The Core Dump Overhead
When an application crashes on a modern Ubuntu system, the kernel automatically generates a “core dump”—a snapshot of the application’s working memory at the exact moment of failure. The systemd-coredump utility intercepts this data, compresses it, and saves it to your disk for debugging purposes. While core dumps are invaluable for software developers trying to trace complex bugs, they consume significant disk space and I/O bandwidth. If a background service is crash-looping repeatedly, systemd-coredump can rapidly fill your hard drive and degrade overall system performance.
Configuring Systemd Limits
If you are running a production server or a standard desktop where you have no intention of analyzing memory dumps, you can reconfigure systemd to stop capturing and storing these files. You must edit the main coredump configuration file using a terminal text editor.
Open a terminal window and type sudo nano /etc/systemd/coredump.conf, then press Enter. Scroll through the configuration file until you locate the line reading #Storage=external. The hash symbol means the system is using the default behavior. Delete the hash symbol to uncomment the line, and change “external” to “none” so the line reads exactly: Storage=none. Save the file (Ctrl+O, Enter) and exit nano (Ctrl+X). Finally, apply the new configuration by reloading the systemd daemon with the command: sudo systemctl daemon-reload. Ubuntu will no longer save crash dumps to your disk.