audit-logging-framework

An audit logging framework for Relational databases.

Prerequisite

You must have following softwares already setup-

1. kafka (Pub-Sub model messaging system, a Big Data tool)
2. zookeeper (It is already shipped with kafka. Kafka uses ZooKeeper to manage the cluster)
3. mongodb (To store audit logs data)
4. node and npm (development and build environment)

Working

This app has been developed to work with Relational databases. Its working can be explained in following steps-

To create an audit-log: you hit an API http://localhost:5000/auditLog

method: POST

request:

{
  "tableId": 1,
  "rowId": 1,
  "oldData": {
    "id": 1,
    "name": "varun",
    "age": 22,
    "profession": "engineer"
  },
  "newData": {
    "id": 1,
    "name": "varun kumar",
    "profession": "software developer"
  },
  "ipAddress": "localhost",
  "user": {
    "email": "[email protected]"
  },
  "miscellaneous": {
    "platform": "web",
    "os": "linux"
  }
}

Here oldData and newData are complete data of a row in table. user is user details and miscellaneous is any other data that you want to store in logs. tableId and rowId are unique identifiers for auditLogs. You can assign all your tables a unique id. This data is pushed to kafka for processing. Processing is basically calculating diff of oldData and newData and creating a new payload. This new payload is then stored to mongodb. Stored data would look like-

{
  "_id" : ObjectId("5baf76b53b471c41857a1a0d"),
  "rowId" : 1,
  "tableId" : 1,
  "updatedAt" : ISODate("2018-09-29T12:57:25.540Z"),
  "__v" : 0,
  "createdAt" : ISODate("2018-09-29T12:57:25.540Z"),
  "changeHistory" : [
    {
      "date" : ISODate("2018-09-29T12:54:20.137Z"),
      "_id" : ObjectId("5baf76b5e8480034fa2197ce"),
      "ipAddress" : "localhost",
      "user" : {
        "email" : "[email protected]"
      },
      "miscellaneous" : {
        "platform" : "web",
        "os" : "linux"
      },
      "log" : {
        "name" : "varun kumar",
        "profession" : "software developer"
      }
    }
  ]
}

If you hit http://localhost:5000/auditLog again with same tableId and rowId, logs would be pushed to changeHistory. When you create a new record in table (in your relational database), you pass oldData as {} (empty object). In that case all of newData would be pushed to changeHistory.

To retrieve audit-log: you hit same API http://localhost:5000/auditLog

method: GET

queryParams: tableId, rowId, count

example-

GET http://localhost:5000/auditLog/?tableId=1&rowId=1&count=5

In response you would get changeHistory of given tableId and rowId with no of results = count.

The data would be sorted according to data i.e. latest 5 results

How to install

1. Clone repo git clone https://github.com/varunon9/audit-logging-framework.git
2. Go inside repo cd audit-logging-framework
3. Edit config/index.js for connection details.
4. Go to kafka installation directory cd /opt/kafka_2.11-1.0.0
5. First start zookeeper sudo bin/zookeeper-server-start.sh config/zookeeper.properties
6. Start kafka sudo bin/kafka-server-start.sh config/server.properties
7. You will have to create kafka-topic (name mentioned in config/index.js)
8. create topic named audit-logging bin/kafka-topics.sh --create --zookeeper localhost:2181 --replication-factor 1 --partitions 1 --topic audit-logging
9. Now come back to project installation directory and start app npm start
10. Log data from your service class i.e. when data is created/updated in table. POST localhost:5000/auditLog/
11. Retrieve data when showing audit-logs. GET localhost:5000/auditLog

Blog

https://medium.com/@varunon9/an-audit-logging-framework-for-relational-databases-f902550cd849