Tom's Notes » Cisco CCNA4

Module 6 – Network Administration

Tom GT — Thu, 18 Jun 2009 21:00:02 +0000

Index

Network Management

Network Management

OSI and Network Management

There are 4 parts to the OSI management model:

Orginizational – describes components of network management and their relationships
Information – structure and storage of information, representation of objects
Communication – method for data transfer between the agent and manager
Functional – Fault, Configuration, Accounting, Performance, Security (FCAPS)

SNMP and CMIP

Simple Network Management Protocol (SNMP):

made up of a protocol, a database structure, and a set of data objects
SNMP v2c was released in ‘93, SNMP v3 is current
TCP/IP adopted SNMP as a protocol in 1989

Common Management Information Protocol (CMIP):

ISO standard
made up of a complex set of standards

MIB structure

MIB I includes 114 objects, and the newer MIB II is expanded, with 185 objects defined.

There are objects administered by the Internet Activities Board (IAB), which are standardized. Vendors may also create custom objects, while it is recommended that they release the definitions of their objects, they are not always.

SNMP Protocol

There are three main functions that can be used by the Network Management Station, GetRequest, GetNextRequest, and SetRequest. SNMP v2c added a Bulk function to these, allowing more than one value to be requested at a time.

SNMP Configuration

Basic SNMP Configuration
Router(config) snmp-server community [string (acts as password)] [ro (read only) | rw (read/write)]

Specify the location and main contact of managed device
Router(config) snmp-server location [text]
Router(config) snmp-server contact [text]

RMON

is an MIB
based on IETF RFCs
gets statistics by analyzing all frames on a segment
RMON1 works on the data link layer
RMON2 works on the network layer
has a different group for each type of information it analyzes

Syslog

based on UNIX syslog utility
has 8 severity levels, from 0 (most critical) to 7 (least critical):
- 0 – Emergencies
- 1 – Alerts
- 2 – Critical
- 3 – Errors
- 4 – Warnings
- 5 – Notifications
- 6 – Informational (default on Cisco IOS)
- 7 – Debugging

Configuration:

Enable logging to all supported destinations:
Router(config)#logging on

Send log messages to a syslog server host:
Router(config)#logging [hostname | ip address]

Set logging severity level
Router(config)#logging trap [level name]

Include timestamp with syslog message:
Router(config)#service timestamps log datetime

Module 5 – Frame Relay

Tom GT — Wed, 17 Jun 2009 21:00:08 +0000

Index

Frame Relay Concepts

General Concepts

Customer end is the DTE
Frame Relay switch is the DCE
The connection between two DTEs is called a Virtual Circuit (VC)
Dynamic VCs are known as Switched VCs (SVCs)
Static VCs are known as Permanent VCs (PVCs), which are the most common.
Frame Relay has no error recovery, corrupt frames are discarded without notification. It is up to the upper layers to detect missing frames and request retransmission
Several VCs can operate on one physical link
DLCIs are used locally to identify where a packet is going, and is put in the address field of an outgoing frame.

Bandwidth and flow

Typical port speed is 64kbps to 4Mbps, though some providers offer up to 45Mbps
Each VC has a Commited Information Rate (CIR), which is the guaranteed bandwidth.
As it is unlikely for all VCs to need to transmit at the same time, lines are commonly over-allocated by up to 2-3x the CIR
Packets that are being sent over the CIR are marked as Discard Eligible, and will be discarded should a link become congested.
The difference between the port speed and the CIR is the Excess Information Rate (EIR)
The load average is worked out by dividing the data transferred by the committed time (Tc)
Idle time can not be saved up for use at a later time of load
The FECN bits are set on all frames downstream of a congested link
The BECN bits are set on all frames upstream of a congested link
When a DTE receives a frame with ECN bits set, it is expected to reduce the amount of traffic it creates

Address Mapping and Topology

DLCIs are mapped to the IP address of the remote router, either with RARP or statically
Frame Relay can be used in a full or partial mesh topology, point to point links are possible, but are usually not cost effective
There can only be <100 VCs per physical interface, so full mesh topologies are not usually cost effective for larger networks

LMI

Link Management Interface (LMI) is used on Frame Relay to provide status updates
Some DLCIs are reserved for use by LMI
There are three types of LMI supported by Cisco routers:
- Cisco – original LMI extensions
- ANSI – ANSI standard T1.617 Annex D
- Q933a – ITU standard Q933 Annex A

Frame Relay Configuration

Basic Configuration

Router(config)# interface [interface]
Router(config-if)# ip address [address] [mask]
Router(config-if)# encapsulation frame-relay [cisco | ietf]

Static Map

Router(config)# frame-relay map [protocol] [address] [dlci] ['broadcast']

Subinterfaces

Router(config) interface [interface].[sub-if no.] [point-to-point | multipoint]
Router(config-if) ip address [address] [mask]
Router(config-if) frame-relay interface-dlci [dlci]

Module 4 – ISDN and DDR

Tom GT — Sun, 14 Jun 2009 21:00:25 +0000

Index

ISDN Concepts

ISDN Benefits

Traditional PSTN is based on an analog connection between the local exchange and the customer site. Analog links are limited in the available bandwidth.

ISDN was made possible by telcos upgrading switches to support digital signals over the local loop.

Some of the benefits include:

Carries several different types of data
Faster to connect than analog
B channels offer more bandwidth than dialup
B channels are able to be used with PPP

The D channel is used to setup calls and signaling, and is either 16 or 64kbps.

Each B channel offers 64kbps of bandwidth.

ISDN standards and access methods

ISDN is made up by three series of protocols:

E protocols – defines telephone network standards
I protocols – defines concepts, terminology, and general methods. I.100 includes general ISDN concepts. I.200 defines service aspects. I.300 defines network aspects. I.400 defines how UNI is provided.
Q protocols – defines how call setup and switching is handled.

Unlike a standard TCP connection, ISDN uses out-of-band signalling, meaning all connections are negotiatied over the D channel, while data goes through the B channels.

B channels can carry digitized speech signals. They use either HDLC or PPP. PPP is more robust as it allows for authentication and negotiation of link and protocol configuration.

The D channel uses Link Access Procedure on the D Channel (LAPD) as its data link layer protocol.

North America and Japan PRI connections are 23B+D (giving T1), and most other countries use 30B+D (giving E1).

ISDN 3-layer model and related standards

BRI and PRI pysical layer specifications are defined in ITU-T I.430 and I.431

ISDN data link layer is based on LAPD and is specified in ITU-T Q.920, Q.921, Q.922, and Q.923

The network layer is defined by ITU-T Q.930 and Q.931, also known as I.450 and I.451, respectively.

With a BRI connection, a standard phone line local loop is used. Although there is only one physical link, the 3 ISDN channels are multiplexed as to not require any extra physical links.

Each BRI frame on the physical layer has:

8 bits from B1
8 bits from B2
2 bits from D
6 bits overhead

4,000 frames are sent per second, giving a total of 192kbps, though 48kb of this is overhead, which makes for 144kbps of throughput.

The overhead bits are used for these purposes:

Framing – provides synchronization
Load balancing – changes the average bit value
Echo of previous D channel bits – used for contention resolution
Activation bit – activates devices
Spare bit – unassigned

The ISDN signaling channel uses LAPD for Layer 2. The flag and control fields for LAPD are the same as in HDLC. The address field is 2 bytes long, made up of the following:

Service Address Point Identifier (SAPI) – identifies the portal that is providing LAPD services to layer 3. One byte long
Command/Response (C/R) bit – specifies if the command is a command or response
Terminal Endpoint Identifier (TEI) – identifies the terminal, ranges between 0 and 63 if static, 64-126 if dynamic, and 127 for a broadcast. 7b long.

ISDN functions

The D channel is always up, so that ISDN calls can be started.

The following steps are used to connect an ISDN call:

Called number sent to local ISDN switch via D channel
The local switch sets up a connection to the remote ISDN switch using the SS7 protocol
The remote switch signals the remote site via the D channel
The remote site’s NT-1 device sends a call-connect message to the remote switch
The remote switch uses SS7 to forward the call-connect message to the local switch
The local switch connects one B channel to the remote site, leaving the other free for another conversation or data link, as both channels can be used at the same time.

ISDN reference points

The following devices are used to facilitate ISDN connectivity:

Terminal Equipment 1 (TE1) – native ISDN device. eg. ISDN router or ISDN phone
Terminal Equipment 2 (TE2) - non-ISDN device. eg. workstation or router. Requires TA to connect to ISDN service
Terminal Adapter (TA) – Converts serial link into ISDN BRI
Network Termination 2 – point at which the ISDN lines at the customer site are aggregated and switched using a customer switching device
Network Termination 1 – controls the physical/electrical termination at the customer premises. Converts from 4 wire ISDN BRI to 2 line used over local loop

The following are reference points used to identify connections between ISDN devices:

R – connection between TA and TE2
S – connection between NT2 and customer premises equipment
T – electrically identical to the S interface, referencing the outbound connection from the NT2 to the ISDN network or NT1
U – link between the NT1 and the telco owned ISDN network

As S and T connections are electrically identical, interfaces are often labeled as S/T, meaning they can perform either role.

Determining the router ISDN interface

For North America and Japan, a router should use an ISDN interface with the NT1 built in, as the NT1 is a customer premises device in these countries. These interfaces are labeled with a ‘U’

In other countries, the NT1 is usually provided by the telco, so the interface should not include an NT1. These interfaces are labeled with ‘S/T’

If the router doesn’t have a way to fit an ISDN interface, a TA is required, this will connect to a serial interface on the router.

ISDN switch types

As there are no fixed standards for D channel communication with ISDN switches, it is necessary to specify the type on the router. The type of switch generally varies from country to country, though it can change within a country as well.

With some switches, a service profile identifier (SPID) is required. An SPID sets the line configuration, and allows for voice and data to both be carried over the local loop.

SPIDs are only used in North America and Japan.

Below is a table of the different switch types by country.

Country	Switch Type
US and Canada	AT&T 5ESS and 4ESS; Northern Telecom DMS-100
France	VN2, VN3
Japan	NTT
UK	Net3 and Net5
Europe	Net3

ISDN Configuration

Configuring ISDN BRI

To setup a BRI connection, use the following commands:

Router(config)# isdn switch-type [type]

Router(config)# interface bri [number]

If required: Router(config-if)# spid1 [spid-number] [ldn (optional)]

If required: Router(config-if)# spid2 [spid-number] [ldn (optional)]

Router(config-if)# no shutdown

Configuring ISDN PRI

To setup a PRI connection, use the following commands:

Router(config)# controller [t1 | e1]

For T1 lines: Router(config-controller)# framing [sf | esf]

For E1 lines: Router(config-controller)# framing [crc4 | no-crc4] [australia (optional)]

Router(config-controller)# linecode [ami | b8zs | hdb3] - Nth America usually uses B8ZS, Europe usually uses HDB3.

Router(config-controller)# pri-group [timeslots range (1-24 for T1, 1-31 for E1)]

Router(config)# interface serial [slot/port: | unit:] [23 (T1) | 15 (E1)]

Router(config-if)# isdn switch-type [type]

Verifying ISDN configuration

Router# show isdn status allows the isdn link to the local switch to be checked. When the link is working correctly, the layer 1 status will be ‘ACTIVE’, and the layer 2 status ‘MULTIPLE_FRAME_ESTABLISHED’

Router# show isdn active will give information regarding:

called number
time until disconnect
Advice of Charge (AOC)
charging units used
whether AOC is provided during or after calls

Router# show dialer shows the following details:

current call status
dial timer config
dial reason
remote device that is connected

Router# show interface bri 0/0 will display interface statistics.

Troubleshooting ISDN configurations

Router# debug isdn q921 displays layer 2 information related to the D channel. This command should be used if show isdn status doesn’t show layer 1 as ‘ACTIVE’ and layer 2 as ‘MULTIPLE_FRAME_ESTABLISHED’

Router# debug isdn q931 shows exchanged messages for call setup and teardown

Router# debug ppp authentication displays messages regarding PPP authentication

Router# debug ppp negotiation displays messages related to the PPP link configuration when the connection is started

Router# debug ppp error shows errors relating to PPP.

The debug ppp commands are for use when there are layer 2 issues and show isdn status doesn’t show any problems with the ISDN connection.

DDR Configuration

DDR operation

Dial-on-demand routing (DDR) is used to make a connection when an ‘interesting packet’ needs to cross the link. Packets that are not interesting are forwarded only if the connection is already active. The protocols of which the packets are ‘interesing’ are set with dialer-list.

The following steps are used for an interesting packet to be forwarded using DDR:

Router recieves packet, checks there is a valid route to the destination, and identifies the outbound interface
If the interface is connected to the next hop router, the packet is forwarded, if the interface is not connected, the packet is identified as interesting.
The router connects the the remote site. Now that the link is up, the packet is forwarded.
All packets to the remote site, whether interesting or not are forwarded over the link.
Once the set time limit has passed without an interesting packet, the call is terminated.

Configuring legacy DDR

Legacy DDR refers to having one connection set per interface.

Follow these steps to set up legacy DDR:

Set the static routes
Define the interesting traffic
Set up the dialer information

Static Routes

Static routes should be used over for connections with DDR, as using a dynamic routing protocol could cause the connection to be dialed more often than it is needed.

Defining interesting traffic

Using dialer-list [no.] protocol ip permit allows all IP traffic, if more control is needed, an ACL can be used. Create the ACL and then use dialer-list [no.] protocol ip list [ACL no.]

Dialer profiles

Dialer profiles allow for a single ISDN interface to connect to different links as required. Several interfaces can also be used, so if, for example, there were three possible connections, but only 2 were ever required at once, the dialer could have 3 profiles in a pool with 2 interfaces.

When configuring dialer interfaces, the dialer pool membership is specified for the physical interface, with the other settings configured for each logical interface.

Debugging DDR configs

debug dialer [events | packets] will help diagnose problems with the DDR configuration.

Module 3 – PPP

Tom GT — Tue, 09 Jun 2009 21:00:30 +0000

Index

Serial Point-to-Point Links

TDM
Time division multiplexing interleaves data so protocols do not have to wait for another protocol to finish before they can transmit. It is a similar concept to Hyper-Threading in CPUs.
Demarcation Point
The demarcation point or demarc is where the responsibility of the link goes to the customer, rather than the link provider.

In the US, this is before the DCE device (CSU/DSU), where the local loop ends. In most other countries, the DCE is provided by the telco, in the form of an NTU. NTUs allow the telco to manage and troubleshoot the local loop.
DTE and DCE
The DTE is the customer device, which is commonly a router, though there are many other devices it could be, including a computer or fax machine.

The DCE is the device which changes the DTE’s data into a form suitable for the WAN link.

There are 4 main specifications defined in DTE/DCE standards:

Physical – number of pins and connector form factor
Electrical – voltage levels
Functional – assigns functions to different signals
Procedural – sequence of data transmission

When two DTE devices are connected together, such as in a lab, a null modem cable must be used. For synchronous links, one of the DTEs must have a clock rate set, to emulate a DCE.

A DTE will usually have a DB-60 or smart-serial connection on it, and the cable will have the connection specified by the telco on the other end.
HDLC
HDLC is the main signaling standard used by WAN links. Below is a list of some of the derivatives of HDLC in use:

Link Access Procedure, Balanced for X.25
Link Access Procedure on the D channel for ISDN
Link Access Procedure for Modems and PPP for modems
Link Access Procedure for Frame Relay

HDLC has three types of frames:

Information frames – for data
Supervisory frames – request/response for when piggybacking is not used
Unnumbered frames – control, such as connection setup

Enabling HDLC on an Interface
Router(config-if)# encapsulation hdlc
Troubleshooting a serial interface
By using show interface [type] [no.], details of the interface will be shown, so the incorrect parameter can be found.

PPP Authentication

PPP layered architecture

PPP is made from two sub-protocols:

Link Control Protocol – establishes a point-to-point link
Network Control Protocol – configures the network layer protocols

LCP is used for the following:

Authentication - authenticates the connection, using either PAP or CHAP
Compression – sets up compression on the link which is then decompressed when it reaches the other end of the link
Error detection – recognizing errors
Multilink – IOS 11.1 and later supports multilink, which provides load balancing
PPP Callback – increases security by requesting the other device calls back to make the connection

NCP has a different protocol to control each network layer protocol, e.g IPCP is used for IP connections.

PPP uses the following fields:

Flag – 01111110 is used to mark the beginning and end of a frame
Address – a broadcast address (11111111) is used, as PPP doesn’t use individual station addresses
Protocol – specifies the type of data being carried
Data - between 0 and 1500 bytes of data
FCS - checksum used to detect errors

Establishing a PPP connection

Link-establishment phase – LCP frames are used to configure the link. Details such as MTU are confirmed during this phase. The phase is completed with a configuration acknowledgement frame
Authentication phase (optional) – if configured, the link is authenticated before the network layer link is initialized
Network layer protocol phase – configures network layer protocols

PPP Authentication Protocols

There are two authentication protocols used by PPP, Password Authentication Protocol (PAP) and Challenge Handshake Authentication Protocol (CHAP). CHAP is the preferred protocol, as it is more secure.

PAP

PAP sends the username and password repeatedly until the other node acknowledges it as correct or terminates the session. PAP is not very secure as the username and password are sent in plain text.

CHAP

CHAP uses a 3 way handshake, starting off with a challenge from the central site router. The remote router then uses the password and the challenge and sends back the checksum, if the checksum is correct, the connection is allowed, if it fails, the connection is dropped. There is a limit on the number of tries from the remote router, to reduce the risk even further.

To set up the hostname and password for CHAP:

Router(config-if)# ppp chap hostname [hostname]

Router(config-if)# ppp chap password [password]