I’ve got the system up and running, and am using the the custom bouncer as my system (Debian 11) is using firewalld. I’ve written a script for the bouncer that successfully adds IPs to an ipset in the drop zone of firewalld.
I’m using firewalld to add to the ipset due to loss of ‘state’ between firewalld and ipset if one adds/removes to the set via the ipset command manually. I’m also issuing a firewalld reload as part of the script.
I’ve got the following collections enabled, and am also using a custom parser for a zoneminder log (web_php). All seem to be working fine.
COLLECTIONS
------------------------------------------------------------------------------------------------------------
NAME 📦 STATUS VERSION LOCAL PATH
------------------------------------------------------------------------------------------------------------
crowdsecurity/haproxy ✔️ enabled 0.1 /etc/crowdsec/collections/haproxy.yaml
crowdsecurity/base-http-scenarios ✔️ enabled 0.5 /etc/crowdsec/collections/base-http-scenarios.yaml
crowdsecurity/apache2 ✔️ enabled 0.1 /etc/crowdsec/collections/apache2.yaml
crowdsecurity/linux ✔️ enabled 0.2 /etc/crowdsec/collections/linux.yaml
crowdsecurity/sshd ✔️ enabled 0.2 /etc/crowdsec/collections/sshd.yaml
------------------------------------------------------------------------------------------------------------
The thing I’ve noticed is that the crowd-sourced IPs that are being blocked – which is obviously data I want – is being fed to the custom bouncer one by one. Is this how it should be working? If so, what is the rate of the blocked IPs coming in? Or is there a larger list that could be front-loaded first, and then only use the incoming IPs after that?
I’ve noticed it’s inducing some load and am concerned about this as it will be running on a vps w/ a low amount of horsepower.
Thanks in advance