Every server. Every change. One timeline.
Argus watches your servers and records everything that changes on them: packages, services, configuration files, deploys. When something breaks, the change that most likely caused it is already on the same timeline, with the reasons Argus thinks so.
Fleet ledger
Illustration of the home screen: each strip is a server, orange pins are changesTwo questions answer most outages
1 · What changed?
Every change on every server lands on one timeline next to its health. Argus scores the changes before a failure by timing, kind and whether they touched the failing server, and shows the score with its reasons.
2 · What's different?
A failing server compared with its healthy siblings: packages, services and configuration. Usually only one or two things differ, and Argus lists them on the server's page.
The verdict, labelled honestly
A change that happened just before the failure, on the same server, is usually the cause. Argus calls it a likely cause and keeps facts and inferences apart, so nobody acts on a guess by mistake.
Changes other tools never see
Most monitoring shows deploys if your CI reports them. Argus also sees changes made by hand, by package managers and by the system, because the agent watches the server itself.
| Change | How Argus sees it | Example |
|---|---|---|
| Packages | dpkg, dnf/yum and apk on Linux; installed programs on Windows | libssl3 3.0.13-0ubuntu3.4 installed |
| Services | systemd units on Linux, the service manager on Windows: stops, starts, restart loops | redis-server.service stopped |
| Configuration files | Paths you choose; Argus notices when they're modified | /etc/nginx/nginx.conf edited |
| Deploys | A signed webhook from GitHub Actions or any CI | web 1.4.3 deployed to web-1, web-2 |
| Reboots and OS updates | Boot time and kernel / Windows build | Windows build 20348.2700 → 20348.2762 |
What you get when you sign in
Fleet ledger
Every server's health over the last hour, day or week, with every change pinned above it. Click a server to see its charts, logs and changes.
Alerts that explain themselves
Thresholds, availability, log patterns and restart loops. Related alerts become one incident, and each incident lists ranked likely causes.
On call from your phone
The Argus Android app pages you, pages again if nobody answers, and protects acknowledge and resolve with your fingerprint.
Deploys from CI
Send deployments from GitHub Actions or any CI with a scoped key. They appear on the timeline and count as evidence.
Teams and roles
Owners, admins, responders and viewers; invitation links; an audit log of every change anyone makes in Argus.
Light and dark
Two themes designed separately for long sessions and control rooms. Argus follows your system until you choose.
Nothing listens on your servers
Outbound only
The agent connects out over TLS with a certificate unique to each server. Revoke one server without touching the others.
Least privilege
Role-based access, single-use invitation links, rate-limited sign-in and an audit trail for administration.
Your data, kept safe
Stored in Germany. Encrypted nightly backups are locked against deletion for 14 days, and restores are tested every week.
Starting with servers, built to grow
- Self-service sign-up with Google and GitHub, and guided setup that checks your first server is connected.
- Agent management from the app: collection policies by label, safe rollouts in rings, Windows agent.
- Kubernetes-aware peer groups and differences as evidence in incidents.
- Kubernetes, CI/CD pipelines and application monitoring on the same change timeline.