Oh Dear is an outside-in uptime product. It pings public sites, SSL, cron jobs, and DNS from a probe network. That is not a mechanic. Tink installs in one command, watches the Linux server, explains what is wrong, and helps you fix it.
Accidental sysadmins land on Oh Dear because “is the site up?” and “did cron run?” are the first questions. When nginx is down, the disk will fill in six days, or SSH is being brute-forced, Oh Dear stays quiet unless a public URL or scheduled task actually fails.
Tink is the other job: detect the issue on the machine, say why it happened in plain English, propose the command, and run it only after you approve. Keep Oh Dear if you need outside-in SSL, cron, DNS, and broken-link crawls. Use Tink if you run Linux servers and want a mechanic.
| Feature | Tink | Oh Dear |
|---|---|---|
| Setup time | 30 seconds (one curl | sh command) | Minutes — add sites, SSL, cron, DNS, and broken-link checks in the Oh Dear UI |
| What you get | Working monitoring, diagnosis, and approved fixes | External uptime, SSL, mixed content, broken links, cron, DNS, and domain expiry |
| Pricing | Free (Scout) / $9 / $29 per machine per month | Per-site SaaS — more sites, Lighthouse audits, and extra check types cost extra |
| Hidden costs | None — fully managed | Site count, application health endpoints, and still no diagnosis when the VPS is sick |
| Monitoring approach | Agent on the server — CPU, disk, services, logs, certs, ports | External HTTP, SSL, DNS, cron, and sitemap crawls from Oh Dear's probe network |
| Configuration | None after install — heuristics and AI | Per-site checks, notification channels, and optional application health URLs |
| Plain-English diagnosis | Yes — AI explains root cause, impact, and fix | Down/up, cert expiry, and cron missed — you still SSH in to learn why |
| Fix execution | Proposes and executes approved commands with an audit trail | Observability only — Oh Dear cannot restart nginx or free disk |
| Alerting | Built-in across 8 channels | Email, Slack, Discord, SMS, webhook, and similar after you wire contacts |
| Predictive alerts | Yes — disk fills in ~6 days, memory and CPU trends | Cert and domain expiry dates — not disk, memory, or CPU forecasts |
| SSH brute-force detection | Built-in — parses auth.log every scan | Not in scope — Oh Dear never sees auth.log |
| Machine offline detection | Agent presence monitoring with multi-channel alerts | Only if a public URL, health endpoint, or cron check fails |
| Public status page | Shareable URL with 90-day history | Yes — Oh Dear status pages are a first-class product |
| Weekly fleet digest | Automated Monday digest + daily brief when issues are open | Uptime and check reports — not a fleet health narrative |
| On-call tracking | Built-in /oncall command + incident acknowledgment | Add PagerDuty, Opsgenie, or another on-call tool |
| Conversation interface | Telegram, WhatsApp, web dashboard, CLI | Oh Dear dashboard and alert emails |
| Learning curve | None — works after install | Low for a single site; higher as check types and sites stack |
| Best for | Freelancers, small teams, accidental sysadmins (1-50 Linux servers) | Indie sites and agencies that only need outside-in uptime, SSL, cron, and DNS checks |
Keep Oh Dear when you already want that exact job:
The green check is already in the dashboard. A working ops loop for five Linux boxes is not:
For a 5-server team, Tink Mechanic at $45/month is cheaper than an Oh Dear plan plus the hours spent diagnosing every red check, and every scan includes a diagnosis an uptime ping will not type.
No per-site tax. No check-type maze. One command install.
Try Tink free — one command installAlso compare: Tink vs UptimeRobot · Tink vs Pingdom · Tink vs Datadog