Welcome to HEAL Docs!
< All Topics
Print

Telemetry Ingestion and GenAI Forensic Analysis

Configuration

Follow the below configuration steps:

1. Before running migrations, remove any existing rows for relevant agents from the table instance_command_mapping.

2. Migrate the database with the shared SQL files.

3. Update the following Consul keys:

service/datareceiver/cache/worker/thread/queue/size 800
service/datareceiver/cache/worker/thread/size 100
service/datareceiver/opensearch/connection/io/reactor/size 2
service/datareceiver/redis/connection/timeout/secs 5
service/datareceiver/redis/socket/timeout/secs 30
service/datareceiver/rmq/queue/max/size 10000
service/datareceiver/sink/rmq/data/interval/milliseconds 5000

4. Update the lines below in haproxy.cfg and reload HAProxy to apply changes.

acl url_dr_http7 path_beg -i /external-data
use_backend data_receiver_http_backend if url_dr_http7

5. Place the provided Running_Process_Details artifact into the Forensic Agent identifier folder.

6. Update the instance_command_mapping table by setting status = 1 (based on agent identifiers) to enable process running metrics collection and store the data in OpenSearch.

7. OpenSearch index naming for process running metrics is heal_raw_external_data_${year.month.date}.

8. Update the ISM policy ism_heal_external_data_indexes by changing min_index_age from 180d to 7d.

Note:

* Approximate OpenSearch CPU overhead for processing process running metrics: 40%

* Approximate OpenSearch memory overhead: 5%

* With min_index_age set to 7d, approximately 856 GB is required for one week of data.

* With min_index_age set to 180 days, storage is approximately 5.1 GB per hour, so 122 GB per day, and 22 TB for 180 days.

* Observed with a process line output of approximately 5,000 lines.

Previous Section
Was this article helpful?
Please Share Your Feedback
How Can We Improve This Article?
Table of Contents
Scroll to Top