TigerGraph-DevLabs / TG_elk

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

TigerGraph Logs with Kibana

  • Used for bare metal installations of TigerGraph
  • Can be used with any 3.x version of TigerGraph
  • Code for selecting which logs are read can be found in filebeat/config/filebeat.yml
    • This will need to be modified if the TigerGraph install is not using the default log location
  • Code for filtering logs can be found in logstash/pipeline/logstash.conf

Requirements

This app requires:

  • Bare metal install of TigerGraph
  • TigerGraph to be a 3.x version
  • Docker to be installed on the server

Commands for running ELK with TigerGraph

Start docker daemon:

$ sudo systemctl start docker

Start Docker containers:

$ sudo docker-compose up -d

Stop Docker containers:

$ sudo docker-compose down

The first time the containers are launched, you will have to reset the Kibana password. This will only be needed to log into the UI.

Reset Kibana password:

$ sudo docker-compose exec elasticsearch bin/elasticsearch-reset-password --batch --user elastic

Kibana Dashboard Creation

Once the app is up and running, a sample dashboard can be used for viewing the logs. To Install the sample dashboard:

  • Open Kibana at https://localhost:5601
  • Log in with credentials:
    • username: elastic
    • password: (password generated from command in previous section)
  • Use the search bar at the top of the page to navigate to the "Saved Objects" page
  • Click "Import" and choose the file found in kibanaDashboardSample folder of this repository
  • Navigate to "Dashboard" page
  • Select"MyExampleDashboard" from list of options

Once on dashboard page, time filter may need to be updated to show a larger time window for logs to appear


TigerGraph Log Sources

This is where you can find the log files within your TigerGraph install associated with the queries in case you wish to read the logs themselves

  • API endpoint logs: log/restpp/RESTPP#1.INFO
  • Graph Studio logs: log/gui/GUI#1.out

About

License:MIT License


Languages

Language:Shell 75.4%Language:Dockerfile 21.2%Language:Go 3.5%