yi-jiayu / telegrammetry

Stats for Telegram

Telegrammetry

Stats for Telegram

Dependencies

• Python (tested with 3.7)
• Elasticsearch (tested with 6.2.4)
• Kibana (tested with 6.2.4)
• Filebeat (tested with 6.2.4)

Setup

1. Fill in your Telegram account details into telegram.ini.

2. Run messages.py and redirect its output to a file:

PYTHONUNBUFFERED=1 python messages.py > messages.log

Note: You need to run messages.py at least once first to set up your Telegram session on a new device with the code that will be sent to you.

3. Start Elasticsearch

4. Modify filebeat.yml and add your message log path under paths for the log input:

 # Paths that should be crawled and fetched. Glob based paths.
  paths:
    - /path/to/your/log/file

5. Start Filebeat and set the config path to the tgrm folder:

filebeat -e --path.config $PWD

The -e flag tells Filebeat to log to standard error.

6. Start Kibana.

7. Inside Kibana, create an index pattern for filebeat-, selecting timestamp as the Time Filter field name.

8. After creating a new index pattern, go to the "Management" tab and select "Index Patterns". The ID of the newly-created index pattern will be in the address bar just behind indices/. Copy it.

9. Inside visualisations.json, replace all IDs under kibanaSavedObjectMeta.searchSourceJSON.index with your index ID.

10. Go to "Saved Objects" under the "Management" tab and import your modified visualisations.json and dashboard.json.