stud - The Scalable TLS Unwrapping Daemon
stud is a network proxy that terminates TLS/SSL connections and forwards the
unencrypted traffic to some backend. It's designed to handle 10s of thousands of
connections efficiently on multicore machines.
It follows a process-per-core model; a parent process spawns N children who
accept() on a common socket to distribute connected clients among them.
Within each child, asynchronous socket I/O is conducted across the local
OpenSSL's nonblocking API. By default,
stud has an overhead of ~200KB per connection--it preallocates
some buffer space for data in flight between frontend in backend.
stud has very few features--it's designed to be paired with an intelligent
backend like haproxy or nginx. It maintains a strict 1:1 connection pattern
with this backend handler so that the backend can dictate throttling behavior,
maxmium connection behavior, availability of service, etc.
stud has one "cool trick"--it will optionally write the client IP address
as the first few octets (depending on IPv4 or IPv6) to the backend--or provide
that information using HAProxy's PROXY protocol. In this way, backends
who care about the client IP can still access it even though
appears to be the connected client.
Requirements and Limitations
libev >= 4 openssl (recent, >=1.0.0 recommended)
Stud currently works on Linux, OpenBSD, FreeBSD, and MacOSX. It has been tested the most heavily on Linux/x86_64.
While porting it to other POSIX platforms is likely trivial, it hasn't be done yet. Patches welcome!
If you're handling a large number of connections, you'll
probably want to raise
ulimit -n before running
It's very strongly recommended to not run
stud as root; ideally, it would
be run as a user ("stud", perhaps) that does nothing but run
$ make $ sudo make install
The only required argument is a path to a PEM file that contains the certificate and private key.
The entire set of arguments can be invoked with
Encryption Methods: --tls (TLSv1, default) --ssl (SSLv3) -c CIPHER_SUITE (set allowed ciphers) Socket: -b HOST,PORT (backend [connect], default "127.0.0.1,8000") -f HOST,PORT (frontend [bind], default "*,8443") Performance: -n CORES (number of worker processes, default 1) Special: --write-ip (write 1 octet with the IP family followed by 4 (IPv4) or 16 (IPv6) octets little-endian to backend before the actual data) --write-proxy (write HaProxy's PROXY protocol line before actual data: "PROXY TCP4 <source-ip> <dest-ip> <source-port> <dest-port>\r\n" Note, that currently only TCP4 implemented. Also note, that dest-ip and dest-port are initialized once after the socket is bound. It means that you will get 0.0.0.0 as dest-ip instead of actual IP if that what the listening socket was bound to)
stud uses no configuration file.
stud was originally written by Jamie Turner (@jamwt) and is maintained
by the Bump (http://bu.mp) server team. It currently (6/11) provides
server-side TLS termination for over 40 million Bump users.
* Colin Percival @cperciva -- early audit and code review * Frank DENIS @jedisct1 -- port to BSD, IPv6 support, various fixes * Denis Bilenko -- HAProxy PROXY protocol support