A server that’s “working, so leave it alone” always breaks at the worst possible moment. Here’s what should be done every month to prevent that.
1. OS and software updates
Windows Server, Linux, the 1C platform, the database, the web server — they all get regular security updates. Install them deliberately: take a server snapshot first, then update, then check that everything works.
2. Backup verification
Not “backups are running,” but “the backup actually restores.” Once a month, do a test restore of the database or a few files to a separate machine. At the same time, check that backups aren’t growing out of control and are stored off the server.
3. Free disk space
A full disk is the most common reason 1C and databases stop working. Logs, temp files, old backups — all of this needs cleaning up, and disk usage growth needs tracking so you can order more space in advance.
4. Login and security logs
Who logged into the server, from where, and how many failed attempts there were. Hundreds of failed RDP logins mean someone is brute-forcing a password, which should be blocked with a firewall and login attempt limits. Former employees’ accounts must be disabled.
5. Passwords and access
Review user accounts: remove unneeded ones, revisit permissions, enable two-factor authentication on admin accounts. Passwords belong in a secure vault, not in a file on the desktop.
6. Load and performance
Review CPU, memory and disk graphs for the month. If the server is consistently loaded above 70–80 %, it’s time to add resources; if it’s constantly idle, you can pay less.
7. Certificate and domain expiry
An expired SSL certificate “breaks” your website and email at the worst moment. Check expiry dates and set up auto-renewal or reminders.
8. Monitoring health
Are alerts coming through? To the right chat? Do the right people see them? Once a month it’s worth running a “fire drill” — stop a test service and make sure the alert fires.
9. Documentation
An infrastructure diagram, a list of services and ports, where each backup is stored and how to restore. If your administrator quits tomorrow, the new one should get up to speed in a day, not a month.
10. Report
A short report to management: what was done, what was found, what we recommend. Without a report, administration becomes a “black box,” and its value isn’t visible until the first outage.
If there’s no one to do this
This exact list is included in our administration packages: updates, backup verification, monitoring, security and a monthly report. The first step is a free server audit, after which you get the same checklist marked with what’s already done and what isn’t.
Need help with the task from this article?
We’ll pick a configuration, calculate the cost and help with the move.