5.3 KiB
Elasticsearch Recon Ingestion Scripts (ERIS)
A utility for ingesting various large scale reconnaissance data logs into Elasticsearch
The is a suite of tools to aid in the ingestion of recon data from various sources (httpx, masscan, zonefiles, etc) into an Elasticsearch cluster. The entire codebase is designed with asynconous processing, aswell as load balancing ingestion across all of the nodes in your cluster. Additionally, live data ingestion is supported from many of the sources supported. This means data can be directly processed and ingested into your Elasticsearch cluster instantly. The structure allows for the developement of "modules" or "plugins" if you will, to quickly create custom ingestion helpers for anything!
Prerequisites
- python
- elasticsearch (
pip install elasticsearch) - aiofiles (
pip install aiofiles) - aiohttp (
pip install aiohttp)
- elasticsearch (
Usage
python eris.py [options] <input>
Note: The <input> can be a file or a directory of files, depending on the ingestion script.
Options
General arguments
| Argument | Description |
|---|---|
input_path |
Path to the input file or directory |
--watch |
Create or watch a FIFO for real-time indexing |
Elasticsearch arguments
| Argument | Description | Default |
|---|---|---|
--host |
Elasticsearch host | http://localhost/ |
--port |
Elasticsearch port | 9200 |
--user |
Elasticsearch username | elastic |
--password |
Elasticsearch password | $ES_PASSWORD |
--api-key |
Elasticsearch API Key for authentication | $ES_APIKEY |
--self-signed |
Elasticsearch connection with a self-signed certificate |
Elasticsearch indexing arguments
| Argument | Description | Default |
|---|---|---|
--index |
Elasticsearch index name | Depends on ingestor |
--pipeline |
Use an ingest pipeline for the index | |
--replicas |
Number of replicas for the index | 1 |
--shards |
Number of shards for the index | 1 |
Performance arguments
| Argument | Description | Default |
|---|---|---|
--chunk-max |
Maximum size in MB of a chunk | 100 |
--chunk-size |
Number of records to index in a chunk | 50000 |
--retries |
Number of times to retry indexing a chunk before failing | 100 |
--timeout |
Number of seconds to wait before retrying a chunk | 60 |
Ingestion arguments
| Argument | Description |
|---|---|
--certs |
Index Certstream records |
--httpx |
Index HTTPX records |
--masscan |
Index Masscan records |
--massdns |
Index massdns records |
--zone |
Index zone DNS records |
This ingestion suite will use the built in node sniffer, so by connecting to a single node, you can load balance across the entire cluster. It is good to know how much nodes you have in the cluster to determine how to fine tune the arguments for the best performance, based on your environment.
GeoIP Pipeline
Create & add a geoip pipeline and use the following in your index mappings:
"geoip": {
"city_name": "City",
"continent_name": "Continent",
"country_iso_code": "CC",
"country_name": "Country",
"location": {
"lat": 0.0000,
"lon": 0.0000
},
"region_iso_code": "RR",
"region_name": "Region"
}
Changelog
- Added ingestion script for certificate transparency logs in real time using websockets.
--dry-runremoved as this nears production level- Implemented async elasticsearch into the codebase & refactored some of the logic to accomadate.
- The
--watchfeature now uses a FIFO to do live ingestion. - Isolated eris.py into it's own file and seperated the ingestion agents into their own modules.
Roadmap
- Fix issue with
ingest_certs.pyand not needing to pass a file to it. - Create a module for RIR database ingestion (WHOIS, delegations, transfer, ASN mapping, peering, etc)
- Dynamically update the batch metrics when the sniffer adds or removes nodes.