nano SIEM

Getting help

Open a support case on app.nano.rs, say what went wrong, attach a support bundle, and see how support works

You can get help in three ways:

  • Open a support case on app.nano.rs. This is the fastest way to reach us about your deployment.
  • Email support@nano.rs (open email). When we recognize your address, the email becomes a case in My cases.
  • Ask the community on Discord. Do not post private details there.

Check the status page first. It shows the health of app.nano.rs and other nano services. See How support works for coverage hours and response targets.

Open a support case

  1. Sign in to app.nano.rs and open Support in the left rail. You can also select Get help at the bottom of any page.
  2. Under My cases, choose Add → Support.
  3. Choose the deployment. We fill in the nano version when we know it. Check it is right.
  4. Choose a severity:
    • SEV1: production is down, or data is at risk now.
    • SEV2: major degradation. For example, ingest stopped or search is unusable.
    • SEV3: minor degradation, or a workaround exists.
    • SEV4: a question.
  5. Fill in the form, attach a support bundle if you can, and choose Open case.

A support case is private. Only members of your organization and nano support can see it. It never appears on the public feature board, and no one can vote on it.

My cases lists your organization's support cases, bug reports and feature requests, newest first, with their status. We email the person who opened a case when we reply or change its status.

From nano's Help menu

In nano, open the Help menu and choose Report a bug or Get support. app.nano.rs opens with the form already filled in: a summary, what happened, the page you were on, the nano version, and recent failed requests. Sign in if asked. The details stay in your browser until you submit. Check every field and remove anything you do not want to send.

If the report was too long for the link, the form says so. Attach the full report that nano downloaded for you.

What to include

  • What happened, in your own words. Include the exact error text or status code.
  • What you expected to happen.
  • When it happened, with the timezone. For example 2026-10-28 20:30 EDT or a UTC timestamp. Say when it started, when it last happened, and whether it is still happening.
  • The exact names of affected things. Use the full name or path of the detection rule, log source, dashboard or saved search. "The Windows rule" is hard to find. Suspicious LSASS access (process_access) is easy.
  • How the deployment runs: open core or Self-run; Compose or Kubernetes. On Kubernetes, the chart version. MSSPs: name the affected client deployment.
  • Request ID: the ID shown in the error, the response headers or the failed network request.
  • For ingestion problems: the source slug, the sender and its version, the protocol, a redacted config, and one representative raw event with private data removed. Say whether the request was accepted and whether the event is searchable or unparsed.
  • A screenshot if the problem is visible on a page.

Keep passwords, API keys, ingest tokens and private client data out of the case text and screenshots.

Support bundles

On Self-run, nano support cannot reach your install. A support bundle shows us its state. Attach one to a support case or bug report. On a Self-run deployment, attaching a bundle also starts an automated first look that answers what it can; anything it can't resolve goes to an engineer.

Make a bundle

On Docker Compose, run it on any one host of the install. On a multi-host install it collects the other hosts over SSH.

cd ~/nano && scripts/nano support-bundle            # the last 24 hours
scripts/nano support-bundle --since 3d              # a longer window

On Kubernetes, run it from a machine with kubectl access to the release's namespace:

python3 nano/scripts/nano-support-bundle --kubernetes --namespace nano

On several servers, run it on the first app server: it collects every server in the inventory over SSH, so that server must reach every other one without a prompt (ssh -o BatchMode=yes <server address> true), which a fresh install doesn't set up. Without that, the bundle holds only the server it ran on and lists the others as unreachable; then run scripts/nano support-bundle --local-only on each server and attach every file.

It writes one .tar.gz file that only you can read. Older releases do not have this command. On those, use the bundle script from the nano-deploy agent skill, skills/nano-deploy/scripts/support-bundle.py. See Deploy with your AI agent.

What it collects

  • Versions, the edition, image locks and the install ID.
  • Host facts: operating system, CPU, memory, time sync and clock skew between hosts, disk use, and whether each data directory is its own mount.
  • Service state, health, restarts and out-of-memory kills.
  • Cluster health, and the roles of the database, cache and load-balancer services on a high-availability install.
  • ClickHouse diagnostics: recent errors, replication, read-only replicas, parts, disks, mutations and dictionaries.
  • Vector component errors and buffer metrics.
  • The nano services' own logs for the chosen window, size-capped.
  • Recent configuration changes: the time, action and object only.
  • On Kubernetes: pods, events, describe output, logs and the Helm release values.
  • The names of your configuration settings. Secret values are removed.

What it removes

Every file is redacted before it is written. The tool replaces secret configuration values, tokens, JSON web tokens, private keys and certificates, passwords in URLs, the license token, Authorization and Cookie headers, key=value secrets and ingest tokens. Logs from the ingest services keep only warning and error lines, with event fragments removed.

What it never collects

  • Your log events, or the raw logs your sources sent to nano.
  • Search or query results, lookup tables or IOC lists.
  • Secret configuration values, private keys or license files.

Review before you send it

Redaction is pattern-based, so read the bundle before you attach it:

tar -tzf nano-support-*.tar.gz
tar -xzf nano-support-*.tar.gz -O | less

Remove anything you do not want to share. Logs can contain host names, IP addresses and user names.

Attach it

On the case form, choose Support bundle. Or open an existing case and choose Support bundle. We accept .tar.gz, .tgz, .zip, .log, .txt and .json files up to 200 MB.

Bundles are stored privately. They are never public. Only members of your organization and nano support can download them, through a link that expires after five minutes. Every upload and download is logged.

We treat a bundle as data. Nothing in it is run.

Retention

  • Support bundles are deleted 90 days after the case is closed. The case keeps a note that a bundle existed.
  • Cases and their comments stay until you delete them.
  • If you delete a case, its bundles are deleted with it.

How support works

Support is asynchronous, by support case and email. There is no phone queue, on-call or 24×7 coverage, and no shared chat channel, unless your order form says otherwise.

  • Open a request: open a support case on app.nano.rs, or email support@nano.rs. Cases are private.
  • Coverage hours: Monday to Friday, 9am to 6pm US Eastern, excluding US federal holidays. For most of the year that is the afternoon and evening in London, the evening and night in India, and overnight in Sydney. Clocks change on different dates, so check the exact hours for your city.
  • AI first, then engineers: every request first gets an AI-driven response. It acknowledges the request, gathers diagnostics and context through the support case the way a tier 1 or tier 2 engineer would, and answers what it can. Anything it can't resolve goes to the engineers who build nano.
  • Support bundles: a support bundle shows us your install's state. The nano-deploy coding-agent skill can collect a support bundle and run diagnostics for you. See Deploy with your AI agent.
  • Dedicated support: under an order form, AI triage acknowledges every request within 15 minutes at any hour, Sev 1 and Sev 2 requests escalate automatically to on-call engineers, and an add-on adds 24/7 paged Sev 1 response. Talk to us.
  • Product questions? Ask the docs AI at nano.rs/docs/ask/.
  • Discord is community help. It has no response target.
  • Screen-shares are at our discretion. We schedule them in advance and keep them to an agreed length.

Response targets

A target is the time to a first response from a person, not the time to a fix. Only time inside the coverage hours counts.

SeveritySelf-runDedicated (order form)
Sev 11 business day4 hours
Sev 23 business days8 hours
Sev 35 business days2 business days
Sev 4Feedback portalFeedback portal
  • We set the severity, using the definitions in the SLA. We may change it as we learn more.
  • Self-run targets apply only to a Supported Configuration: a supported version and topology, the documented minimum requirements, and unmodified software.
  • Self-run Sev 1 means a production install can't ingest or search, there is no workaround, and the cause is in the nano software.
  • Your order form may set better targets.

What Self-run support covers

Support covers:

  • using the documented installer, Helm chart and upgrade steps;
  • licensing;
  • configuring nano;
  • finding and fixing defects in nano.

Support doesn't cover running or troubleshooting your own environment, even when it is what blocks an install or upgrade:

  • hardware, the operating system and its patching;
  • the container runtime and Docker networking;
  • Kubernetes;
  • storage and disk mounts;
  • network, firewall, proxy and DNS;
  • TLS certificates;
  • your identity provider;
  • backups and data recovery;
  • cloud accounts;
  • your AI provider.

If the cause is in your environment, we tell you and point you to the relevant docs. We may then close the request. Hands-on help with your environment is available as paid professional services, under a separate order form, at our discretion.

See the Service Level Agreement and the Self-Run Addendum for the full terms.

Onboarding help

On this page