Until self-calibration is included in the VLA imaging pipeline it is sometime necessary to run it outside of the workspaces environment.
Step-by-step guide
- Copy the calibrated ms and flagversions table to the working directory in into a work area e.g. /lustre/aoc/sciops/<your dir>/<ws_request id>/working
- Edit the run_casa.sh and selfcal_script.py (see info box below) to use the current working directory and ms name. These should also be stored in the working directory.
- login to nmpost-master and submit the batch job (from inside the working directory: "sbatch run_casa.sh")
- wait a few days
- The job should email you when done (it may claim the job has failed when in fact to it ran OK).
- As vlapipe, move the contents of the project directory from the original workspaces directory for the imaging run somewhere safe to e.g. /lustre/aoc/sciops/<your dir>/old_<ws_request id> (just in case..). Note you need to keep them away from the paths workspaces uses (spool, image-qa in /lustre/aoc/cluster/pipeline/dsoc-prod/workspaces/...) otherwise ingest can fail.
- As vlapipe, copy the contents of the new (with selcalselfcal) products directory to the original.
- In the original directory, gunzip the weblog.
- Make a note in the weblog by manually making and editing the qa_notes file in the weblog that the image has been self-calibrated. Report the CASA version that was used for the self-cal process. Also add an internal note that self-cal was performed.
- Submit for phase-2 review as usual
- Clean up your working area once the job has passed QA and been ingested.
Info |
---|
run_casa.sh: #!/bin/sh #Don't put any commands before the #SBATCH options or they will not work # casa's python requires a DISPLAY for matplot, so create a virtual X server selfcal_script.py: # This CASA pipescript is meant for use with CASA 6.5.3 and pipeline 2023.1.0.42 |
Related articles
Content by Label | ||||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
...