Documentation

Ping (ICMP Echo)

Sends native ICMP Echo Request packets (IPv4 or IPv6) to test basic reachability and measure round-trip latency.

• Ping count: packets sent per check — default 3, range 1–10.
• Package size: payload bytes per packet — default 56, range 0–65,527. Useful for MTU/fragmentation testing.
• Stop on success: sends pings one at a time and stops as soon as one succeeds, instead of always sending the full count.
• Results: minimum, maximum, and average latency plus packet-loss percentage.

TCP Connect

Opens a raw TCP socket and performs a connection handshake to verify a service is listening on a given port.

• Port: destination port — new tasks default to 22 (SSH); commonly used for 80/443 (web), 3306 (MySQL), 21 (FTP), 25/587 (SMTP), and any custom application port.
• Connect count: attempts per check — default 1, range 1–10.
• Stop on success: same early-exit behavior as Ping.

HTTP / HTTPS

Performs a full HTTP(S) request and evaluates the response — the most configurable check type, with an entire advanced-validation layer described below.

• Status codes: any 2xx response is a success by default; supply a custom comma-separated list (e.g. 200, 301, 302, 401) to override.
• Redirects: followed automatically up to 10 hops unless the task is set to treat the first response as final.
• Timeout: defaults to 10 seconds for connect + read combined.
• Custom User-Agent and request headers are supported (see Advanced Validation).

SNMP Telemetry

Polls managed network devices (routers, switches, NAS, printers) for diagnostic data using standard SNMP OIDs.

• Versions: v1, v2c (default for new tasks), and v3. Default port 161.
• v1 / v2c: authenticated with a community string you supply (there is no built-in default community).
• v3 security: username, security level (noAuthNoPriv / authNoPriv / authPriv), authentication protocol (MD5/SHA), and privacy protocol (DES/AES).
• Reboot detection: walks sysUpTime and flags the device if it has restarted since the last check.
• System resource monitoring: optionally walk and track additional system OIDs (CPU, memory, etc. — device-dependent).
• Interface monitoring: select specific network interfaces (via ifDescr/ifType/ifAlias) and get alerted if a monitored interface goes down or disappears from the device.

Check interval

How often the task runs, in minutes. Default is 30; there's no upper limit, so you can run checks as infrequently as once a day or as often as once a minute.

Only Wi-Fi / Specific SSID

Restrict a task to run only while on Wi-Fi, optionally locked to one specific network name. Ideal for home-lab or NAS monitors that are only reachable on your local network — checks are simply skipped (not failed) while off that network.

Notifications

Enabled by default. Choose whether to be alerted on Failure only, or on any Change in status (up→down and down→up).

High priority

Switches the task to Android's exact-alarm scheduling so it isn't deferred by Doze/battery optimization, and can trigger a distinct, louder alarm sound instead of a normal notification when it fails.

Tags

Free-text labels for organizing large host lists. Filter the dashboard instantly by tag, and a red "errors" chip appears automatically whenever any tagged task is failing.

Stop-on-success / stop-on-failure limits

Optionally auto-pause a task after N consecutive successes or failures (default 0 = unlimited/never auto-pause) — useful for one-off diagnostics or to stop hammering a host that's been down for a long time.

SSL/TLS certificate checks

• Ignore SSL errors: bypass certificate validation entirely — useful for internal servers with self-signed certs.
• Certificate expiry alert: fail the check when the server's certificate expires within a configurable window — default 14 days.

Regex content assertions

• Match regex: fails the check if the pattern is not found in the response body (case-insensitive).
• No-match regex: fails the check if the pattern is found — catch strings like "Internal Server Error" or "Database connection failed" even on a 200 response.

Hyperlink crawler

Parses the HTML response and recursively checks every <a href> link it finds, up to a configurable crawl depth (default 1 hop, up to 3), reporting any broken (non-2xx or timed-out) links on the page.

JavaScript validation

Run a custom JavaScript snippet against the response for complex, multi-step validation logic. The response body is available inside the script as the response variable. Return a string to report a specific failure reason, or null/nothing for success.

DNS / connect-to override

Route a request for one host:port to a different target address:port, similar to curl's --resolve. Useful for testing a specific backend node behind a load balancer, or hitting a staging server while keeping the original Host header.

Custom headers & authentication

Add any number of custom request headers, of three types:

• Generic: a plain header name/value pair.
• Basic Auth: username and password, sent as a standard HTTP Basic Authorization header.
• Generic Auth (secret): for bearer tokens or API keys — the value is masked in logs.

A default User-Agent and Accept/Accept-Language header set is applied automatically unless a task overrides them.

Max retries

Number of retry attempts after an initial failure, before the task is marked down — default 3.

Retry interval & exponential backoff

Base delay between retries — default 30 seconds. Enable Exponential Backoff to double the delay after each failed attempt (e.g. 30s → 60s → 120s), reducing load on a host that's already struggling. Retries are skipped automatically if the device itself has no network connectivity.

Notify after N failures

Debounce alerts so a single blip doesn't page you — HostGenie waits for a configurable number of consecutive failures (default 1, up to 20) before notifying.

No-internet notification

A separate, repeatable alert (default every 60 minutes) fires when the device itself loses connectivity — so you can tell "my phone is offline" apart from "the host is down."

Suspension intervals ("quiet hours")

Pause notifications during defined windows without pausing the checks themselves — logs and history keep recording. Enabled by default with a 22:00–04:00 window; add as many custom windows as you need.

Global pause / resume

A single switch in the app bar suspends scheduling for every task at once — handy during planned maintenance windows. Manually starting any task flips it back on.

Encrypted JSON export & import

Export every task and setting to a single .json file, protected with a password you choose. The password is run through Argon2id key derivation and the payload is encrypted with AES-256-GCM before being compressed to disk — strong enough to store the backup in general-purpose cloud storage. Import restores the exact configuration on any device running HostGenie.

Android Auto-Backup (opt-in)

Off by default to keep credentials from leaving your device. Enable it in Settings to let Android's own backup service sync your configuration to your Google account.

Log retention

Each task keeps up to 100 log entries by default. Optionally set an auto-clear age (in days) to prune older results automatically — disabled by default, so logs are kept until the limit is reached.

Bulk management

Multi-select tasks from the dashboard to pause, resume, or delete groups at once — handy when managing dozens of hosts by tag.

Material You dynamic color

On by default on supported devices (Android 12+) — HostGenie's palette adapts to your wallpaper-derived system colors. Turn it off in Settings to use the fixed HostGenie palette instead.

Light / dark / system theme

Follows the system theme by default; force Light or Dark independently of your device setting. Card style and loading-indicator style are also configurable for a more personalized dashboard.