status-im
/
es-retention-graphs
mirror of https://github.com/status-im/es-retention-graphs.git
2
0
Fork 0
Code Issues Projects Releases Wiki Activity
es-retention-graphs/README.md

1.7 KiB
Raw Permalink Blame History

Description

This Python script generates graphs visualizing peer retention in Status app.

Details

The script queries an ElasticSearch endpoint for logstash-* indices and aggregates counts of instances of log messages with set peer_id field.

This data is then analyzed using Pandas and graphed using Seaborn in the form of

This was built from a combination of a CSV generating script and a cohort analysis done by @jakubgs and @bgits.

Usage

The script provides a number of options:

Usage: main.py [options]

This generates a CSV with buckets of peer_ids for every day.

Options:
  -h, --help            show this help message and exit
  -H ES_HOST, --host=ES_HOST
                        ElasticSearch host.
  -P ES_PORT, --port=ES_PORT
                        ElasticSearch port.
  -i INDEX_PATTERN, --index-pattern=INDEX_PATTERN
                        Patter for matching indices.
  -f FIELD, --field=FIELD
                        Name of the field to count.
  -F FLEET, --fleet=FLEET
                        Name of the fleet to query.
  -m MAX_SIZE, --max-size=MAX_SIZE
                        Max number of counts to find.
  -d IMAGE_DPI, --image-dpi=IMAGE_DPI
                        DPI of generated PNG images.
  -o OUTPUT_DIR, --output-dir=OUTPUT_DIR
                        Dir into which images are generated.

Example: ./unique_count.py -i "logstash-2019.11.*" -f "peer_id"

Example