Hi Ben,

We don't have general-purpose cron-like scheduling for running pipeline jobs within LabKey Server.

The way that R scripts are automatically able to authenticate to the server based on the user that launched them is unique to R. It's most likely possible to enable this for other external tools, like Perl, but since the scripts call to the server using HTTP requests, it requires work to get them to use the correct HTTP session ids.

Thanks,
Josh

ok. so when i set an r script to 'run in the background as a pipeline job', what exactly happens?

In this case, the credentials of the user that created the script are used to run the job. A temporary subdirectory is created off the pipeline root that is used to run the script and to save any script outputs for rendering at a later time.

what determines when it's run and how often does that occur?

It gets tossed onto the pipeline queue when a request is made to render the report. There's currently no scheduling component to it.

It was originally added as a way to render potentially long running R reports asynchronously and was envisioned to be used in conjunction with the R report caching checkbox (which is only available in studies).

ok, i think i get the idea. is the idea behind the pipeline/result caching approach that some large script will run, then other scripts will consume these results?

if so, how do i best get other scripts to use the result of the pipelined script? in our case, I have a time consuming script that runs to produce a large matrix describing kinship coefficients across all colony members. this data rarely changes. I'd like to cache these results such that other scripts can simply take this matrix and perform calculations with it. how would i best go about this? can i save the matrix to a file then get other scripts to import it? where do cached results get saved?