Short answer: Check the size with journalctl --disk-usage, trim it once with sudo journalctl --vacuum-size=500M (or --vacuum-time=14d), then set SystemMaxUse=500M under [Journal] in /etc/systemd/journald.conf and run sudo systemctl restart systemd-journald. Vacuuming deletes only old archived journal files, but that log history is gone afterwards.
1. See how much space the journal uses
Run journalctl --disk-usage. It prints the total size of the archived and active journal files. The files normally live in /var/log/journal (persistent) or /run/log/journal (kept in memory only).
If the directory /var/log/journal does not exist, your system may keep the journal in memory only, and the size you see is limited by RAM, not by disk.
2. Trim it once
To keep the journal at or below a size, run sudo journalctl --vacuum-size=500M. To keep only recent entries, run sudo journalctl --vacuum-time=14d. Both remove old archived journal files; the logs they contained cannot be recovered afterwards.
Run journalctl --disk-usage again to confirm. Vacuuming removes whole archived files, so the result may be a little above or below the target.
3. Set a permanent limit
Open /etc/systemd/journald.conf (or create a drop-in file in /etc/systemd/journald.conf.d/) and set SystemMaxUse=500M under the [Journal] section. If you also want to keep some free space, SystemKeepFree=1G is another option. SystemMaxUse limits the persistent journal; RuntimeMaxUse does the same for the in-memory journal.
Apply it with sudo systemctl restart systemd-journald, then check journalctl --disk-usage after a while.
4. Decide what you need to keep
A smaller journal means a shorter history for debugging. If you need a longer history of one service, ship its logs to another place instead of keeping everything on the server.
Look at which service writes the most: journalctl -b -p warning --no-pager | tail shows recent warnings, and a service that logs thousands of lines per minute is worth fixing at the source.