Difference between revisions of "Fail2ban"

From ArchWiki
Jump to navigation Jump to search
(→‎Filesystem Access: Add tmp directory)
Line 16: Line 16:
Use the service unit {{ic|fail2ban.service}}, refer to [[systemd]] for instructions.
Use the service unit {{ic|fail2ban.service}}, refer to [[systemd]] for instructions.
You might need this package.. community/python2-systemd
== Hardening ==
== Hardening ==

Revision as of 22:37, 9 July 2014


Warning: Using an IP blacklist will stop trivial attacks but it relies on an additional daemon and successful logging (the partition containing /var can become full, especially if an attacker is pounding on the server). Additionally, if the attacker knows your IP address, they can send packets with a spoofed source header and get you locked out of the server. SSH keys provide an elegant solution to the problem of brute forcing without these problems.

Fail2ban scans various textual log files and bans IP that makes too many password failures by updating firewall rules to reject the IP address, similar to Sshguard.

Warning: For correct function it is essential that the tool parses the IP addresses in the log correctly. You should always test the log filters work as intended per application you want to protect.


Install fail2ban from the official repositories.

If you want Fail2ban to send an email when someone has been banned, you have to configure SSMTP (for example).


Use the service unit fail2ban.service, refer to systemd for instructions.

You might need this package.. community/python2-systemd


Currently, fail2ban requires to run as root, therefore you may wish to consider some additional hardening on the process with systemd. Ref:systemd for Administrators, Part XII


For added security consider limiting fail2ban capabilities by specifying CapabilityBoundingSet in the drop-in configuration file for the provided fail2ban.service:


In the example above, CAP_DAC_READ_SEARCH will allow fail2ban full read access, and CAP_NET_ADMIN and CAP_NET_RAW allow setting of firewall rules with iptables. Additional capabilities may be required, depending on your fail2ban configuration. See man capabilities for more info.

Filesystem Access

Also considering limiting file system read and write access, by using ReadOnlyDirectories and ReadWriteDirectories, again under the [Service] section. For example:

ReadWriteDirectories=/var/run/fail2ban /var/lib/fail2ban /var/spool/postfix/maildrop /tmp

In the example above, this limits the file system to read-only, except for /var/run/fail2ban for pid and socket files, and /var/spool/postfix/maildrop for postfix sendmail. Again, this will be dependent on you system configuration and fail2ban configuration. The /tmp directory is needed for some fail2ban actions. Note that adding /var/log is necessary if you want fail2ban to log its activity.

SSH jail

Edit /etc/fail2ban/jail.conf and modify the ssh-iptables section to enable it and configure the action.

If your firewall is iptables:

enabled  = true
filter   = sshd
action   = iptables[name=SSH, port=ssh, protocol=tcp]                                         
           sendmail-whois[name=SSH, dest=your@mail.org, sender=fail2ban@mail.com]
logpath  = /var/log/auth.log                                                                    
maxretry = 5

Fail2Ban from version 0.9 can also read directly from the systemd journal by setting backend = systemd.

If your firewall is shorewall:

enabled  = true
filter   = sshd
action   = shorewall
           sendmail-whois[name=SSH, dest=your@mail.org, sender=fail2ban@mail.com]
logpath  = /var/log/auth.log                                                                    
maxretry = 5
Note: You can set BLACKLISTNEWONLY to No in /etc/shorewall/shorewall.conf otherwise the rule added to ban an IP address will affect only new connections.

Also do not forget to add/change:


in your /etc/ssh/sshd_config. Else, password failures are not logged correctly.

See also