Docs / Monitoring

Monitors

#HTTP monitors

Pulsefast requests the URL and treats any 2xx or 3xx response as up, unless you set expected status codes (for example 200, 204). Redirects are followed, up to 5.

Setting	What it does
Check every	5 seconds to 1 hour. Your plan sets the fastest: 60s on Free, 30s on Starter, 10s on Pro, 5s on Team and Business (and during the 14-day trial).
Timeout	1 to 30 seconds. Slower responses count as failures.
Retries	Failed checks in a row before you're alerted (0–5). Failures are rechecked every few seconds and confirmed from another region.
Keyword	Text the response must contain, or must not contain (for example `"status":"error"`). Case-sensitive.
SSL warning	Days before the certificate expires to warn you. 0 turns it off.
While it's down, alert again	Repeat the alert every 5 minutes to 6 hours until it recovers.

New monitors start at your plan's fastest interval. Checks every 5 or 10 seconds default to a 5-second timeout and no retries: every failure is confirmed from another region before anyone is alerted, so retries add delay without adding safety. If a check is still waiting after 3 seconds, the other regions start checking right away, so even a site that hangs is alerted within about 15 seconds on 5-second checks.

#Request options

Open Request options on the monitor form to check APIs that need more than a plain GET:

Method: GET, HEAD, POST, PUT, PATCH, DELETE or OPTIONS.
Headers: one per line, as Name: value. Use them for tokens, such as Authorization: Bearer …. Header values are encrypted at rest and never shown again; leave the field empty when editing to keep them.
Body: up to 16 KB, for POST, PUT and PATCH. Add a matching Content-Type header.

Checks never reach private or internal addresses (10.x, 192.168.x, localhost and so on), even through redirects or DNS tricks.

#Port monitors

For databases, mail servers, game servers or anything that isn't a website: Pulsefast opens a TCP connection to host:port and closes it. Refused connections and timeouts are failures.

#Heartbeat monitors

For cron jobs, backups and background workers. Your job calls a URL when it finishes; if the call doesn't arrive on time, you're alerted.

Expected every: how often the job runs (1 minute to 7 days).
Grace time: how late a ping may be before it counts as missed (1 minute to 6 hours).

At the end of a cron job:

0 3 * * * /usr/local/bin/backup.sh && curl -fsS -m 10 --retry 3 https://pulsefast.io/hb/YOUR_TOKEN

To report a failure right away, call the URL with /fail at the end:

/usr/local/bin/backup.sh && curl -fsS -m 10 https://pulsefast.io/hb/YOUR_TOKEN || curl -fsS -m 10 https://pulsefast.io/hb/YOUR_TOKEN/fail

#Plan limits

If you move to a plan with fewer monitors, your oldest monitors keep running and the newest are paused until you upgrade or delete some. Intervals faster than the plan allows are slowed to its fastest.

View as Markdown · Something unclear? Tell us