cpo", but could also do as psreldev, although psreldev might need "anaconda login" as another user so "anaconda upload" will work) upload and build the environment from the "pinned" YAML file. 

...

The yaml files for creating the environment are in /cds/sw/ds/ana/conda1/manage/jenkins/.

To create the env's without changing the pinnings, clone the previous environment and install the newly upload psana version like this:

conda install -c lcls-i psana=4.0.44 --experimental-solver=libmamba

To create new pinnings (which may require also changing conda_build_config.yaml in the py2/py3 recipe repos) you can create cpo created the two env's like this:

source /reg/g/psdm/etc/psconda.sh -v2 (v2 to get experimental mamba solver)
conda env create --name ana-4.0.44-py3 --experimental-solver=libmamba --file=/cds/sw/ds/ana/conda1/manage/jenkins/ana-env-py3.yaml 
conda env create --name ana-4.0.44 --experimental-solver=libmamba --file=/cds/sw/ds/ana/conda1/manage/jenkins/ana-env-py2.yaml

LCLS2 Github Actions Approach (Deprecated)

Currently deprecated since conda forge changes versions more rapidly than we would like, so we use the "Pinned Approach" described below.

...