CentOS-6.6 Fail2Ban And Postfix Selinux AVCs

Home » CentOS » CentOS-6.6 Fail2Ban And Postfix Selinux AVCs
CentOS 2 Comments

I am seeing these in the log of one of our off-site NX hosts running CentOS-6.6.

type=AVC msg=audit(1421683972.786:4372): avc: denied { create } for pid”788 comm=”iptables” scontext=system_u:system_r:fail2ban_t:s0
tcontext=system_u:system_r:fail2ban_t:s0 tclass=rawip_socket
Was caused by:
Missing type enforcement (TE) allow rule.

You can use audit2allow to generate a loadable module to allow this access.

SELinux is preventing /sbin/iptables-multi-1.4.7 from search access on the directory .

***** Plugin catchall (100. confidence) suggests
***************************

If you believe that iptables-multi-1.4.7 should be allowed search access on the directory by default. Then you should report this as a bug. You can generate a local policy module to allow this access. Do allow this access for now by executing:
# grep iptables /var/log/audit/audit.log | audit2allow -M mypol
# semodule -i mypol.pp

I presume that the following is somehow related to that host sending out mail, possible by fail2ban, since we run postfix on that host and the sendmail SMTP package is not installed.

type=AVC msg=audit(1421683972.826:4376): avc: denied { read } for pid”796 comm=”sendmail” path=”inotify” dev=inotifyfs ino=1
scontext=system_u:system_r:system_mail_t:s0
tcontext=system_u:object_r:inotifyfs_t:s0 tclass=dir
Was caused by:
Missing type enforcement (TE) allow rule.

You can use audit2allow to generate a loadable module to allow this access.

SELinux is preventing /usr/sbin/sendmail.postfix from read access on the directory inotify.

***** Plugin leaks (86.2 confidence) suggests
******************************

If you want to ignore sendmail.postfix trying to read access the inotify directory, because you believe it should not need this access. Then you should report this as a bug. You can generate a local policy module to dontaudit this access. Do
# grep /usr/sbin/sendmail.postfix /var/log/audit/audit.log |
audit2allow -D -M mypol
# semodule -i mypol.pp

***** Plugin catchall (14.7 confidence) suggests
***************************

If you believe that sendmail.postfix should be allowed read access on the inotify directory by default. Then you should report this as a bug. You can generate a local policy module to allow this access. Do allow this access for now by executing:
# grep sendmail /var/log/audit/audit.log | audit2allow -M mypol
# semodule -i mypol.pp

We are nonetheless receiving fail2ban email notifications from that host. Therefore I am not sure what this avc is telling us.

We use Fail2Ban on a number of other Inetenet facing hosts. We have not yet detected anything similar on those, although until this morning’s descovery we never really looked for this situation specifically.

To check and see if anything was off kilter in the contexts I ran restorcon -Rv on /sbin and on /usr and restarted Fail2Ban. When I
reviewed the log file I found these in /var/log/messages.

setroubleshoot: [avc.ERROR] Plugin Exception restorecon #012Traceback
(most recent call last):#012 File
“/usr/lib64/python2.6/site-packages/setroubleshoot/analyze.py”, line
191, in analyze_avc#012 report = plugin.analyze(avc)#012 File
“/usr/share/setroubleshoot/plugins/restorecon.py”, line 99, in analyze#012 if avc.tpath[0] != ‘/’: return None#012IndexError:
string index out of range

Checking back I discovered that these first appeared in our log on January 4. Yum history indicates that there were updates on Jan 2 and Jan 8. The Jan 2 update was to Webmin alone. I doubt that had anything to do with this.

I am not sure what to think at them moment. Is there something wrong with SETroubleShoot?

I can work around these avcs with a local policy but in the event the issue is not our error this may have wider implications so I am posting the details here.

I am also seeing avcs relating to bash accessing ldconfig.

2 thoughts on - CentOS-6.6 Fail2Ban And Postfix Selinux AVCs

  • It appears that the starting date of these errors corresponds to the day on which we first began to jail SSH attempts on that host.

    We eventually ended up with a custom policy that looks like this:

    #============= fail2ban_t =============allow fail2ban_t ldconfig_exec_t:file { read execute open getattr execute_no_trans };

    allow fail2ban_t insmod_exec_t:file { read execute open };
    allow fail2ban_t self:capability { net_admin net_raw };
    allow fail2ban_t self:rawip_socket { getopt create setopt };
    allow fail2ban_t sysctl_kernel_t:dir search;
    allow fail2ban_t sysctl_modprobe_t:file read;

    allow system_mail_t inotifyfs_t:dir read;

    I am not sure whether this issue is the result of something that we have done or left undone. We have another host configured in much the same fashion as this one and it does not display these errors. On the other hand the second host was installed several years ago and has a number of custom polices already applied. It is possible that this problem was dealt with piecemeal or is submerged due to other customisations.

LEAVE A COMMENT