(print servers), or network traffic (network
servers). A server receives a request and, after any necessary
processing, the requested file is returned to the client.
Typically, multiple client programs share the services of a
common server. Figure shows client-server examples.
Content 7.3 Troubleshooting TCP/IP Application Layer
Protocols 7.3.3 Terminals and consoles
Telnet is the standard terminal emulation protocol in
the TCP/IP protocol stack. Telnet is defined in RFC 854 and
operates over the TCP port 23. Telnet and FTP were the first
two services available on ARPANET. To understand how Telnet
works, it is necessary to first make a distinction between
console and terminal. A console refers to a keyboard and
monitor that are directly connected to the computer system. In
mainframe computing, a console was also referred to as a dumb
terminal since it only operated by using the resources of the
remote server. A microcomputer is now more commonly used as a
console. All consoles require a terminal connection to enable
users to log in to remote systems and use resources (for
example, CPU, applications, and storage) as if they were
connected to a local system. A terminal is a console that
artificially emulates the physical hookup of a console. The
destination host assumes it has a direct connection to the
client since the terminal just provides a communication channel
for the user's input and output. A terminal program is commonly
used to connect to a central server over the network. The term
"terminal emulator" refers to a terminal application
that is implemented in software. Clients can use the Telnet
program to establish a terminal connection. Other software such
as HyperTerminal or TeraTerm can also be used and they
typically offer more advanced features. The following lists
commands that could be used to troubleshoot Telnet problems.
Note: Debug output is normally sent to the console port.
This means that if the connection is established through the
Telnet port, the debug output will not be seen. Use the
terminal monitor IOS command to redirect the output to
any of the VTY ports. Keep the amount of debugging that is
enabled to a minimum. Note: RDP Protocol (Terminal
Server) is a little known Telnet port (port 3389) used by
Remote Desktop Protocol (RDP). RDP is the remote Windows
terminal protocol used by Microsoft Windows NT 4.0, Terminal
Server Edition operating system and Windows 2000 Terminal
Services. Keep this in mind when troubleshooting. Telnet for
Troubleshooting
Network administrators often overlook
Telnet as a troubleshooting tool. However, Telnetting to a host
allows better verification of network status than just using
ping. Telnet runs on top of the TCP protocol, so it establishes
a more reliable indication of accessibility than ICMP echo
requests can. It also tests higher-level functions of the
destination host system. A server may be inaccessible for
application layer functions, but still answers pings since
those are handled by the lower layer protocols. Telnet also has
an additional feature that makes it valuable for
troubleshooting application layer protocols. Telnet client
applications allow the user to select the destination port
number to be used. It can be used to connect to other TCP ports
on destination hosts to test out other functions. That means
that Telnet can contact network application programs other than
a Telnet server. This can be useful as a substitute for a
client application program. For example, Telnetting to port 25
(SMTP) will verify that the e-mail server is answering. Telnet
to port 80 (HTTP) to verify the Web server is answering.
Source Telnet Interface
Finally, a useful IOS
command to use when testing an access list is the ip telnet
source-interface command. This specifies the IP address of
an interface as the source address for Telnet connections. To
reset the source address to the default for each connection,
use the no form of this command. By default, Telnet will use
the IP address of the closest interface to the destination as
the source address. However, sometimes another interface may be
preferred as the source. Conceptually, this is similar to
specifying another source IP address when using an extended
ping command. The following example forces the IP address for
FastEthernet interface 0/1 as the source address for Telnet
connections: Router(config)#ip telnet source-interface
FastEthernet 0/1 Figure shows some IOS commands used to
isolate Telnet problems.
Content 7.3
Troubleshooting TCP/IP Application Layer Protocols
7.3.4 Web traffic Hypertext Transfer Protocol
(HTTP) is the protocol used to transfer the files that make up
web pages. Although the HTTP specification allows for data to
be transferred on port 80 using either TCP or UDP, most
implementations use TCP. HTTPS is a secure version of the HTTP
protocol. Aside from the initial connection and setup, HTTPS
and HTTP are basically the same. The difference lies in the
initial setup between client and server. HTTPS uses the Secure
Socket Layer (SSL) protocol. SSL was created in order to secure
credit card purchases over the Internet. It requires that both
sides of a connection be authenticated and that data be
encrypted and decrypted. It uses port 443 to initiate a secure
connection. HTTP connectivity can be tested using any Telnet
application that allows a port number to be specified by
Telnetting to the IP address of the destination server using
port 80. - If the connection failed, a message will
display stating that the Telnet application could not open a
connection to the host on port 80.
- If the connection
was successful, a hello message may be displayed or a Telnet
window will open, but there will be no response. This indicates
HTTP connectivity to the server. To have the Web server
respond, type GET / HTTP/1.0, then press the enter key
twice.
Note: By default characters are not
echoed in MS Telnet upon successful connection to a Web Server
unless 'local echo' is enabled in the preferences. The
following table presents commands that can be used to
troubleshoot a World Wide Web network application. A
troubleshooter uses the information from these commands to
isolate problems at the application layer that are related to
the Web and the HTTP protocol.
Content 7.3
Troubleshooting TCP/IP Application Layer Protocols
7.3.5 Electronic mail Simple Mail Transport
Protocol (SMTP) is used to transport e-mail messages in ASCII
format using TCP between clients and servers. Other protocols
such as Post Office Protocol (POP) or Internet Access Message
Protocol (IMAP) are used to retrieve e-mails from mail servers.
For example, assume User-A wants to send e-mail to User-B. When
User-A clicks on the Send button, e-mail is sent to the
local e-mail server using the SMTP protocol. The e-mail server
will then send the e-mail using SMTP to User-B’s e-mail server.
It remains stored there until User-B collects it. Later, User-B
connects to the local e-mail server and downloads the e-mails
using either POP3 or IMAP 4. POP v3 is the current version of
the protocol and it is incompatible with earlier versions. POP3
downloads user e-mails to the local computer. For this reason,
POP3 is best suited in situations where users retrieve their
e-mail from the same computer. If users use different
computers, their e-mails will likely be spread around several
computers. IMAP v4 is another alternative that lets users
download their e-mail at any time to any computer. Because
these different protocols are used to send and receive mail, it
is possible that mail clients can perform one task and not the
other. Therefore, when verifying the configuration of a mail
client, both the mail relay (SMTP) server and mail (POP or
IMAP) servers should be verified. SMTP, IMAP, and POP
connectivity can be tested using any Telnet application that
allows a port number to be specified. Telnet to the IP address
of the destination server using ports 25, 143, and 110