Free multicast software for linux

Emcast project is a generic multicast toolkit.
Emcast is a multicast toolkit for distributed/peer-to- peer applications that require multicast communication.
It includes the program "emcast", a generic multicast utility (like netcat), and the library "libemcast", a generic multicast library.
Emcast supports IPv4 multicast (IM) and can easily support almost any end-host multicast (EM) protocol.
The EM protocols supported are Banana Tree Protocol (BTP) and Internet Relay Chat (IRC).
Enhancements:
- Copy from inet address, not string. Interesting that mcast was still working before in Linux.

multisend project uses multicast to send files to multiple machines at the same time.
I wrote it for syncying VMware images out to a bunch of lab machines; 8gb * 15 takes awhile with http or scp. It listens on the local machine and uses ssh to start the receiving end, which connects back for the tcp stream.
Its pretty functional; the default mode takes a source and dest, so its like scp with a bunch of targets. Or you can use it in sync mode (-S), which sends files to the same location as on the local machine. Very handy for setting up one lab machine, then syncing all the relavent files out to the rest of the lab with a single command.
It sends md5sums and doesnt overwrite anything until the transfers been verified, so itll never corrupt data. It also falls back to tcp if for some reason multicast isnt working.
It could use a little more polish, but mostly it needs some real documentation; I havent worked on it awhile, and Im the only person who really uses it. At some point Ill include some example scripts that make it easy to use.
Usage: multisend [OPTION] localfile remotefile
or: multisend [OPTION] localfile remotedir
Copies local file to remote file, or multiple files to remote directory
-t < timeout > specify a timeout for clients to connect, in seconds
-n < nclients > wait for a specific number of clients to connect before transferring
-r < rate > rate to transfer at, in kb, mb or gb per second
-S, --sync sync mode
-s, --script file to execute to connect to clients
-g, --group multicast group to use
-a, --archive same as -dpR
-p preserve ownership, permissions, timestamps
-R, --recursive copy directories recursively
-x stay on this file system
-v, --verbose give more info
You must specify either -t or -n
You should also specify a rate as well; otherwise most packets will be dropped
Enhancements:
- The code is functional and has been tested, but could use more polish.
- There may be latent bugs, but it will not corrupt data.

UDPcast project is a file transfer tool that can send data simultaneously to many destinations on a LAN. This can for instance be used to install entire classrooms of PCs at once.
The advantage of UDPcast over using other methods (nfs, ftp, whatever) is that UDPcast uses Ethernets multicast abilities: it wont take longer to install 15 machines than it would to install just 2.
UDPcast is released under the GPL 2.0 license.
UDPcast can be started from the included linuxrc boot disk for OS installations, or from the command line when using it for other purposes.
Enhancements:
- The kernel was upgraded to 2.6.21.3 and busybox was upgraded to 1.5.1.
- The code was adapted to Solaris 10 x86 by adding includes and configure checks.
- A patch was made to fix parallel make and make DESTDIR=/tmp/xxx install.
- gcc4 warnings were addressed.

libasyncns provides a C library for Linux/Unix for executing name service queries asynchronously. It is an asynchronous wrapper around getaddrinfo and getnameinfo from the libc.
In contrast to GNUs asynchronous name resolving API getaddrinfo_a(), libasyncns does not make use of UNIX signals for reporting completion of name queries. Instead, the API exports a standard UNIX file descriptor which may be integerated cleanly into custom main loops.
In contrast to asynchronous DNS resolving libraries like libdenise, skadns, adns, libasyncns is just an asynchronous wrapper around the libcs synchronous getaddrinfo() API,
which has the adva ntage of allowing name resolution using techniques like Multicast DNS, LDAP or NIS using standard libc NSS (Name Service Switch) modules.
libasyncns is compatible with IPv6 if the underlying libc is.
libasyncns is very tiny, consisting of just one header and one source file. It has no dependencies besides libc.
By default libasyncns spawns a number of worker threads (LWPs) to process name queries. Alternatively or when POSIX Threads are not supported, libasyncns can fork() off worker processes instead.
Installation:
As this package is made with the GNU autotools you should run ./configure inside the distribution directory for configuring the source tree. After that you should run make for compilation and make install (as root) for installation of libasyncns.
Enhancements:
- This version properly detects res_query() on Linux/AMD64 and supports older autoconf versions.

This is a unified diff (you probably need GNU patch to apply it) to lame 3.58. Please note that this diff is now obsolete as recent beta versions of lame include a version of this code. Just type "make mp3rtp". The code was broken, but the CVS code is reported to work (and interoperate with playRTPMPEG) as of Feb 21 2000.

Please note that the output stream will only have the correct speed if the input is live recorded stream from your sound card or lame encodes exactly in real-time on your CPU!

RTP is the Realtime Transport Protocol as defined in RFC 1889. It is the transport mechanism of your choice to multicast an mp3 stream.

mrouted project is a DVMRP multicast routing daemon.
mrouted is an implementation of the DVMRP multicast routing protocol. It turns a UNIX workstation into a DVMRP multicast router with tunnel support, in order to cross non-multicast-aware routers.
Enhancements:
- IGMP could report membership in local-only groups (i.e. 224.0.0.X)
- IGMP could get confused by hearing its own new membership reports, thus a router would never perform fast leave.
- IGMP could reset timers for the wrong interface.
- mrouted put a bogus value in the maximum timeout field of IGMPv2 query packets.
- Non-querier mrouters would respond to IGMP leave messages
- mrouted was not performing fast leave properly
- If the last member goes away on a transit network, the upstream router would stop forwarding even if there are downstream members.
- Kernel hash function improved
- Eliminated possibility of panic(): timeout in cache maintenance
- Reordered resource allocation when sending upcall to handle failure properly
- some endian-ness bugs squashed in mrouted, probably more to go.
- Multicast traceroute could send a reply on a disabled interface.

MUSCLE is a somewhat scalable, robust, cross-platform client-server messaging system for dynamic distributed applications that runs under any POSIX-compliant operating system.
MUSCLE has been developed, used, and refined as the networking component of BeShare, CueStation, CueConsole, and various other audio control applications at Level Control Systems for over four years.
Main features:
- Use BMessage-like muscle::Messages under any operating system.
- Send flattened muscle::Messages from one computer to another over TCP streams, eliminating unnecessary compatibility and protocol-versioning hassles.
- Run a "muscled server" on a central machine, and write (or download) client programs to log in to the server and communicate through it. Once logged in, client programs can discover who else is logged in, send muscle::Messages to other clients, and store muscle::Messages in the servers RAM for other clients to download later. Clients can also "subscribe" to selected data on the server, and be automatically notified whenever it changes. Unicast-style, multicast-style, and broadcast-style Message routing are all supported via an intelligent hierarchical pattern-matching routing mechanism. Writing multiplayer games, IRC style chat applications, SETI style distributed calculation apps, or any other type of distributed software is made easy because MUSCLE handles all the dirty work for you!
- Customize the included "muscled server" by defining your own session logic or message-streaming protocol. (Note that this is only necessary for certain specialized applications--the standard server provides sufficient functionality for most things)
- Write your client code in C++, Java, or Python. Single-threaded and multi-threaded messaging APIs are provided for all three languages.
- Or just use the included message, string, dataIO, hashtable, dequeue, string-tokenizer, reference-count, regular expression and pattern matching parser, and object-pool code by themselves, as handy cross-platform utility classes. All source code is included, and you are free to use and abuse it any way you wish.
- See the Beginners Guide for a more detailed description of the system and its capabilites.
Enhancements:
- This release includes various minor bugfixes to the Java and C++ code, as well as a few minor optimizations.