Skip to main content

Periodic maintenance

Work through this checklist regularly. The intervals are a starting point; adapt them to your plant.

Before you begin​

  • You can sign in on the home page as root, the system administrator who manages the users of every project.
  • You can sign in on the server with an administrator account, to read the log files in the installation folder.

Steps​

1. Check the processes and the logs​

Weekly. On the home page, open Processes and look at the Restarts column: a process that is restarted again and again has a problem that the restarts hide. See Manage the YTCC processes.

Read the warnings and errors in the logs in the logs folder of the installation, for example C:\ytcc\v15\logs, one file per day named scada-manager.YYYY-MM-DD.log. The files are kept for 10 days by default, so a weekly check sees every day. See Logs and diagnostics.

2. Check the disk and the history​

Monthly. In the configuration editor, open the archive table planner and compare the size and growth of each table with its limits. Look for disk and history alarms in the alarm history. See Archive tables and assignments and Disk and history protection.

3. Back up and try a restore​

After every configuration change, and at least monthly: back up each project with Back up… on the home page and its history database with the PostgreSQL tools, and keep the files off the server. The disk protection can delete old backups when the disk runs out of space.

Try a restore from time to time on a separate computer with the same YTCC version: import the backup there with Import on the home page, and check that the project opens and its screens work. Do not test a restore with Overwrite on the running server. See Backup and restore.

4. Review users and passwords​

Quarterly, and whenever someone leaves or changes role. In each project, disable the users who no longer need access, check the profiles of the others, and check that no default account still has its user name as password. Read the user audit log for failed sign-ins; it keeps about 10 MB, and older entries are replaced. See Profiles and users, Review audit logs and Server administrators and recovery.

5. Check the clocks​

Monthly. Check that the server still synchronises with its time server and that the device times of the RTUs are not marked invalid. See Time synchronisation (NTP).

6. Plan version updates​

Check which version you run in Settings › About on the home page. Plan each update with a backup and, for important plants, as a new version line installed next to the old one, which can be rolled back. After each update, go through the Security hardening checklist again. See Which version you are reading and Upgrade to a new version and roll back.

Result​

You have a record of each check, current backups stored off the server, and a restore that is known to work.