Processing Node Installation

Processing nodes need access to the data, as well as being able to run a processing script via ssh given LSDC collection ID and additional parameters, retrieve relevant experimental metadata, then run processing using standard data reduction programs.

Pre-requisites

  1. The most recent RHEL should be installed

  2. The following directories should be visible:

    • /nsls2/software/mx/daq

    • /nsls2/data/<tla> where <tla> = three-letter acronym for beamline (amx/fmx/nyx)

Procedure

Use the “Schedule Remote Job” item in Explorer for the selected node(s) to run the following roles:

  • Conda/”Conda - Install custom conda env (lsdc-processing)”

  • Miscellaneous/”mx_software”

If this computer has not been set up before, firewall configuration may be necessary for communication to happen to the amostra, analysisstore, and conftrak services on xf17id2-srv2.

  • Test this by doing the following while on the NSLS-II network - if a connection is made, then port is already available:

    telnet xf17id2-srv2 7770

  • Ask your AST lead to update firewall settings to allow communication from your new node to ports 7770, 7771, 7773 on xf17id2-srv2