For example, something that is too complex for your comfort level, a security concern, or maybe your hardware can’t keep up with the service’s needs?
Gladly, fail2ban exists. :) Note that it’s not just smtp anyway. Anything on port 22 (ssh) or 80/443 (http/https) get constantly tested as well. I’ve actually set up fail2ban rules to ban anyone who is querying / on my webserver, it catches of lot of those pests.
This method supposedly works great too.
CrowdSec has completely replaced fail2ban for me. It’s a bit harder to setup but it’s way more flexible with bans/statistics/etc. Also uses less ram.
It’s also fun to watch the ban counter go up for things that I would never think about configuring on fail2ban, such as nginx CVEs.
Edit: fixed url. Oops!
Thanks for mentioning it, I didn’t know about it. Protecting against CVEs sounds indeed awesome. I took a more brutal approach to fix the constant pentesting : I ban everyone who triggers a 404. :D Of course, this only work because it’s a private server, only meant to be accessed by me and people with deep links. I’ve whitelisted IPs commonly used by my relatives, and I’ve made a log parser that warns me when those IPs trigger a 404, which let me know if there are legit ones, and is also a great way to find problems in my applications. But of course, this wouldn’t fly on a public server. :)
Note for others reading this, the correct link is CrowdSec
