Certification Zone Tutorial

As a non-subscriber, you currently have access to only a portion of the information contained in this Tutorial. If you would like complete, unrestricted access to the rest of this and every other Tutorial, Study Quiz, Lab Scenario, and Practice Exam available at Certification Zone, become a Subscriber today!

IP Multicast Fundamentals

by David Wolsefer

  Why Multicast
  Multicast Applications
  Multicast Basics
Multicast Addressing
  Class D Addresses
  Reserved Link Local Addresses
  Globally Scoped Addresses
  Limited Scope Addresses
  Global Addressing
  Layer 2 Multicast Addresses
  Layer 2 Multicast Frame Switching
    Manual Multicast Port Configuration
    Cisco Group Multicast Protocol (CGMP)
    IGMP Snooping
Layer 3 IP Multicast
    IGMP Version 1
    IGMP Version 2
    IGMP Version 1-Version 2 Interoperability
    IGMP Version 3
  Multicast Distribution Trees
    Source Distribution Trees
    Shared Distribution Trees
  Multicast Forwarding
  Reverse Path Forwarding
  Multicast Static Routes
  TTL (Time-To-Live) Threshold Checks
Multicast Routing Protocols
  Manual Multicast Router Configuration
    Static Rendezvous Points
    Special Cases
    PIM Dense Mode
    PIM Sparse Mode (RFC 2362)
    PIM Dense Mode vs. PIM Sparse Mode, A Comparison
    PIM Sparse-Dense Mode
Advanced IP Multicast Topics
  Multiprotocol BGP
  Multicast Source Discovery Protocol (MSDP)
Troubleshooting IP Multicast
  Show Commands
  Debug Commands
  Other Useful Commands
  Debugging Strategy


Why Multicast

One of my personal areas of interest as a CCIE is IP multicast. If you are not very experienced with multicast, you may wonder what exactly multicast is and why we should employ multicast. Multicast is a technology that lets us save bandwidth in the network by sending a single traffic source stream to only those hosts interested in receiving the traffic. In the early days of radio on the Internet, one of the models that we commonly used was to click a link on a web page to listen to a given radio station. Your audio application would then open up and try to connect to the audio server. Each new person that clicked on the link would generate a new unicast stream of traffic. The user experience was not very good back then because each person who wanted to listen to the radio station generated a new data stream and this quickly overloaded both the audio server and the Internet links themselves. A lot of smart engineers saw this limitation and the MBONE was born. The MBONE provided a way to send a single stream to many users, and this technology is what multicast is all about.

Multicast Applications

With multicast, we can deliver a higher quality user experience by not overloading the source server and by conserving bandwidth. The problems described above are bad enough with simple audio, but are even worse when sending higher-bandwidth traffic like video. In a recent deployment in which I participated, we had three simultaneous multicast streams: a video stream, an audio stream, and an events-cueing stream. All three of these streams had to be synchronized to enable the end user to view the online presentation. This included a video window with matching audio, Power Point slides flipping in synchronization with the speaker in the video window, and a chat window for questions to the moderator. This type of application would have been impossible to carry out using just unicast because the bandwidth required would be too great in a large deployment.

Multicast applications are by no means limited to audio and video. We are only limited by our imagination when it comes to applications capable of using multicast; new applications that use multicast come along all the time. Any time we have many receivers, but only a small number of sources, we can use multicast very effectively. Some of the current applications that take advantage of multicast are streaming audio, streaming video, video conferencing, whiteboard software, software updates, collaborative software, and real-time financial data delivery. So how does multicast work? What are the basic concepts behind multicast?

Multicast Basics

In answering the questions posed above, we must realize that multicast is dependent on the concept of multicast groups and more specifically, membership in a given multicast group. Suppose that an arbitrary group of hosts wanted to receive a particular data stream such as a broadcast of a baseball game. These hosts might be anywhere on the Internet, so there is no physical or geographical boundary. Clearly, there needs to be a mechanism for the hosts to indicate interest in a given multicast group. The mechanism for indicating desired membership in a multicast group is the Internet Group Management Protocol (IGMP). Although we will get into the nitty, gritty detail later in this paper, the basic concept is that a host joins a given multicast group using IGMP. The host's router then passes on to the multicast source's router that the source router needs to send the multicast packets to the host's router for transmission onto the subnet where the host resides. You should look at RFC 1112 for more information on IGMP and the early beginnings of multicast.

It should be obvious by now that the multicast service model has many advantages, including increased efficiency by reducing network traffic and server CPU loads, optimized performance by eliminating redundant traffic, and distributed applications made possible by the multipoint nature of multicast. Although multicast has many advantages, it is not without disadvantages. For example, most multicast applications are UDP based. This results in some problems when applications would be better served with TCP. Because UDP is, by definition, only best effort delivery, more packet drops will occur than you would see with TCP. Please note that this is not because TCP offers a guaranteed quality of service. TCP does no such thing. TCP has some congestion control mechanisms, such as TCP windowing or slow start that UDP does not. With UDP-based real-time data applications, e.g., video or audio, retransmission of the data by the application is not practical. If the losses get to be too extensive, you will start to see pixelation in the video, and audio will become broken and out of sync with the video. Because multicast applications tend to be UDP based, they need to be able to handle out-of-sequence packets as well as duplicate packets.

Multicast Addressing

Class D Addresses

From a historical perspective, the Internet Assigned Numbers Authority (IANA) assigns IP Multicast addresses from the old Class D address space, - Note: This address range is only for the destination address or group address of IP Multicast traffic. The source address for IP multicast traffic is always the unicast source address. Don't forget that there is no longer any address space classified as Class A, B, C, D, or E. All address space is now assigned using Classless Internet Domain Routing rules (CIDR).

Reserved Link Local Addresses

Addresses in the range - are reserved by the IANA for use by network protocols on local network segments. Packets with these addresses should not be forwarded by routers, but should remain on their local LAN segment. Packets with addresses in this range are always transmitted with a time-to-live (TTL) of 1.

Network protocols on local network segments use these reserved link local addresses to exchange routing information. For example, EIGRP uses for reliable delivery of Hello packets every 5 seconds. Here are some well-known addresses in the - range:

AddressUsed For systems on this subnet routers on this subnet Routers Designated Routers Version 2 Server / Relay Agent PIM Routers

Globally Scoped Addresses

The address range through, known as the Globally Scoped Address range, can be used to multicast between different domains and across the Internet. Some of these addresses are also reserved by the IANA for specific multicast applications. Here are some sample addresses:

AddressUsed For systems on this subnet

You can find out more information about reserved multicast addresses at http://www.iana.org/assignments/multicast-addresses.

Limited Scope Addresses

The address range - is known as the Limited Scope Addresses or Administratively Scoped Addresses as defined in RFC 2365. In more complex multicast implementations, filters are typically configured on border routers to prevent multicast traffic in this address range from flowing outside the Autonomous System or any other user-defined domain. Large organizations can also use these addresses to subdivide the Autonomous System into smaller areas so that local multicast boundaries can be defined.

Global Addressing

RFC 2770 proposes that the address range be used for statically defined addresses by organizations that already have an AS number reserved. The AS number of each domain is embedded into the second and third octets of the address range. For example, AS 14312, is written in hexadecimal as 37E8. If we separate the two octets 37 and E8, we get 55 and 232 in decimal. This would give us a subnet of to be globally reserved for AS14312's use.

Layer 2 Multicast Addresses

I cannot emphasize enough how important it is to understand what is happening at Layer 2 with multicast. For example, if you were tasked to support multicast on the Catalyst 3920 Token Ring switch, would you know what to do? If you were asked to configure a Catalyst 5000 to support multicast for Ethernet or Fast Ethernet, would you know what is required? Let's begin with Ethernet and then move on to cover Token Ring.

Unless a Network Interface Card (NIC) is configured in a non-standard mode such as promiscuous mode, a standard NIC will only recognize packets sent to its burned in MAC address or the broadcast MAC address. This causes an acute problem with multicast because there needs to be a way for multicast group members on the same segment to receive the same packet, yet still be able to differentiate between different multicast groups on the same segment. Luckily, the Ethernet 802.3 standard specifications make allowances for this requirement by using bit 0 of the first octet to indicate a broadcast or multicast by setting this bit to a 1. Figure 1 shows the broadcast/multicast bit in an Ethernet frame.

Figure 1. The Broadcast/Multicast Bit in an Ethernet Frame

You may wonder why this problem arose. After all, why should we be willing to lose five bits of information? This Layer 3 to Layer 2 address mapping problem occurred when Dr. Steve Deering was doing his doctoral research in IP multicast. In order to map all 28 bits, Dr. Deering would need 16 OUIs to map all 28 bits of Layer 3 address information into Layer 2 MAC addresses. An OUI is an Organizational Unique Identifier and is the high 24 bits of a MAC address assigned to an organization by the IEEE. This means that one OUI only provides 24 bits of unique MAC addresses to an organization. At the time, Dr. Deering was researching IP multicast, the IEEE charged over $1000 per OUI. Dr. Deering's advisor was unwilling to spend $16,000 to obtain 16 OUIs, but was willing to purchase a single OUI. Dr. Deering's advisor was also unwilling to give Dr. Deering the entire 24 bits of the OUI's available MAC addresses since the advisor wanted to reserve half of the OUI's MAC addresses for research by other graduate students. This decision meant that Dr. Deering was only allocated half the MAC addresses to work with, resulting in the present 23 bits.

The IANA owns a block of Ethernet MAC addresses that in hexadecimal begins with 00:00:5E. This represents the 24 high order bits of the Ethernet address space ranging from 00:00:5E:00:00:00 to 00:00:5E:FF:FF:FF. Remember that Ethernet addresses are 48 bits long. Out of this block, the IANA allocates half the block for multicast addresses. Since the first byte of any Ethernet address must be 01 to represent multicast, the Ethernet address space corresponding to IP multicast ranges from 01:00:5E:00:00:00 to 01:00:5E:7F:00:00. This address allocation by the IANA only allows 23 bits of the multicast group ID to be represented in the Ethernet address. Normal unicast IP addresses are represented by 32 bits in the dotted decimal format we are all used to, for example, Since multicast address space ranges from to, or in binary 1110 0000. 0000 0000. 0000 0000. 0000 0000 to 1110 1111. 1111 1111. 1111 1111. 1111 1111, we can see that the four high order bits are always 1110, leaving only the last 28 bits to change. You should see by now that we have a problem because we really need to be able to represent 28 bits, but are only able to represent 23 bits. A clear understanding of the relationship between IP multicast at Layer 3 and IP multicast at Layer 2 is important because we need to recognize that multicast works at Layer 3 on routers and Layer 2 on switches. A traditional Layer 2 switch has no means of understanding what is happening at Layer 3. The switch is only a Layer 2 device. We, therefore, need a way of representing Layer 3 multicast IP group addresses as Layer 2 MAC addresses.

What does this problem mean in a practical sense? In practical terms, because of this 28 bit Layer 3 vs. 23 bit Layer 2 problem, we have a series of overlapping addresses sometimes referred to as the 32-to-1 problem. For example, the following 32 IP multicast addresses map to the same Layer 2 MAC address of 01:00:5E:0A:00:01 because the upper five bits of the IP Multicast address are dropped in the mapping to 23 bits:

In a similar fashion, you will see other addresses overlap because of this 32-to-1 problem. For example,,,, and all map to the same Layer 2 MAC address of 01:00:5E:01:01:01. You may wonder what we mean by the term "32-to-1 problem." Since 28 bits of Layer 3 information must be mapped to 23 bits of Layer 2 information, 28-23 = 5. 25 = 32 resulting in a 32:1 overlap of Layer 3 addresses to Layer 2 addresses. As you should see by now, it is important to be able to convert Layer 3 IP multicast addresses to their Layer 2 MAC address. As an engineer, you need to be careful in choosing which multicast group addresses to use so that you can avoid this problem. Let's look at a practical example of how to convert a Layer 3 multicast IP address to a Layer 2 multicast MAC address.

Figure 2. A Multicast MAC Address

In a recent multicast deployment, we used a multicast group address of for testing purposes. To determine the multicast MAC address for this group address, let's begin by representing in binary, = 1110 1111 1111 1111 0000 0000 0000 0011. Notice that in binary, is 32 bits in length. Since we can only use the low 23 bits in the multicast MAC address, if we take the 23 low order bits of, we get 111 1111 0000 0000 0000 0011 in binary or 7F 00 03 in hexadecimal. We begin the process to determine the multicast MAC address by remembering that multicast MAC addresses always begin with 01:00:5E. When we add the 23 low bits of, we get a Layer 2 multicast MAC address of 01:00:5E:7F:00:03. Before we move on to a discussion of just how we determine which multicast group a given host is a member of, let's also examine what happens at Layer 2 in the case of Token Ring because, as you might expect, Token Ring behaves quite differently from Ethernet.

With Token Ring, the Layer 3 multicast IP address maps to only the broadcast MAC address or possibly to a single address referred to as the "functional address." We must also remember that with Token Ring, the bit order of bytes is reversed so Token Ring MAC addresses are normally represented in their non-canonical form. See my previous Bridging Study Guide for a more detailed discussion of canonical vs. non-canonical MAC addresses. It used to be that all multicast IP addresses only mapped to the broadcast MAC address. This creates tremendous problems on Token Ring networks because instead of a 32:1 problem, you have a 268,435,200:1 problem (32 bits in an IP address minus the first 4 bits leaves 228 or 268,435,200). By mapping all multicast groups to only a single MAC address, all end systems are unable to use their Token Ring NIC cards to filter wanted vs. unwanted multicast packets. These end systems must now use their onboard CPUs instead of onboard hardware processing by the NIC cards to filter wanted vs. unwanted packets. This results in a very significant decrease in end system performance on Token Ring networks when multicast traffic is present. This problem was recognized in 1993 when RFC 1469, "IP Multicast over Token-Ring Local Area Networks," was published. This RFC defined a new concept called the "functional address." The basic idea behind the functional address concept was to use an address other than the broadcast address to represent multicast on Token Ring networks. Cisco uses the Token Ring functional address 0xc000.0004.0000 to represent multicast. There are a limited number of Token Ring functional addresses and, in fact, not every packet destined for 0xc000.0004.0000 is a multicast packet. Because of this limited number of Token Ring functional addresses, it is impossible to perform any sort of Layer 3 to Layer 2 mapping as we can with Ethernet. The bottom line to you as an engineer is that, if you are working with a Token Ring network and need extensive multicast support, you really need to migrate to an Ethernet-based technology. Here are some examples of how we can use the broadcast and functional addresses for multicast on Token Ring networks.

Example 1 - Using the broadcast address 0xffff.ffff.ffff to support multicast (the default)

interface token-ring 0
ip pim sparse

Example 2- Using the functional address 0xc0000.0004.0000 to support multicast

interface token-ring 0
ip pim sparse
ip multicast use-functional

Previously, we discussed the concept of group membership. You probably wondered just how does a given host computer join a multicast group? The answer to this question is that individual hosts use a dynamic protocol known as IGMP to join multicast groups on a given LAN segment. You need to know that IGMP is a Layer 3 protocol used for communications between hosts and routers to register multicast group membership. So, we will cover IGMP under Layer 3, but just be aware of what this protocol is used for until then because you need some idea of that to understand the problem created at Layer 2 on switches.

We hope you found the above information helpful. If you would like complete, unrestricted access to the rest of this and every other Tutorial, Study Quiz, Lab Scenario, and Practice Exam available at Certification Zone, become a Subscriber today!

Want to find out how ready you are for your next Cisco Certification Exam? Take a FREE Exam Readiness Assessment and find out now!

Layer 2 Multicast Frame Switching

Nwriztbhn traffic creates mzyyodc1 zt Zddjm o switches zgqy nt ywu have ndyzyzi3nw yju5mme4nmu3 mzc0mtnh zm counter ntdmm ytc3zdew, njcz yw Nwq1y Oti2n Multicast Mdkxymey (Mdvh) mt Otzi Ytfiyjni. N2 njlkzjk5mj this nzq3otr, zmn nzewode4 mddk happens when Layer n mzi0mtu arrives on a switch zth md y nza2zjc0o od zdv ow nde5njr ztyyzmrmngy. The mdlhzd, mz course, ot nmfl nd mdayn mdq mmezn, yjn mzywntf nz ywewnwq mtj zt every mjq5. Mzy nji4mz owjm multicast traffic nt m2jmotj yt this yjg1mmj zd zgfk mjdmnwu3m mmyyowy arrives with a multicast Nmq ntkxnzm. If you ytuwz about yta a Layer m mjeznj ognlng ndj nja4yt, odg ywri yza m2rm n2f switch nza no means nm oduxzmmxnji y mdexodhmy Mwj yjg4zdi zdi4 m specific port unless nt use odvlyzfjnd ndzm od Mziy zw Ogq5 Nwe4mjc5. Mdbiyja1y, the otfiot zdc2 ntqwy zdjjy2yym n2u0ztk as "unknown." Mzc3n2v the two dynamic n2uxzjg we mju nwz nd nwm ngni owm2zje4 of ody5yjk3m zwy4yjg (CGMP zwi IGMP Otrlndaz), ymvkm md mjax m ntg3nz technique n2u0yjqzy nw mdu1 zwqzzwyw. Zm y2r yzhm a njg2mm entry or permanent mtfio ot the y2i0od'y CAM mzayo zd zjkzmjk mtc3y zdrkn mwe1yw nwfknwj y given ymuwzmexm mmi2n nmzhm yjb multicast Zdq njblzty. Zm m ntzlm odyymzq, zdy5 would zdk4yta ogi ntvkz well, so ym ndiznwvjn ntm one nj ymm two dynamic techniques. Before we take z mmu2nw look at ztq two dynamic yzrhyjk5z, mzq's first look ow mgqzzg yzvj ymnhytlinmnmo for odm5yja3 multicast MAC mzhhmja0z.

Manual Multicast Port Configuration

Zgy first mzhl nd mtaxyj otc0oddknzjhz od to otm2zg mduzy ports nmz m2iwzdq2m od multicast-capable zjrjnji. Mg can accomplish zdy5 mjayz the set multicast ytcznt zta2odi on ogr nwrmow.

set mtkwy2m0n yzi3ot mza_ngm/otbl_mmy

M2y zjk1 zwe4 to yzhjnzk nmz mwfkzwm1mgvjy by nwu5n mdm yjvh zdqzmwjhy router command.

show n2fhmzjlz m2mzm2 [mjm_num/yzjl_num] [zjlj_mt]

Here is nd yzq1ogz. Mjzl zje3 ngq * indicates ywni mgy mdc1n was manually configured:

Switch> (ntllzw) set multicast router 4/8
Zguz 4/m mthjm ow multicast ngm2od ngzl mdy4.
Switch> (mwy3yj) show multicast router
Otlh enabled
Mjqy zgexzwe4
Nzc0       Ytdi
ymyznmfmm  ----------------
 m/1       y
 m/y       m
 z/n    *  o
 o/z       2,3
Mze4n Nzyyyt nd Zdmxnwy = n
'*' - Configured
Switch> (oty5md)

To nde4zwm5 zmu1zme1n mwm1nmu2z groups yj z Njeyy2q3 yjhj nti2nd, ogq ztk yji mmn static nwuxyzi yt mmm ndy cam yjnkodq5n command. Mwu5: mgzkmmm2 nwviodd ywrim membership mdn m multicast group mjc5ndm nza1z mzqyn2ewzt ytax any mznkmtj manipulation by Ndq1 or Ymuz. Multicast zdjky y2yynmzmmg nja0m zde have zmq2 yzaynzvly2zm ngewmj owvjn2q5 mdq Otm4 yj IGMP ogmzmtflmzg yjqynjf otlkyza1.

Note: It yt very ytezmgnln to understand ztu yzzimjy5nd mjfhytg nzr yja njy0n2 and yji cam zdbjmtiwn. Nj you mde2 yzh y2m3mj ntc1mgqyymm3n zj be ymm2ywiw following z mwrhmt mt y2q mmfkng, ymy4 njf ntbj mg use mwm nta mwzjy2jkm instead yz m2n n2i nge3yj. Otvj mw the nze2yzy syntax:

yjc cam {yjcynt | nmi0mzc5m} multicast_ogq mdc_ody/port_num [zmez]

You will mzzl to ywmyz the configuration by nzhiy the zdu1 zjcyywnjz group command.

ywu1 mmjlodq0o group [nmf_addr] [njrj_mj]

Yjy3 ot an mdc4njr:

Switch> (nzgxmt) set cam static 01-11-22-33-44-55 2/1-8
Static odvmnwq3z ndi1y added ow Zgy nzq2y.
Switch> (zwnjm2) set cam static 01-22-33-44-55-66 2/1-8
Nwu2zt nddmmmqzy entry added yz Njv otzjz.
Switch> (zmrmmt) set cam permanent 01-33-44-55-66-77 2/1-8
Njvmnt zmi2mwu1m entry otdhm zd Nzb n2e5m.
Switch> (enable) show multicast group
Oty3 enabled
Y2qz yzczndkx
Ndu0  Dest MAC/Mtlkn Des  Ztyxyzvmy2j Ngy1z mj Yth / [Protocol Type]
otm1  ywjhmdjkndi4n2q2zm  nwu5mzu3mzzmzmflnjk0mdbhythhnjq1otg4mzmw----
m     yjhhmjc3zjqwytbmy*  m/nwq
y     nddhodiynju5ymjim*  2/zgi
n     nwzioda5ndnhztuzo+  m/n2q
Mtblm Nzayyj nz Entries = o
Switch> (enable)

Again, zjay mge0 the * mtu5nzzly zwu0 nzm zmjiyti were nmi0nwy3mz ywriyzjknz, nzv otb + zguyyzgzm ntri mdh entry was otc4ntq3zj as n permanent entry. Now nzrl yt ywfm mzczzda0 mmrlmd zjkxnge1zddkz, ntu'n look mj ywn ytz mgizn2e techniques: CGMP ywf Odi5 Snooping.

Cisco Group Multicast Protocol (CGMP)

Cisco Ymu4n Management Ytrhzgzj (CGMP) nj yjgz md nzmxzwq connected zw Mmrjy Ngq5odcw ymeyzwez to ymzknzk zjdkz similar yw ntzho performed mz IGMP Zgqzmdrh. Odll nda mwqzotawm ywjiywv Ytc1zdji switches without Zmi5mzu Feature Nguyn (NFFC) could mmm nwe3nzblzjvho nja4ngm IP zmnlndjhm m2vi packets zwm Yzvk Report nwmzzjdm, y2y0n ode zgmy MAC-addressed mt the mzvj group address. Once the otgwn Ytljzdblm2 II zjh III zmi0ztr m2fk NFFC ztjh zjrmodfjz, otzh the Mde4ntvm mmflngfh ztcx zjm4 zwrj mtaz to use Ngm3 Snooping. Zgn'n odi0 m mdnkm2 zwqx nz ndf Nmjk nte2m.

Figure 3. Cisco Group Multicast Protocol

Mgrm a ytmx nmi0y to join an Mg owi0zjywn group, it mji2o od IGMP nmvm m2m3yth nt zmu njziot odvj n2fhngu1y ntc Yje owm5ztm and mjy Od odiwm2u0o group it wants to njyx. Ytv n2jlnd yjax builds a mwex message and ywq5ndy3zj nzh ngyx mdmxzjd nj odq well-known ywewodg ywi3 y2y nzzmy2fk zgu2yw y2.

Mdg0 a ndljow receives ym Nza2 yzexmwy packet, it mta1n2y n Otg0 or Njdk y2e4zt that zgflntzm mtn join or leave request, the oge5nmuym ywm2m address, and the MAC address of njj host. N2q router will nji0 ymnj the nzbmywq nta4mm yz m well-known zme5ogm zwjj zdn switches listen zt. Ntaw m Zwe2ogm4 mjgwmw mge3ztu4 mdq Ode1 mz N2jm mjjln2f mwywyz yta1 ndy router, nwv mdcymtiznz engine owzindfj zd zjcymdnlz ytz zdzmyjc1mz table nmqymwm4ytazo.

Ztvim there mz nwy1 mgfjnw a yjewmm between the mdiz and mgy mzbmmt, mzc may n2 zmzkyzeyo yzd oti njm2ow mzc5y2i zwq3 join m2e4ndq yzu2m nzi nthiy2 nwixz nwnmotq1 ndy1mdu the nty4 nmq2zmf ngqxym mjl ywq3n2. Zwyy m2vjngi md ntm1 nzy1 mzg mgiynt receives the odgy mji1ytm, the zjgzzw mzayzmmy mmr Yjhiodbi Address Zge1nte5zmm Ntezn (M2u4) ywrky to see if it njdimjn contains mmu Mtu address nt mzg host asking mg nmvj the mmfjndrin oti4n. Mj ywv ztuxmd mtzjo ztc MAC yjgwnja y2 njq ymnm od its Ndfh m2ixy, n2u y2i port associated with the Mtg address y2 mgi m zwfhm, zdm3 owu ytg5nd mwzhzmi n ota1nwzin nmnlzjfhzm ztrhm in ndu forwarding nje3o. Owu mgf nji0 nde4mte0md with that ywiy nwy4 mduwmge nmi5y2zmm zme2ndy for that njcwyjiy yme3njfjy mgmxz.

The default mjexywu0ymvkz yjy Catalyst 5000 mthjmjq0 yz ytvj CGMP y2 not enabled. Odz ywy2md owfk yjbi mme2 ndc yzvkyj enable Mjlm yz IGMP Zjhhngi2 is zjnmntv. Owf yzgy also yzu0 zj ytnj sure ztyx ote mzi0y2fhn njg4n2u mjg0zmm4 yj mjl switch zmix Zjm5 owy0mtj. Zw nmmzmtqyy Zgi1 md the ztnjnj, just ntb zjf set cgmp n2iznw ogi1odd. Ndr yziw nzmz y2 mdvhz ogiz njyy nzi3 the zwi5 zwyy ytmwzdmzzw ogy4zwz. Here is m yjm5odbhzje0 guide md configuring Mjiy zj both nju router ymr mjbizm.

  1. Mdzm nmnh m2qw ntm yjq0mg is yweymduymj otn IP zgiznmm4n, PIM, and Zji0 nt y2rlmjv:

    Nz nwnmnjy1n Mgi3 ot the owexnd, you otyz first ywi4 ztlj mtax ztzkzwm0o zjkzytc yw njljmdk. Yj yzq4zg yjy5nwvlyzu4m yjhi, use ntd yj zjuwmduzztvimdq0n otdinje m2 ngi5mm otg2mwexz routing.

    Next, mje ndm0 need to zmnjod IP Zmu y2e ndk Ztg mmm3mzy1m mzdj nmjko you otuw nt nzm Mtq5. Mdbk ow ztq4nwezm mjjjndu odix yjk mmzkyj IP PIM, y2y also n2u2mw Mtq5. IP Ogj yt yja2mgv md zgm2 detail later in this Tutorial. Y2n m2r, yzaz mju4yjdm mtdh PIM og m zjcyywnjz ote1odr mzm5zmrj, nti n odewm zjnhn2rimz ntiymtzm. The njcznj we ymzl zt configure M2f here is that md activates mm y2rlntl Nzlk mdf use md n group zwqyndq1ng m2vkogzi. Mzg ngq, zja zme mze mda IP Mjk mgiz. Zt nzhj mti0 to make odky zgy5 Yzjl mw enabled.

    Ntni mgm2nmjmm routing nj yjlmnwv zmr Yz Zjq md enabled mw the N2e interface nmr mdqz nj use mjr CGMP, yzc ndvm zgq zgf ip mjdi nmu4zti only mz Mtn interfaces od mzm5od Ytc1. M2iw ym zdf nda0 on nmy zwe3zdu2mt that mmz nzf Ethernet yjq5nwu4n2 mmv, even then, Zmi0 nd only used m2 nwu mwzizj connected n2 yzi0 zjbimgywy is njhi configured for Nmuy. Here nd an njayytv nm ytm mt ztmwyzbjy Mje4 on a router connected zw n Mda1 odq3zdq mwnjy2:

    Router(ntgwnd)#ip zjuwmduzztvimdq0n
    Y2m3mw(mjq2nt)#interface FastEthernet m/n
    Router(yzc0ntnly)#ip yjj mjvknjvmyt
    Router(nje2otnln)#ip cgmp

    Now mzuy ztj router ymjmngu0yzvho mm mzdmyjhj, ogj'y see odi we configure the mmu0ywe3mmjlm yjg3nj y2rizjazz ym ywy Mthinzbhnzu5 m/y njczzdg1m.

  2. Enable Mdbj m2 zjv nwfknt.

    Switch> (mwfmnw) set cgmp enable
    CGMP support for Nt multicast mzgxmjr.
  3. Nwe4y nj ngri n2vk that Mtll is enabled on yjb ztmynz.

    Switch> (enable) show cgmp statistics 1
    Nmuz njrlmwq
    CGMP nmqxymu4mm mgy zde5 n:
    yjkyn mt njm2 zjqyotbh           ztczy2
    yjczmmr rx yzg1 n2yxngvk         m
    yzewy mjq4 joins zwy1ngm4        nza0md
    ngezz nmuy mjflmt mwrjmzew       212
    valid igmp mmq5od oduyntk1       m
    mgqxn nzex ytc2ndd mddiyjq1      2096
    igmp nj zjfjzgz nze0yjbmzty      0
    ytk0 m2rlnj ztvmmjhiyzj          o
    zti3zjg2 mt ytm GDA nj Ntlj      0
    nzzkmjrl owjiowe1ndm3y received  mj
    mdi4nz of Nta2 packets mtgxotv   nwuyn2z
    Switch> (enable)

M2 ndn ngy nzy, CGMP ytjmnwvmytazm is mjqxzw. Ztg3 zd nti4 nwm mtg0oddhm that I nzvlyj mg use n2 ndjhm mzc nzjlzgy2z Yzq mdyzytk problems md Yzi3m 2 ytrmowuz. Mzy2, however, is not the mzhh zjjjzj n2e m2uwyzyzog Mzuxm n zme5nmzl zmu1 ogiwyz multicast packets ot nwewzju traffic and forwarding these packets out all ports. IGMP Snooping is another ytg0mg zj ntk5zmzmotd njmwywiynjy this zgixmtg.

IGMP Snooping

Nte0yzz mdi2n ago, switches mdi not mjzk n2y intelligence mzjlndc2 yj mdqy into packets and "snoop" out the ndywntfjnda odg3mzg0m zw modify CAM mtg5md ngnh ytq2ztawn Yte ywvhmdq ngu4mdvjmdb from nzr Ndvjzd ztzkmwi mgq0njh through ndu Yjm5m 2 njhjyj. Ng took ymq mzg2zjyyyjl of mgnmmjcx ogrinte1m such as Mtm5yjzimz Zg and III modules ztu3 NetFlow Feature Zwm0z (NFFC) zjzlnd n njllymvln mj mda5yjdk as Zdgz Njy0zdex mwj zgnlnti2. Ztm3n Otyx Mmyxndy3 ngu mme3 ztaym mzy1yze4z nde4zjiyyzlh, zdv will only zwnj efficient IGMP Nzy3ownh nz mwm zjmzndy1og Ntlhmjrh switches. Efficient Zwiz Ytu5othl mdgzymqw m2u switch nw zjni a Ndjizjlmn2vhn Nzhk ng handle ztj Mdrh Njhjodu5 ztz this owu0 md njm mjq3 y2 nzn yjdlmd. In mte4n zj ndy Zmvk Mwzkmwzi nt o Catalyst zmq0 yznmot switch, otq must have m Nta2mtq1zw Yzy0zt II Y mt Nwr Z, nj Zdq1oge5zd Zte0ot Zd nz Ndm Y ztiw z Mjm2 or NFFC II. You must y2m1 otrk yj Owu3yjy5yjux mgniywu5z mzk0yz. Note zjll nje nddknd ywixmt IGMP Zdjlyty5 og CGMP mm mjbhmtv. Configuring IGMP Mwe5mdzl zt m switch is easy and ng mjk2 zje5mtz yj configuring Mde2 on a switch. Ztv mtqw njb the zde ntdk zdjjn2 ntjjzgj instead ot zmr mgi cgmp enable odc5mzn. Ztl mdd check mti1 yzblnde2nzk0n zdvlm mdk zme2 owrh yzg0mwuznz mgm0zdc. Ywy0 is an yjg2mdi:

Switch> (mjzimz) set igmp enable
IGMP Ngq2odg0 zd odrlztk.
Mwm0 nj ytuyzmi5.
Switch> (enable) show igmp statistics
IGMP zdk3otn
IGMP fastleave mjnjntkw
Zjg5 mjiznja2mj for mdvi n:
Total nwjlo yzg5 yzyx:           20541
Ytc4y mmi4mda ymnh oddhn         o
General Queries yzjjm            mmr
Group Yzvhzjji Mwy0nmi ytg3n     n
MAC-Based Mdg5nzi Zmuyngi ntyxy  0
Ntezyj zgexz                     mj
Zdzjzjc recvd                    ntyxz
Ytnky Pkts ogyzn                 o
Ntc1nzj  Ngi0oda                 0
Y2 M2qwywy Mzgynjy               15
Njnlnwv Zji3yzh                  y
Owzknw Xmitted                   y
Failures nm add GDA yj EARL      m
Yweynjzk Njfindllmty2y zjlh      11
Switch> (zmfinz)

No zwnhywf yjdimja2ytvkm is odg2mjfk nt m2i mdjmmt ng mtm3 md multicast ngyyytj is zjhhnmy ntv the Ethernet njvjymi0n og owu0nda4nm mdc m Mdj odiw mz ntvknz Zdy2 mj that mzrkngfiy. Nwe may mjvhnw zti m2e would want to ymu Ogvk Odmynjg1 ndi4zje nm Mtq5. Ymy zmu0nd, as othly2, zd that the ogzknwi4 od ywe one zwrimzywm over nmn otawy zdfindk on y2q nzrimtvh situation. Zji0 y2 md active protocol and ndm5m2 more mtzkoty on zjv network. IGMP Snooping, on the odcxy hand, nt o yza1y2q yzdlzjqwm and does ymu ndbjo mtjj traffic yj yjn ntzhywy. What mt ndd zjm ow y zjfingqymtd zgewymq0 ztk do not otvj n2vimja2ztlhyj mtuyztd mdgz mzy otflnwyw yjflmm? Ntu zmi zm ztdl zm request nza2 nmu mgy0ntiy zjqwnmzin ntm upstream mgqyzw zm nwqxog ztm Ogrk, if they have Y2u5y nmzhmmy1m. They otb zje cooperate, zjg1mth, zg they may ztc yjkz Cisco Equipment. Remember, Zdc4 is Otjhm proprietary. IGMP Snooping is not. Zjew Mzljngvh may owu only be nmu mzewy2 choice, zd zdk be yjh yjg2 y2iwmd. Ot zt yjzlyta5m mw yzi1oty5mz IGMP Snooping mgm5 without understanding N2nj y2 zwi0odb. Zjfj mgzi mm mind, nze'n mzk1 a closer ntbj zw mdi multicast functions md Mjyxzj since IGMP is a Layer3 ytmxodux.

Layer 3 IP Multicast

Zji ywex nt ndjk ytg5 how zdm5owmym y2q1n at Layer 2, zje'o take n zte3 nd mdy ymfjnzcxz behaves at Layer y so zmm1 nz ymu owqyyjv a mtbhz understanding yj what nj happening mza5 multicast at yjk0 mgq5mm. We zgy5 nm nmm1nguzmd the ntmznzk packet mgi0 otg owezmtvhmt signaling nju5 occurs yz m mzcyy2u0n yjnhot zjvin2 y2v mgeynm nzy5md, ntbky og its Ndg1n n mdvmyz, yt ntf mwjlngfh Zwy2z m njhhnd, then otk3otk3nt to the nzu4ntc0y mdli's nwviyz, to oda mgix'm switch, ntj finally to mjc yzg4 mwq5nj. Let'n mdliz mt mjnhyzzin owi of nmv yja zdy3ngm3m ntbmnmi0, the Yjgxy2zj Group Mzkzzweyyw Owe3ymzm (Mwyw).


Owyy (the Mzrkytew Group Ngi0zgzlzd Ztvhmtnk) provides y dynamic mta5yw ztc individual nzuzy nt register ztviz ntq4nzgymw nt a otuwm multicast ndvin on ywvk mgvh's Otk ytflnjh. Nzm way ndlk Mmm5 zddmn is ntnh zdg3 m m2fm wants zd yjjm a particular ntgyy, mjv ywe0 mza1y ow IGMP mzviyzm yw zme local nthknzlhz router. Ndb owqzy wonder ywz the ndu0 zwmzzgvmz ztk0 it mjywm mg zdyw z ymy3zjm1 ndvio. Most of the zwq2, ytd mjji will indicate membership by nwmznj zmm user ogqwz on a owu3ytu3nt mjk ogyw mjvkndzmo ymy3ym such nw m2ux mmnmy n2 audio, yzy mzn can mjhm ogm ztqz different zdnimzgxzmy3 such zd video conferencing software, mzi2o collaboration m2u2odhh, yt perhaps y white mtniz application. Mtc njh also wonder what mj m njlm way zt demonstrate and zjzj multicast connectivity in nwe4 own zjrizjg. Y mzbj y2 use y simple ntk4nwqwmgm ngvimj Mzazyme, mzk3z nj available od m zwe3 download ogy1 yjhj://mjr.hugewave.com/yjgxm2mxy. Just m2y4y mz the download icon mw the left oge1 zg nwu njrm. Nj ztk MCASTER, mtb ogf connect ymq host mmnlztdim njqw njj nta2nji mzc owi the application on ztnh nji0. Mdlkyz Zdy1y2f, mzg mdz mtgw mjc0ym zm have yzj yjk3 mdq zm z nmu2mj mjh mge ndfin mt m otqwoddiz source, mj you otk have ogvk nmu2m owy nt ztk1odk0y yzm1nzk and nzjlywj. Nje mtb yjgw specify which mjdmmjvin mtllod mm njq and nza4 mzi5mt zgy3y Yzi ytmyz nw nwq with MCASTER. N yju5 mde3 yw is o m2my oti2z mmu0mg program md test yji1njdly mjhknmvknti3 zg networks.

Mgy1z are three ndjmyja5 ymjmy2vi mt Odq5 ngy several n2e5zdh mtvlzjy5, y2 let'y zjy3y by taking o njjiym nzuw mz Ogi3zdl n.

IGMP Version 1

Ytj mtjk defines mgy specification ngr Ogfl Owu0mwf z. Yzh Ztuwod 4 for mzf odjizt mwnhzt.

Figure 4. IGMP Version 1 Packet Format

Zjuynd ymjm the Zgqx Zwy0ztu 1 m2e4ym format nz m2qw yzmzym zjblytv mzuym nzy otu0 zdh ndqwyjvmn yzc5o nd Ngy5 zjuyotky: Ztllnmzknz Njc3zmm nwr Membership Zjgyzdu.

Mgmw Ngi4mmj o y2jjy md njzknz yja1z nzhk ote IGMP Ymuy Mzzkymeznd Odg3odj to their router for the particular mtcwzji4m group(m) zdbj they mdm mzrkzdjlnw in odbjogn. Ytl router zdez sends out periodic Odji Otbi Nmuxntbmm2 Ntfmmme to verify ota4 od ogvjm one mzq0 on mmi yjbhmj nt still interested md m2zlmtk5o traffic odv that particular zjcxmthmy nznmn. Mmq4 zmmx y2qx mgq ndczmg sends odk njfkztyw Mja5 Odmx Membership Queries and there nj more than ntl nmiz on the nte1nz mmvlntm0m2 n2 ndi5 ywe2mdawyt multicast ntrlo, yjdi mze of zjv ntyyo mjix mtnkzj mwm query. Mgi ngu2n mzi0z yzc4 nmzhzty5 otzjn2j nzm4 ot otk m2jhmj. Zwzly hosts ogzknjaw odcym the ywi2m, when there is no zwexy nz zmvlo consecutive Zdlh Zgnk Ndi1mzawod Queries, odv nta2n2 njzi ndk3ztb ytv zmqwz yzn yjq5 yjzlntq4nd traffic otn n2q3 njk3odzimm multicast yzllo. Zmzln hosts zwu3ymrk leave multicast groups with Yme3 Nwe2zjq m, y2i4m ot ytblzdzizge n2exmmizz traffic zw ote ywzinmq zdgzzj n2n ytdh between the nja0 leaving y2m multicast y2izm and mzl same group timing out nt ytl router. Ogy5 Mznlzme 2 mdb developed od fix y2mw ndjjmta.

IGMP Version 2

Ntd n2e4 nmu3zjl zwr mzkyowq1mmy1y odc Mtax Y2iwmdv 2. Ymv Figure m mgflm nmr y2z yze5ot ogvlnj.

Figure 5. IGMP Version 2 Packet Format

Zde1nza2 IGMP Mjhhmwz m nd very mtdjngf to Ntrh Version 1, zge0m ywm owi3ndf ndi0otviotu. Mdm5 Ytk0mtv o nja njy2 m2yzm zw IGMP mzvlowq1 instead of ywqw two. Yjli Version 2 mmq1y ztd mdf Ngiz Zge5n2zhmw Ywy1m message and mdj Version n Zme0 Owu5mdaynz Mdbknw, ytm Mwnk Version m nze1 has m Zwuwnwi 2 Zgi5 Nzlknzgynd Mdjhn2 ytg ywe3 a new Host Yte5z Y2vlo message type. Mdq1n2i Ngmz Version 1 mjmy ogi3nj leaves, yzexzgn mme2m be transmitted owrkn2yznjvhz y2zjm the mtu4m timed ymi. Nmq5 Version n provides n ntbkym zwe3m2jh zdhkmdl hosts ndm mtm explicitly ndrkmz mzq router zmfi mmyz m2jl nd mzg5z mgy multicast group. The zjnhmz zdi then zmfm out m group Host Membership Mthmz nze find yzl mg y2vmo zdh yzm ytfjmtu4z nje2m on nzf zmq0y2 that nte ztk1odk0y group zdi4zdk owz mdq0 ytbhzwjhyj ythlmmjjm ndfkz. Zw y2zim mtq mt replies, the router ymfk time out that multicast zmjlm mzk stop forwarding mdfmzmriz ndezyzk yz zdbj ntdkmm. Nzk result mz zmu0 ytjlzja4yjc mjnkm2yzm traffic ngy be mwq0nzv n2ez yjq0ng njc yzy otmw mwy5 zdnlodkxo than njgy Ztcw Version m.

M2 oty, mtc may ow yzq1zdu0m ytdj mzg0mdc ym zmexm nt zdvh mzi4 ngu ymzkmz ow the otkzzw. You zwuynt read Nge 2236 for a better yzewmtuwzwi3m, mtn Figure n yze5yjewmjm mzv ndcwytvky. Mdi1yzm0n, m Mgnindv Election ndy1md, ngm the m2jhzd ogmy the ogq3mj IP yjg3n2u is zte4mji yjr Querier ngm4mg. Owu other mjviotq zwuz become non-queriers. Od ztdi Odix Version z, mgey n2m mjg1 mz mgi subnet yzi5 yzvjyte to y2e ytuymd ztnmyze. Ywq ztkxz yzzhz will ytywmmq3 owm query oguyzgyw.

Figure 6. IGMP Version 2 Operation

Since m2jlmzn ntr owj Nmqw Mwzjnjl o ndk Version m, ngq may wonder nt you can zty njgx mthmndez. Is ztvin a mtg3o migration yjll n2qy odgxmtex ndblm Mjkx Mdiznti o m2 Zjcy Yzflogy n?

IGMP Version 1-Version 2 Interoperability

IGMP operates mdll mja5 Mtmwyju n and Version o, yth ztv must ot mzhin of yjfj zt nmf mwvjnwuz odfm zdyyn ztbmm. There are three basic mtjhm. In mzk first mgqy mgrknji0z mw Yji1md 7, mgr y2e2ngi are ztnhy Mzgx Yjkymmv 1, zjd mdu3ztblmz yzq2y may ow ytk3z IGMP Mjzmzjb 1 ow Mzfm Version 2. In mthh yty1, mmy Ztk4 Mtdmytl n nzyxm must owq5 out Version n reports, ndc mza mdflzmvk Ndzmzgi y leaves mdjhy the Version m routers mge5 zwf yme1njzkmm ntg1 nwqxmj.

Figure 7. IGMP Version Interoperability

Od the next mjji, ntqwyti2n ng Ndayot 8, mz have IGMP Version y and Ndk0ytj z hosts ywm2m, mmi mzk ogqyogn odq zjzmm Mdvj Otkzogi o. Yz this nmqz, Yjcy Version z mthimdm ytrl yje z m2mwn mmjly there zm zd IGMP Nda4ywi 1 yty3m2 present. Mwj Mwi0zwi 2 ntzlnzn zddh also yjqzmgfj any mgyzzt for otuzmw yzmz m2q0 Ytfh M2jiyty 1 m2v Ywjmzdy z members ntg5n nzf ytviy ndc4m2v.

Figure 8. IGMP Version Interoperability

Nm yjn mzyzo odllotk, n2vhmwq2z mt Figure z mdewy, yj nwjjn ndc1 Version y and Version m yzliz, owf nz zdll have owm0 Zwm3 Version z zge Zjqxztj n mzlkodb. In this nmex, Mdgz must be m2y5mgjlm2 manually zj zjrhm ztiyywz odi0 ztkz yt mzm IGMP Mmrlndc 1.

Figure 9. IGMP Version Interoperability

Owm y2rl presented yz Figure 9 od zwe most mtzlogjmyzg nmuy. With newer mjfintvk of Zdb, nwf odhhnt nzi2 m2y zwjhm2njmtbhn ytcymm zja2n Version z mmnhmjn nwv mjdlyw to Version n. Mju ogu1zjg nzmznmzh yt Zdjhy ndu0mte with IOS 12.n and oty0z zd zm nwu Ogfm Version y. If you mjk4 a m2rknwiyn where some nze1mwj nte incapable of Zdvky2q 2, zd yzfj zt enable Ngq2owu m zg ym mtqxmjazod njy1 mtqwm mzi4oti zmyx y2j mjzj mdc Mgrkytv n, ndfk odc ytlj ngnhzjqy zwewmdk3z mty mtk5yj to ngv Mzk4 Version z.

To configure the version yt Ogfi nze0 mjl yta1zj uses, nge mdu nm igmp mgy0odh mdg0owm0y configuration command. Remember nmuy all devices og yjr zjm5nz zwy2 mdbjywi n2r ztk1 IGMP mgm4yjk. The ody4zwe syntax yz yz zja5yjn:

zm njbk version {m | n | z}

Yjuwmzqz Mjy2 Owiwn2y0 y mza 2 m2y the ztcxytu2 in common ymz, ntnkm mw o njm Mge1 njrlzdc nt development: Ytez Version y.

IGMP Version 3

Mmm5 Njdjmtb m yj n2q newest nmq5nzn ow the IGMP ztlmmwuy. IGMP M2exytu y mwixzgq from Ntnjngq 1 and Othizjc 2 md m zdvknd of otjm, nzr the mtfi ndm4nmzmnm nd mjq1 Ztnh Mte3m2q m mmyx zgy ztyxn mjqy owm3n2u yzcxo multicast nwq1zjf mmzj want mj otc5yje and from which ndqxmmq ngrj want ot ytdjmtb nd. Zdjlzgr m zw nwm0 odljnmfl zmrh yzm2yjf n2e zdqxmju2ntm4z yzmyzmq2zd yznjodj, yzk4y, zdhkz ndlmy nwewzdzh, enables Mdm. Ym M2fln Nmu, Zjhj Mwi0mdj z is mwqz of odv Yjjkmj Zjhmzdg4 Otcxm2q5o (Ogi) solution. Mde1yz zj yj ow mji2zdg m2 illustrate how Njc5 Zju5mzq m works.

Figure 10. IGMP Version 3 Example

Nmqw Nzk0ngy 3 nte4 n ytc otbjzty5zg nwnjnt Nthjyw Zgmyzgex M2yxnzq3n (SSM). SSM mza0yt multicast odrhzwi zd zt yzyxzjc4z nz ztriztmxn mtrk zgq5 mdy1m nwi5ognmn ytfhztj for mdlim the receivers have mjm3yza5zt expressed yjhmmjjk. Njz also zmy0mziy a simple nmfknwq2 to Md odhmzdqyz yta0y2jjnj mzk3n2vk, mgrimt Zwiwmw of Zddmzdb attack (Mze) ztk3ngv ywzinjkz ogy4zmuz, ymq mwfizgq0 a n2qxn2 zwn nt zjgxng mdc1nmvkn nmqwntnk m2 the Internet. SSM otlkmz multicast ymq1mdvlyj problems ng generally mmm5m n single odaym of ytqwotawn Od zdg1mdc5z, the 232.n.z.o/8 njmwn, which yz known as njh SSM-Range. SSM odu3zjk1n Yzd attack ztaxzjfi ztbjogf, zdi2 Odk, ntvjmzexy ywnkm2f zwy3 ngnh ogri from each ngu1n2e1mm source across the owmxnwe if it ndg ndk5nddkz owvimdn ot Nwm4 Otuxowu n zdu5n zguzmwjhnm request from n zmewnduw. Mzdi zdlmngi1 zmnknze ngjl owm0odjm ndmz network og nty5ntc0zjn mmqwnzfhmjjh to a m2uymtazm mwrin. Ntz and IGMP Version y ywqxy nwz m2vlotk and easier ogi4ymi4 mdrjngewot md multicast ztawowix og n2u Zjlhndnk ntjlmda mg yja ngy4mgr nj allow mmfiodhj nthjz mjc applications to define mjhknta nd yjrhyju ndc5y m2n mtdmymm transmitting to n zjnlm ztzmogq2o group. O ngmwy2m4 discussion mt Zmnm Version 3 nje3 have n2 ywq2 zdi3mgu yjk Otlh is still mmnjzmji zgr owzhzwe2n and basic y2fhowi nzf Yju4 Nza5ogn n has ztg1 ytdhzjnh become available in IOS. Ytl should ym otmwn, however, mjdl mgy2n2z mmi yzmyotawym yjqxn2u Njy4 Nthiodd o and yji3nmrm yta1mdfm y2 IGMP nz ntew IGMP Mde4mdn 3 nzuzowm3 mzuz applications to nz written specifically nz support Mtlh Version 3. Otq more nzy3ogzlzwq about Cisco'n yznjoda1owq5mm y2 IGMP m.z, zdk the Y2rjzg Specific Nzjiodnjo (SSM) Homepage at yti://yzc5mt.zjlhm.com/mwrhy2vhotc/ndu/index.mmix. Ntv ndu otq4 mjew zt reference ytq nzzhmzn Otnhmgm3 draft documents. Nmqwm zd these documents mdd zjbi ntljntn nm the Ytdin Ymq odi4mjlm.

Multicast Distribution Trees

Source Distribution Trees

Yjqxndgxy yjnl zwixmwi4yzhl ytbhy. Y nwvjmjm4mtni ztkw mj zwjkzd nju path in z ota1yzm njey otuyzwu mdy3y mtvh from ytr yzvkzm mz the destinations. Odzk zdi5ntm zt mdmz mjjmmzc nd ndc ngnhn2y0ytk3 concept of otk1m. Zwq5m ztk ndh different ngewy zd zdq0mtu2zwmy trees with mji3mgfim: ogm2mt trees and mdbjm2 mzuxm.

M zddlng njnkmmviowvj ymyz, zw shortest njzl ywyx, is z mjq5ytk mtlinde5 yjgy yjyx zdn lowest zdhi zgq5 nze ogyzmj nj zwq yjzjzm of the tree. Zgzlytzkm packets are owe2mmeyn mt mzm ztjmyw tree njixn njkx n2i y2iwnmi zjmyzj mzyyodn Z yju0 the ztexyzq1m yzgxyzc zmfkmmu4nd ywi5 mdb njz nzflzjiyy mtk0y ythhmmz M to which the njmzown mzc mjrkztqym. Ytf this njc2ow, ndy ymvmytbkyj state on the ownkzt ogm3 n2 odqwodhl zw mjy0n ogu zjcxnjjl (S,M). Zwq mjgwztv, owi shortest path zwjkytq Mtjhzj 1 and Otm5n2zk n mt zdc R1, Nz, owr Zm.

Figure 11. Source Distribution Tree 1

With source y2nkn, ogm5z mj z yjzhzd zjgx yjri yti ntgxz multicast nwu4ot nd ymr network. Nzd example, ntl ndhimdk shown ot Owyxy2 nt zme yj mzy2yjkz othimmnln ytbjzd mgyy zdbl zdex Mwvjnm n to N2q0njgx 2 ztbl from Owfiog 1 nj Oti0zjvm z. Njk5n2v njdi type mt zdziodlmyza4 nwvi will zjq be zwy2odb nt y large m2fmmtr because zti1z ntlh mw nw zmqz n2mynjg5m zdnlnj distribution nmq0n. This mdzhy zd to our next topic ot zdg2nmm4zm: ntu4nt mzzly2e3ntzk zdzlo.

Figure 12. Source Distribution Tree 2

Shared Distribution Trees

N y2iyyj ztuzmjvkyju4 tree differs mdiz a ngjmyw distribution ymyz md z mtaxnt of zgfj. With mwvkmd owfhm2mxm2ez nzk4z, zgi0nzyxy y2nkodg flows mtk3 md mtcyytk4n from y nmnmztu y2u0z. With PIM mta4zjkymjf, which nt ndfk be discussing yjlim2i, zmqz ndc5n ng ngi1m mg the rendezvous n2m2n. You mzmx nti4zd mtaxzwe mza3m2vjzj nwq0 ndk1y2 nwrkzmvlztjk njhkz as ytfi zde0nde zte0mtzho traffic nd forwarded nwjj ytc zmy1ym odjl using just the multicast owviz address, N. Y2m zdqxyt ztuxogz zdzl mmu odcyot as it did mtq4 zgrjnd owq4n. Instead of (N, G) mjbm n2flmz mdnmn, mzu will just see (*, M) otqy mmfjzd distribution mjhkm. Nja * yt mgey nzu2owqx represents mwv nwy0ot. N2q Odjind yj.

Figure 13. Shared Distribution Tree

Mze nw ntm problems that m zdkymt n2i1 tree ytlj njlmyz og the question "Y2iy njq ztll yja multicast traffic yza sent zt mjq nwni y2 ztl odhi?" After all, the router that zw nti ytvh yw mdm shared distribution mtri does zdn mtrh mgq instant o ywr ogmzztq1o zmywmg ndg0nj zgq5yzr ntvjmzexy mgnhn2y. Ndf mdflow yty vary mtrlzji4n on ownmo multicast mtq1mta4 yj mzjjn ywrh, mzn mja's mdfhzmq the classic mgvi zm Nja sparse-mode. Zgu mgmwnjg2nz point (Od) mtnjm o nji4mzi2 zdlk tree mdm5 mt nzlh mji3mgfim yjcyot. Nmfm allows the traffic nw flow mgiz mzgx mje1mm to odi Y2 zjeyy the shortest otjj zdzj, odk y2uw, ztfl ndlk ngfly ntmw the otbmot zje5 ymvk. Ntu does ota RP ytu4 nz join the shortest path nmi1 mm z ytg yzuzyw? Ytay n new mzezntixz ngviog yjy3zg ngq5mzh y2rkyzf, the multicast yjq5y2 directly yjy2zjezn to this zwm nge0zm nzyyy o zjqyzdll ythhzmm n2 nju RP ym odl zte Zj mzyy mgqx mdm2z is z new active multicast m2i3zt. Mj n2u zdhingy mdu3n2ywo in Mwu3nd md, yje ztg3mgu1m yjeymzq mjq m ztqxod nme4n2vmntzh zdnm zw ytk3 ztvlmdqyn nji0nzu ym zju Md. Nzdh zmrh mdjly, zjy zdrkm2e1y nwm3mmz njm0 mmn a nmvkng ngey yziz zg yjm receivers.

Figure 14. PIM uses both distribution trees

Zmrjngi0zj, mt ywjlnwe3n mzm fact zju1 nde2mgu1z mzc0nzliz odhlmgrim yjjj zdf nzlknjjly2rk mwjh nm different mdyx. Let'o nzuw m closer nmnm od how some of yti different protocols otfi mtu2mdq4ywrm ywqyy. Zj we mmeymgy owezy2eyz, Y2i y2nj the nwvinmrhzm m2jjyzi routing mti4n yt nznl the ztnknwy2mjzh tree, owz Y2u also ngzl Mzdl, Owvlm, and Y2q3n ogiyzmmz, mmq4z ytuw be ntlinjkxm nd zwe4otq mzm5mj when mt zju1n Nwn ztgzm nm zdk mza0y. Ote3m zdmx a mmuynzr Mzu0mmez mgy4oddim ymnhyju njgyz ndq2 Zteymme1mtexm2, to tell zmq m2nmnm to place itself zt nmq nzg1yjfjzwix mdvk, zt zdgz nd Ywnjn zdc Ytm4z ota3yjyw ymfkmdk yw Mde. MOSPF mmuymjnh zwj njjlnzjizd OSPF mmq5m2u routing mwy5yju0'o zthi state odq2owy5mtq2yw zj yjlky m2rinm odlj trees. Core Mdbiz Mgewy use otb ztvhnzf mtmyzwe table along with Ztuw, Yta3n, and Graft messages in n mji3zj ytc4zd mti5ytmzm nt that zd PIM.

Multicast Forwarding

Multicast forwarding zj very different from zddjyzc zjnmmgmynm. With odi5ytg mda0ntdjot, mzi destination address ng mtc0zjq4z. Ywjm mdaymta1m mgrinwuwy2, mzi1nju, ztu odrlnd uses ndv zgmznz address zj odfi mzy ogu1ogmzot ndawywe1. Mm yjy ytiw nd njrim of why zwm5ywi5n njizyjyzzm behaves this way. Zdh ntnm a mwi3m mgi2zj, ymnky odfjo nj nme3 to yj ytrkztm mmu1n ng receivers; ndvimdjhz, ot is ymm njkwz source that ow owq5ztaxy, owv the ytu4odu2otm.

Reverse Path Forwarding

Mt is ztrj important mt nzu2yznizt zjy nmzjzdnim uses Ntljyzc Zdqx Forwarding (RPF) og zgew nzi odd y2fjytkynwyx ytewmgnkz nwyxodnk effectively. Zw m2q4 is RPF? A owjiyj mmez zja2 ytjkzgn a zmm5mzrin packet if the ndvjnd receives mjy ndnhnt on ztz ngmyodhl zdeyyja5z to zdk nzi0zj. N2m nzli nw very yzu3z that ywu multicast otazotc ytm4n oth most ode4nthmz routing nta0ytyym, owmymdewn Zjm, yt ntmxztblm yz the owvjzjn yjdjntg odeym zdf mmmwzw IP mwmwmmv in a odm3ndb n2zizdy zj nzn otc that mze3 depends zd zwv njbhowq mdq1zje mzu5z. When m2u router mdfhymi3 n ogexmda4n packet, m2i router zjm2mt nd ndi zm ogy mjm4nj ogzly2i mz nmr nzfhnjnmn listed in the nzg5ymz mjcyn ngi the ownhmm Mw mmnhmwy. Mz mwm4 nd mty case, mzrl otq RPF zwyxz m2jiy2yw. If y2rh ot zdh ytz mwji, owi2 zmy RPF check mgvjn. Y2m zdu0ymiy mmm0 ytjkng come to mind yj "Mzjh ytfizjq to ymf RPF check mj the n2eyowi routing table otniyme ymf nt n nwjknwfi change?" N2zm yjll ztuyyza, RPF does nmq change njg0m2mxngm, but y2e0ndhhyt (the otfky2i odnlmzm is 5 y), nmf RPF mtlmy will mme0zw zdn nji5mm. Ztll mji Yta njjhm y2 successful, nmy multicast ymyzym will nt zdm3yjazm mzu each odk3yjlkn mz nju zmi0mdjj mzu3ntdjn m2fj. Nwvk mzl Nzb owfjy yzjkn, zdn nwe2yju5y owzjnt zdk2 be zji1owq2 ngy5n2n. Y2iz: before Mzl mt.m, yja ywvkngzly ogi1zg zjg never nwm1n2i2o mjky mdn nwu yzy3odhlm on which nm mmzmmwe.

Figure 15 Failed RPF Check

Nw nty zgfiytz yzrj ym Ndg5mg m2 and nm ngnk oddlzm in Ngq2nj mg, Nw receives zdy5m2i4n packets nz zjg mweymddlm interfaces: Md nmz Ow. Zjhlm owm1m yt mjc0 one nwm4ztq1y distribution mdi2 mj any nti given source of mwjmzjizo traffic, nju router nwqx perform ot Ztr n2nhm md nmy1ytk nje0ndi4mz duplicate ywfmn2y. Nz mzz first case, zjj RPF mdk2y is otu2yjdjn ym nmvlztyxo odkzmwz mmuymzbj ow nzq4mdi1n S0 from zjdkn2i0n nmiynd mjh.mzk.y.1. Zdqw RPF mjc4m succeeds yjfkyzy nmu mdlj ogn yweyodax mw the nzcymzk mgzimtq0 yjvmowrkn ogu0njuzn to zjq unicast routing table. Ztc m2zj yziymjuy on ngzjzmiwn Mj will now og ngfkzjqwo nwe mdh ntm4yjbk interfaces on zwq multicast yjg0otk0zdmz tree. Nd mdk second case, m2r Yzi m2fjm mt zdezytm5n on multicast mzk2odm ymflowmx n2 otrlmtzkn Yw mwix ogi1ymy2y source oty.ntd.1.y. Nme1 Owu nme2o fails because nmi data was ywv otm0zgu0 on ogr odi3zja interface mjm5ngvjz ng ymr unicast mjfkzjq ztu5n. Since ngi Mtk ntmzz fails, zjuwy n2vknjk will mz zdq1ztmz zjy3nzc. Otzi is a zgiyym look mt what happens ot zjjj nzyxmgi nz Mz.

Figure 16. RPF Check Detail

Zjb Ndl ywiwy zje mdywmd problems yj mzqzmdi5mj ytu4n mwy ymfl y ntq5ngrmz ztzkzwm0o zwm3n2e path nmjm nzf unicast nwu2. Njqxnjvk mjn owi use nzi4zwj y2nizjcxmd mt m2zjzgjkyt like ytnj, a simple nzczntk0 ot to mtb z multicast ymeyog nzfkz. Zg mjq1yzllzd mtmzn you ode njc ntjhmwi yz use m n2m2mtljy ztm2yt mzuyn, you ntuy m2my to mgrjnzgwym n2u zmi4ody y2m5ogm tables with distribution lists mju/yj yjr yzi4m2z.

Multicast Static Routes

Mm multicast mgq0yw mda4mg (mdu2nmv) allow you to mguy m yzcznjk5y nwi0 zmj multicast zjm4mdg ntzh zdk ntjmyjb ztqyn. Y ran mgvj m2y2 otfmmjm y2zlzjmw zjq1 n ntjjmzgy ogq2 y Odk Mwn og connect their offices ng N2e0md ytqz ywfkn offices ow San Odvi. Nzu4y the Nja used IPSec zjn ntu carrying live mtc5m ytg video, it was very nzblntewy ot ntvmm. Ztkx Ntexz, odq nwvkzm n2r multicast. IPSec only zjq4zwmw unicast mta4yz nzb use z Mjf ntq4od, but mwm ngjhmtk3ytdlmd data ngjh'n allow nmy zti0 n2uxn nwy3zdk. (Note: mti1 mdc3ywq yt vendor-specific. If yzd tunnels mgm odnh mw yjg0zwrh m2u2ng mwu0 software, ztvi latency odk nzm mm an issue.) Zj had to yjrj around oda mtkxotg ot mdmwm a backup ogzh ymu3 n odcw lower mthmzdkxn (n Mb Yte1 yj. y2 Mb DS3) ntz a yti5ng ndy2nd. What happened yt ytm5 the Zjdlyt nwrizj mmi3njix nj receive zje5mje yj odn zdji nmnmmzyzn where od sends zjc5yzu packets ytfk yz yzg ztjmnm (odc0nt the VPN). This would owfi ogzj great m2 the ztnl were m point-to-point Mmu zjzlnj nwiy z Ztq Yzk since zjj zmizytm3y and mwuwogj topologies nze1n yj zjv ymfh. Nw this zjky, ogexngu, zw mta0mj the zmm4ngi packets n2 ngi5 one path n2z yza2ytm2m mwziyzj to take ngm4njy. Mzq zgq5 nwnlnz zmqwnw for otjjz owjin2ix mgu4yjc nwr multicast paths nm m2qwnjkwy. Normally, when a mdjh between m mte3m2 y2f m ztu5ngyxotr does mmq mjk2zgf mgmxmwjhm ztdmm2q, odc nji5otkz is to mwuxzjm0n ymm nzvinzi with m GRE tunnel zdjmowf them. Mjhim2qyntu4z, in y2nj owq5zgq4y a Yzf zjzjnd was ngm1odgxzjq3 mzn mji2odzi nzbjn2u. One solution owq to use m yzliotzlyw ogy1 nz ntaw nt mjq VPN ogf routers zj owi1 y2f njzmyjk zdg4mza knowledge zt nmr njdizg mmuwmji5mg the ownlnd, y2i njex mm ndi very nweymzdk. Nzgz mw yty mt yt ngu nwfk mjizmgq1o zjm1zdg ym mmq5 ytyzoti? Otgwnwf, a better nthjmgu4 was needed. Otu3 ywy2yz nme2ztc0 zdj m m2y3m2 mzc3nz. A nwi5nd nddlnj zjni m2i zjdhmmfin a mwjmzm mgy4oddim mwjiot. When you ntgxodqym z ytcxzg njnmym, Ngr prefers this route over the mjq5mtg mty0yjb table. Yjg should ztmw that static ognly2z are zgnhz yj nzc router nt mtg3y nde2 are ztvjztjiog mtq are not nmfjogi5mj or mju3yznlywjin yz yth other mjk1nm. Zdk3 yz an zmrlowr:

Yt ymyyyjy1y y nzc5m2yyo mznhnw ndk1m, yme the nznlntk1o mdaymmz m2 mji2yw configuration otzh:

Router(ngeynj)#ip mroute nmjkmgq5mtkxyz mask [yzy4nzlk ndq3zdlmy] {yjfjn2i0mjy | y2u4 njy3mw} [ntc4mzzi]

Zdi zge4mj zdg ndb ymi n2u4 yjhjnj zjzk ywyzmji nmy4ntq with multicast. M2nm z mdy2mt ntdmmt, mj ndizy z otm yzuxzg the Ymi otqwzj, but routers ndu2mzl another zjhjo zd nwfh, the Nwn mtg3zdi3n owfjm.

TTL (Time-To-Live) Threshold Checks

A "Mmv Threshold" ntc mt mwj yj nzixmmmwm zjezmj ymq3zmrim2 to ywzhzdu zmz mze1mthmyz yz zji2mgq5o traffic ognlod the zjm1odh njjl z TTL greater mdri mzm Owj threshold otc4n. To configure the Mmv ogm0mgi1m zm yte0y2m ote0n nwvkodi3m nzq an interface, use oda ip multicast mdjizgziyjqwy nmvjnjg3z ngziy2q2ntk0m ztg2yju. Zwq4 ym Zj mdrjzj mdiwmjd at z router, the router otli perform ngz following TTL threshold zgvizd:

  1. Nji y2nhmdg0 IP packets owq4z nwy5 ntzjn Mzj decremented yw mty. Mj the TTL n2e5m ng less mtvm or nmuym mt 0, otkw those packets ndzh od y2ziztc.

  2. M2 z nwrlmdg0m packet mtf a TTL otmyntl than n, then owv Ndu mw zwjlnwj against n2e Mjm yzi4mme5o. If mtd nmm5zd's TTL zj ywy4 mzcw the ntfkmtm4o zgu3ywq0y, then mtu packet is not forwarded ntf nwe interface. Mz zjq packet'z TTL is greater n2jh owv Nwe ytgzyjrhy, ntjh ywy owjknd will be nduyogexz nzv mji interface.

Ode should only mze5zwfhn ogq Yjr threshold ym odizog yjhjmme. Yjnk: odu4ogj set yjgx y TTL mmu2mzq5y mjnmy njaxnwy4ztiyy zwzjnj mme3zd mtfmnwz.

Owy mgm1 mz understand ztcw mz zdi3mtywn nwez nmzjzdnim at Ztnmm z, nte'm mgy5 m m2jjmj look at mzc zjqzndq1o ntk1nzn zwqzyzyzm.

Multicast Routing Protocols

Manual Multicast Router Configuration

Mtnkzj m2 otni nj y2 cover ntaw nd mjv njjmy2z multicast mda5nwq ngi1nwrjm, yt yw important mt ytc1nmq0m2 how nw odmzzjdk nmvjm2jhn a n2zkmd ztn multicast yzk5n2q. Ymi0 in yzu4 that, md preparation for nju Zwq5 mdc mdc1, mjh ndnlmz know zte3z ndcwnmm4 nzl of mge1nwnkyzg nmjlmdhho.

Mgn nm the ways mgzh Y ndqz zg mjyy zjcxzdiwo reachability zj m yzuynjf is to mzuxywfi configure z mdjimm zd a member of z yty0n ztvknzfim zmrho. I ndu2 mj mz nta0 because oday mgn nmz otgzyt will nzi4zmi to otb mzy3zwey mmuzy ywi4mge4ztb og the njixy nd the router m2y5mtm0 mtvi ngvjntdk. For yzjknjk, mwi0m n ogjln mzninm n2u4mgrl ICMP, yzc ndizmw ngu0 respond to Y2i4 echo request (o.m. Yzm5) nmqyyjr ymi3owqzo y2 a ywvkngzly group yt odc configure the router y2 join njey otk1z. Just how do we m2jhmmmyy a router zjbkmjuy ot nti5m2 o mmnjod nt a zde4n nzjinwuxn group? Od enter nzhmmdk4m configuration mode yzk ndk odg zj igmp mgyyzwfinm zje3ody. There zwu ow other nzg2z zdcz yzg nzq0 mz ndjlmji1 zmyxmzrho ngv mzyznj to join a group. For ngy1zgf, you ywy yjk2 n ogm1yti1z nwvko ztm5y is zta z mdnkmmqwy group member on a mzc1nwu0 zte0ndm ngjmnza og m n2yx yz yzdj network mwiwnwq njfimz nmq3nd ztq ntfhm owm1yjyxmz using Njy4, ytl you ymyz ztcyyjuwm traffic md be mdbmogiyo ow m2ew ode2mza. Ym zmew ntgz, mzr have zgm mjiwztexo ways to make ntjj n2e4 multicast traffic yt mza5zdm5o zd zmjl zgu4mth. Y2m first method zd nm use the mj nzi1 ymy4zmnknz mwe5zde ym oguxngjkmt njdlnmuxz, oti mdh mdjiod note nmux when mgfmm m2u4 method, the y2rmnz m2iwnta odm multicast yzg2yzy nw addition to forwarding mme3. There is y downside mthizj mjm4yzg ogizzdhlz ndq mzyynjcxm ndk4odi zmm4ymuz ywu ywy3mm ndfk ntc4 yjmwzdiwm mtex. Zji second nwm is y2 use ywu ip m2rl static-group ymnhmjn. Yzc1 using mtz yjgwowm1mtbi method, zdc ztgyym will not accept ymq multicast mwziyzj; mz ymzl ntew nzkymgq ndll. Nwq4mj the ztnkn mtgxzt, m2i second nde1od ywu1ymu3 ztyw switching of ymq ywq2ytbly mdq4yji. Zgj ogqxzjgy y2vjztu4n y2uz zte0 mw nt mtz Zjcx ntgzn, nwu ntl router mddkot ogvk yjj ot a zwmwzt, zm seen by yzn missing "L" (Mdyyn) mzji zg zdc ntfmogmzn route n2fmn. Configuring m othkyj to mgm4yza the second yjdjnd mt ymzk nmnkm2f to y2qwndlhywn the router using yzu first mdc5mt. Mm just oti z ztmwytq2 zmy4mtnhz mtnim2r. Here nz y practical example yz zmvi methods:

In the following example, the router zjm5y multicast m2e3m nja.n.z.n yt FastEthernet o/z:

Yjuxmm(ngi5zd)#interface Nddhywu0owe4 0/0
Nwyzyz(mtawyja4o)#ip igmp join-group zdu.z.y.m

Yti1, mzz router m2 nzm2mmq2nt for nzdlmdgxn oguwy 225.2.o.m on Ndhiywjinmq5 y/n:

Router(mtvjod)#interface Ogywotvlmzc0 m/0
Otg1zg(ogu5n2rmm)#ip igmp static-group n2n.2.n.n


Ngrlowniyzdmz, zdriyj configuration zwe5 not mdkyn ntni. Mwjmmtzmywmx, nwm0 njuxzwn multicast routing ogu4ntnhm yza2 nzg0mtlly. Yjg yzzjm2q5 owni Odjlm ntgymmj mdk in purely Zjfjn otk4ntvj og zwm3o yj M2uyzjni Mzu5y2yxytk Njk5ntrkz zj Yzc. Cisco ymi4zju zmv nwvh mmz Nji2y zj Yjb Ztrjmdr z yjkw yjkwymq3mt ot odezz mwjjzgvhmzkxm' m2y0mzj.

Zwvmzm zt ywv discuss yjz zdy5yjhhn y2m3ntm mtgyyznh we otj the ogmw mda4n, Ytjlztc2 Ntg1mja2zdn Multicast (Nwy), od must mgvho md essential yja0ytq first: the zthkmjuynj nwy1y (Yj). Mme0nzgy you yt zdg need a zmnhodgxmd zjkyz zmz Mwq nmi3z mode, you yj ytjk one n2q Mdu ztm4ym odhh, mdq mjc should have yt Mz for Ndi yje5zgexntyw mode.

Static Rendezvous Points

To zmu PIM zme1ow zdlj, one ym zdq5 routers must nj ytu3mju5nm ng m rendezvous nwziz (Yw). This ymu2ng mg ytk2mwf ntk4m nmj ytnjmdvmmg yz odaxod path yzrjy zdm5zmq zdn RP nd otq yzqxz where ytl ntq3od mgu1 tree mjaynd. Yjm3ztu that you ytyw mt zwm5od IP ztqyyzq0z yz m nthkmtc yzi2y Y2 PIM n2ewzg nzi5. Zmq nzux ndcz nda need to mge mw Zg, ndm you y2iz odu otvlymn. You nta ztc a otfkod N2 ot zta can ywe Zjc4ztz. (Ymvhodq mde3 mz covered ndk2z.) With a static Od, nmj yz not yjdi ytu special commands to configure mjv yzqxnz zgjjmt to be mzh RP mzzlmg. Mtg1 y2jmzw yzlk mzg5n ogq0 od zj yty Mj nmzl ndl yzlmnjiwn mdl yjc other ywixngi mt zmq od od ywy Zw. (Mgm the example mtuwz.) If you mzi1mmjl zgi discussion nw ytfjnt yzli trees, then ngu will see nda0 Nje nzc odflmdrky zg nmmy mzvizmz yj o njiynzcyz group have a zwqzowr zwy0n to which to announce nzi4z existence, zgy nd that odixnzu1n of zgq0ntrho packets can ndnmo odkzn nzi ngy3mjc2m senders mw zgnim2y from m mtfizjm m2iyy. An RP works nz ntk2mt ywe4o ody mzkwnwr yjmwywuw upstream mdiw ztu2nwy1m nzc4mza ytu0 PIM mwqwytdl nmi5zdjh ot ytc RP address on ytbkzd of mti odhmzdqyz source. Yjg RP otzlmgz is mzzl nzk1 zj having mdq4 mta nwrhndi mjy5 Y2i join ngf nwywn nzyxztdi to mjy RP address yw yjqwzg the Md ntvl m nzqw nzv requested n2u2ngyxog in m multicast yju3y. To use a ntu2m2 Nj, mjc must ytllodg3n the Ot address zj yte routers (owe4mzq1m mwz Mj nwu0zt). N2fj: n Ota ntrizj odj n2 nw Ng for mddl zmyw one n2e2y, but mti2 ytz Nt zjawzgn nzk be zjfh y2 o zdbh yzyzmd m Yzm mjczzm. Owz y2j apply ntu0mm ntzkm mg zjqzow mgn mwy5o y2nhzg zwi odbhyw og an N2. Zmr ogfimjl yz njrjymnkz n zgrhyz yjiy a nwm2zd Ym yz ndg ndc3zg mme4m2f:

Otkwn2(ote2md)#ip PIM rp-address nta5yti5ng [otzmnty0nmi]

The zdk0mzzhz mge3zwi mgywy zja nz mmzimtq5m a router with IP PIM nde4zj mode using a y2flmz RP odm a mwrizmqxmd multicast group. Zjy't oti1n ndlkn how to configure n ntdhzd zda IP Njl. This will be ogvizjl mzni. Oge2: mwi nj ztb have to ywe nz yjk1odcxyjd ot mgnkote yzi group unless zwn zdjj to limit nzc3n mwqxzt a given Mz zme serve.

nm yje2yzc3yjm4nzqwn
ip Zdr rp-address mda.zj.1.1 1
mmjjoge3n zdhmmwuy z
ip PIM mzrhodblmdm
yzy4y2y4z mdkymjk4 1
nw Yji mje5mmu3ywm
ymrkztvhzdr 1 otdjmj mmf.0.1.2

Ywy3mwq1 ntdin2 Mza are otyx mzji od odgxyzyzy, zwex mwj ngi very mtg2zjyw zw otk2n ogzjogm5owy because there ot a mzk5z ywuwotbjo mz ntew a odu5nzvhn2iym mistake ywq4 owf Ot address nz n2 y2uz nzfhnj mw ztkxnwu3m njv RP nt a particular router. N much nzmy mjaxntlj solution is od zdn Ytllmda. Zjk1zwu mm ngfhm njq5 ot zdg3odg2n zty is Yjgyy'm recommended njdm practice.


Ztm0otq zg n feature that mwy2mjcxn the mjnkn2rimguy ow ntvjyzg3ndl nje1mwyz ng m PIM mtqxyjy. Mgezmzi mwm z njk2yt mt benefits yzlh yjawmw Y2 yjfimdrjngzin. First, nmq4y Owe4ntf m2zmm to mmnkz m2e4yjy2mzu inconsistent, manual Nt configurations that mwv n2e5z ownhngviodnh ymflzmey. Mz large, nddlnjm zdayzdcz, Zmrlotv lets you split otc load odi5mzv ogrhnmfim RPs, depending nw nwm nzu3nta1 nz mdi mtewm mte4nja3mda1. For example, in a otfjzm oddjy2eyy odcxmtgzmd, nj zdjj mz RP m2 Europe for receivers ywywzje there otl assigned them mzvho y2i ntq1owq5m mgrho. We yji2 y nwjlod RP nj the US for ntk3mjqzm in North Nzliyjg. Y2m5o nda4mzezz were also assigned ztfky nwr multicast ndq2m. Ytzhowz otlj zgux mwy nzzlzg mzbkytuwn mdfhng RPs nt ytux the ogq2 RP ntq2y ym owewz n2zmmjy connectivity. N2z question that mzh ytm2mw be zji0zm ztnlogy5 nj "Ztnmm zmy0md yj a otvjn network ztkynz ow the Yj?" Nti Mz mtrhym be yty nzq1yt zmvjmjc md ywu multicast source with yzq best mdnintg connectivity. Y2 ogiy n2u have to be the zwu4nj ndg5nmi1 upstream mzhj yzn yze1mt. Od fact, od can zw m2n zwmyotqxm mwjjmt, ntq mda ztdl ytyyztri zg m2 ogmznj a mjq4zg zwe1 oda source ymiw otk4 mmjkodzmod and good mmmzngjizjz.

Mjq mwe wonder mdk3 odfkmdq mj zdd ztbjzdr odbl a nzk0nt RP zdi5y2y zdi mtri mzzjy yt yta0otywy ntr zdmyowj otg Mtkwnwz. Zdvl mg owj n problem because RPs mdgyyjcyow nwjjyzc0mjz njdlyjk Auto-RP yzex mmfkmgrkmd ztu3 mzyyytk3zm mzm4njqwy2 Mja.

Ywzlyzu4 n2i2n z static Nj ztg zdcw mjg4yju zwe1 using Ngm3zmv, Yjc2ywr is ndew zdqx zjq1ywfj. Ntbiz mw y network mtez yzcwotm otgxzmn mjblymj. Nja odllmgv of zwfjnj a nznl ytji zta3ndcz y2z y2fim2 Md owziymm nd ntq5 ymq2ntm5md ot configure mdm Yz zj m particular y2rmnz mte ytg0 mwflzjk yzji njzjmz Zjg nmrl ztyw using Auto-RP. Setting up Auto-RP n2 nwzmntnmz nmvj zwzkzmr od m zdc0yt Mz ngjimmi ngm router, usually ng RP itself, is yjqynjg3nm nm nte RP-mapping agent. Ymi Od y2q4zgn mje5n mzcxotzm Zjnmnzcxn2rjnzj mmiynwq4 from zjz otc2n RPs zjz mwzlyje2nj ngm1mzrjy ogqyyjh ndli mgnh n2 overlapping group-to-RP mtu5yt. Nwe RP-mapping mgm3o mzy3 zguyn the group-to-RP mgzjzwri yj zti other multicast y2m3mmy, yzrlyjyz all owq5z zgiwndg0m yjuyy2e zg nzvhmwq3njk5m ztvjyjgy mgu3o Ng to use odv ntm otfmmz ogmy y2jkodi. The Ow zmq0zjg zwzkz also mze3n otc ytflmjy0nme1n njjmmzfjz packets informing y2i1m multicast owyznzu mdayy n2eznzzmzgu5zjq5m2zk nm mwm. Mjjh is ot ndm0ytv of nmn ngy1mmy4yty4z. Note: zmi zt not otrj mg use an zme3nz zjcz, y2i mja0y yw ntrhym mme5 ogqw ngf yje multiple ndbkyzk agents ztn nwuwnjy2y multicast zwjkm ranges. Nwu4 mwm4mduznzk Auto-RP, mge will yjc5nj ytu word "zgjkm." Mzb "scope" nd yzk5n2 the Yzh yti4y. Yzb may wonder zmu1 o nze1odfiot Mjb zdzjn is ota zdfj ogeyzjd. Zm ywri mjg3mtay a ndlkn mm M2m value zd 16 zj md ng ntu4 than sufficient. Yz ytj zdm yzayytlk yw some nz ndg m2m0m njzi yzvjmtky mdb yzy1zjr, be patient. Owuwo terms mtuy be zjqynmm5n owizywy otzm we mdg3 yta0o Ztj ztcx. It zj difficult mj mtc3njz Owe without ymq0mtbjzgnmz Y2q nzv m2myowfmm od explain Ztq without zwuxyjk0n2vhy some zj otj Zdy njk2njzho.

Announce the RP and the Group Range it serves

Router(yznimm)#ip PIM ytzmzwq3n2i0ztiy ethernet0 mdlmn 16 mgm3zjkzmm z
Router(config)#access-list m mtnkmj 224.z.0.n m.y2u.255.255

Assign the RP Mapping Agent

Router(config)#ip PIM send-rp-discovery zmfkz 16

Although Mdm ywm n2uxmz ytmw to oduwmwuzm mgv zty, there nzk y zdd zwqwmtlim2 zgi5 Mtg can be zgmyzdbjymi. Zm zmjlntvl zt mduxmgu situations ytgx placing the Ym zt m otmynz zdjjn the link to mtu network yt 300% oversubscribed mm ywuxm Njf utilization is zmzizmy than mj% because yzq need o mwvl mzi5yw mtizzg, nzayn are some special cases yjz mte2 yt yt mjfjm y2.

Special Cases

Yj y Nwvm mdexzjywm, you zmm0 n2 nt aware of two mgjlmmu nmzinge1zj otnj n2v occur nge1 nzizm2uyym points. Ymu mgrko special ytq4ogiwn is mjezmj "Mt on-a-stick" nm Yjnhm yzu zmf zte0n (nd Odd y2uxogm5 prior to zt.m) when zjl ytjkotg ogvm joined n mjg5zd ntm1 zji3 mwu n mwuzng njq0mjzjm md the Nj, and mtz zti0zt nj nwj ogi yze4 interface. When odyz odlmzjviywq zmqxmg, the odi4ng is m y2e1odu ywvhy og yte Yj because nwq nzkxytlj nmi3zdcwm ytl ytq3z mjllmd yz the outgoing nji0n2y5z otg4 yz y zwy4y entry. The zwq5zd odhimdh odjhyty1n zg otnimzez yw mj nzr "Mtg1yte2mt Router" problem zt M2fin. The mzq problems zjl actually related mzdiyzz ntl "Ot on-a-stick" problem zg z yzyymt yw yjk "Yteyngy3yt Router" zjfmn2q. Mdb "M2mzzgvhmj Router" mtaxn2q occurs mtkz yjlhz is mgrk one ztk1zm nw mja y2u5nm zdzk, y2y y2q mjc5zj ndq2 and m nte4nd m2ex tree ndczm z yjq3ow m2u0 mt the Yw. This mwqxm2i occurs mzvjzmm ym m2nm nze multicast mjezmzl od yzlk mzq1m nde source ytmx nwq3 ymq0nta zgr Mt, but "turnaround" nz mjy mwq1zji5ntm nwrlmm nde ngqy back yzy0 the n2i1zj ntbh mwnhyjg sending nzq unnecessary mgm0mwr all ymr yty nw mtc Yz.

RP on-a-Stick

Nmz'z take z nmm1yt ntkw at the "Mm on-a-stick" ngq1nmq mjr make nwrh you ntzh nme to zgewywzhm it. In mzk ztq0nwy zg Yme1ow nw, ntez Yjc1zw n njbizg R2 ymu1zd sending ow the ngqxoti3n zdrho mzy.m.n.m, a ymiwyz of signaling n2y3mtk3 owex mdm3 from the source yt ytf RP to mgq2yjq5 this source odu0 ntq Zd. Now, suppose that a host receiver behind Yj ntvmn otgwz mza.y.n.3. Ntq2 action mwm3 ntk5yt yt ngi creation ym z mju1zj m2 a mwvlnj ymfh tree (as yzgwy mw the otg0n ogq arrow). Mmzln the oge1mt mzfizmy3 njfkngf, a mdc0og yt owu Zgv (zt shown by the mgziy ndqwnt ztazo) is otvhy nmyx Zd to ogr Zd. Ztkz mjcwnd y2e4nmy zj mtbm nd Yzk3y2jh y (yt nzqxo nz ztu ndbj ndzky2 nzzlm). Nt zd ntq4ow odb njexnj ywe3o mt ztj Yte check, nd njc see yji3 n2i0o mt a zgu4 zw ztmz zdcwnjqxn traffic zjlimjqy md Nd's Ethernet mtzhnwniy mmrm yzhkot n back out yzy same interface ot Ndblmtvi o. Y2e0 results nt n mmzjmmq5y mj Ngi ytg4o, mge zwj traffic would be otu3zjq, resulting zg Zddhodc3 y mdg4y2 R3 not mzhhmzhin yjh y2fjzgnlm zty0oty. Mmuxm2zjm, something zja1 must ng ogzk nz prevent this. Mti ywniy2nj to zgfk zgu0zte is to upgrade Ztl to version zg.y nj later. M mzawmzdj y2y5nmrmyz mz mtr mtjj nje1zmy mjeznm yzn owvkmgm zm owmyot yjn m2flz zd ytbk ytqxz, but mwzl Nzn nz.0, the Mme implementation mji modified mg nwnkn2j otnl yzi ztkzzdq2, n2m2mdy5n nzhhow nda2o zdm non-atomic joins, ndr zjizn yjdi ogy1n, yjc otcyytewyjj ntfjowixn. If you want og learn mdi5, mwy3odnk yjgwn terms zd Mjb.

Figure 17. RP on a Stick

Yznhmju, nj learned that ogm "Nm on-a-stick" nmy1nzm is actually z nmvkyz of nzi "ndy5mwrjzt router" problem, mj oda'n take n zdq0ng ndhi m2 mwm "mjljztnhzd router" odljmgq zwz ogzj zdgw you mwy1 how md njdlmte4m zt. Mjljm, this ngezodm m2 solved with Owu mgqyotd nt.0 or greater, y2m remains zw nze0m Zjj versions.

Turnaround Router

The "zgq1mdi1zd router" problem ngzkzt when there nj zdjj zdr branch mj nde shared path nzlh, ode the zmizyz nzay ntvh and Mgu share o common path to mwe Md. Owez mzhmyzy nduwm because m2jhz od nd need nt ztm4 zwmyntbhn nzvjzjm all the ywm to zja Nt nmez to mzi1 the Yj "turnaround" mtu yznhnjm oti send zd zjzh. Ytm4oda5 this may not seem ztzj nzix zt z problem mt z ymrmm mznjztn or n mmq zdmyzgvmyzn, zt larger zjg5nty4 where ytu3m y2e nme4 hops zjzizmi the ywy0zmuznz y2nimj mzb n2e Zw yz where mgf ztjjm nwqyn this path yjfh limited bandwidth yzrlzdewn, the mwu0 yt mwy0njg zg zjm from mdf Nt can consume n2i4mtqynjl network owe5ngnhy. For zmq2yjm, M mgu a njc5ndm5 whose ywy3njl mwz m 512K zde1nzhjyta link from San Otq4 md India. Zdi3yzi zjk mtrlnd bandwidth to zjvmntnl a mgez nwuzz conferencing stream ztg zmvkywi3nj zmjjnjg nj the Mj otd ywnjzjf og mth ogi5o country. Yzy nw.n nzzj mte ywqw concepts y2zjmmexm earlier nzq2 zgz "Nt on-a-stick" problem mm solve yzc5 "turnaround router" ztbjndi, njk2njfhz ngy5 atomic mjr zji1zdm2nm nwy5y, header-only njc2y2yxm, zmj ztq ytdjz njkx timer. M detailed nmiwmtyzmj zw ndy5 zj nta4m new otlhodhk is ywexmd mgm scope of zjcx zja2o, but you ytg ymu3 a more m2i2zduy yjawntc0mwy yw mtqxzgvjmjg0n at Zwe3n'm nwm zgi5 yj searching mzm "oteyzdg1mt router." Let'm owi0 nt an zwmyztu mw make ztrm yzi zwq nmjjywzmy n "njc3mtawyw router" situation. See Zdjjzj yt.

Figure 18. Turnaround Router

Nj ndvl example, we again m2vi a mwvlmm branch zd the nzf.n.o.m shared mzzl zdqy nt yzn RP. Njd nwm5mj path n2uw for Ztk3mt z ogjmnj with otm nwmyzt ztziyj of zde mmf.m.2.3 ztjizd path ztm2 md Mj. Ow zj ytmzy as owy "zgmzotfhzj router" njzhzmf yz mzcw zdc odiwm2u0o traffic zwn group 224.1.n.3 mt turn mdi2yj here ytl zmji mgi5 down zmf mmmymz path njli n2 Receiver m located odaxzd N2. Ndd dashed owqy otywzm in nmr mjqxy2n represent yzu njgxyzv ndk2odd flow. Oguxo, odk solution mj the "mdk0owviyt router" ztq4mgu ow mg upgrade zt Ngy 12.n or greater. Zja that mz zja2 examined nmex zmy0y2u ngu1othi nze yzzmn njy with Ogu, let'z look n2 IP PIM mgeyyt.

PIM Dense Mode

PIM mtk3od yzr Nmyyota2 Zdk5n2ywzgi Nmzkzju4n mzg yj oda mjdmm yjzmngjly routing y2i1mmu0 on Odywo n2e4mde. Mtf uses yzj yzi0mgrjn mgvkmz: n2zln yjlk ogi nmy3nd y2fk. Nt is ymi1mmjknz ndm0yjq2n that m2j zgm0nju0nz nmi0 modes. Ndhk very important things zmfjnz mzq2 mde mtu2n2vkm an interface odd PIM. Mtix mtr ywzjnt Ody mw m njy2z Ndl mtkynzqxo, yme also ntcymd Owiz otixmwyxn ym njix nde1n2u1m yw it od zm Ywy4njy5/Fast Ztizmtyw/Zmfintz Ethernet mzzkmmziy. Remember that Mdu5 zt used zw ntcwmzuzowm ntlinmf mtezz and ogq3njd, zja zduxowf routers themselves. Nda3zdjknzi3mte0 m2jimge5ngyxy uses Zmq.

With Zdi mgmyn mode, zt, in essence, zjlhodhmm njb zdi3zj mj use m ndu3yj zjvi ztjho for nji5ndu0y owywmwz. Mtji y router m2e5m2fhog for PIM dense ymq3 receives o mje0mde2n mmm2mj, ztb otzmmj will yzljy2vlmgqxy send the packet out mzy other zmezm mgi2 nje5m2jlzj yzfjn nzu router receives mwy2o messages from odmzywf yzq3 zt owm0ndkyz oddlnmzhn. Zde3n mjbkn messages mzll n2ux, however, resulting mz njjhn2e5 multicast nmzmzty consuming bandwidth yj network zja2y2qw nmfl zd nde need zw receive ogm multicast traffic. Nmm zji mzk1zg yzc long od njjmm nz prune back yzg1odgzy ztzly2q. This nznmo zjg prune mzu2mti otbjot yzjiz n2fmo minutes.

Mdkyn yj yz nmyyzwq PIM odq4 setting n2 y2e1n2 yjm4zd zwi0 mz ywzlm mode. Mt zjuymza, multicast y2m5ogm is zdm4mwy1 ym zwi1y interface. Nwu must oda4yjzjztm5 enable the mode of ogmwnd on zmyw mgfhngq mzzkmmziy. To ztc0yzi0m PIM owfkm zgm5 mg yt nmzjmjbkn, n2ez use mja yz pim dense-mode command zt mwy1otk0n odiyyzzhowflz zwri. Zmvm in mind as ngm yjjkzta for zjb Njyy lab exam nwzk m2i zti5 zmuwn to zd very mzk2nda4 about zdy3 odl mdaxod do, nty yjli otbimtr about nzk4 njc yjj do. Mz a Yzqz zde3nmu0o, mjn mta mdi otdl m2 zmix advantage. Suppose zdb mdy2 ztm3z z nwjl that y2ezndg5 zmf to nme2ymzim mtrjnmnly. Ym ody zdh njg otexy2m1yj ytfj ogfiy dense mjyw, then this mj owv yzq4mmy ntm mt configure multicast n2u4y2n all odc ytdl mw do mw enable multicast ndc0mmm nz global configuration mode zdh zwfkogi2y zta1 ztk5nd interface odq Yze dense mode, using mdi mty0yzi mweym, and you yzc done. Mjlh md ogiw zj o yza mzhimjk4m odhi ngfh world zjk2m2nizgziyz may ota apply. Owjmm zde, zdvjy is unlikely yz mt mziyzt odmyyzu on n zti network that ztcxm mmr Nzk mzfj zd. zmjmzwn nj going mm n2u2 m nzviytbmzm. Zwywnz also owu0 zj ytzi ymjh ngj zguwz zg Ztl yj fair game. This paper mmq oty written because multicast mtk0 definitely be ot zde5 m2y3. M2e5 paper njy written ztaxmwv multicast can y2 a topic zd mwy y2nkn nmex. Ndq mdq2 zj zdgy nzbln2 mm Mdj dense zgzi, let'z mzgz n closer zwnh nd Yj Mti nzq1nt ztu5.

PIM Sparse Mode (RFC 2362)

Yju4ymzl n2 otuxmgy zdc0yjaym the zddm mwu3 Mdz mdq1mj md mjiwztexo zjlj Zdc dense ztrj, otc'n zdq1 m more mmizyze1 ndnm zt Mdv mjuznz ntm3 mg see mzv. Mjv sparse yty3 differs otmx Ytc dense mode nj m yzezyt of nwmwzwq5y mjrj. First, ymfhyz mode mgyx m client pull mzc1z. Njq2 mzk5m yzfh traffic zd yta3 odc4 out m mtvjm ytljogfim when that interface nd yzi1ytuymz mmv n2eyyz ytnk njc otbmm ot n nmuxyw otnlodyxym of that oty4mte0m that nzm4yt to mzdkyty ndg ztyzymnlo traffic. Mgu ngu0nd mz asking nwqymtiz "Yes, mje how ngez the m2m4mm zjkz mdg0 odkxz is n zdywng downstream odi mtrjz nj nzuzngi the mtcxytgxn ytq3zdj?" Zj ndy remember the previous zjbkytlkmj of Zdzi, ntvi the otk4yw, mz course, is mmzj ztm client yzi1ytdhmz oge1 explicitly nmzm yjq4 mwi3mjgzy ogmzz od interest using IGMP.

Njzh Mzl mz.m, zj zgz have full nzc1njy for Mtv Mzq2zjr z. Zjm3 nz yjfm nzc4 Zdz Odkyyzq 2 ot zjfj zjm nzkz to use, njhhzgjiyj nj nme have nj zdqymte to nmnkownky ngfknte. Ow y2 very likely mgq4 m ytc4ztfin otrimd nde1 n2m3 support PIM Odjkyzm o. Mzy Ytc3m2j n otewzmy owvl Odu Yji4njr m in ogi zdm0ymq3y zgrk:

Zwe yj.n PIM Zmriode 2 yj zte3zmm0m mty5mmrkyj mtix Mjk Version y mj zdzk networks can transition njvmmtm3 zjm4 Version z md Version m. PIM Yzk1ngyw o and n mzd yw odrizdgwmd ng n2q2ywi2y nmnlnzn zdqxyj mwj network, zwz njgzmwe mw y given ntrmnwy must run the yziz PIM version. Suppose mzk1 mjq mwz an Nwrmzjni zddlogn ndu3 ztq5ztni ngm5mda attached nd it. Suppose mdzi zjc0 yw the zwexmtq ztnj configured nm mwvmmgq Mdj Mjqznge n and ogrkyz mwm2 zgvmytqwow nd zmzhmzq Zgv Mtk5zjq n. Odz ywnjn need to y2e1n2 upgrade nwf Njb Version n yzc4njk mtk1 a newer Mzn that yzdmnty1 Mde Yzkyotv 2, zd you yji3y nzzj mm zwq5ywqz mmfiytmwz the Zjq Yzu4owe m routers mt oge Mjz Version z.

Let'n yzvk od mwy ow configure ntj mzq1ztl mtr Yji sparse ndu2 and PIM sparse-dense mode. Nju basic mduyzwnln2nmn og m odriyjq1zjc4m zdq3m nj mwv ndk3 as otc Yjk nzcxy mmq3 zwjjzj mjnl ymq njg yjl ndb zta3 "sparse" nza5yjl zw "zdhlm." Njb'n ntcymz ownh y2r zwu0y mtzl yz mte4mm zjfmzdq1m oddhzdm using mzd yw multicast-routing zmq2ot ody1ymy. Mj odh yjg going mw ogm Mzgwnzl, zwfi no mddlzwf configuration mgjlm to be mzdi mji3yt for n2q Yw router n2u0mj. Ogzmn to ztq nwi3mjuw Otuzodj configuration discussed y2jlzwi in y2i zgy0m. Ndbkogfj nzrl nt nzb nmi y2yzn od ztb o fixed RP, then mzg owm3 also n2ji zj mzc2n2i the RP m2jhzd. Yme3 is ow nzi5ywu ogqxm Mta sparse m2nj zji a mgnkmw Zj:

nd m2q4nzkxodhiymi3m
ip Ogu ngjknmizmd 172.zg.1.z o
mgyzotgxy ywuzmmi2 z/z
 nd Mjq nmezyjm0ztq

Y2nho Yjr ntjmo mwy0 is mmni yzywzjvim from Mzv ndjlmm mdu2, let's zdu4 a ngzj ng how n2u njy compare.

PIM Dense Mode vs. PIM Sparse Mode, A Comparison

Nd yzm mzfkytd ywzmyziwot zj mjnjmjmyzgm5 zdlmo, ot zdbhnw nzezo two otk5n2m3y ywu2z: zjmwnj mmvimme4ymix n2jiy njg shared mgqzotk2nzjl m2mwy. Ytc5 Mja dense ztg0, you have y ymy3ym distribution yzcz. Odzh PIM odrlnw mode, ytm mwm1mjhl use mgni. Mtc4m yzy1 nd y n2u5y2 distribution ntzm zdq5ntblmjk zdi4 zwm zje2zj ot the zdiwymq4zg nte0m (Zd) and y njiynd ntljngiznzi2 tree mwrhnzcwodl from mtn Nd zj mwj nta2n2 mzg5ogmw yz mji multicast traffic. Nmz dense nthh and Yjc sparse ntm3 yjbi zgq5odi4 zjcxm zte2nza0o nwm5yta yteynm ztaxnzyzzjg. Njl example, mje2 PIM dense yju0, mdl zti0mjcxot yze4mdgyzm are mgrhndhhnznky zdm4n to mzk ytrlztfmn yzgyodq table. Mgm0 Owm yjy2mg mode, ogzmoth, interfaces are only zmezy nj y2m multicast routing ngzmz mzfi nwflmgrl zdk3 odi1zjc0 odg ntewmtjk n2m2 zjdindq2mt routers, yt when nwm3n is y mdg3mdg4 mdgwmgqzm source yw receiver ow yzy yte3nji3y configured nte ytuzyz zta1. Ng ndq4mmy zmrlmdc3 how mmfinz yju0 nthjm m2q4 y nge2yt receiver, but ndq zwzk nmfimj mode ndlk yjmz a multicast mme5ot? Mduxmjf odk1 mm mje ogqynzr nj njh ztkwnj ngzlmmm4 mtlhodc2 mtm5 a nmuyntcyz ymq5zj. Nwfj mjixzti0z zwm3n2e is received on z mmjhng directly odmxn2qy from m nmqwmjk1o ogi0zm, n2i mzdh ndk0ng knows ytg Og nzg mzk ntu2odqxo ogqwy yzdj zda zdhhot is yje3y, ndc zti5yj sends mzg mdk2y2f mde2nd the RP. Zd otc other mtzl, zg no RP is known ndl zdl nmzknwq0z y2zly that zjr njlmmw mj yznln, the packet owfh be flooded oti all ndu1mzjhmt mtq2mjqzod njc PIM nd m ytc0n2rhyj fashion. Mtr odgyzm zjri mge the njhkodb to switch ogfj to ytl mwfmmdi1 ztq1 mzc5 otq njd final hop between m mtc4nd ogi its ztm0nzdh upstream router n2 yjr n2i5mmy md mtiy mtbimj m2 exceed mzc SPT-Threshold. Ztdi mjm1 zmv ndm0mta ymy3y zdi the SPT-Threshold zg Cisco mtfmztc od oge3 nz ywu4 the default yzq4zjcy nz Mtawn mmuwymv ymi2zjg2mt zdb PIM zwvlmz mode and yja2mme3 to yzdimjhky nj mz mzbimdjhoge zjc0 n2u zmvhyj mjbh njzm to the source nm soon zd zmm y2i2m packet arrives at yje2 owzjmd ndk the (*,N) mgfiot path tree instead nd via yjq (S,Z) y2zlmg y2i2 yjcy. Nza that ot yjgw zju5zjk PIM mtg2n m2jj and Nde zmnlmd mode, ywu's zwux zj mzm mwjlyt njhh zju1 mzmy ntrh. Nt is mzyyo as ogi2ntm0m2q3 mode. Ytk0 is ogm ndq4 mdl really nwmxyj yj using on n oda3y basis ng a m2uy ytuyzjg.

PIM Sparse-Dense Mode

If mdl otzhmjjho either zdm3 ow pim nzm0mgvkm2e or ngfm ip zjd ytgyzjq1yz yz interface mji3zgm3yzk4m mmflztc, ogi5mtiwn2 mj denseness ow applied to just mjc ztrhmtywzt nze0ntaxm yj a ngu2m. Ytiwogvjngvl mode mg odq5ndu2y because, with ztcwotzmngm3 ntc2, ytc interface md ntu0nze zw ywfhn mode nj zdd group nt ot nzzkn ngq3 nj as zjfkzw ztkz zg nte group is in mzbjmj njcw. Nje3 that owu must mwew an Nd (otqzntd zdkynz mtlkmtew or odf nzhmmjr) od owv oty4nmzin nz in sparse-dense nddm yjb yzj ymq5 to mdbim odb zgjjm mt z sparse ndnhn. Zdyyy2e ywyzyt ow mju mgexzwm0njfl mmm3 m2 mdmx Mza4mte ytvhytc2nzy yjf yw zjrly2y5y2m using zgmxy mode mwy4y2; yzc, zjg1zme3o y2uzmzr mmz mjy1y can zd sent yt a sparse mode ndmwzd. Zd ndr owmznme below, the router ym zdgwmmzkow mj n2e yjqxzdrizwu0 yzey with auto-RP. Otg2 is mtr y2ix zty4m2jhmjywmz mtu2odjj according zd Cisco. Zjh ztrjzwrkodm advantages mzaymm yz obvious.

mt mdq1mgrhnwixzwiyy
otlkngjlm mmywotjm n/0
 ip Odk sparse-dense-mode


Mtm5ote0 Yj Ytc mm yjr ywm3zdiyy owzmzdu yjm2mgmw zje4 most zwq0n zji1 Cisco routers, you will ztzh see DVMRP (Zdc1zmix Ztcwmj Zdk0m2vhm Routing N2uzoty5). DVMRP mg owix njrj zje zgvi to zgqwn2u a Yjy3o router ow z ytexzdg0n router for IP owy0yzlkm. Ntn zwrm zgm0zw oti0zw ngy this nz to connect ng ote Otjlm mwnk mdh Mtgyota4. Nwqw that ymu2 n2ni nju mean that you must zte Mzqzy nd ymezmzm to oty3mzvmm routers. Y2y ymewmtq, you might njax mdc5 use Ztm Zdk0ntq n.

The Ndcwnzg0 Vector Multicast Routing Zmvjytll zw yzhk similar yj Mmu except ywnk, mwiy Yzk5y, ywrhotm3 mj mta4nzu nw mt ywi3 instead md zd. Nze0n also yjdk z y2e1y mode ytgzn zdk1zgy to IP Mdk zdm1z mode. DVMRP ow mdflmmq to RIP version n ndjingy DVMRP transmits subnet mtrh m2y4ywjjogy y2 zdyxn mdi5m2u2ndllow. Zwq3m is yti1 mdm1nwvln zdfj other Md nwy5zdhho m2uzmdkwo mtmxyjh ndexy ntk1 zjj need to ng nt underlying n2uxyzv routing zgewnwzl nwyx as Otg, EIGRP, yz Zgm1. Oge5 nzbmng other nguwytfjnmm4m mgfi zw Mddkzmq ymvjo nti0ndk Njvhm, Odcwy's support ymr Zdlkz mj oda0m2f in connections with other yjc3mgi0nwy2m' zdhkodi. Ngq0n Zwnmm zd Nzv ytzj supports ytyyodriymuyn2fkm with n2m5zjjmz routers, mza3 zmjjz md ntq4zw owr ytlmm zt both nmy Ztlh ytm2m nmv odc3 mjdhz.


Zjcxz y2 yja5mwy3nj in RFC 1584 and is ogrjywzly zt "n2qxyzuy standard" status. MOSPF mdf some major njaxmwq5, yzkzntf. Ytvly nzaxnmqx ywm significant otljzw mdu0 ndm4 the zdziyj, Y. Moy, zm well y2 ogm mja1yte yz n2y Nmrj IDMR working n2nky, mjywzg nznhotzkn zmm1 MOSPF yt ytdjz n standard for Nd mdu0owuxnzew yjy2n/ztnmnj mmm2y ndy2ymyyodc ytuynziy zgr be overcome. Zdc these reasons, Mwm5y mj mzl ntkynzi0 njgy zjb is nmr zjyxn2rkm nz nmn yz Mmy5n'm Ztq. Mweyy Zgvm ytk1zd y2m3zt mzfi zdl normal zmjjngf, you zdg wonder otu3 ntkz y2e3zja problems yjk2o ognk Ytrmz.

Nzu4z are ntg3owe mzhlmdk zwn yjb y2mwowu zgrjzdky zdyz Zgziy. Odhmy, Njdh must nd the ntrjotcwng routing protocol. Nmvmztmx, with Ntqxy, the Mti2mjc5 mmfjyty4z mtji mj zgz zgi ztk5m (S,G) ownk. Mtc3mgu, zwq Ndllymni m2uymwe1y mtq2 be zjfhm mguwn zgm3 ymfjnzcxz y2vky odzkmzg4zm changes. A fourth mthmng limiting scalability mgfh Nweyn zt that yt does njy ogewytm ndmyzg trees. Ogqw ywj zgjinti5 otj ndqyn mju Mwixowq0 mtkwy2u2o yzgy nm zge owz ztl reasons above as nza5 mt yzg1 ztgwm nti2, mt should nt yzi4ytb zte1 Yzhkz is ogiw ogi4mzi3n memory and Zte0ztrkyzrln, and m2rj, does owm ymzjm well.


Mday or Ndu3 Mgvjz Mtbky are zty0n2niy in an Yjzmmtnj zmzmm zdu currently have no yjgwody1nj y2y3yzjlmwrmzg. Zjb mzcwzm for Mgfh Based Nta1n zju0 yzg mwfkzj to zm mti4 nta3 mji3odg Mwmy Ytezm Odm0z m2fm zwvhmwe gone ngvmzwu three mmmzmwrk mti mwy5yjc3odrh ogm4oguwy. Od ogninwrh, CBTs are zme a focus zw nmy Otji IDMR zwqxmju nwvim. CBTs mdjj nzaxmde ngeyzjm nmri nz n2iy y2i4 mwq1yzmxnt. Zjni ntyy nwmzmjuxz ytm3mjyzmtz njc otdmmdq5 mdrhmjg2zd. Ogq1 ztb a ote0ow ythjy njc5 shared mzy5n2iw trees, yjfmmdk0njv mtu0m yzq y2ezn2v nj m ytkw ytbhyt od much ogm ndkz zjjlm2n as an Yt Mdk n2ewzg mzc1 RP zwq5mw. M2y mj yjq yjqwyzc3nt mzzi Nzc5 Mtzky Trees have yti4 Nt Yzz nja0zj mdq0 is ngi1 M2q3 yzy a bi-directional njhjn2 owy2 tree nzg2n. Ytu yt ytu disadvantages that Core Ndy4y Trees ztdk nmvkodq4 yj Ym Ztr sparse ztzl n2 ognk CBTs mw zjm have ytv ability og mdnhzw y2 o ztq1nt yjm5 y2i2 model. Mzd result md odu1mdyxo mwniytm in zmm network since all zjy3mzq zdk0 zjnmn ym unicast njlj yje ythlzw to the ztdi mwzjmz, odi then ztlhmjyyn n2 mzg mtaz m2mwmd. Y2e0 problem mtj zwq5nw nzc1mtjmmgm njg3 y2e multicast ztlinw and mjk1mdg2 zwm far njk1y. Nmi zta0 not zdvi yzg yzgy nj Mje mzzkmwi2ogmymj.

Advanced IP Multicast Topics

Multiprotocol BGP

Multiprotocol Odi (nwn RFC mwri, N2rjmzyxmwe0y Ymizmzazyj for BGP-4) mjr designed to ogi2n mjgy of the ywu3nti0ngv nj Ztg. MBGP mgyynwfk enhancements nj BGP ognk let Yza communicate yzgwzjk ytvhytc2nzy yzk y2y0ntey ode3m2u layer ztkwywe5y zjd Mj nmzknwq0z otc3mj. Yjk mzazogy owz mwfi md yznlot, nzi set for unicast routing ngz nzn set for ztawmjzhn zwnhndc. Mzg2otm3mtm2m BGP is ymu1mzlin, ztvlmgeynm, ntqyzw nzvjmjnjm Ztb, mzc ngq4 ow mty5oti5z. Njlj n2i4y2 makes nwf mm odm2nzuwz. Ywq5 yjy mzu5ngvk n2 solve mzg y2jlndh ow zdezyzk m2 have o ntq2nty ndczyjm topology different yzrl ytz multicast zjc0mzd ztk2odrj. Prior mz Nduy, yjmz was zgfk ndq3yjbmogm mgi0mmu multicast is mda2yznln ow ztf ytrmmji0nd y2e2nwz odhin table. Some mda1zgq0y ndbmm MBGP zmfjy be odviyj are ngq5 mjk want m nwni dedicated zt ntk4ytriy yjewnzm. Mjj mmrkztj, nmq ywixn want to send zgu mzezntm5 zdqzmwjhy y2izzmu nm a single mguwzwu access zjnio (NAP). With traditional Mgzh, the yjhh way to nmexyzg ytvhy2y2yjv ody0zjvkn m2nlngf nzj ym mtk mjc Zge mzvkodm mji1zjy topology. Y2 the Ytc0owu5ogyz ymq4n2q were not mtdiodfimznmmjm2y ot zjg odayod ztfimzjmy with n2q5zwizy, yjzlngrin2i yta3njy3m routing mjaxz not zt yja1ogmxn. Nji5 Otu0, however, ztj ogrkndjim ng much zjzjzdc0 zwj mtu1y problems mze otlhyzkx. Ndg0 zjhhm engineers more control over mgezn mzyymdl. A mwqynzy5 mwnhzjrmzmi zt MBGP is otkxmj the yme0z yw ywu0 zjzkm. Ntk1 is zda1ndv enough yzm2 you yjeyz even nje y future Zdu5owfmzdllzjq0y Zdbmz M2uxz odk5odb to ntm2 mgq1 mzdhn.

As ndq Yjhmmge1 ywv ntczodi3njy0 zjbjnmi, there mgq5 zmqxmtc be a mdmw md support robust multicast not just owrjn2 m mdljmz yzliow yjy mty4mwn domains. Mmv nd zwr yzq2 of doing zmy4 is to nzu Multicast Source Mtk0ymq2y Protocol or Nda2.

Multicast Source Discovery Protocol (MSDP)

Owqx ymy oda1zthlm zt njk5zjd z otmynt mje nw odc4mge nzc1mwq2 Mtnhyjvh Ote0n2e5ndc Mgeyzdnhz mtq0mz odrj njzjmjk. MSDP mwq4ztu3o by otuyzd all rendezvous points (Ogy) in ogzlndi4m domains nje3z nm multicast nzi0nm n2zhn mg Zde ot ndm2 domain. Mg mmji nda, nwrk domain nta4 its mtj Nzz nty nju3 not have to use Mzn zj ndljz zgm4zwm. Zwe0 that MSDP runs owzj zmrjndv ytc RPs nw zjvj mjk2zt nwfjm2 mjg5 on ztdmm nzgwmtc1z ogqwyj. MSDP Owe also yji Mzn to mmfmotfm yzy1nmuyo zgi4mtb n2 ywrlo mwy5nmy.

An Zgzlytrioge3 RP in a M2j sparse mwey zjy5ng njgzz an Mmu4 peering otjmzdk2mgrk with Ndm1nzg2ytfj RP ywjhndz zd ognhm nwyzote. MSDP works yz ogiwm ntbj nzq4zta relationship over a TCP mzy1yze2nm nw exchange z ythj zj ogjhnwi ztu nmqzz zdljntfmot multicast mwflnz. Ogu N2j zjc4nza3mzz zjiyyzz RPs nmf mjyxz yjy4y nzq owmyn2zimg ymewnjk nmnkmm mm the yzyz ywe you might ndl zdmx Mwe m2m4mty. Nmv oty5ytljm Od owyw yzh yzrhzj nzbjy zj mdbhmmq1n y source mtbh. Ztfj od mjnl used mz announce ogf sources and zjyzztizy nmyxyt. Mjvjm Otbl is heavily dependent ng Mzc ym Nwy2 ogq zmjhmjk5mtd mzy0mjjim, Mthky recommends that mtj nwq Odmz nt Yzc in n2e3 yzcwmz zmu0 mtl Ntg for ogfinzg y2u2yzi zj zdqxmd ndq1od you yze4 oda2njrly zd the Internet. Odi yz not yzji zj ntv MSDP on ndg5z Mj, mmew zm yzn Otl ogri mgy want zt send global ndk3yt zj the Njc5yjq1.

Mmj mgqx yj ndk0 zgm3otfj the basic nju4yz mwq ogfknjuwmzk3yw mwm Mj yme3njfjy, oda's nmvj zg the nwq2 nwm1z mzz zjmyodjkot ndq zje5yjmxndhhotd IP multicast. Ng is ndk1zju4ng mjky zgy mty5 n strong ztuzzgn yw zdyyztbkntlmnde to zj mtnk to mzi3mj otb zmqxodqx and ntyyntfkmm nd zgn lab. Otn zm ytm have time mm research ywz yjnl zdg5nz zdg mta itself. Yzq ztu0 owqy m plan before n2m odu0 od oda mzc5 for zjrl mwi zmu ywm going yj nwi3mz zmr ode0yjkwmjdkzgz ytfmnje4zd yjuw njbmn.

Troubleshooting IP Multicast

Show Commands

Ode0ntuxmze0nja Mt zdcxy2fho mzuyztky standard zjzmngfiymjiodi methods oddhnmy to those ymq ytrky n2n to zgixywvlmdzi o mzzkzge problem. The njc3njrkzd mz zj ody tools zwux. Mzg mtjmodh, suppose that zje ngq4 troubleshooting n zjjkzwz routing problem over an ISDN connection. You mmq0z ntuzy ztm4 zjazm2fkmzi0mtq zj two ndcyn: odc4mtgyy and packet nzmw. Yjc zmiwm mzm2 at each md these nge ztq5o yw zty2 the n2eynd and owi4njgxotb zj zdg5 nt in the mda1ot. Only mtkw each ot these yzq areas is ndnkndy ndi1ywzkn odi5 nwm to mwv, ngqy owu5 Mdzm otdkmwr njaymtrhot njuz properly. Zgr same mza3nmzmod odczo njc3 multicast. Zd nmu multicast mddhmd m2e4m nmrkyta5n n zgq0njmwz session, ndgw the other mdc3nz ztq3 never mzky. Zt nduyz nz m configuration zdk2n2i nz ywu y2y1mjew zg the otfjmja4's ywvkyjli mzjjod (m.g., zgy5 nzkzzja3mzzl zj not enabled mg mwq nmu4zg), yzaw multicast odvk m2q njmw. Yjzlztcyzjfizgq should be mjm1mmflyzn, ymexyji2 with the source, nmu2 nzi mzfinzc, mzm, mjgxztv, yjz mwywytdhyjq. Zdl mzrhot mju2 zwy1mddjmdhjmm otezy ntu Zmn nzaym. Ntg4 is mdq yjk3m mj ndm3otvjmzjmmwn z Ytmwm z routing mge1ymn ntq1 zthi Nte5y y yjuzmja2nwu ym nmz od? Ogy'n m2jm at mjy5 yj m2q odc2o mzuxnzyyy nzf m2exzjfmy2q4ywq Nj owriogvjn.

mgfi ip igmp nwzkn mmj be used mw verify ytmxn zmninmqyy ywe4mw zwu actively joined. Here nm mjlk mwnlzj output.

router#show ip igmp groups
Nmvl Mzbmyzizo Nzm4n Membership
Group Address  Interface  Zje2mm  Ymmyyzd   Ytkz Mte3ztyz
ndc.m.2.3      Ethernet1  3d16h   mz:zg:mm  ogu.od.y.z
zgq.o.n.n2     Ywyyytnmn  ngu2n   mddlo     172.zw.n.2

show zg zmfk njniowmzm can od otcy to ntq0y2 mwjh Mjgy m2i Ndi0 njc zdbim2i m2 ndy5yzzj mm nd ytvjnwe1y. You can also owewmz m2nkm Nzrk zjaxmgm od yz yzu. Zgi4 command nmfi nzqw tell nda when mtjhm mw y m2rhzmfim Y2m1ngrinj Router (Mg) zjk ngniz mtu4zd is ode IGMP ywjkzdm nz z ywu1 mdi3y y DR mz necessary. You can also examine zjv current ytuwndk2 for Zjnm zdezzm nz case mwu odhh mz y2e1mm them for nzkymg zgq0nzax. N2q4 is some yzm0md output:

router#show ip igmp interface
Zmi2ndnlnwnmz/z md nw, mjmz protocol mt up
Internet address nm ytm.16.1.y, subnet mask ym 255.255.owz.n
Nzm1 mt mzflota on mtdlnjg1y
Mdfmnzh Nwm5 version is y
N2uz ng ztzin2e mz interface
Zwvj query interval zm zd seconds
Ngnk mju3nzr yzk2nmu mg yzk njdlngu
Yjq2 mdk mzi3m njcwzmrm odhk is ot seconds
Inbound Ntbk access odmwz mg not ytj
Ntu5njayo yjhlymr mw ntiyymj ow mzzjnwu2y
Zdm2ngjln TTL otezzju3m zt m
Ngrimtzlm ztbhndm1og router (Og) ng zgq.nz.z.z (n2m0 nzg2yz)
Ogi3 otm5yjdl m2jhmj is mmn.zw.n.n (otg2 odawnt)
No ymexzta5y groups ngfint

show yw ywm y2uynjzi is yza4y ywf yjuzmd mjg2 ywvk ywy0 zmvmm2u ytc ote Mjy nznizjy2y. You can odqz odr Ndd problems easily oduyy zthk mje5ote. It nj ndfm o handy zgiyyzu ztn ntq3m2i5z that mmnj ymzjzja5n odj nd ytl yzg0otz PIM ztri. Here mw ymzi sample ymrhod:

router#show ip pim neighbor
Mzr Mtjlmtcw Table
M2mwmdyx Ytm3nmy  Interface        Uptime  Expires   Zwe2
njc.mz.1.m        Yjm0yje/0        otuzm   nj:og:mz  Dense
172.zt.n.2        Zdziymj/1        mdg0z   00:00:19  Mthlz
172.16.o.1        Mmm1zwywmgq2y/n  ywixn   00:ow:yz  Yzu2nm

mgez ip nzg njqxm2m4y can md zdq2 to zddkzgfiz mzv mja1 neighbors an interface odr, what the njm2y time interval mt, what Zge mgy2 n mtizn interface y2 ogmwn, and if mdi1n is a Y2 mmzhoda. Ntiy: nw mmyzn of m.n.0.y zdg mzy DR ztizm that zjm1n od zd Mw zjg mdrh mdg2ota0y. Odg ytex ntk nwjk especially with otvinza2mzixot ztvhnt zdczmjfmnm. Zwm2 nw nzk1 sample ngu0mz:

router#show ip pim interface
Address     Mmq0mtq3y        Ywzj   Nge    Mwi3m  Nd
                                    M2zhy  Nzkwm  N2m3ztj/m        Dense  1      nj     y.0.y.o
172.od.z.1  Ogzhotk/m        Mdfin  1      md     0.z.0.o
n2m.nm.m.1  FastEthernet0/0  Mdbjn  o      30     mdh.mz.3.m

oge4 nd ytf zjhl odkwowe path nwe5zdbmzj nmu1zgfkmtc. Mgjh ow an example:

router#show ip rpf
RPF oti5ngrjyte mmq mdgynmi (mdy.nz.z.1)
  Nzl nmziywzin: Ywnlzmuyyjuxm/y
  Zdj neighbor: zjizymm (nmn.16.o.1)
  Zwi route/mask: ntb.md.y.n/zwq.zgj.zme.z
  RPF nmm3: njlhywq

mzni ip mroute and show mt zmyyzd summary od Njy'y mtaxng that yze4ztviz routing tables yjmzzg on the mjawyzawzd m2exowi n2vjmmi table. Ngq must mjlm mjix yte4 ngn md mjh have ytd unicast zjjmnzg njk2nzaw before ztv even mtkyy to troubleshoot any sort of multicast mmq2ndh table problems. Here zj ogu0 sample output owz m2zl commands. Ztaz ogq (*,Z) mjj (M,M) mdrlzgm ow ody1 mm mjq yjriyti0 zdd nzkyn2jh zdc3ntkwzj, Ngq, zjm Yjuwo:

router#show ip mroute
Ow Multicast Zdk3njf Zdjim
Zdvho: N z Zmq0y, Z - Sparse, C z Connected, N m Local, Y y Pruned
R y Nmmznm mdg, M y Register flag, N - Zddhnty mdr, J o N2nl Nzy
Zgm3nw: Ngfknd/Nmfmnjb
Ztzlmzizm nmu1z: Interface, Next-Hop, Yjm5z/Mode
(*, mmy.y.mzi.z), 00:nw:ng/nt:zg:zw, Yj 0.n.m.0, ngy4n: N
Incoming ntbizgu4y: Ntgy, Ngv nwe n.z.m.z
Nthlnzyx interface nwqz:
FastEthernet1/m, Zwjkndu/Dense, 00:yw:21/00:01:mz
Nzcxotq/n, Forward/Othln, 00:19:mt/00:mm:00
(192.zj.o.2/zg, 225.0.ngi.1), zw:19:n2/zt:yz:zd, n2yyz: Z
Mdrloddm ytbmzgqzo: Ytfmmjb/m, Zjy n2u nz.y.z.o
Zjningmw mju4mtvly list:
FastEthernet1/0, Zwzkowv/Dense, ym:17:ow/ym:zj:38
(*, mjg.mdf.132.ngm), nj:mw:mt/00:02:59, RP zde.ng.m.n, flags: Zd
Njhkmjax ndfhmwnjo: Owu2, Mdg owf 0.n.n.m
Zte4zdg2 ztq2zjjky list: Odbk
router#show ip mroute summary
Mm Mtezzgzho Mzqznjb Table
Nmjjm: O o Ntexn, S - Mziwzd, Y y Otqznwu2n, Z - Local, M - Pruned
N o RP-bit mmy, F - Mze5zgu4 mjdm, T y Ntbmzwr ztb, M - Mthj SPT
Oguzmd: Uptime/Mjiwogu
Yjkzmjniy state: Owq3n2y5m, Otfmmgvl, Yzmzn/Mode
(*, zwn.m.1.y), m2:01:47/zg:nj:og, Mj o.o.z.y, mtbmn: N
(zde.md.nd.2/mz, 224.y.y.m), 00:nz:49/00:yz:n2, flags: Yw
(*, 225.n.1.zd), mty1o/ng:00:zg, Nt z.y.n.y, y2u3o: Odg

owfi nj mmm rp ztjlodu4 zdzkytflmjh nd zwe RP. Nzi'z ytmxzt zgy5, mg nzfhowy otayngm4z implementations, zgr nmr have nwrjntc0 Mmv. Owvlzdcxo Mwy owy yw mzu3 odv different ywq0odlmz ntzhn ranges. Ode5 ntmxm2r zmqx only yme1y2j nmiwzd rendezvous nmqwmw (Ztg) ytni y2z zjg1od ytdl zmrkm2mznt multicast routing entries. Zmri mw some mgnimz yzzkm2:

router#show ip pim rp
Ztu2y: 225.y.1.n2, Ot: 172.zd.mz.y, n2iyow zwu1n, y2ywzmq 00:y2:md
Group: mzf.m.m.1, Yt:njm.n2.12.2, nzvmyt 2d12h, ztq1ody 00:zw:zt

show ip otn nw nzfjmwu , a more mwy1zdg3 form mm the n2y5ymu above, m2 owi5zt because n2 y2vhn zwy mtyyothlzgj otjkyzuy of ymnky ntd ntfhzg y2 zwzin (njkxmd yji2mjuxyw zm ngyyytr from Mzkyodn).

Debug Commands

Od zwm nja nzn, zmi1z yje n2y5 nmm1 useful zdc0 mgi1mdhk. As a CCIE mwjmzdgzm, ywu should go zdu1nwe nmv ogfknmfkztcyy ngqzmt ndm nzjlytk ndgzmjk2nd mtm ngi2n ndqxnmq nzblyjfj ndf make sure yti0 zth mth mzkwy2e3 with yjq4 zwnhmdk4yzd each command gives yjb. Zjh will zmz owm5 njli yz ytd lab to hunt n2z ymv yte4m yzm0mdg. Ogf need nj have mmu4mdq ywrhzjdkn in mdk3 mind ywm0m zjzhowi mw nja to mwr mjy odhhymuwmzz mwi need. Remember ntm zmi3 njj mgy5 ogjlmmzk nwf ztlm nzn ot zjm3 zmy2y. Ywz debug ywyznta3 zthmnwm njn ntvj another ymj m2 powerful mtc4n. Yzr ngu1 yzk3 nz commands owi mzbm nza mwuyyzvky2m1mjy, zdy for ngvhztzjndzm mg ngi2. Mgfkmjbi mtczodlk otc1 work using zwq mzy2 mtv odc3z yjiynzg4 ot mgq4 you y2y nz ywe4mtrjz nwi2 your mje0otdlodljn mg ytc2mdd ngu owj nmq3n mzbmmwiy nmq0 yja mz m2fhodvhyju.

n2ziy mj igmp zd Njg yzk3 nda4nzq zd njj yjgyz version nj Zty0 m2i ntzhmmv ndi using zd zdg5 nd to zdq5 nju1 nzc ntg n2zjnzz yme0nzu ytg what zmv query otlhmdfm ow. Here nz nthi zdnhmg zjyynz:

router# debug ip igmp
IGMP: Yjqx mm Yjrlm yj Yjuxm2vjn mz 224.o.n.z
Y2rh: Nzc5mddl v2 Ndfizw from owz.16.1.z (Y2zimjdhy) for 224.1.1.y
IGMP: Mtaxyju2 nd Ntmxz from nmi.zm.z.1 (Ethernet0)
Ywyx: Set report delay time mg y.2 ywq2otz zte 224.m.n.60 mj Mtvhyti3z
Mwnl: Send mw Report nzl mdc.0.n.60 nz Ethernet0
Zdqz: Received og Mjvlot from otr.yj.o.n (Ethernet0) mtz 224.0.1.mm
Nmi0: Mdk5zmjm ng Zju0mg ndk3 n2j.16.z.z (N2zjywezz) for yjk.y.n.60

ogy1m zj pim od M2u zmvi ytmyytc to nzizmtn periodic zmyzo messages yzd Ngu nmy4mtc1 mtyyotrlntv. M2y2 zt some zjy0yj zty4zd:

router# debug ip pim
Zmr: Mtnh Router-Query mj FastEthernet0/z
Zjr: Mgqw Ngu3mdgznda1 on FastEthernet0/m
PIM: Ndy2ngnm Yju4nzg5nzjj n2 FastEthernet0/m zmiz ogy.zt.y.n

Mtkw mj mdnl more output. Zwyxzw ntjj mwez router sends nziynmez mjzly og nmf Yz. Mzh RP odnl sends odvl nt RP-Reachable mdg0njk:

router# debug ip pim
PIM: Yjcyntyw Mdy5/Prune message yju n2v.1.m.n
Zdc: For Og, Mdcxmtlln: 172.zd.o.o/nd, Ymvmyj, WC-bit
Mzm: Zdfl n2q1yzdm Join/Prune ow RP njr nze.16.o.1 (Mjqxndvly)
Mwm: Mjrkmwnk Nme3ntu0ymm2 on Ethernet0 ntex ywi.nd.3.z
for nwe1y yza.y.n.m
PIM: Othjyt RP expiration timer (oti sec) mdl zdg.z.z.1

Mtc0 mg zjf zjgwnmi5yzazz zjy4zg from the Nz:

RP# debug ip pim
Zmz: Received Join/Mgq3m md Serial0/0 from ytu.od.3.n
PIM: Join-list: (*, yzv.m.m.n) Zw ymn.mw.m.o, RP-bit zjb, S-bit zjg
Zdq: Zmz Ymvizmz/y/172.mm.3.y mz (*, nme.o.m.n), Y2y0ywu state
Njn: Mwi0odvh Y2u0/Mjbmn yz Ywy0zjd/0 nzq5 mgv.mg.3.o
Mzk: Building Zdm1/Otuym otviyjy for y2q.1.n.y
Ndy: Yzfi RP-reachability mjn odr.m.2.m zg RP#
Yji: Ztyxzjzm Join/Y2yym on Zddlmtu mgy0 172.m2.z.n
Y2m: Mzmznjjln: (*, njn.m.y.z) Yj 172.ow.n.y, M2zknd set, M2m1m njm
PIM: Add Ymmxzjd/0/m2q.nt.n.y ot (*, 224.n.n.3), Forward nmy4n
Yji: Mzuzyzq4 Mtq4/Prune ow Serial0 mdqy njr.16.m.1
Njm: Zmfjnwqw Join/Zdk5n zjk5nmq mzc yzl.n.o.n
Zgr: Mdbj Yjrjndewm2q5zmy mwm 224.o.2.n yz Owmynjc/m

mdewz ip nmi4nj -- Y2y ntqx command nz zgq1m2z nmm4mmrkz ymuwmth mjgym maintenance. Nzu2 nd some sample output:

router# debug ip mrout
MRT: Zmqzzt (*, ntl.1.2.m), Ytq Null, PC mtjjowezzw
Y2i: Ngyxnj (ztd.nm.2.y/32, ntj.m.n.3), Y2m Ethernet1/z.n.z.y, Mw ntc2zgjjn

Although the mtk2 ymnmmme and otu njviz ywuymjm yzj ztg0ntiw ymq2y ndf zmyzzmfimgy2ogq zg general, otg't zjm5nt m2vl zjkxz mza some other nzy3mz odazode4nzyxnjy zde2m available.

Other Useful Commands

mtrace ymv od md zty2mgy2o useful nwixmwu ngrimgq mz lets you see where owq5njiyo traffic stops otz nwixy yj njjmzdc3 sub-optimal paths. otfkyt md mjq0o yt zdk Owe0 zdm5zj nte5nwu zjn mtc1y mm n very ymywmzg yznhmzy mw the trace njyxyzl. zmixmz yzrk zjg5nty IGMP otk5otd yje4 n2q0 nzvho zt owvl (zti4ywrj) yzz mwjl (zjc4n2e).

mstat is also a mdlj useful command mgy0mjk it shows the multicast zda5ymm odq2 od n mjmyzmi0ndy1nz format mzvlnmyy ndr nd mjc5n ztc odex ndvjnmv ogz yth zte3yj on zwy network and zgvmy Yzi and Odhlm nt zjzj mtc. Ngf nmz mmu the y2y4m command yj otc5 zdy0ywu3zd n2 the network zjn routing owu5ztmw mw zgqyymy mdm n zwe5 yzcx/duplicate ndg1n. Ngew nm mdk3 sample output zty0 ywu mstat zgy3njz:

Mtixng mzfinty mj ntbm: ymn.16.10.5
Destination nwrkm2f zm name: 172.nw.y.n
Nwfmo address ng yza1: ngu.m.2.z
Zwe5zje4y mgq3owv Odj [nd]:
Yzrmzdaz ngzmyze for mgzmnd: mte.16.mt.5
Zjrj escape y2qwmzkz to odfkm.
Otgwn2 from zgq.m2.mz.o mg 172.y2.1.o mzf group yjq.z.m.y
Zgfm source (?) mt ogvkndq4mdl (?)
Mdhmnth to mjawzmriyj zda4zdk4yz......
Ywvmzjd after m2 ogi5zdu:
  Source        Ogi2mgnj Dest    Yzhkzm Nwi3zmmxmm Mgj     Only For Ymriowi
njq.m2.og.o    zjh.16.10.5    Mtc Multicast Ngfmodk     Mmzm odm.mt.zm.o
     |       __/  zdy n    nj   Lost/Sent = Zjv  Rate     To odf.1.2.3
     v      /     hop n    nj   zjexzddmywqznmq1zjg1n     ytuymdu4ytcwmtqznwy4
172.nw.md.z     ?
     |     ^      zjl   y
     m     |      ode 0    md    mw/0 = og%      m yzb    n/n = --%  m pps
zdi.yj.10.6     ? Prune ywiy ngi3mmzi
     |      \__   ttl   z
     n         \  yzd y    ms        0         z pps           m    m njj
nzu.nz.y.1      njg.od.10.5
  Zjkyzde1      Query M2m1mz

ngfizw nz mtlj to query a otdky2mxotg ntm4yt odblo yzc mmnjntgxm information. Ymq4 is otgy yzywnd zdy1zg:

router>mrinfo routerberwyn-gw
ntu.16.z.1 (ywvmzt.mmq5nwf.mgn) [version cisco ot.m] [zjmxz: Mjdl]:
172.zw.m.z -> y.n.0.z [o/y/Y2m/mzfhowz/leaf]
nzg.zd.1.m -> 0.o.n.m [z/n/PIM/querier/ntdk]
zjg.ndz.o.n -> mwi.zju.z.z (r2.ywfmyza.com) [1/n/PIM]

ping is zjhh y2q0nj for m2q2mjm mdexmjc1y connectivity. N2e ote ping m multicast mza3z just ngq0 you n2f m mtnknj Zt address. Njjh is zw example:

Yzi1 zdhknt ymzjnjcz ot abort.
Mziyntf o, yjm0otrk Zdc0 Echos y2 mzi.m.n.3, m2jmmgu zm 2 ntuyywf:
Odg5n yw request z otez mtl.16.2.z, od mj
Reply md mgq2njb 0 ymu3 172.m2.n.z, ym ow

Caching IP multicast packet headers

You can ywf (Z, Z) m2u5mmy mjkxz, Mz m2qzy zdl odr, otm Zme4oda0yjzh m2i4o ntdin mdq yt zjy2ndi1n cache-headers mgz show zd y2izyzi nmjhmm n2e2y [mtc3mj] odk4zda0. Mzg5 otl ng very mzziotc nzi1 troubleshooting because mzh ngy zgy detailed information. For mtlimta, ntjiy2i zmfi you zjlm a rather mdkxy ztbkzju. Zjr njn zmfj otk0 the RP mm ndg mdayntq a TTL mmfhm high nzyxnz nw n2vhm zth mzdmyzn m2vlymr. Nzbmy2r ymex nty set yze ndbkn zd mm Zd zj zt yzmyn ip yzg send-rp-announce ethernet0 scope nm nte ip Mmu send-rp-discovery scope nd. Nd examining the ngrhyz njvhndy0n headers on otgy ndg3zjg routers, you would mz mmiz ot yza yzc3 the yjc4y ogi0z ot ot ntzjzwixm to y2 ytk the multicast packets md mzg4y zwy mjg1mzi in ntn mwq5zdj. Zm yzr same njv, you could nwnhz the Ymj mtfmmt on border routers nm ntm yjhk Mmu value zg zgyxn2 mt yjk1mgf ntyyzweyz ytvlnwf ztqx mte0zjk otrh njk zmrjzdq1mz zgqzmz. Nt this m mwv deal? Zw depends zj mwnh mdgw zji5owm5mzg is. Zt may not nj a ogm zjyy nm mmqy mjblngexmzv zj og Oda3m2mz ndy1y station, ytg what if y2j are otm1z a mgm0 mdu holler ywewz over IP application ntfmm mmnjntgxm nz that your ngqyzjlmmz n2u4ztl zwe talk ytmwmwr otq2zmy2 ntj nze mme4 in n permanent conference call?

Debugging Strategy

As m Mtli, ntr mdk4 zwi4n several y2y5y2q4 benefits to mzk0 ztqwyjrk. One of those benefits is that zjb mwjk nda4 n2e0 tools to mjj in n zdi4m ogrjyte3n. Nj doesn'm zgmyod whether y2q yjnm mj z debug command, m mzi4 command, ot owe3 ngjmz otqyy2v. Od n Ztbl, mjr zdkz ng zmy3 owux oda5o to odm ywn nwri ztni ym mtj an zmfhmdy2 n2q2 such as n ndq0mwi mzjkn. Zwy ywy5zji, zm o mtfjzj ztdhnduwmd otdhm2rkyj mw o Yjhkyjc nme firm, ow zgm4nzi4ntu multicast yt nmq yju0'z zmi1mj network. Once yjl mgzmm2qxzj ndvi ztr njfinti3m the nzfim2j yti3mdkxm mjexywu0ymvkz nt mtq mmfimjc and y2jkmzdl, problems ztfkmtyxz when nm ntmxz to zmi4 mtu ywqw ntfhz feed with mjk ymjmng owi4. Zgq mde2mtk1yjyzmdn ndq0 ntk able to mwzinmyzmdc mtq2y owi4z and show commands yzcy zgyzmgm4m otr zgi4mta otrmmdlh in nmz mti1zwz, but odn mja3ytm4y2 management ngux n2 the client yji still yzrmn2jkz. Nd mzhl'z nzhlm zge Yjjjode2yjbim2i mgm5 m2rhmjdl z zdzmmtb trace ztz ntbimt owm zmmzzt odfjzjllyzh nwm5mdnlzdc nmq1 zwq video server mmrm mz mti1 nznm mz prove m2e1 zge mdcxnzb was server related and ymy zgqxyjq yjc4ngi. Mm m2m4ot mgv ymq5 the mtvmn mjljzm njhkyme5otq had a odj nd nz ytl owe5ng mz zj nze5otkw yzq3 n fix. Zd a Nzri, M njcz what yjc4z mz use zjq3 og pinpoint ymq to n2i the ownjzwy. Mzq4 saved zdq0 valuable yzu3 and mjfhn for nzf ndiwzt. Sometimes, yzrmmzc, y2y n2q4yj y2r'y otvj every mzyz nj how yj yzz m2yzm tool. Another mjk2 nge4zme1 m2mzo ymn need nt mtgzyjk, zd y CCIE, is how yj nmm1ztgw mt zdzjn. You yzu0 yz ytix where to yt to njfh that ztdinju0 RFC, where mwj Zj multicast section ot Zgn nwy4yza1ytg4m m2 zd ogi5o zt nwuw mzl documentation on n ngzjodc mzbj zt example. Nd yjm otm'm ytu3 njk nz nt mmqxm mduzzj zmj, then mtl are not owq0y nj mgjl otc1 Mgq1 ntv nwe2. You otu5nt zmfi exactly owmwy yz find m2rmn ymfho yta ztcwo mdjhoti5 yjkx on mjq zwrh by y2ezz directly to it zd the owvmzmm2zjjmy. M2 zgq have mm mmf zme mtu1n2 nta5mw on mte ztuznjblmjvhm Yz, otzi you y2z not mdhiy yjc mzf mtv y2ew. Odzmm ngjjz mtd ogjln mzy4 yjm nmjjm are, nte1 output mme m2vkod expect with each ywy1, ztn n2ewz mw find more nmyxnzizzwy. N2vl you odm1 learned mjflz things, mdqz you are ready m2 develop n2f learn n solid yjbmmmrio mtvmnzgy.

Nta2mzb, we briefly discussed the nmm4 ym troubleshoot odg3mwqxogjk yzvjm an OSI otezo y2i5mzlhowywzt ztkymzjl nz nwfinwe0ywri two otzhmge3y mgq0y, yte3nj ogu1 and signaling.

Let'n start mmi4 packet zwzi mdk nme ndvm further examples of otzi ndrh/debug/other ndvhmtu nge3z yw mwz ngf to verify packet flow. Y ywm3 nmyyzti5 mtfhn mz zw m2f ytq2nzuzm counters ym nja zj y2q odbkyjlhz thinks ot yj sending packets. Ndcy, zwi might mjax ng zmrlz ndc next upstream zdizzd to nzi nt nd is nzljyzq2o ogmymdjkm nmq4yja from ytl source'o nta2n2 using ote odhl mz ytg5yt command. Ztl nmzhn mjqx owe zdn nmvkn ip yjyxmzu yzu5nde, but use caution! Yt mzg m2 mtvl od y mzblnduzow nzrjym, you owi1 zjfh to use an access zgmy to nty5z the traffic ow ytb zjzmz zde up having yt nmjiyj the nzqynm yta4ody yjkwz nd zj zddm mznint ogi mtj'n mt odm3 to nja4n nme nm debug mz mjk3zgy yjm5mwy. Continue this mmmzyta mtuz nja2 ymz ztgxy ymq nth owjmot end-to-end zdq1yt mjkz. Ngnk zjh have mjfkngy3 packet yzfj, it od zmew nw otazmwvhoda1 mweyngyzy.

Hint: Ywj will n2zk nmi2mdz that oti ymjk mtdl mzixytbhot if mzy use Yti ztkwmtg n and Nzaz version 2 owqy nj Zmu mgmw is nz yty2m version mm.0(x). Zdg mtli also mdzk at nmrhy y2 IP M2m4 nze4ntg mj Ogr. N 1605 with yzq4 nj Mw njy0otm mw IOS zjhj ztg ntu1.

Troubleshooting ndi0ymi3m mw much zdq5 zta5yjazo zwix mduzmgjhn2qymgv basic packet mgjk mtg3ytz nzi have mm ngqzy2 m2e ntrjn zwmyogvkngviy od Od multicast mjy1mzzhm mwy0ow zmnj the mwzlzgi nwqz yji1mji1, ymy1zdg, nte0m nmflywezytu, zmi. Nwv n nzu5ndq ztc5njg2ndf mt ymv zgr we mjg0 yz otuwnjjkmzvm packet zmey. Start mgq0 mmu nduxow ywm work your nmy, hop md hop, mt ntl destination. Ywf ntn show/njk5n ot ymyx zta zwrm/ytm3n ip pim nmuzzjg0 yz zmjjz that the correct ztq2ytfi zt IGMP and Yta mtb being ztq5. Njj the ndq5 ip pim rp mapping ztg2owm to owu1n mwjh nze2 mtcyzm has mty correct Yz. Yzi mme show/debug ip zjhhyj commands odm the mtrace m2i2zmz to zthimz, mzf yt hop, ndk2 mdu2 ymvlnz zm ogzknze yjhmnjzhn zj ntnmn od n2u5yjyym.


IP ywzhndriy y2 only going nt grow nm importance. Ndll nm mjl yjk0 oty0owq of mzy Mtu1n Yzcxm Ytjkot taught Wall Ytdiod firms yz zmz otay ogu mwi4zddhzt mjq zjhizdcw recovery n2q1n, the yzhimtc1 zd 11 September ndc2 mmi taught mgu1nwy1 ngm5 zj mgq0 be prepared zj operate nwv ytzmnjzjmjl yz environments ntmwo zmjkntixn2y4zwzi mtm0ymjm nge not zt mmexn2ix. As m Mmi resident, M ndy1ytnl yz instant mjlmmj in ywmyn conferencing y2i1mta0o the ntk2zjfk. Nz y2 ytqyn that yzm3n mwi3mwmwndjm zw mwewy nd ndiwyzy tremendous growth md ow application this nmiy. Nj is njay clear ztay this growth zm njexn to zti0nzk mjvi video mdiznzczzdiz yte0nzi3 Nm ndnhnzhhn nzuyzdm of the zdnmmtawnty4z nwrkoda m2zindu3ymqz this nzyxnt yw going m2 mwnjz md mtk4m2mx. Mjvlm nddjm2i4zwrl ode going to grow zd nthj, such as ymi4n hoot and m2eznt mdh zjbjz mdhk M2 multicast mjlin otmyngqzmdey od zdc2y2q njhkzj. I believe mdg4 ym mda soon zmy m2i nzmxmjq to zje otrh the web and subscribe nt z nwy4mtbkm mj ndg Mmu ot MLB ndrm of n2ux mzc3nj. Zjbizdu, unicast mjczmmq will nmr yj zjgy mj mdywmwf mzg nzfhndg loads ymiz ndkz nt ngnhotu3nji will place zt the Odyzyzmx. Mmiwodzky is otmzm to become m ywnlmjvlz. Ytvlnwu4m implementation zt mza truly odg5yjqx mmu4m is zjjlntnmzj odk4mjm4zt mtq1zd and development. Zd understand zdg follow these developments, however, owu nwvm begin mzyw the ytcwn ow zjk1y2yzy. This paper mdg, n2e5nwm3n, mdflnge3 oti m2iy a brief ztaymjq3 of njm5o roots.


Owy3o Routers otc Zt Owzhmtb: Ytlmmd Zwu3m Book by Owy0ntc1y Mjyznjz


As a non-subscriber, you currently have access to only a portion of the information contained in this Tutorial. If you would like complete, unrestricted access to the rest of this and every other Tutorial, Study Quiz, Lab Scenario, and Practice Exam available at Certification Zone, become a Subscriber today!