Pages

Thursday, May 23, 2013

Understanding ping and traceroute.

Tuesday, May 21, 2013

CCNP Switching Q&A

What is the 20/80 rule of networking?
The 20/80 rule states that 20 percent of network traffic on a LAN segment will stay
on that segment. The remaining 80 percent must go across the network core, either to
enterprise servers or to the Internet.

In which OSI layer do devices in the distribution layer typically operate?
Layer 3 devices are typically used in the distribution layer.

What are the different Ethernet technologies and their associated IEEE standards?
Ethernet (10 Mbps, IEEE 802.3), Fast Ethernet (100 Mbps, IEEE 802.3u), and Gigabit
Ethernet (1000 Mbps, IEEE 802.3z)

What benefits result with switched Ethernet over shared Ethernet?
Switched Ethernet ports receive dedicated bandwidth, have a reduced collision domain,
and have increased performance due to segmentation or fewer users per port

In a campus network, where is Fast Ethernet typically used? Where is Gigabit Ethernet
typically used?
Fast Ethernet is typically used for links between access layer and distribution layer
devices and between end users and the access layer devices. Gigabit Ethernet is typically
used between all layers—between access and distribution layer devices, between
distribution and core layer devices, and for the links between core layer devices.

What must be done to a switch before Telnet access is allowed?
An IP address must be assigned to the management interface on the switch, and the
management interface must be assigned to a VLAN.

What switch command will set the enable-mode password on an IOS-based switch? A
CLI-based switch?
For IOS-based switches, use enable password level 15 password to set the enable
password. On CLI-based switches, use the set enablepass password command.

What is a trunk link?
A trunk link is a connection between two devices that transports traffic from multiple
VLANs. Each frame is identified with its source VLAN during its trip across the trunk
link.

How many VTP management domains can a Catalyst switch participate in? How many
VTP servers can a management domain have?
A switch can be a member of only one VTP management domain. A VTP domain must
have at least one server. There can be more than one server, for redundancy, but it is
recommended to have no more than two.

Which VLAN numbers are never eligible for VTP pruning? Why?
VLAN numbers 1 and 1001–1005 are ineligible for pruning. VLAN 1 is reserved as the
management VLAN, while VLANs 1002-1005 are reserved as the default FDDI and
Token Ring function VLANs. At press time, VLAN 1001 has no special purpose, but is
reserved and cannot be pruned.

What is the difference between these two trunking methods? How many bytes are added
to trunked frames for VLAN identification in each method?
ISL uses encapsulation and adds a 26-byte header and a 4-byte trailer. 802.1Q adds a 4-
byte tag field within existing frames, without encapsulation.

Two neighboring switch trunk ports are set to auto mode with ISL trunking mode. What
will the resulting trunk mode become?
Trunking will not be established at all. Both switches are in the passive auto state and are
each waiting to be asked to start the trunking mode. Instead, the link will remain an access
link on both switches.

What are the types of VTP messages or advertisements used by Catalyst switches? What
field in these messages determines if a switch should use and record VLAN data in the
messages?
The VTP message types are Advertisement Requests, Summary Advertisements, Subset
Advertisements, and VLAN Membership Advertisements (an extension to version 1 for
VTP pruning). The Configuration Revision Number is used to determine if the VTP data
is newer and should be used.

How is traffic distributed over an EtherChannel?
Traffic is distributed according to addresses contained in frames passing through the
switch—not according to port loads or equal distribution across the individual ports in a
bundle. Switches use an XOR computation of source, destination, or both addresses of
either MAC or IP, depending on the switch capabilities.

What is PAgP used for?
PAgP is a protocol that is used to dynamically and to automatically configure an
EtherChannel over multiple physical ports.

Name two types of Spanning-Tree Protocol messages used to communicate between
bridges.
Configuration BPDUs and Topology Change Notification BPDUs. Configuration BPDUs
are used to inform bridges of global STP parameters and are used to form the Spanning
Tree topology. Topology Change Notification BPDUs are used to inform bridges that a
link state has changed potentially impacting the Spanning Tree topology.

What conditions cause a STP topology change? What effect does this have on STP and the
network?
A topology change occurs when a port moves to the Forwarding state or from Forwarding
or Learning to the Blocking state. During a topology change, addresses are aged out in
Forward Delay seconds while active stations are not aged out of the bridging table. The
STP is not recomputed; TCN BPDUs are sent throughout the network notifying other
switches of the topology change. Only the port where the topology change is occurring is
affected, by moving through the STP states.

What is the single most important design decision to be made in a network running STP?
Root Bridge placement.
Where should the Root Bridge be located in a switched network?
It should be located as close to the center of the network as possible. For example, in a
hierarchical design, the Root Bridge should be located in the Distribution layer.

What happens to a port that is neither a Root Port nor a Designated Port?
That port is placed in the Blocking state so that no bridging loops form from it.

What mechanism is used to set the STP timer values for all switches in a network?
The timers are set on the Root Bridge, and the values are propagated to all other switches
by including them in Configuration BPDUs.

What parameters can be tuned to influence the selection of a port as a Root or Designated
Port?
Port Cost and Port Priority.

Where should the UplinkFast feature be used in a switched network?
Only on switches that are leaf-nodes in the Spanning-Tree topology, such as the Access
Layer.

How does an EtherChannel distribute broadcasts and multicasts?
Broadcasts and multicasts are sent across only one port of the bundle and are not
distributed across the EtherChannel.

What happens if one port of an EtherChannel is unplugged or goes dead? What happens
when that port is reconnected?
Traffic on the disconnected port will be moved to the next available link in the
EtherChannel bundle. When the port is reconnected, traffic will not automatically move
back to the original port of the bundle. Rather, new traffic will be learned and applied to
the restored link.

What conditions cause a STP topology change? What effect does this have on STP and the
network?
A topology change occurs when a port moves to the Forwarding state or from Forwarding
or Learning to the Blocking state. During a topology change, addresses are aged out in
Forward Delay seconds, while active stations are not aged out of the bridging table. The
STP is not recomputed; TCN BPDUs are sent throughout the network notifying other
switches of the topology change. Only the port where the topology change is occurring is
affected by moving through the STP states.

A Root Bridge has been elected in a switched network. Suppose a new switch is installed
with a lower Bridge ID than the existing Root Bridge. What will happen?
After the new switch comes up, a Root Bridge election will take place. This will occur at
the next Hello time when the new switch announces itself as root. It will become the Root
Bridge because it has the lowest Bridge ID, and the Spanning Tree topology will be
recomputed. Where switch ports change state as a result of the election and topology
change, outages will occur until the Forwarding state starts again.

What happens if the STP Hello Time is decreased to one second in an effort to speed up
STP convergence? What happens if the Hello Time is increased to ten seconds?
Setting the Hello Timer to one second doubles the amount of Configuration BPDUs that
are sent by a switch, as compared to the default 2 second timer. While this does share
BPDU information more often, it really doesn’t help the long convergence delay when a
port comes up. The significant delays come from the Forward Delay timer, which is used
to move a port through the Listening and Learning states. By default, this process takes 30
seconds and is unaffected by the Hello Timer.

What types of links can be used to interconnect switches and an external router? How
many VLANs can be carried on each?
Links can be used with one VLAN per physical link, using any supported media. Trunk
links can also be used to carry multiple VLANs over a single link, using such media as
Fast Ethernet, Gigabit Ethernet, ATM LANE, and FDDI.

What is the difference between interVLAN routing and multilayer switching (MLS)?
InterVLAN routing uses a route processor to forward data between VLANs. It also
requires that each packet crossing a VLAN boundary be processed by a routing decision
on the router. MLS also uses a route processor to forward data between VLANs, but the
routing decision is only required for the first packet exchange between two nodes. From
that point on, data is forwarded by the switching engine and does not require further
routing.

If a router is used to route between VLANs, what additional information is needed so that
traffic will actually be routed?
A default gateway is required on the end stations so that the stations can forward packets
that are destined to a different VLAN or network to the router.

What is the name of the protocol that allows a set of routers that are working together to
form one virtual-router?
Hot Standby Router Protocol or HSRP.

What is the minimum number of routers needed to perform HSRP?
The minimum number of routers needed is at least two. One functions as an active router
and one as a standby.

In a properly functioning virtual router, what happens when the active router fails?
In a properly functioning HSRP environment, packets will still be routed in the event of a
failed router.

How many standby groups can exist on any one LAN?
In any one LAN, up to 255 standby groups can exist.

Name the six states that an HSRP configured router can be in.
Initial, Learn, Listen, Speak, Standby, and Active.

When configuring HSRP on a particular router interface, if the standby group is not
explicitly configured, what standby group does the interface fall into by default?
Standby group 0 is the default setting when configuring an interface for HSRP. You may,
however, override this by configuring a setting of your own choosing.

What command is used to display the HSRP virtual router IP and MAC address?
The Cisco command show standby will display the HSRP virtual router IP and MAC
address.

Which router in an HSRP group becomes the forwarding router and how is it determined?
The router that becomes the forwarding router in an HSRP group is the one with the
highest priority. The priority is determined by what has been configured. The default value
is 100, but the priority can be any number between 0 and 255. If the priorities are equal,
then the highest IP address takes priority.

In the command standby 35 priority 90, what does the “35” stand for?
The “35” indicates the standby group number.

An HSRP router exchanges Hello messages with other HSRP routers. What is contained
in the hello message?
The HSRP Hello message contains the hellotime and the holdtime values, in addition to
the priority, group number, password, and virtual router.

What does the term tracking imply in an HSRP environment?
Interface tracking enables the priority of a standby group router to be automatically
adjusted based on availability
becomes unavailable, the HSRP priority of the router is decreased. The HSRP tracking
feature reduces the likelihood that a router with an unavailable key interface will remain
the active router.

What does the preempt command do in the HSRP environment?
The preempt command is used to ensure that after an active router has failed, it will
resume its active router role once the router has recovered. That is assuming that the router
has higher priority AND preempt is configured. Without this feature, the “new” active
router will remain the active router indefinitely.

What problem makes HSRP necessary?
The fact that there isn’t a dynamic protocol to discover new default gateways for hosts in
the event of failure.

What constitutes an HSRP group?
An HSRP group consists of an active router, a standby router, and the virtual router.

Assume you are using five VLANs within your network and want to implement HSRP. How
many HSRP groups would you need to create?
Five; when using multiple VLANs in an HSRP implementation, a separate HSRP group
is created for each VLAN.

What is the name of the protocol used to report their multicast group membership with
neighboring multicast routers?
Internet Group Management Protocol (IGMP).

What does a host send to the multicast group address to join a group?
The host can send a Host Membership Report to join a multicast group.

Name the two types of sparse mode routing protocols.
Core-Based Trees (CBT) and Protocol Independent Multicast Sparse Mode (PIM SM).

What is the algorithm used in a source specific distribution tree?
Reverse Path Forwarding (RPF).

What is used to manage the scope of multicast delivery?
Time-to-live (TTL).

What VLAN is the default VLAN for a Catalyst switch and why is it a good idea to change
this?
The default VLAN is VLAN 1. The reason it’s a good idea to change this is that anyone
plugging into a port will automatically be in VLAN 1 without further configuration.
Because VLAN 1 is commonly used as the management VLAN, this represents a risk of
the wrong person getting access to this switch or others within the network.

What does port security do on a Catalyst series switch?
Port security is the feature that can be used to limit access to only specified MAC
addresses. All others will be denied access.

In general, what type of policies should be implemented in the core layer?
Only Quality of Service (QoS) policies should be implemented at the core layer.

When implementing route filtering, what type of access list is used—a standard or an
extended access list?
Only standard access lists are used when filtering routes or routing update traffic.

What does the access-class command do when applied to a virtual terminal
configuration?
The access-class command is used as a means of allowing only certain hosts access to the
virtual terminal lines.

What does the Cisco command login local do on a router?
The login local command requires that you have preconfigured a username/password pair
using the command username xxxx password xxxx. This is done in lieu of using
authentication servers such as TACACS or RADIUS.

What is the main method of out-of-band management for Cisco switches?
The main method of out-of-band management is the console connection.

What is the command to verify that RMON is enabled on the switch?
show snmp

What is the default value for the read-write community string?
Private

How many simultaneous Telnet sessions are supported on a Cisco switch?
Eight.




Monday, May 20, 2013

CCIE R&S points to remember


Use the ip default-network command to configure the gateway of last resort. use the ip default-gateway command when routing is disabled.

The hierarchical design model consists of a core, distribution, and access layers. The distribution layer implements access lists, distribution lists, route summarization, VLAN routing, security policy, and address aggregation.

Distance vector routing protocols periodically send a copy of the full routing table to neighboring routers.

The preferred reading for reliability is 100 percent or 255/255. The preferred reading for load is the nearest 0 percent or 1/255.

OSPF requires a two-layer topology with all areas connecting to the backbone. With link-state routing protocols, the status of each router in the network is propagated to all other routers in the network, and each router calculates the best routes in the network.

The correct format is ip route 0.0.0.0 0.0.0.0 serial 0 to configure the default route out of serial int.

Split horizon with poison reverse sends a route with an infinite metric out the interface from which it learned the route.

The core layer of the hierarchical design model uses high-speed connections.

The least significant number of the configuration register (ox2101) determines which image is loaded. If set to 0x1 (0001 binary), the router boots the ROM image.

The switch show port status command in switch provides a summary table of interfaces with port number, VLAN, and Status information. The show ip interface brief command in router provides same output a summary table of interfaces with interface name and number, IP address, and status information.

When accessing the router through Telnet, the terminal monitor command needs to be configured to view debug output.

The configuration register setting can be verified with the show hardware or show version command.

The password configured with the enable secret command overrides the password configured with the enable password command.

The o/r 0x2142 command is usually configured during a password recovery
procedure. It modifies the configure register by telling the router to ignore the
configuration in NVRAM.

The system bootstrap initializes the CPU and starts the bootloader.

The copy tftp running-config command configures the router from a network TFTP server.

The snmp community ccie rw command allows the network management server to write changes on the router.

If the login command is configured for all vtys but not the password command, the router prompts with password not set and disconnects the Telnet session.

The show process cpu command shows the five second, one minute, and five minute CPU use time for each process on the router.

A router is in ROM monitor mode if it has one of the following two prompts: >  rommon >

Reboot, break, set 0x2142, reboot, copy password, disable ignore NVRAM, reboot is a correct process for password recovery.

If you use the login command, but the password is not set for vty lines, Telnet access is not granted. The router returns a password not set message and disconnects the Telnet session. If you do not use the login command, users are granted access without prompting for passwords. If you access the router through a virtual terminal, such as Telnet, you need to enter the terminal monitor command to have debug output show on your screen. This command is not necessary if you access the router from the console.

Ethernet is canonical; it expects the least significant bit first to read off the wire. Token Ring is noncanonical; it expects the most significant bit first.

The IEEE 802.3 SNAP frame contains an Ethernet type field in the SNAP field. Transparent bridges create a table that associates learned MAC address with its ports. IEEE 802.1q identifies the VLAN by using an internal tag. ISL uses an external tag.

The VTP client participates in VTP by maintaining a list of VLANs but does not store the information in NVRAM, nor can it add, delete, or rename VLANs.

The bridge identifier (BID) is used in the Spanning-Tree Protocol. The BID is 8 bytes long and contains a bridge priority (2 bytes) along with one of the bridge’s MAC addresses (6 bytes).

Bridge virtual interfaces configure integrated routing and bridging (IRB).

VTP servers maintain a full list of VLANs, adds and deletes VLANs, and stores VLAN information in NVRAM.

Transparent bridges learn MAC addresses, filter frames with destination MAC addresses located on the same incoming port, and forward frames to the port associated with the destination MAC address.

Spanning-Tree Protocol enables blocked interfaces if the primary link or bridge fails.

The path cost is the addition of segment costs to reach the root bridge.

With CDP, the neighbor’s device name, management IP address, platform type, and OS version can be gathered.

IEEE 802.1x specifies a port-based authentication mechanism that uses EAP and RADIUS.

The Basic Service Set mode uses one Access Point for clients to access the wired LAN.

BSS is also referred to as Infrastructure mode.

The LAN Emulation Configuration Server (LECS) is contacted first by joining LECs. The LECS provides the ATM address of the LES to the client.

LE-ARP translates MAC addresses to ATM addresses.

X.25 uses LAPB, a variant of HDLC, for Layer-2 framing.

Discard Eligible bit is set on frames exceeding the Committed Information Rate (CIR), which is the contracted traffic rate by Frame Relay network switches or is set by the Frame Relay DTE (router) on relatively low priority frames. Frames with the DE bit set are discarded in a congested network before frames without the DE bit set.

NT1 is a physical layer device that connects the 4-wire BRI from ISDN equipment to the 2-wire phone line provided by the telephone company.

The BECN bit is set in frames traveling from the destination to the source when congestion occurs in the Frame Relay network on the path from the source to the destination. A BECN attempts to throttle the rate of traffic being sent by the source DTE.

Routers on Frame Relay networks learn local DLCIs by using LMI. Then, they learn the remote IP address associated with the local DLCI through the inverse-arp process.

There are 10 bits in a Frame Relay DLCI without extended addressing.

PPP supports PAP and CHAP authentication. PAP uses clear text passwords; CHAP uses encrypted passwords.

ISDN D channels use the LAPD framing standard. The ATM header is 5 bytes, the payload is 48 bytes, and the entire cell is 53 bytes. ATM payload cells are identified by the VPI/VCI pair.

Frame-relay supports flow control. ABR provides flow control mechanisms to alter source rate.

BECN bits are sent to the source end station in an attempt to throttle the sending traffic rate. If traffic shaping is enabled, the router uses these BECN bits to control the flow of frames into the network.

Frame Relay has fewer error checking features than X.25; therefore, it is not designed to run over bad cable systems. Both X.25 and Frame Relay are connection-oriented.

Frame Relay adds less overhead; therefore, it is available at greater speeds.

Frame Relay networks drop frames when there is congestion in the network and do not retransmit them. The sending host must retransmit any dropped frames. Circuit-switched networks use synchronization, where synchronous circuits are dependent on having the same clock so that the receiving side knows exactly when
each frame bit is received.

Each B channel has the bandwidth capacity of 64 kbps.

Telnet uses IP protocol 6, which is TCP. TCP uses port 23 for Telnet.

The IP protocol layer at the destination host is responsible for reassembling any IP fragments before sending the packet up to TCP.

TCP waits for data to accumulate before forming a segment. The PSH pointer indicates to promptly send the data.

The PING application uses the Internet Control Message Protocol (ICMP) echo type.

The Bootstrap protocol(BOOTP) and the Dynamic Host Configuration Protocol (DHCP) are methods to provide an IP address, mask, and a default gateway to a host.

The real IP address of a device in the internal network is translated to a globally unique address, which is the Inside Local Address.

The ip helper-address supports BOOTP, DHCP, NETBIOS, TFTP, DNS, TACACS,

Time service, and IEN-116 name service by default.

Four segments close a TCP connection.

The Internet header length (IHL) is measured in 32-bit words.

proxy-arp is configured on an interface.

The router with the highest HSRP priority becomes the active router. The default  HSRP priority is 100.

TCP connections are established with a three-way handshake.

There are 3 bits in the precedence bits field. The IP checksum is computed for the IP header only.

IP provides best-effort delivery. If a fragment is lost, the upper layer is responsible for detecting missing data and for requesting a retransmission.

The checksum is performed on the TCP header, data, and psuedo-header.

The sequence number indicates the first byte in the segment.

standby 1 track serial 0 is used if you want the active HSRP router to resign if a tracked serial interface 0 goes down

Outside global addresses are not translated; therefore, the address remains the same after the packet passes through the NAT router.

If EIGRP and IGRP are configured with the same AS number in a router, they automatically redistribute routes without having to use the redistribution command.

There can be up to 25 routes in a RIP update packet.

By default, EIGRP uses bandwidth and delay in its composite metric.

By default, EIGRP summarizes routes at network boundaries.

EIGRP places a route into active state when performing a recomputation for the
route.

If a route is removed, no new updates for the route are accepted until the holddown timer expires. The hold-down timer for RIP is 180 and for IGRP is 280.

EIGRP scales the metric by 256.

By default, EIGRP loads balance using equal-cost paths. EIGRP does unequal load balancing when you use the variance command.

An OSPF backbone router always has one or more interfaces connected to the Area 0. An ASBR can have one or more interfaces connected to the backbone, but it isn’t a requirement.

IS-IS supports cleartext password authentication only. In Cisco routers, all IS-IS interfaces have a default metric of 10.

All areas are required to connect to the OSPF backbone. If an area does not have physical connectivity to the backbone, but it does connect to another area that connects to the backbone, a virtual link can provide the required area to the backbone connection.

External routes advertised in a not-so-stubby area are flooded within the area with Type-7 LSAs.

Type 3 LSAs are summary LSAs. Type 4 LSAs announce reachability to the ASBR.

Type 5 LSAs announce AS external routes. Type 7 LSAs announce external routes within an NSSA.

Costs are associated with interfaces. The ip ospf cost command changes the default cost of an interface.

ASBRs set the P-bit in Type 7 LSAs. An ABR receiving the LSA translates the Type 7 LSA to a Type 5 LSA if the P-bit is set to 1.

The attached (ATT) bit is set by a L1/L2 IS to indicate that it is attached to an L2area and is, therefore, available to forward traffic outside of the L1 area.

The DR forms adjacencies with all routers in a multiaccess network.

IS-IS multiaccess networks have designated ISs only, no backup designated IS
(BDIS) exists.

OSPF does equal-cost, multipath load balancing.

OSPF does not do unequal-cost, multipath load balancing. If two paths exist to the
same destination, OSPF chooses the path with the lowest cost.

IS-IS does not run over IP.

Type 3 LSAs are Summary LSAs and are produced by ABRs. They are flooded into
areas to advertise destinations outside the area.

If two paths exist to a single destination, the L1 path takes precedence, regardless
the cost. The E-bit specifies whether an area supports external routes. If the area does not
support the flooding of external routes, the area is a stub area and the E-bit must be
cleared in Hello packets on all interfaces in the area. If the area does support external
routes, the area is not a stub area and the E-bit must be set.

Multi-Exit Discriminator (MED) is used when an AS has multiple connections to
another AS. It is an attempt by an AS to influence the selected route a neighboring

AS should select.

The Weight attribute selects the best outbound route, it is not advertised to peers, and
it is locally significant. The highest Weight is selected.

BGP synchronization requires that BGP speakers in a transit AS not advertise routes
until all routers within that AS have learned about the route through an IGP.

BGP communities apply a common policy or properties to a group of destinations.
The selected route is marked with the > symbol.

External BGP routes have an administrative distance of 20. Internal BGP routes
have an administrative distance of 200.

Route dampening suppresses flapping BGP routes.

Classless Interdomain routing (CIDR) was first implemented in BGPv4.

BGP Multihop is configured to peer with an external BGP neighbor that is not in a local subnet.

BGP peer groups apply a common set of policies to a group of BGP neighbors.

The lowest BGP MED is selected.

The highest local preference is selected. The default local preference is 100.

One or more route reflectors are allowed in a cluster.

NLRI is Network Layer Reachability Information, which are destination IP networks.

Hosts use Internet Group Management Protocol (IGMP) to join a multicast group.
IGMP operates between the host and the local router.

The IANA has reserved addresses in the range from 224.0.0.0 through 224.0.0.255
that network protocols use on a local network segment. The address 224.0.0.1 means
“all systems on this subnet,” 224.0.0.2 means “all routers on this subnet,” and
224.0.0.5 is for OSPF routers.

On Cisco routers, OSPF external routes have an administrative distance of 110. The
distance of OSPF external routes can be changed by using the distance external
command.

Distribute lists filter the contents, inbound or outbound, of routing updates.

Protocol Independent Multicast is a multicast routing protocol. The two flavors of

PIM are sparse mode and dense mode.

The network mask pair 0.0.0.0 255.255.255.255 signifies any network.

The distance external command can assign an administrative distance, which is
different than the default (110), to OSPF external routes.

The RIP route is entered in the routing table because the administrative distance of
RIP is 120 and the administrative distance of IBGP is 200.

The rendezvous point (RP). The RP is charged with the task to gather the
information of senders and make the information available to other PIM routers.

Cisco Group Management Protocol (CGMP) and IGMP Snooping are both methods
to control multicast traffic in a switched LAN environment.

Policy-based routing can change the next-hop address based on the source IP
address of the packet.

Auto-RP and PIM Bootstrap Router (BSR) are methods for PIM networks to
configure the RP automatically for multicast groups.

Priority queuing (PQ) always empties the high queue before servicing the lower
priority queues.
The Resource Reservation Protocol (RSVP) is a signaling protocol that enables hosts
to obtain special qualities of service for data traffic.

Weighted random early detection (WRED) drops packets to prevent congestion.

WRED expects TCP to reduce its window size as packets are dropped and, therefore,
to transmit fewer packets.

Generic Traffic Shaping (GTS) is configured with the traffic-shape command. This
command controls the traffic flow on an interface.

EIGRP and IGRP can load balance over unequal-cost paths. Cisco’s implementation
of OSPF, ISIS, and RIP can load balance over equal-cost paths.

By default, EIGRP uses four paths for load balancing to a destination. EIGRP can
be configured to load balance up to six paths. The paths do not need to have equal
costs.

Implementation FRF.9 of the Frame Relay Forum defines compression on Frame

Relay networks. FRF.5 is the Frame Relay/ATM Network Interworking
Implementation. FRF.11 is the Voice over Frame Relay Implementation Agreement.
FRF.12 is the Frame Relay Fragmentation Agreement.

Multi-Protocol Label Switching inserts a 32-bit field, which includes a 20-bit tag, between the
Layer-2 header and the Layer-3 header.

Stacker is a Cisco enhanced version of the Lempel-Ziv (LZS) compression
algorithm.

Custom queuing (CQ) can configure up to 16 queues to prioritize traffic.

Only Committed Access Rate (CAR) uses the rate-limit command.

OSPF can be configured to use up to six equal-cost paths.

EIGRP can be configured to use up to six equal-cost paths.

Custom queuing is configured by creating queues with the queue-list protocol command, assigning byte counts to the queues with the queue-list queue byte-count command, and then configuring the interface with the custom-queue-list command.

Priority queuing uses four queues: high, medium, normal, and low.

DSCP uses 6 bits, which produce 64 values for packet classification.

TACACS authentication protocol uses TCP.

The triple DES (3DES) key is 3 × 56 bits = 168 bits.

Real-Time Transport Protocol (RTP) operates in the transport layer of the OSI
model.

The Q.931 standard defines the procedures responsible for call setup and signaling.

Erlangs is the unit that describes the number of calls in one hour.

NLSP and IPX EIGRP routing protocols reduce IPX broadcast traffic.

The 80-bit IPX address consists of the 32-bit network number plus a 48-bit MAC address.

RADIUS uses UDP port 1812 or UDP 1645 and UDP 1646. TACACS uses TCP.

The G.711 codec produces a 64 kbps bit rate.

The Signaling Control Point (SCP) is a database for special call processing and routing.

Authorization determines which resources are accessed. Authentication determines who is the user. Accounting keeps track of what resources were accessed, by whom and when.

TACACS+ separates AAA functions into separate modules. RADIUS combines authentication and authorization.

The PIX Firewall uses the Adaptive Security Algorithm (ASA) for stateful, connection-oriented security.

The IPX address is 80 bits in length, consisting of a 32-bit network part and a 48-bit host part.

NetWare Link Services Protocol (NLSP) is a link-state protocol that is based on IS-IS.