|Main index||Section 8||Options|
The tcpd program can be set up to monitor incoming requests for telnet, finger, ftp, exec, rsh, rlogin, tftp, talk, comsat and other services that have a one-to-one mapping onto executable files.
The program supports both 4.3BSD-style sockets and System V.4-style TLI. Functionality may be limited when the protocol underneath TLI is not an internet protocol.
Operation is as follows: whenever a request for service arrives, the inetd daemon is tricked into running the tcpd program instead of the desired server. tcpd logs the request and does some additional checks. When all is well, tcpd runs the appropriate server program and goes away.
Optional features are: pattern-based access control, client username lookups with the RFC 931 etc. protocol, protection against hosts that pretend to have someone elses host name, and protection against hosts that pretend to have someone elses network address.
In order to find out where your logs are going, examine the syslog configuration file, usually /etc/syslog.conf.
tcpd verifies the client host name that is returned by the address->name DNS server by looking at the host name and address that are returned by the name->address DNS server. If any discrepancy is detected, tcpd concludes that it is dealing with a host that pretends to have someone elses host name.
If the sources are compiled with -DPARANOID, tcpd will drop the connection in case of a host name/address mismatch. Otherwise, the hostname can be matched with the PARANOID wildcard, after which suitable action can be taken.
In order to monitor access to the finger service, move the original finger daemon to the "other" place and install tcpd in the place of the original finger daemon. No changes are required to configuration files.
# mkdir /other/place # mv /usr/etc/in.fingerd /other/place # cp tcpd /usr/etc/in.fingerd
The example assumes that the network daemons live in /usr/etc. On some systems, network daemons live in /usr/sbin or in /usr/libexec, or have no `in. prefix to their name.
In order to monitor access to the finger service, perform the following edits on the inetd configuration file (usually /etc/inetd.conf or /etc/inet/inetd.conf):
finger stream tcp nowait nobody /usr/etc/in.fingerd in.fingerd
finger stream tcp nowait nobody /some/where/tcpd in.fingerd
The example assumes that the network daemons live in /usr/etc. On some systems, network daemons live in /usr/sbin or in /usr/libexec, the daemons have no `in. prefix to their name, or there is no userid field in the inetd configuration file.
Similar changes will be needed for the other services that are to be covered by tcpd. Send a `kill -HUP to the inetd(8) process to make the changes effective. AIX users may also have to execute the `inetimp command.
ntalk dgram udp wait root /some/where/tcpd /usr/local/lib/ntalkd
Only the last component (ntalkd) of the pathname will be used for access control and logging.
The program does not work with RPC services over TCP. These services are registered as rpc/tcp in the inetd configuration file. The only non-trivial service that is affected by this limitation is rexd, which is used by the on(1) command. This is no great loss. On most systems, rexd is less secure than a wildcard in /etc/hosts.equiv.
RPC broadcast requests (for example: rwall, rup, rusers) always appear to come from the responding host. What happens is that the client broadcasts the request to all portmap daemons on its network; each portmap daemon forwards the request to a local daemon. As far as the rwall etc. daemons know, the request comes from the local host.
The default locations of the host access control tables are:
hosts_access(5), format of the tcpd access control tables. syslog.conf(5), format of the syslogd control file. inetd.conf(5), format of the inetd control file.
Wietse Venema (firstname.lastname@example.org), Department of Mathematics and Computing Science, Eindhoven University of Technology Den Dolech 2, P.O. Box 513, 5600 MB Eindhoven, The Netherlands
|Main index||Section 8||Options|
|“||… one of the main causes of the fall of the Roman Empire was that, lacking zero, they had no way to indicate successful termination of their C programs.||”|
|— Robert Firth|