> For the complete documentation index, see [llms.txt](https://docs.layeronecloud.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.layeronecloud.com/network-monitor/architecture/retention-and-capture.md).

# Retention, capture and extension points

trimmed to 60 minutes by default with a per-site entry-count safety bound.

## Retention <a href="#retention" id="retention"></a>

* Two-second live interface and active-device values: compressed Redis streams, trimmed to 60 minutes by default with a per-site entry-count safety bound.
* Sparse one-minute aggregates: PostgreSQL, 30 days.
* Typed application usage and coverage minutes: PostgreSQL daily partitions, 30 days; there is no backfill from legacy port-only aggregates.
* Application client/service IP pairs and their classification summaries: 24 hours of queryable detail, expired rows removed by hourly maintenance.
* DDoS charts: up to seven days through the UI, backed by the same one-minute aggregate retention; live points follow the Redis live retention.
* Conflict and health events: PostgreSQL, 90 days.
* DDoS incident/source detail and packet evidence: 7 days.
* Device current state: retained until an operator archives it.
* Agent outage spool: 24 hours or 512 MiB, whichever is reached first.
* Voice alerts and their call attempts: PostgreSQL, alongside events.

Outside explicitly enabled DDoS evidence windows, raw packets and full hostnames are not durably stored or transmitted. Application IP pairs are minute aggregates with a 24-hour retention window, not a log of individual connection events. Normal application telemetry contains only minute aggregates, locally reduced registrable domains or canonical destination IPs, short-lived client/service IP pairs, ports, evidence types, and coverage. Evidence frames are truncated to 256 captured bytes and are bounded to 250 MiB or ten minutes per incident, four concurrent captures, and 1 GiB of sensor-side disk usage.

## Capture technology <a href="#capture-technology" id="capture-technology"></a>

The first release uses libpcap. On Linux it captures through the kernel packet socket interface and installs a BPF capture filter, giving a small, widely supported agent without a kernel module. eBPF and AF\_XDP can reduce copy cost at very high packet rates, but introduce kernel, verifier, driver, queue, and deployment constraints that are not justified before real Proxmox load tests. The aggregation and protobuf boundary lets the capture backend be replaced later without redesigning the control plane.

## Flow-exporter extension point <a href="#flow-exporter-extension-point" id="flow-exporter-extension-point"></a>

Arista sFlow/flow ingestion is intentionally not part of this first release. A future collector should normalize exporter samples into the same site, segment, device, host address, application, and port dimensions while preserving a distinct source and sampling-coverage value. It must not scale sampled flow data into billing records or silently mix it with Linux interface-counter totals.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://docs.layeronecloud.com/network-monitor/architecture/retention-and-capture.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.