GRACC Request Daemon

This daemon listens and responds to requests for replays on a AMQP queue.

Installing

It is easy to install the GRACC Request Daemon with virtualenv

virtualenv gracc-test
. gracc-test/bin/activate
pip install -r requirements.txt
python setup.py install

Docker Installation

A docker image with gracc-request installed in available as opensciencegrid/gracc-request.

Corrections

The configuration for a correction is

[[Corrections]]
index = 'gracc.corrections'
doc_type = 'project'
match_fields = ['ProjectName']
source_field = 'CorrectedProjectName'
dest_field = 'ProjectName'

Each correction is required to have the above fields. You can imagine the logic as:

• If the match_fields match the incoming record
• Take the value in source_field and put it in dest_field

The Elasticsearch document for the above configuration could be:

{
  "_index": "gracc.corrections-0",
  "_type": "project",
  "_id": "AV0ItRBqFBezTrseiOVF",
  "_score": 1,
  "_source": {
    "ProjectName": "osg.mab",
    "CorrectedProjectName": "mab"
  }
}

In the above example config and document, the correction would look for records which have a ProjectName == "osg.mab", and set the new ProjectName to "mab".

Regular expression matches

The corrections also have the ability to perform regular expression matches. The configuration for a regular expression:

[[Corrections]]
index = 'gracc.corrections'
doc_type = 'host_description_regex'
match_fields = ['Host_description']
source_field = 'Corrected_OIM_Site'
dest_field = 'OIM_Site'
regex = true

And the ES document would look like:

{
  "_index": "gracc.corrections-0",
  "_type": "host_description_regex",
  "_id": "asldkfj;alksjdf",
  "_score": 1,
  "_source": {
    "Host_description": ".*\.bridges\.psc\.edu",
    "Corrected_OIM_Site": "PSC Bridges",
  }
}

In this case, it would match the Host_description field of the incoming record with the regular expression in the ES record. If it is a match, then it would set OIM_Site to the value in Corrected_OIM_Site.

ID Mappings

InstitutionID

All project and facilities have a InstitutionID assigned by the OSG. You can find these values in the OSG Institutions Webpage.

These IDs hold the pointer to the facilities accepted name and some metadata that can be pulled in later.

A sample of this metadata can be seen below.

{
    "name": "Albert Einstein College of Medicine",
    "id": "https://osg-htc.org/iid/yzcm7hs9f1d0",
    "ror_id": "https://ror.org/05cf8a891",
    "unitid": "385415",
    "longitude": -73.846327,
    "latitude": 40.852847,
    "ipeds_metadata": {
        "website_address": "www.einsteinmed.edu/",
        "historically_black_college_or_university": false,
        "tribal_college_or_university": false,
        "program_length": "Four or more years",
        "control": "Private not-for-profit",
        "state": "NY",
        "institution_size": "1,000 - 4,999"
    }
}

FieldOfScienceID

Field of Science IDs are pulled from the NSF created SED-CIP list. This list is a comprehensive list of all fields of science that are recognized by the NSF along with three tiers of categorization.

You can find the Excel mapping spreadsheet here -> https://ncses.nsf.gov/pubs/nsf24300/assets/technical-notes/tables/nsf24300-taba-005.xlsx

In the future if the SED-CIP list is updated there is historical precedent for them providing a mapping file to the new ids.