N2T Python Implementation

This is a reimplementation of N2T using python.

N2T is an identifier scheme resolver. Its role is to redirect clients to the registered handler for an identifier scheme. The handler is expected to continue interaction with the client for identifier resolution.

Manual Deploy on Amazon Linux 2023

The steps to deployment are broadly:

1. Setup Nginx Unit as the web server
2. Install the n2t python application
3. Configure the n2t application with identifier schem information

Setup Nginx Unit as the web server

On server instance:

Given the YUM repository description file unit.repo:

[unit]
name=unit repo
baseurl=https://packages.nginx.org/unit/amzn/2023/$basearch/
gpgcheck=0
enabled=1

Add the repository and install Nginx unit:

sudo yum update
sudo yum install yum-utils
sudo /usr/bin/yum-config-manager --add-repo unit.repo
sudo yum update
sudo yum install unit
sudo yum install unit-devel unit-jsc17 unit-python311 unit-wasm
sudo cdlsysctl start unit

Note: The service is installed as disabled and needs to be enabled, but cdlsysctl doesn't allow that so needs to be done by an administrator, e.g.:

sudo systemctl enable unit

Is the service running?

sudo cdlsysctl status unit
● unit.service - NGINX Unit
     Loaded: loaded (/usr/lib/systemd/system/unit.service; disabled; preset: disabled)
     Active: active (running) since Wed 2024-02-28 04:32:23 PST; 4s ago
   Main PID: 1893665 (unitd)
      Tasks: 3 (limit: 1055)
     Memory: 5.7M
        CPU: 16ms
     CGroup: /system.slice/unit.service
             ├─1893665 "unit: main v1.32.0 [/usr/sbin/unitd --log /var/log/unit/unit.log --pid /var/run/unit/unit.pid --no-daemon]"
             ├─1893667 "unit: controller"
             └─1893668 "unit: router"

To enable access to unit by the ezid user, it is necessary to override the default systemd startup for unit. This is done with an override.conf file in /etc/systemd/system/unit.service.d:

[Service]
Environment="UNITD_OPTIONS=--log /var/log/unit/unit.log --pid /var/run/unit/unit.pid --user ezid --group ezid"
Group=ezid
ExecStartPost=/usr/bin/sh -c 'sleep 3; chmod 660 /var/run/unit/control.sock; chmod 640 /var/log/unit/*.log'

Show the current unit configuration:

$ curl -X GET --unix-socket /var/run/unit/control.sock http://localhost/config/
{
    "listeners": {},
    "routes": [],
    "applications": {}
}

Install the n2t python application

Setup python and application. Installing the application is done by cloning from the GitHub repository. From the application home (i.e. /ezid on UC3 systems):

git clone https://github.com/CDLUC3/n2t.git
cd n2t
python3.11 -m venv venv
source venv/bin/.activate
python --version
  Python 3.11.6
python -m pip install -U pip
python -m pip install -e .

Run the n2t cli to verify the application is installed:

$  n2t
Usage: n2t [OPTIONS] COMMAND [ARGS]...

  Management commands for N2T.

  This script is used for initializing the JSON representation of scheme
  records from the original YAML and creating or updating the sqlite store
  used by the resolver application.

Options:
  -c, --config FILE
  --help             Show this message and exit.

Commands:
  info       Print application version and basic status.
  loaddb     Load or update the scheme database.
  yaml2json  Generate or update JSON record from YAML source.

Configure the n2t application with identifier schem information

Before operation, the N2T application scheme database must be initialized. This process creates an sqlite database containing the scheme records to support resolution. The scheme records are loaded from the JSON sources, and the database will need to be updated when the JSON sources are modified.

The N2T application is configured using a .env file. The available properties include:

*: Default folder locations are relative to the n2t folder in the cloned repository.

A typical configuration file for a staging environment will be:

N2T_DB_CONNECTION_STRING="sqlite:///data/n2t.sqlite"
N2T_JSON_DIR="schemes"
N2T_ENVIRONMENT="staging"

Those settings indicate the name of the environment "staging", that the JSON scheme definitions are in the folder schemes and the application database is data/n2t.sqlite. The folders are relative to the location where the n2t command is run. These properties may also use absolute paths, for example:

N2T_DB_CONNECTION_STRING="sqlite:////ezid/var/n2t.sqlite"
N2T_JSON_DIR="/ezid/n2t/schemes"

Any of the settings are overridden by an equivalent named environment variable if set.

If the schemes folder is not present it needs to be created from the legacy N2T prefixes file:

wget "https://legacy-n2t.n2t.net/e/n2t_full_prefixes.yaml"
n2t yaml2json

The application database may be created and populated when the JSON scheme records are available:

cd /ezid/n2t
mkdir -p data
n2t loaddb

To verify setup, use the info command. The output should be something like:

$ n2t info
{
  "version": "0.5.0",
  "environment": "development",
  "status": "initialized",
  "description": "n2t_full_prefixes",
  "created": "2024-03-06T16:01:43.245204",
  "schemes": {
    "total": 2774,
    "valid": 1596
  }
}

Managing Scheme Records