csf bogon bug

This forum is only for reproducible bugs with csf and lfd (i.e. not iptables problems, lack of understanding how to use a feature, etc). Posts must be accompanied with full technical details of the problem and how it can be recreated. Any posts not adhering to this, or not considered bugs, will be moved to the General Discussion (csf) forum.
Post Reply
marcele
Junior Member
Posts: 215
Joined: 17 Sep 2007, 17:02

csf bogon bug

Post by marcele »

With bogons enabled (LF_BOGON) it blocks interfaces even if they are in the interface "skip" list (using ETH_DEVICE_SKIP). So with it enabled it blocks connecting to internal IP's on eth1 even if eth1 shouldn't have rules applied to it.

Using csf version 4.36 on centos 5
Buccleuch
Junior Member
Posts: 16
Joined: 15 Jan 2009, 16:26

Post by Buccleuch »

Sorry, but it's not a bug per se...

From the config file:

Code: Select all

# Do NOT use this option if your server uses IP's on the bogon list (e.g. this
# is often the case with servers behind a NAT firewall using ip routing)
The only way bogons work along with using something like a 10.0.0.0/8 private network is if you place your public interface in ETH_DEVICE and place your private interface in ETH_DEVICE_SKIP.

This will cause csf to order the rules such that any traffic destined to the ETH_DEVICE_SKIP takes precedence over the bogon detection.

If ETH_DEVICE is blank, bogon detection will take precedence over ETH_DEVICE_SKIP.

At least that's how it worked in my limited experience, been testing and playing with CSF/LFD since only last night. :)
chirpy
Moderator
Posts: 3537
Joined: 09 Dec 2006, 18:13

Post by chirpy »

Buccleuch is correct, however I will look into the placement of the ETH_DEVICE_SKIP rules placement in the chains.
marcele
Junior Member
Posts: 215
Joined: 17 Sep 2007, 17:02

Thanks for the info

Post by marcele »

Thanks for the info guys. I wasn't aware that you had to explicitly add the device to ETH_DEVICE... I always had it blank and just added my internal eth1 device to ETH_DEVICE_SKIP. This was always working in the past. I just had this issue after updating csf .. something must have changed.
Post Reply