Monitoring: servers, sites and certificates under watch
Know that a server or a website is no longer responding, or that a certificate is about to expire, before your customers notice.
Technical overview of a completed project, not marketed. La Passerelle was built and validated in-house; no access, purchase or subscription is offered. See the work.
Three kinds of monitoring
Servers: ping
Checks that a server answers on the network.
Websites: HTTP request
Queries a page of your site and interprets the response, to tell a healthy site from one in error.
Certificates
Tracks the validity of your certificates to avoid a surprise expiry.
Ping: checks that a TCP port answers, with an optional maximum latency. HTTP request: expected status codes (any 2xx by default), expected text in the response or the value of a JSON field. Certificate: alert before expiry (21 days by default) and invalid certificate. Frequency from 5 to 60 minutes; an outage is only declared after several failures in a row (2 by default).
In pictures
What it prevents
- an outage found by your customers or your teams rather than by you;
- a site that responds but shows an error;
- an expired certificate that triggers a security warning in browsers.
Cost in credits
| Check | Cost |
|---|---|
| Server ping | Free |
| HTTP request on a site | Free |
| Certificate check | Free |
Checks are free; only the alert messages sent use credits. See the credit model studied.
Frequently asked questions
How am I notified?
Through the flow you choose: at each change of state (down, degraded, restored), a message follows your steps and destinations (Slack, email, ticket, WhatsApp…).
Is the data of the monitored sites kept?
Message content is erased after processing. A monitor’s detail shows availability over 24 h and 30 days and the latest checks.
A similar need?
I design custom integrations and tools: describe your need in a few lines.