OECG – Chapter 19


Remember that a multicast SOURCE address will be the unicast address of the sender. A multicast DESTINATION address is a representation of a multicast group. Receivers that wish to receive traffic for a specific multicast group will all have the same multicast destination address.

Components of a Multicast IP Address

  • The first 4 bits of the first octet are always 1110.
  • That translates to a usable Class D range of to
  • These addresses represent groups, you should remember by now. So there is no concept of a subnet.

Well-Known Multicast Addresses

  • Permanent multicast groups – –
    • –
      • Use locally only; this range is not forwarded by routers.
      • – all multicast hosts
      • – all mutlicast routers
      • – DVMRP routers
      • – all OSPF routers
      • – OSPF designated routers
      • – RIPv2 routers
      • – EIGRP routers
      • – PIM routers
      • – IGMPv3
      • – RGMP
    • –
      • This range is forwarded by routers.
      • – Cisco proprietary Auto-Rendezvous (Auto-RP) Announce
      • – Cisco-RP Discovery
  • Source-Specific Multicast (SSM) – – SSM allows applications to choose what source will send them the multicast. The idea is that a recipient using SSM will end up with a better quality multicast if he is getting it from a source of a better quality. This feature relies upon IGMPv3; there’s practically no application using this yet because of how new it is.
  • GLOP – – RFC 2770. GLOP is not an acronym. A GLOP multicast address is one that has been registered by an ASN holder. So if you’re a private company wanting your own global multicast addresses for your applications, you can use a 256-address GLOP block. The GLOP will be formed by converting your ASN number into a 16-bit binary address. Split that 16-bit binary into 2 8 bit chunks; each 8-bit chunk will form the second and third octet of your GLOP.
  • Private (conceptually like RFC1918) – – You can use these locally on your network, but they aren’t supposed to bleed out onto the Internet.

Other multicast addresses not defined in the well-known ranges are considered transient ranges. Anyone is free to use them. It’s left up to good netizenship to use the address and then release it when you’re not using it anymore.

How To Convert A Multicast IP to a Multicast MAC – the quick summary – the first half (24 bits) of the MAC will always be 01-00-5E. The second half (24 bits) will always be a binary 0 + the last 23 bits of the multicast IP. More details follow.

  • Convert the IP to binary. Remember that the first 4 bits are always going to be 1110.
  • Replace the 1110 with the 6 hex digits (2 bits) 01-00-5E. Your multicast MAC will always start with 01-00-5E.
  • The next 5 bits of the binary IP translate to a single binary 0 in the MAC.
  • The last 23 bits of the binary IP translate to the last 23 bits of the binary MAC.
  • Now convert the last half of the binary MAC (the binary 0 + 23 other bits = 24 bits total) into hex.
  • Take the first half 01-00-5E with the second half you computed in the step above, and you’ve got it.

Note that this method does not always provide a unique MAC. There is a small chance that 2 applications could translate to the same MAC on the same segment, in which case the receivers would know from L3 that the multicast was not intended for them. But this is not desirable behavior and something to avoid if you are in control of multicast addressing on your network.

About the author

Ethan Banks

Most people know me because I write & podcast about IT on the Packet Pushers network. I also co-authored "Computer Networks Problems & Solutions" with Russ White.

Find out more on my about page.


Most people know me because I write & podcast about IT on the Packet Pushers network. I also co-authored "Computer Networks Problems & Solutions" with Russ White.

Find out more on my about page.

Subscribe via Email

Receive complete, ad-free posts in your inbox as I publish them here.