Network monitoring

A basic set goes a long way: is the equipment responding, how busy are the links, are errors building up on ports, and is there room where logs are written. That catches most problems early, and the list grows after every unexpected incident.

An on-call engineer, never a shared mailbox. An email read once a day is no better than no email. We tune thresholds to avoid a stream of false alarms, because once that starts people stop reacting.

Usually not. Monitoring runs on our own platform and is included in the managed IT agreement. If you must retain logs yourselves, for example because of NIS2, the data can also be forwarded to a system you own.