PROBLEM
A service “kind of works” on one box, nobody can rebuild it from notes, and logs land wherever they fell.
PODEJŚCIE
System baseline, network service configuration, log rotation and a procedure you can repeat after reboot or on a new host.
SCOPE
- install and baseline hardening (Linux / Windows Server)
- network services (DNS, web, files, mail in scope)
- logging, rotation and artefact location
- rebuild notes
DELIVERABLE
- configuration in agreed scope
- step-by-step rebuild note
- log path and verification method
PROCESS
- inventory host and dependencies
- agreed baseline and hardening
- service and functional test
- reboot verification
System administration starts where “it works on our machine” ends: kernel, time, disk, service, privileges, logs — in that order. Without that, SIEM and scripts only decorate mess on a single host.
I work on agreed Windows Server / Linux machines: baseline, network services, log rotation, rebuild documentation. The engagement goal: after reboot you return to a known state, and the next admin does not have to reverse-engineer chat history.
This is for deep work on a system, not whole-company IT on retainer. If you need order across many hosts and accounts, see IT administration.