DumbMonitFeatures
Everything it does
Alerting, agents, teams, status pages and the rest, in one container. What each kind of monitor reports is on the Monitors pages.
Alerts without the noise
Useful rules are on right after install. The work goes into sending fewer, better messages.
- Dependencies
Declare a switch as the parent of what sits behind it. When it goes down you get one notification, not thirty; the others read “Suppressed by parent”.
- Grouped by host
A NAS whose array degrades sends one message per cycle, not one per disk. Reminders, escalation, acknowledgement, snooze, per-monitor ignore and clearing resolved history are all built in. A bell on each monitor counts its active alerts.
- Baselines that learn first
Anomaly detection with no threshold to tune: 168 buckets, one per hour of the week. It stays silent for its first 14 days and only shows what it would have fired.
- Forecasts
“This filesystem will be full within four days” is listed as a forecast, apart from what is broken right now.
- Maintenance, silences and a notification policy
One-off or weekly windows, hysteresis, flap hold, quiet hours, batching and an hourly cap, so a bad night does not become two hundred pushes.
The path of an alert
34 notification channels
Each one has a Test button and a setup page in the documentation.
- Discord
- Slack
- Telegram
- Microsoft Teams
- Matrix
- Mattermost
- Rocket.Chat
- Google Chat
- Zulip
- ntfy
- Gotify
- Pushover
- Pushbullet
- Bark (iOS)
- Web Push (browser, PWA)
- Apprise
- Home Assistant
- Email (SMTP)
- Signal
- SMS via Twilio
- WhatsApp (Cloud API)
- WhatsApp (CallMeBot)
- Threema Gateway
- Webex
- SMS via Vonage
- SMS via Telnyx
- SMS via ASPSMS
- PagerDuty
- Opsgenie
- DumbMonit on-call
- Jira
- GLPI
- Zammad
- Custom webhook
An agent for what does not speak SNMP
A small binary that pushes CPU, memory, disks, network, services and uptime to the server, plus temperatures, fans, SMART and ZFS pools where the machine exposes them. It calls the server, never the other way round: no port to open on the machine.
# the UI fills in the token for you
curl -sSL http://server:8080/install.sh | sh -s -- \
--token=dmon_xxx --url=http://server:8080A PowerShell script is served at /install.ps1 for Windows. The agent is also an image, ghcr.io/laupernoe/dumbmonit-agent.
- LinuxSupportedx86_64 and aarch64, systemd or OpenRC. Docker containers and OS health (updates, pending reboot, failed units).
- WindowsSupportedx86_64, installed as a Windows service. SMART through smartctl.
- macOSPreviewlaunchd, Apple silicon and Intel; binaries attached to each release. Not yet run on real hardware.
- FreeBSDPreviewrc.d, x86_64, ZFS pools. Not yet run on real hardware.
Remote sites, outbound only
In relay mode an agent runs the server's probes (SNMP, Proxmox, HTTP…) from its own network — a second site, a client's network or a VPS behind NAT — with nothing inbound: no VPN, no port to open. Monitor a remote site
Remote site
The connection is opened from the remote site, outbound only: no VPN, no inbound port.
Teams, clients and on-call
What a small IT team or an MSP needs, without an extra product.
- Multi-client (MSP) mode
One instance for several clients: each has its own sites, folders, alerts, dashboards and SLA reports.
- Accounts
Admin, operator and viewer roles, TOTP and passkeys, an audit log, single sign-on over OpenID Connect or SAML, SCIM provisioning, and sign-in with LDAP or Active Directory. Account creation is off by default.
- On-call and escalation
On-call schedules, cover swaps and multi-step escalation per person. Teams and permissions per folder.
- Deploy with Active Directory
A ready-made kit installs the Windows agent on your domain servers, with the server address and port already filled in.
Multi-client (MSP) mode
Status pages for the people you serve
Public pages at /s/<name>: groups of services with 90 days of history, uptime, incident and maintenance notices, email subscribers with double opt-in, RSS, README badges and a compact embed. A page can also be private, behind a password or single sign-on.
Only the labels you choose leave the server. Monitor names, addresses and types never appear on the page, in a badge or in an email.
- On a phone
Read the state, acknowledge an alert, start a maintenance window. Installable on the home screen, deliberately without an offline mode: it never shows yesterday's state.
- Wall mode
A full-screen board for a room monitor: the verdict and the clock, the fleet figures, the health strip, open alerts and who is on call. Readable from across the room.
- Topology map
One picture of the whole infrastructure, drawn from what DumbMonit already knows: dependencies, relays, hypervisor guests and LLDP neighbours. Nothing to type in by hand. Read more in the documentation
Dashboards and reports
Drag-and-drop dashboards, ten templates that fill themselves with your devices, a metrics explorer, monthly SLA reports as PDF or CSV, and a short scheduled email with availability and incidents.
Security score
Every supported monitor gets a 0–100 score and an A–F grade from the vendor’s own best-practice checks, PingCastle-style: category, severity, evidence, remediation, and a worst-first Security page.
The score is a guide, not an audit: it only reads what DumbMonit already collects, so it never replaces a security review.
API and assistants
- Open, documented API
Every route behind
/api/openapi.json(OpenAPI 3.1): scoped tokens (dmt_…, read or write) that can expire, be restricted to a network, and are rate-limited. API reference - An assistant over MCP
27 tools for ChatGPT, VS Code or Cursor: read status, monitors, alerts, metrics and agents; a write token can also silence, acknowledge, add a monitor or restart a container. Secrets never come back. Connect an assistant
- Plays well with what you run
/metrics,/federateand a Prometheus-compatible endpoint let an existing Grafana read DumbMonit's data.