I use fail2ban and add detection (for example I noticed that after I implemented it for ssh, they started using SMTP for brute force, so had to add that one as well.
I also have another rule that observes fail2ban log and adds repeated offenders to a long term black list.