alertloop watches your uptime, servers, containers and apps. When something goes down it reads your real logs, names the root cause, and restarts what it safely can across systemd, pm2 and Docker.
unknown directive "proxy_bufer_size" in /etc/nginx/conf.d/api.conf:42, introduced by the 02:11 deploy.You get the answer, not just the alert. Routine failures are fixed before anyone is paged; anything risky waits for your approval.
Checks from US and Europe, plus agent signals from systemd, pm2 and Docker.
AI reads your real logs and names the likely root cause.
Safe restarts with loop protection. Risky fixes wait for one-click approval.
Root cause by email, WhatsApp or push, and in the auto-heal feed.
Trusted by teams that keep the lights on
alertloop is the boring, opinionated stack that does the SRE basics well. Agents, monitors, patching, on-call, and observability, with every customer isolated in their own database.
5-minute install on Linux, Windows, macOS. Watches systemd, pm2 and Docker, and restarts what goes down. Signed commands, <25MB RAM.
HTTP, TCP, SSL, ping, DNS, process, log, JMX, SNMP — with multi-window burn-rate alerts and noise-killing dedup.
KEV-aware patch runs with maintenance windows, canaries, and automatic rollback on probe failure.
Rotations, overrides, follow-the-sun. Push, SMS, voice, Slack & Teams — fail over in 30 seconds.
Every customer gets a dedicated MySQL database, not a shared tenant_id row. That keeps audits simple.
Traces, spans, real-user metrics. Sub-second queries over months of telemetry — billions of events.
Uptime from US and EU, page speed tracking, bot detection, and SSL and domain expiry alerts, well before anything lapses.
Errors from your sites and apps, grouped automatically. See what's new, what keeps coming back, and what's getting worse.
Mobile app push, WhatsApp with deep links to the incident, and email through your own SMTP server and domain.
How alertloop compares to a plain uptime monitor, and what you get on the free tier.
Yes. You can monitor uptime, servers, SSL certificates and more on the free tier with no credit card. Unlike a pure uptime checker, alertloop also patches and auto-remediates, so it replaces several tools at once. See the full UptimeRobot and Site24x7 comparison.
Website and API uptime, HTTP, TCP, DNS and ping checks, SSL expiry, and servers (CPU, memory, disk, network, processes, logs) — plus full APM and real-user monitoring, all in one console.
Most tools just tell you something broke. alertloop reads your logs to find the root cause with AI, and can fix it automatically, with maintenance windows, canaries and automatic rollback.
No. The free tier installs in about five minutes and needs no card. You only upgrade when you're ready to scale to a bigger fleet.
Yes. Public and private status pages, SLA tracking, on-call rotations and escalations, with alerts over push, SMS, voice, Slack and Teams.
Every customer gets a dedicated database, not a shared table, and agents are mTLS-pinned. See the Security page for the details.
Yes. The agent tracks Docker container health and crash loops, discovers pm2 apps automatically, and shows every systemd service in one grid. If one goes down, alertloop finds the root cause in its logs and can restart it automatically.
Yes. Website monitoring covers uptime from the US and Europe, page speed, bot traffic, and SSL certificate and domain expiry alerts, so nothing lapses by surprise.
OpenAI, Anthropic Claude, Google Gemini, Groq and self-hosted Ollama. Free provider options are available, and you can use a different model for analysis and for chat.
Yes. Set a token limit per account, get alerts at 80% and 100%, optionally enforce a hard cap, and see what each analysis cost on a live usage meter.