Most operating systems limit an amount of TCP sockets that can be used 
simultaneously. It is possible for a busy Tor relay to run into these
limits, thus being unable to fully utilize the bandwidth resources it 
has at its disposal. Following system-specific tips might be helpful
to alleviate the aforementioned problem.

Linux
-----

Use 'ulimit -n' to raise an allowed number of file descriptors to be 
opened on your host at the same time.

FreeBSD
-------

Tune the followind sysctl(8) variables:
 * kern.maxfiles - maximum allowed file descriptors (for entire system)
 * kern.maxfilesperproc - maximum file descriptors one process is allowed
   to use
 * kern.ipc.maxsockets - overall maximum numbers of sockets for entire 
   system
 * kern.ipc.somaxconn - size of listen queue for incoming TCP connections
   for entire system

See also:
 * https://www.freebsd.org/doc/handbook/configtuning-kernel-limits.html
 * https://wiki.freebsd.org/NetworkPerformanceTuning

Mac OS X
--------

Since Mac OS X is BSD-based system, most of the above hold for OS X as well.
However, launchd(8) is known to modify kern.maxfiles and kern.maxfilesperproc
when it launches tor service (see launchd.plist(5) manpage). Also, 
kern.ipc.maxsockets is determined dynamically by the system and thus is 
read-only on OS X.

OpenBSD
-------

Because OpenBSD is primarily focused on security and stability, it uses default
resource limits stricter than those of more popular Unix-like operating systems.

OpenBSD stores a kernel-level file descriptor limit in the sysctl variable
kern.maxfiles. It defaults to 7,030. To change it to, for example, 16,000 while
the system is running, use the command 'sudo sysctl kern.maxfiles=16000'.
kern.maxfiles will reset to the default value upon system reboot unless you also
add 'kern.maxfiles=16000' to the file /etc/sysctl.conf.

There are stricter resource limits set on user classes, which are stored in
/etc/login.conf. This config file also allows limit sets for daemons started
with scripts in the /etc/rc.d directory, which presumably includes Tor.

To increase the file descriptor limit from its default of 1,024, add the
following to /etc/login.conf:

tor:\
	:openfiles-max=13500:\
	:tc=daemon:

Upon restarting Tor, it will be able to open up to 13,500 file descriptors.

This will work *only* if you are starting Tor with the script /etc/rc.d/tor. If
you're using a custom build instead of the package, you can easily copy the rc.d
script from the Tor port directory. Alternatively, you can ensure that the Tor's
daemon user has its own user class and make a /etc/login.conf entry for it.

High-bandwidth relays sometimes give the syslog warning:

/bsd: WARNING: mclpools limit reached; increase kern.maxclusters

In this case, increase kern.maxclusters with the sysctl command and in the file
/etc/sysctl.conf, as described with kern.maxfiles above. Use 'sysctl
kern.maxclusters' to query the current value. Increasing by about 15% per day
until the error no longer appears is a good guideline.

Disclaimer
----------

Do note that this document is a draft and above information may be
technically incorrect and/or incomplete. If so, please open a ticket
on https://trac.torproject.org or post to tor-relays mailing list.

Are you running a busy Tor relay? Let us know how you are solving
the out-of-sockets problem on your system.