Build against what's arriving.
Tap a source for up to 60 seconds or 1,000 events, before your steps run, and use those events as input while you build. Live capture
Send your app, server and cloud logs to one place, search them fast, and keep them as long as you choose, on your own infrastructure. Teams leaving Splunk keep their HEC senders, and nothing is priced per seat or per query.
Use the senders you already run. nano parses what it recognizes, keeps every field of what it doesn't, and stores it all on your own disks.
Every log source has a pipeline: ordered steps that run before events are parsed and stored. Clean out the noise, strip secrets, fix formats and add context on live data, and nothing changes in production until you publish.
{"user":"ana","msg":"login ok","password":"hunter2""[REDACTED]"} Redact: password masked before storage user=ana msg="login ok" password=[REDACTED] team=identitylevel=ERROR msg="db timeout" host=pay-1 team=payments×3app=sshd user=root src_ip=203.0.113.9 owner=platformSee what's arriving
Tap a source for up to 60 seconds or 1,000 events, before your steps run, and use those events as input while you build. Live capture
Editing a draft doesn't touch production. Review a validated snapshot, publish it as a new version, and roll back to an earlier one if you need to. Build a pipeline
Drop health checks, dedupe repeated errors, sample chatty sources and redact secrets before anything reaches disk. Step reference
| nano | |
|---|---|
| Collect | HEC, OTLP/HTTP, JSON over HTTPS, syslog, Kafka, AWS S3/SQS and GCP Pub/Sub. Vector, Fluent Bit, Cribl and the OpenTelemetry Collector all forward to it. |
| Shape | Pipelines clean and reshape events as they arrive, before storage: filter, sample, dedupe, split, redact, mask or hash, rename, route and enrich from lookups. |
| Parse | Catalog parsers for nginx, Apache, systemd-journald, syslog, Kubernetes audit, AWS CloudTrail, ALB, S3 access and VPC Flow logs, GCP audit logs, OpenTelemetry logs and generic JSON. Anything else goes in as Generic JSON or syslog with every field kept, or you write a parser. |
| Search | nPL search on ClickHouse, field stats, saved searches and lookups. |
| Visualize | Dashboards and reports on any search. |
| Alert | Scheduled rules on nPL searches. |
| Keep | On disks you own, for the retention you set. |
| Pay | Open source is free. nano Self-run is per GB/day of ingest, from US$149/month at 5 GB/day. Nothing per seat, host or query. |
One number decides the price: GB/day of ingest. Add users, servers and searches without a new quote. Pricing
nano runs on your servers or in your cloud account. Log data is never sent to nano. What your install sends to nano
The same install runs detections today. nano Self-run adds cases, SOAR, hunts and pivt AI when a security team needs them.
Yes. Search, dashboards, saved searches and alert rules work on any logs you send, security or not. The security features stay out of your way until you want them.
The catalog has parsers for nginx, Apache, systemd-journald, syslog, Kubernetes audit, AWS CloudTrail, ALB, S3 access and VPC Flow logs, GCP audit logs, OpenTelemetry logs and generic JSON, and it keeps growing. Every install starts with zero log sources, and you choose which parsers to install. For anything else, use Generic JSON or Generic syslog, or write your own parser.
Send them as Generic JSON or Generic syslog. nano stores them and keeps every field that doesn't map under unmapped, plus the original line in raw_data, so you can still search them by field.
Yes. Every log source has a pipeline: ordered steps that run as events arrive, before they're parsed and stored. Filter out noise, sample, dedupe, split, redact or hash fields, rename, route, and enrich from lookups, or write a step in VRL. Capture live events to build against, and publish each change as a version you can roll back. Pipelines are in both editions; pivt's suggested changes need nano Self-run.
Yes. nano receives the HTTP Event Collector protocol, including the event, raw and acknowledgement endpoints, with a per-source ingest token. Point your HEC clients at nano and change the token.
The open-source edition is free on one server. nano Self-run is licensed per GB/day of ingest, from US$149/month at 5 GB/day, with every feature at every size. There's no charge per seat, host or query.
In our published benchmark at 100 GB/day, search p95 stayed between 253 and 318 ms on a small cluster while analysts, detections and dashboards all ran at once. The raw results are published with the benchmark.
No. nano installs with Docker Compose on one Linux server. nano Self-run adds high availability across several servers and Kubernetes when you need them.
Nowhere but your own infrastructure. nano runs on your servers or in your cloud account, and log data is never sent to nano.
From US$149/month at 5 GB/day, US$279/month at 10 GB/day (list price, billed monthly; 10% off billed annually). Every feature in one license, and one license covers two installs.