Recommended Action The following steps are suggested for
this symptom:
-
Step 1 Use a serial analyzer to isolate the source of the
errors. If you detect errors, it is likely that there is a hardware problem
or a clock mismatch in a device that is external to the router.
-
Step 2 Use the loopback and ping tests described later
in this chapter to isolate the specific problem source.
-
Step 3 Look for patterns. For example, if errors occur
at a consistent interval, they could be related to a periodic function
such as the sending of routing updates.
Table 3-2
details the meaning of CRC errors, framing errors, and aborts. These fields
appear in the display shown in Figure
3-1.
Table 3-2 Meaning of Key Input Errors for
Serial Line Troubleshooting
-----------------------------------------------------------------------------------------------------------
Input Error Type (Field Name) Possible Causes and Suggested Actions
-----------------------------------------------------------------------------------------------------------
CRC errors (CRC) Meaning:
CRC calculation does not pass; some data is corrupted.
Possible Causes:
1 Noisy serial line
2 Serial cable is too long; cable from the CSU/DSU to the router is not
shielded
3 SCTE mode is not enabled on DSU
4 CSU line clock is incorrectly configured
5 Ones density problem on T1 link (incorrect framing or coding
specification)
Suggested Actions:
Step 1 Ensure that the line is clean enough for transmission
requirements; shield cable if necessary.
Step 2 Make sure the cable is within the recommended length (no more
than 50 feet [15.24 meters] or 25 feet [7.62 meters] for T1 link).
Step 3 Ensure that all devices are properly configured for common line
clock. Set SCTE on the local and remote DSU. If your
CSU/DSU does not support SCTE, see the section "Inverting the
Transmit Clock" later in this chapter.
Step 4 Make certain that the local and remote CSU/DSU is configured
for the same framing and coding scheme (for example, Extended
Superframe Format [ESF]/Binary 8-Zero Substitution [B8ZS])
used by the leased-line or other carrier service.
Step 5 Contact your leased-line or other carrier service and have them
perform integrity tests on the line.
Framing errors (frame) Meaning:
Detected packet does not end on an 8-bit byte boundary.
Possible Causes:
1 Noisy serial line
2 Improperly designed cable; serial cable is too long; the cable from the
CSU or DSU to the router is not shielded
3 SCTE mode is not enabled on the DSU; the CSU line clock is
incorrectly configured; one of the clocks is configured for local
clocking
4 Ones density problem on T1 span (incorrect framing or coding
specification)
Suggested Actions:
Step 1 Ensure that the line is clean enough for transmission
requirements. Make certain you are using the correct cable.
Shield the cable if necessary.
Step 2 Make sure the cable is within the recommended length (no more
than 50 feet [15.24 meters] or 25 feet [7.62 meters] for T1 link)
Step 3 Ensure that all devices are properly configured to use common
line clock. Set SCTE on the local and remote DSU. If your
CSU/DSU does not support SCTE, see the section "Inverting the
Transmit Clock" later in this chapter.
Step 4 Make certain that the local and remote CSU/DSU is configured
for the same framing and coding scheme (for example,
ESF/B8ZS) used by the leased-line or other carrier service.
Step 5 Contact your leased-line or other carrier service and have them
perform integrity tests on the line.
Aborted transmission (abort) Meaning:
Illegal sequence of one bits (more than 7 in a row)
Possible Causes:
1 SCTE mode is not enabled on DSU
2 CSU line clock is incorrectly configured
3 Serial cable is too long; cable from the CSU or DSU to the router is
not shielded
4 Ones density problem on T1 link (incorrect framing or coding
specification)
5 Packet terminated in middle of transmission; typical cause is an
interface reset or a framing error
6 Hardware problem---bad circuit, bad CSU/DSU, bad sending interface
on remote router
Suggested Actions:
Step 1 Ensure that all devices are properly configured to use common
line clock. Set SCTE on the local and remote DSU. If your
CSU/DSU does not support SCTE, see the section "Inverting the
Transmit Clock" later in this chapter.
Step 2 Shield the cable if necessary. Make certain the cable is within
the recommended length (no more than 50 feet [15.24 meters] or
25 feet [7.62 meters] for T1 link); ensure that all connections are
good.
Step 3 Check the hardware at both ends of the link. Swap faulty
equipment as necessary.
Step 4 Lower data rates and determine if aborts decrease.
Step 5 Use local and remote loopback tests to determine where aborts
are occurring (see the section "Special Serial Line Tests," later
in this chapter.)
Step 6 Contact your leased-line or other carrier service and have them
perform integrity tests on the line.
-----------------------------------------------------------------------------------------------------------
Inverting the Transmit Clock
If you are attempting serial connections of greater than 64 kbps with a
CSU/DSU that does not support serial clock transmit external (SCTE), you
might have to invert the transmit clock on the router. Inverting the transmit
clock compensates for phase-shifts between the data and clock signals.
On a Cisco 7000 series router, enter the invert-transmit-clock interface
configuration command. For Cisco 4000 series routers, use the dte-invert-txc
interface configuration command. To ensure that you are using the correct
command syntax for your router, check the Router Products Configuration
Guide and the Router Products Command Reference publications.
Note On older platforms, inverting the transmit clock
might require that you move a physical jumper.
Evaluating Output Drops
Output drops appear in the output of the show interfaces serial number
command when the system is attempting to hand off a packet to a transmit
buffer but no buffers are available. The output drops count is illustrated
in Figure
3-1.
Symptom Increasing output drops
Possible Cause Input rate to serial interface exceeds
bandwidth available on serial link
Recommended Action The following steps are suggested
for this symptom:
-
Step 1 Minimize periodic broadcast traffic such as routing
and SAP updates by using access lists or other means. For example, to increase
the delay between SAP updates, use the ipx sap-interval interface
configuration command.
-
Step 2 Increase the output hold queue size in small increments,
using the hold-queue out interface configuration command.
-
Step 3 On affected interfaces, turn off fast switching
for heavily-used protocols. For example, to turn of IP fast switching,
enter the no ip route-cache interface configuration command. For
the command syntax for other protocols, consult the Router Products
Configuration Guide and the Router Products Command Reference publications.
-
Step 4 Implement priority queuing on slower serial links
by configuring priority lists. For information on configuring priority
lists, see the Router Products Configuration Guide and the Router
Products Command Reference publications.
Note Output drops are acceptable under certain conditions.
For instance, if a link is known to be overused (with no opportunity or
way to remedy the situation), it is often considered preferable to drop
packets than to hold them. This is true for protocols that support flow
control and can retransmit data (such as TCP/IP and Novell IPX). However,
some protocols, such as DECnet and Local Area Transport (LAT) are sensitive
to dropped packets and accommodate retransmission poorly, if at all.
Evaluating Input Drops
Input drops appear in the show interfaces serial number EXEC command
when too many packets from that interface are still being processed in
the system. The input drops count is illustrated in Figure
3-1.
Symptom Increasing number of input drops
Possible Cause Input rate exceeds the capacity of the
router or input queues exceed the size of output queues.
Note Input drop problems are typically seen when traffic
is being routed between faster interfaces (such as Ethernet, FDDI, and
Token Ring) and serial interfaces. When traffic is light, there is no problem.
As traffic rates increase, backups start occurring. By design, routers
drop packets during these congested periods.
Recommended Action The following steps are recommended
when this symptom is encountered:
-
Step 1 Increase the output queue size on common destination
interfaces for the interface that is dropping packets. Use the hold-queue
out interface configuration command.
-
Step 2 Reduce the input queue size (using the hold-queue
in interface configuration command) to force input drops to become
output drops. Output drops have less impact on the performance of the router
than do input drops.
Evaluating Interface Resets
Interface resets that appear in the show interfaces serial number
EXEC command are the result of missed keepalive packets. The interface
resets count is illustrated in Figure
3-1.
Symptom Increasing interface resets
Possible Cause The following causes can result in this
symptom:
-
Congestion on link (typically associated with output drops)
-
Bad line causing CD transitions
-
Possible hardware problem at the CSU, DSU, or switch
Recommended Action When analyzing interface resets, you
must examine other fields of the show interfaces serial number command
output to determine the source of the problem. Assuming an increase in
interface resets is being recorded, examine the following fields (illustrated
in Figure
3-1):
-
Step 1 If there are a high number of output drops in the
show interfaces serial number output, see the section "Evaluating
Output Drops" earlier in this chapter.
-
Step 2 Check the carrier transitions field in the show
interfaces serial number display. If carrier transitions are high while
interface resets are being registered, the problem is likely to be a bad
link or bad CSU or DSU. Contact your leased line/carrier service and swap
faulty equipment as necessary.
-
Step 3 Examine the input errors field in the show interfaces
serial number display. If input errors are high while interface resets
are increasing, the problem is probably a bad link or bad CSU/DSU. Contact
your leased line or other carrier service and swap faulty equipment as
necessary.
Evaluating Carrier Transitions
Carrier transitions appear in the output of the show interfaces serial
number EXEC command whenever there is an interruption in the carrier
signal; for example, when there is an interface reset at the remote end
of a link. The carrier transitions count is illustrated in Figure
3-1.
Symptom Increasing carrier transitions count
Possible Cause The following causes can result in this
symptom:
-
Line interruptions due to an external source (examples: physical separation
of cabling; Red or Yellow T1 alarms; lightning strikes somewhere along
the network)
-
Faulty switch, DSU, or router hardware
Recommended Action The following steps are suggested when
this symptom is encountered:
-
Step 1 Check hardware at both ends of the link (attach
breakout box or serial analyzer and test to determine source of problems).
-
Step 2 If analyzer or breakout box are unable to identify
any external problems, check router hardware.
-
Step 3 Swap faulty equipment as necessary.
Using the show controllers Command to Troubleshoot
Serial Lines
The show controllers EXEC command is another important diagnostic
tool. For serial interfaces on Cisco 7000 series routers, use the show
controllers cbus EXEC command. For Cisco access products, use the show
controllers EXEC command. For the AGS, CGS, and MGS, use the show
controllers mci EXEC command.
Figure
3-2 shows the output from the show controllers cbus EXEC command.
This command is used on Cisco 7000 series routers with the fast serial
interface processor (FSIP) card. Make certain that the cable to the CSU/DSU
is attached to the proper interface. Check the microcode version to see
if it is current.
Figure 3-2 show controllers cbus Command Output
The show controllers EXEC command is used on access products
such as the Cisco 2000, Cisco 2500, Cisco 3000, and Cisco 4000 series.
Figure 3-3
shows the show controllers command output from the basic-rate interface
(BRI) and serial interfaces on a Cisco 2503. (Note that, in the interest
of space, some output is not shown.) The show controllers output
indicates the state of the interface channels and describes the whether
a cable is attached to the interface. In Figure
3-3, serial interface 0 has an RS-232 DTE cable attached; serial interface
1 has no cable attached.
Figure 3-3 show controllers Command Output
Figure
3-4 illustrates the output for the show controllers mci command.
This command is used on AGS, CGS, and MGS routers only. If the electrical
interface is displayed as "UNKNOWN" (instead of V.35, EIA/TIA-449, or some
other electrical interface type), a bad applique or a problem with the
internal wiring of the card is likely. This might also indicate an improperly
connected cable. In addition, the corresponding display for the show
interfaces serial number EXEC command will show that the interface
and line protocol are down. (See Figure
3-1.)
Figure 3-4 Output from the show controllers
mci Command
Using debug Commands to Troubleshoot Serial Lines
The output from debug privileged EXEC commands provides diagnostic
information concerning a variety of internetworking events relating to
protocol status and network activity in general.
Caution Throughout this and other chapters, the use
of debug commands is suggested for obtaining information about network
traffic and router status. Use these commands with great care. In general,
it is recommended that these commands only be used under the direction
of your router technical support representative when troubleshooting specific
problems. Enabling debugging can disrupt operation of the router when internetworks
are experiencing high load conditions. When you finish using a debug
command, remember to disable it with its specific no debug command
or with the no debug all command (the undebug command is
also accepted).
To minimize the impact of using debug commands, follow this procedure:
-
Step 1 Issue the no logging console global configuration
command on your router. This command disables all logging to the console
terminal.
-
Step 2 Telnet to a router port and enter the enable
EXEC command.
-
Step 3 Issue the terminal monitor command and issue
the necessary debug commands.
Following this procedure minimizes the load created by using debug commands
because the console port no longer has to generate character-by-character
processor interrupts.
Following are some debug commands that are useful when troubleshooting
serial and WAN problems.
-
debug serial interface---Verifies whether HDLC keepalive packets
are incrementing; if not, a possible timing problem exists on the interface
card or in the network.
-
debug x25 events---Detects X.25 events, such as the opening and
closing of switched virtual circuits (SVCs). The resulting "Cause and Diagnostic"
information is included with the event report. Refer to the Debug Command
Reference publication for more information concerning this command.
-
debug lapb---Obtains LAPB or Level 2 X.25 information.
-
debug arp---Indicates whether the router is sending information
about or learning about routers (with ARP packets) on the other side of
the WAN cloud. Use this command when some nodes on a TCP/IP network are
responding, but others are not.
-
debug frame-relay lmi---Obtains local management interface (LMI)
information for determining whether a Frame Relay switch and router are
sending and receiving LMI packets.
-
debug frame-relay events---Determines whether exchanges are occurring
between a router and a Frame Relay switch.
-
debug ppp negotiation---Shows Point-to-Point Protocol (PPP) packets
transmitted during PPP startup, where PPP options are negotiated.
-
debug ppp packet---Shows PPP packets being sent and received. This
command displays the low-level packet dumps.
-
debug ppp errors---Shows PPP errors (such as illegal or malformed
frames) associated with PPP connection negotiation and operation.
-
debug ppp chap---Shows PPP Challenge Handshake Authentication Protocol
(CHAP) and Password Authentication Protocol (PAP) packet exchanges.
-
debug serial packet---Shows SMDS packets being sent and received.
This display also prints out necessary error messages to indicate why a
packet was not sent or was received erroneously. For SMDS, dumps the entire
SMDS header and some payload data when an SMDS packet is transmitted or
received.
More information about the output of each debug command is provided in
the Debug Command Reference publication.
Troubleshooting Clocking Problems
Clocking conflicts in serial connections can lead to either chronic loss
of connection service or generally degraded performance. The following
discussion addresses five issues regarding clocking problems:
Clocking Overview
The CSU/DSU derives the data clock from the data that passes through it.
In order to recover the clock, the CSU/DSU hardware must receive
at least one 1 bit value for every 8 bits of data that pass through it
(this is known as ones density.) Maintaining ones density allows
the hardware to recover the data clock reliably.
Newer T1 implementations commonly use Extended Superframe Format (ESF)
framing with Binary 8-Zero Substitution (B8ZS). B8ZS provides a scheme
by which a special code is substituted whenever 8 consecutive zeros are
sent through the serial link. This code is then interpreted at the remote
end of the connection. This technique guarantees ones density independent
of the data stream.
Older T1 implementations use D4 (also known as Superframe Format) framing
and Alternate Mark Inversion (AMI) coding. AMI requires that the sending
device maintain ones density, because it does not utilize a coding scheme
like B8ZS. This restricts the type of data that can be transmitted because
ones density is not maintained independent of the data stream.
Another important element in serial communications is serial clock transmit
external (SCTE) terminal timing. The SCTE is the clock echoed back from
the data terminal equipment (DTE) device (for example, a router) to the
data communications equipment (DCE) device (for example, the CSU/DSU).
When the DCE device uses the SCTE instead of its internal clock to sample
data from the DTE, it is better able to sample the data without error even
if there is a phase-shift in the cable between the CSU/DSU and the router.
Using SCTE is highly recommended for serial transmissions faster than 64
kbps. If your CSU/DSU does not support SCTE, see the section "Inverting
the Transmit Clock" earlier in this chapter.
Clocking Problem Causes
In general, clocking problems in serial WAN interconnections can be attributed
to one of the following basic causes:
-
Incorrect DSU configuration
-
Incorrect CSU configuration
-
Cables out of specification (longer than 50 feet [15.24 meters] or unshielded)
-
Noisy or poor patch panel connections
-
Several cables connected together in a row
Detecting Clocking Problems
To detect clocking conflicts on your serial interface, look for input errors
as follows:
-
Step 1 Use the show interfaces serial number EXEC
command on the routers at both ends of the link.
-
Step 2 Examine the display output for CRC, framing errors,
and aborts.
-
Step 3 If either of these steps indicates errors exceeding
an approximate range of 0.5 to 2.0 percent of traffic on the interface,
clocking problems are likely to exist somewhere in the WAN.
-
Step 4 Isolate the source of the clocking conflicts as
outlined in the next procedure, "Isolating
Clocking Problems."
-
Step 5 Bypass or repair faulty patch panel.
Isolating Clocking Problems
After you determine that clocking conflicts are the most likely cause of
input errors, the following general steps will help you isolate the source
of those errors:
-
Step 1 Perform a series of loopback and ping tests,
both local and remote, as described in the section "CSU
and DSU Loopback Tests" later in this chapter.
-
Step 2 Determine which end of the connection is the source
of the problem, or if the problem is in the line. In local loopback mode,
run different patterns and sizes in the ping tests (for example,
use 1500 byte datagrams). Using a single pattern and packet size may not
force errors to materialize, particularly when a serial cable to the router
or CSU/DSU is the problem.
-
Step 3 Issue the show interfaces serial number EXEC
command and determine whether input errors counts are increasing and where
they are accumulating.
If input errors are accumulating on both ends of the connection, clocking
of the CSU is the likely problem.
If only one end is experiencing input errors, there is likely to be
a DSU clocking or cabling problem.
If you see aborts on one end, it suggests that the other end
is sending bad information or that there is a line problem.
Note Always refer back to the show interfaces serial
number display output and log any changes in error counts or note if
the error count does not change.
Suggested Clocking Problem Remedies
Table 3-3
outlines suggested remedies for clocking problems, based on the source
of the problem.
Table 3-3 Serial Lines: Clocking Problems
and Suggested Remedies
-----------------------------------------------------------------------------------------------------------------
Clocking Problem Cause Suggested Actions
-----------------------------------------------------------------------------------------------------------------
Incorrect CSU configuration Step 1 Determine whether the CSUs at both ends are in agreement
regarding the clock source (local or line).
Step 2 Configure both to agree if not already correctly configured
(usually the line is the source).
Step 3 Check Line Build Out (LBO) setting on CSU/DSU to ensure
that the impedance matches that of the physical line. For
information on configuring your CSU, consult your CSU
hardware documentation.
Incorrect DSU configuration Step 1 Determine whether the DSUs at both ends have SCTE mode
enabled.
Step 2 Enable SCTE on both ends of the connection if not already
correctly configured.
(For any interface that is connected to a line of 128 kbps or
faster, SCTE must be enabled. If your CSU/DSU does not
support SCTE, see the section "Inverting the Transmit Clock"
earlier in this chapter.)
Step 3 Make sure that ones density is maintained. This requires that the
DSU use the same framing and coding schemes (for example,
ESF and B8ZS) used by the leased-line or other carrier service.
Step 4 If your carrier service uses AMI coding, either invert the
transmit clock on both sides of the link or run the DSU in
bit-stuff mode. For information on configuring your DSU,
consult your DSU hardware documentation.
Cable to router out of specification Step 1 Use shorter cable if longer than 50 feet (15.24 meters).
Step 2 Replace with shielded cable.
-----------------------------------------------------------------------------------------------------------------
Using Extended ping Tests to Troubleshoot Serial
Lines
The ping function is one of the useful tests available on Cisco
internetworking systems (as well as on many host systems). In TCP/IP terminology,
this diagnostic tool also is known as the "Internet Control Message Protocol
(ICMP) Echo Request."
Note The ping function is particularly useful
when high levels of input errors are being registered in the show interfaces
serial number display (see Figure
3-1).
Cisco internetworking systems provide a mechanism to automate the sending
of many ping packets in sequence. Figure
3-5 illustrates the menu used to specify extended ping options. This
example specifies only 20 successive pings; however, when testing the components
on your serial line, you should specify a much larger number, such as 1000
pings.
Figure 3-5 Extended ping Specification Menu
In general, perform serial line ping tests as follows:
-
Step 1 Put CSU or DSU into local loopback mode.
-
Step 2 Configure the extended ping command to send
different data patterns and packet sizes. Figure
3-6 and Figure
3-7 illustrate two useful ping tests, an all-zeros 1500 byte
ping and an all-ones 1500 byte ping, respectively.
Figure 3-6 All-Zeros 1500 Byte ping Test
Figure 3-7 All-Ones 1500 Byte ping Test
-
Step 3 Examine the show interfaces serial number
statistics and determine whether input errors have increased. If input
errors have not increased, the local hardware (DSU, cable, router interface
card, and applique) is likely to be good.
Assuming that this test sequence was prompted by the appearance of
a large number of CRC and framing errors, a clocking problem is likely.
Check the CSU or DSU for a timing problem. Refer to the section "Troubleshooting
Clocking Problems," later in this chapter.
-
Step 4 If you determine that the clocking configuration
is correct and operating properly, put the CSU or DSU into remote loopback
mode.
-
Step 5 Repeat the ping test and look for changes in the
input error statistics.
-
Step 6 If input errors increase, there is either a problem
in the serial line or on the CSU/DSU. Contact the WAN service provider
and swap the CSU or DSU. If problems persist, consult your router technical
support representative.
Adjusting Buffers to Ease Overutilized Serial
Links
Excessively high bandwidth utilization results in reduced overall performance
and can cause intermittent failures. For example, DECnet file transmissions
may be failing due to packets being dropped somewhere in the network. If
the situation is bad enough, you must add bandwidth; however, adding
bandwidth may not be necessary or immediately practical. One way to resolve
marginal serial line overutilization problems is to control how the router
uses data buffers.
Caution In general, you should not adjust system
buffers unless you are working closely with your router technical support
representative. You can severely affect the performance of your hardware
and your network if you incorrectly adjust the system buffers on your router.
You have three options to control how buffers are used:
-
Adjust parameters associated with system buffers
-
Specify the number of packets held in input or output queues (called "hold
queues")
-
Prioritize how traffic is queued for transmission (also called "priority
output queuing")
The configuration commands associated with these options are fully described
in the Router Products Configuration Guide and Router Products
Command Reference publications.
The following discussion focuses on identifying situations in which
these options are likely to apply and defining how you can use these options
to help resolve connectivity and performance problems in serial/WAN interconnections.
Commands are discussed as appropriate.
Tuning System Buffers
There are two general buffer types on Cisco routers. These are referred
to as "hardware" buffers and "system" buffers. Only the system buffers
are directly configurable by system administrators.
The hardware buffers are specifically used as the receive and transmit
buffers associated with each interface and (in the absence of any special
configuration) are dynamically managed by the system software itself.
The system buffers are associated with the main system memory and are
allocated to different size memory blocks. A useful command for determining
the status of your system buffers is the show buffers EXEC command.
Figure 3-8
shows an example of the output from the show buffers command.
Figure 3-8 show buffers Command Output
The show buffers command output in Figure
3-8 indicates high numbers in the trims and created fields for Large
Buffers. If this is the case, you can increase your serial link performance
by increasing the max-free value configured for your system buffers. Use
the buffers max-free number global configuration command to increase
the number of free system buffers. The value you configure should be approximately
150 percent of the figure indicated in the Total field of the show buffers
command output. Repeat this process until the show buffers output
no longer indicates trims and created buffers.
If the show buffers command output shows a large number of failures
in the "(no memory)" field (see the last line of output in Figure
3-8), you must reduce the usage of the system buffers or increase the
amount of shared or main memory (physical RAM) on the router. Call your
router technical support representative for assistance.
Implementing Hold Queue Limits
Hold queues are buffers used by each router interface to store outgoing
or incoming packets. Use the hold-queue interface configuration
command to increase the number of data packets queued before the router
will drop packets.
Note The hold-queue command is used for process
switched packets and periodic updates generated by the router.
Use this command to prevent packets from being dropped and to improve serial-link
performance under the following conditions:
-
You have an application that cannot tolerate drops and the protocol is
able to stand longer delays. DECnet is an example of a protocol that meets
both criteria. LAT does not because it does not tolerate delays.
-
The interface is very slow. (Low bandwidth and/or anticipated utilization
is likely to sporadically exceed available bandwidth.)
Note When you increase the number specified for an
output hold queue, you might need to increase the number of system buffers.
The value used depends on the size of the packets associated with the traffic
anticipated for the network.
Using Priority Queuing to Reduce Bottlenecks
Priority queuing is a list-based control mechanism that allows network
administrators to prioritize traffic transmitted into networks on an interface-by-interface
basis. In a manner that is analogous to Cisco's access list traffic control
mechanisms, priority queuing involves two steps:
-
Step 1 Create a priority list by protocol type and level
of priority.
-
Step 2 Assign the priority list to a specific interface.
Both of these steps use versions of the priority-list global configuration
command (with the keywords protocol and interface, as appropriate).
In addition, further traffic control can be applied by referencing access-list
global configuration commands from priority-list specifications.
For examples of defining priority lists and details about command syntax
associated with priority queuing, refer to the Router Products Configuration
Guide and Router Products Command Reference publications.
Note Priority queuing automatically creates four hold
queues of varying size. This overrides any hold queue specification included
in your configuration.
Use priority queuing to prevent packets from being dropped and to improve
serial link performance under the following conditions:
-
When the interface is slow, there are a variety of traffic types being
transmitted, and you want to improve terminal traffic performance.
-
If you have a serial link that is intermittently experiencing very heavy
loads (such as file transfers occurring at specific times), you can use
priority lists to select which types of traffic should be discarded at
high traffic periods.
In general, start with the default number of queues (altered with the queue-limit
keyword option of the priority-list global configuration command)
when implementing priority queues. After enabling priority queuing, monitor
output drops with the show interfaces serial number EXEC command.
If you notice that output drops are occurring in the traffic queue you
have specified to be high priority, increase the number of packets that
can be queued.
Note When bridging DEC LAT traffic, your router must
drop very few packets, or LAT will not function correctly (that is, sessions
will terminate unexpectedly). A high priority queue depth of about 100
(specified with the queue-limit keyword) is a typical working value
when your router is dropping output packets, and the serial lines are subjected
to about 50 percent bandwidth utilization. If the router is dropping packets
and is at 100 percent utilization, you need another line. Another tool
to relieve congestion when bridging DEC LAT is LAT compression. You can
implement LAT compression with the interface configuration command bridge-group
group lat-compression.
Special Serial Line Tests
In addition to the basic diagnostic capabilities provided with routers,
there are a variety of supplemental tools and techniques that can be used
to determine the conditions of cables, switching gear, modems, hosts, and
remote internetworking hardware. Although complete discussions of these
tools are beyond the scope of this publication, some hints about using
these alternative tools are provided here. For more information, consult
the documentation for your CSU, DSU, serial analyzer, or other equipment.
CSU and DSU Loopback Tests
If the output of the show interfaces serial number EXEC command
indicates that the serial line is up, but the line protocol is down, use
the CSU/DSU loopback tests to determine the source of the problem. Perform
the local loop test first, then the remote test. Figure
3-9 illustrates the topology of the CSU/DSU local and remote loopback
tests.
Figure 3-9 CSU/DSU Local and Remote Loopback
Tests
Note These tests are generic in nature and assume attachment
of the internetworking system to a CSU or DSU. However, the test is essentially
the same for attachment to a multiplexer with built-in CSU/DSU functionality.
Because there is no concept of a loopback in X.25 or Frame Relay packet-switched
network (PSN) environments, loopback tests do not apply to X.25 and Frame
Relay networks.
CSU and DSU Local Loopback Tests for HDLC or PPP Links
The following is a general procedure for performing loopback tests in conjunction
with built-in Cisco system diagnostic capabilities.
-
Step 1 Place the CSU/DSU in local loop mode. In local loop
mode, the use of the line clock (from the T1 service) is terminated, and
the DSU is forced to use the local clock.
-
Step 2 Use the show interfaces serial number EXEC
command to determine whether the line status changes from "line protocol
is down" to "line protocol is up (looped)," or if it remains down.
-
Step 3 If the line protocol comes up when the CSU or DSU
is in local loopback mode, it suggests that the problem is occurring on
the remote end of the serial connection. If the status line does not change
state, there is a possible problem in the router, connecting cable, or
CSU/DSU.
-
Step 4 If the problem appears to be local, issue the debug
serial interface privileged EXEC command.
-
Step 5 Take the CSU/DSU out of local loop mode. With the
line protocol down and the debug serial interface command
enabled, the debug serial interface output will indicate that keepalive
counters are not incrementing.
-
Step 6 Again place the CSU/DSU in local loop mode. This
should cause the keepalive packets to begin to increment. Specifically,
the values for mineseen and yourseen keepalives will increment
every 10 seconds. This information will appear in the debug serial interface
output. If the keepalives do not increment, there may be a timing problem
on the interface card or on the network. (For information on correcting
timing problems, refer to the section "Troubleshooting
Clocking Problems," earlier in this chapter.)
-
Step 7 Check the local router and CSU/DSU hardware, and
any attached cables. Make certain the cables are within the recommended
lengths (no more than 50 feet [15.24 meters], or 25 feet [7.62 meters]
for T1 link). Make certain the cables are attached to the proper ports.
Swap faulty equipment as necessary.
Figure 3-10
shows the output from the debug serial interface command for an
HDLC serial connection, with missed keepalives eventually causing the line
to go down and the interface to reset.
Figure 3-10 debug serial interface Command
Output
CSU and DSU Remote Loopback Tests for HDLC or PPP Links
If you are able to determine that the local hardware is functioning properly,
but you still encounter problems when attempting to establish connections
over the serial link, try using the remote loopback test that follows to
isolate the problem cause.
Note This remote loopback test assumes that HDLC encapsulation
is being used and that the preceding local loop test was performed immediately
before this test.
-
Step 1 Put the remote CSU or DSU into remote loopback.
-
Step 2 Using the show interfaces serial number EXEC
command, determine whether the line protocol remains up, with the status
line indicating "Serial x is up, line protocol is up (looped)" or
if it goes down, with the status line indicating "Line protocol is down."
-
Step 3 If the line protocol remains up (looped), the problem
is probably at the remote end of the serial connection (between the remote
CSU/DSU and the remote router). Perform both local and remote tests at
the remote end to isolate the problem source.
-
Step 4 If the line status changes to "Line protocol is
down" when remote loopback mode is activated, make certain that ones density
is being properly maintained. The CSU/DSU must be configured to use the
same framing and coding schemes (for example, ESF and B8ZS) used by the
leased-line or other carrier service.
-
Step 5 If problems persist, contact your WAN network manager
or the WAN service organization.
Troubleshooting Access Server to Modem Connectivity
This section offers recommended procedures for properly setting up an access
server-to-modem connection, and presents a number of symptom modules that
describe access server-to-modem connectivity problems and suggested actions
for resolving them. This section does not cover hardware problems. For
information on troubleshooting your hardware, see the "Troubleshooting
Router Startup Problems" chapter. See the "Troubleshooting
AppleTalk Connectivity" chapter for modem troubleshooting information
that is directly related to AppleTalk Remote Access (ARA) dial-in sessions.
The first part of this section, "Initiating
a Reverse Telnet Session to a Modem," describes the procedure for establishing
a reverse Telnet session with your modem in order to set the proper speed
and configure it at that speed. The rest of the section includes the following
troubleshooting symptom modules:
Initiating a Reverse Telnet Session to a Modem
Establishing a reverse Telnet session with your modem allows you to configure
the modem at the speed at which you want it to communicate with the Cisco
device. As long as you lock the DTE-side speed of the modem (see Table
3-6 for information on locking the modem speed), the modem will always
speak to the access server or router at the desired speed. Be certain that
the speed of the Cisco device is configured prior to issuing commands to
the modem via a reverse Telnet session. (See Table
3-6 for information on configuring the speed of the access server or
router.)
To initiate a reverse Telnet session to your modem, perform the following
steps:
-
Step 1 From your terminal, issue the command
telnet x.x.x.x 20yy
where x.x.x.x is the IP address of any active, connected interface
on the Cisco device that is currently up, and yy is the line number
to which the modem is connected. For example, the following command
telnet 192.169.53.52 2001 would connect you to the auxiliary port
on a Cisco router with IP address 192.169.53.52. A Telnet command of this
kind can generally be issued from anywhere on the network that can ping
IP address x.x.x.x.
Note On a Cisco router, port 01 is the auxiliary
port. On a Cisco access server, the auxiliary port is last_tty+1,
so on a 16-port access server, the auxiliary port is port 17. Use the show
line EXEC command to make certain you are working with the correct
line.
-
Step 2 If the connection is refused, there may already
be a user connected to that port. Issue the show users EXEC command
to determine if the line is being used. If desired, the line can be cleared
from the console using the clear line privileged EXEC command. When
you are certain the line is not in use, attempt the Telnet connection again.
-
Step 3 If the connection is again refused, confirm that
you have set modem control to modem inout for that line. See Table
3-4 for information on configuring a line on a Cisco device for modem
control.
-
Step 4 After successfully making the Telnet connection,
you are ready to configure the modem. Make sure that when you enter AT,
the modem replies with OK. Figure
3-11 shows a typical Hayes-compatible modem command string. Again,
be certain to check the documentation for your specific modem to verify
the exact syntax of these commands.
Figure 3-11 Typical Hayes-Compatible Modem Command
String
No Connectivity Between Access Server and Modem
Symptom: Connectivity between a modem and a Cisco access server
or router is nonexistent. Attempts to initiate a reverse Telnet session
to the modem have no result, or the user receives a "Connection Refused
by Foreign Host" message. Table
3-4 describes possible causes and suggests actions when modem to access
server connections are unresponsive.
Table 3-4 Modem: No Connectivity Between Access
Server and Modem
-----------------------------------------------------------------------------------------------------------------------
Possible Causes Suggested Actions
-----------------------------------------------------------------------------------------------------------------------
Modem control is not enabled on the access Step 1 Issue the show line EXEC command on the access server
server (modem control on auxiliary ports is only or router. The output for the auxiliary port should show
available in Software Release 9.21 and later). inout or RIisCD in the Modem column. This indicates
that modem control is enabled on the line of the access
server or router. For an explanation of the show line
output, see the "Interpreting show line Output" section
later in this chapter.
Step 2 If you are running software prior to Software
Release 9.21, and therefore do not have modem control,
perform these steps and do not proceed to Step 3:
· Disable echo on the modem. This is typically done
with the E0 modem command. Check your modem
documentation for the exact syntax of modem
commands.
· Disable result codes on the modem. This is typically
done using the Q1 modem command. Check your
modem documentation for the exact syntax. See
Figure 3-12 for a modem command string that disables
echo and result codes on a Hayes-compatible modem.
· On the access server or router, configure the line to
which the modem is connected with the exec timeout
line configuration command. This command tells the
access server to end the EXEC session after a specified
period of time of no activity.
Step 3 If you are running Software Release 9.21 or later,
configure the line for modem control using the
modem inout line configuration command. Modem
control is now enabled on the access server. The debug
modem output should indicate the change.
NOTE: Be certain to use the modem inout command in
favor of the modem ri-is-cd command while the
connectivity of the modem is in question. The latter
command allows the line to accept incoming calls only.
Outgoing calls will be refused, making it impossible to
establish a Telnet session with the modem to configure it.
If you want to enable the modem ri-is-cd command, do
so only after you are certain the modem is functioning
correctly.
Incorrect cabling configuration Step 1 Check the cabling between the modem and the access
server or router. Confirm that the modem is connected to
the auxiliary port on the access server or router with a
rolled RJ-45 cable and an MMOD DB-25 adapter. This
cabling configuration is recommended and supported by
Cisco for RJ-45 ports.
Step 2 If you are using a rolled RJ-45 cable with an MDCE
DB-25 adapter, or a straight RJ-45 cable with an MDTE
DB-25 adapter, you must dismantle the connector on the
EIA/TIA-232 side and move pin 6 to pin 8. This turns the
MDCE or MDTE adapter into an MMOD adapter by
wiring the DCD output of the modem to the DSR input of
the access server or router.
Step 3 Use the show line line-number EXEC command to
verify that the cabling is correct. See the explanation of
the show line command output in the section
"Interpreting show line Output," following.
Hardware problem Step 1 Verify that you are using the correct cabling and that all
connections are good.
Step 2 Check all hardware for damage, including cabling
(broken wire), adapters (loose pin), access server ports,
and modem.
Step 3 See the "Troubleshooting Router Startup Problems"
chapter for more information on hardware
troubleshooting.
-----------------------------------------------------------------------------------------------------------------------
Figure 3-12 Hayes-Compatible Modem Command String
for Pre-Modem Control Software
Interpreting show line Output
The output from the show line line-number EXEC command is useful
when troubleshooting a modem-to-access server or router connection. Figure
3-13 shows the output from the show line line-number command.
Important fields and their meanings are noted following.
Figure 3-13 show line Command Output
When connectivity problems occur, important output appears in the Modem
State and the Modem Hardware State fields.
Note The Modem Hardware State field does not appear
in the show line line-number output for every platform. In certain
cases, the indications for signal states will be shown in the Modem State
field instead.
Table 3-5
shows typical Modem State and Modem Hardware State strings from the output
of the show line line-number command and explains the meaning of
each state.
Table 3-5 Modem and Modem Hardware States
in show line Output
---------------------------------------------------------------------------------------------------
Modem State Modem Hardware State Meaning
---------------------------------------------------------------------------------------------------
Idle CTS noDSR DTR RTS These are the proper modem states for connections between
an access server or router and a modem. Output of any other
kind generally indicates a problem.
Ready -- If the Modem State is Ready instead of Idle, there are three
possibilities:
1 Modem control is not configured on the access server or
router. Configure the access server or router with the
modem inout line configuration command.
2 A session exists on the line. Issue the show users EXEC
command and use the clear line privileged EXEC
command to kill the session if desired.
3 DSR is high. There are two possible reasons for this:
-- Cabling problems---The DSR signal from the modem is
connected to DSR from the access server. The proper
signalling is DCD (modem) to DSR (access server).
Check the cabling configuration as described in
Table 3-4.
-- Modem configured for DCD always high---The modem
should be reconfigured to have DCD high only on
carrier detect (CD). This is usually done with the &C1
modem command (see Figure 3-11), but check your
modem documentation for the exact syntax for your
modem.
You might have to configure the access server line to
which the modem is connected with the no exec line
configuration command. Clear the line with the clear
line privileged EXEC command, initiate a reverse
Telnet session with the modem, and reconfigure the
modem so that DCD is high only on CD.
End the Telnet session by entering disconnect and
reconfigure the access server line with the exec line
configuration command.
Ready noCTS noDSR DTR RTS There are four possibilities for the noCTS string appearing in
the Modem Hardware State field:
1 Modem is turned off.
2 Modem is not connected to the access server properly.
Check the cabling connections from the modem to the
access server.
3 Incorrect cabling configuration (either rolled MDCE or
straight MDTE, but without the pins moved). See
Table 3-4 for information on the recommended cabling
configuration.
4 Modem is not configured for hardware flow control.
Disable hardware flow control on the access server with
the no flowcontrol hardware line configuration
command. Enable hardware flow control on the modem
via a Reverse Telnet session. (Consult your modem
documentation and see the section "Initiating a Reverse
Telnet Session to a Modem," earlier in this chapter.)
Reenable hardware flow control on the access server with
the flowcontrol hardware line configuration command.
Ready CTS DSR DTR RTS There are two possibilities for the presence of the DSR string
instead of the noDSR string in the Modem Hardware State
field:
1 Incorrect cabling configuration (either rolled MDCE or
straight MDTE, but without the pins moved). See
Table 3-4 for information on the recommended cabling
configuration.
2 The modem is configured for DCD always high.
Reconfigure the modem so that DCD is only high on CD.
This is usually done with the &C1 modem command (see
Figure 3-11), but check your modem documentation for
the exact syntax for your modem.
Configure the access server line to which the modem is
connected with the no exec line configuration command.
Clear the line with the clear line privileged EXEC
command, initiate a reverse Telnet session with the
modem, and reconfigure the modem so that DCD is high
only on CD.
End the Telnet session by entering disconnect.
Reconfigure the access server line with the exec line
configuration command.
Ready CTS* DSR* DTR RTS If this string appears in the Modem Hardware State field, it is
likely that modem control is not enabled on the access server.
Use the modem inout line configuration command to enable
modem control on the line.
See Table 3-4 for more information on configuring modem
control on an access server or router line.
---------------------------------------------------------------------------------------------------
Remote Dial-In Sees "Garbage"
Symptom: Attempts to establish remote dial-in sessions over a modem
to a Cisco access server or router return "garbage" and ultimately result
in no connection to the remote site. User might see a "Connection Closed
by Foreign Host" message. Table
3-6 describes possible causes and suggests actions for remote dial-in
sessions seeing "garbage."
Table 3-6 Modem: Remote Dial-In Sessions Seeing
"Garbage"
--------------------------------------------------------------------------------------------------------
Possible Causes Suggested Actions
--------------------------------------------------------------------------------------------------------
Modem speed setting is not locked. Step 1 Issue the show line EXEC command on the access
server or router. The output for the auxiliary port should
indicate the currently configured transmit (Tx) and
receive (Rx) speeds. For an explanation of the output
from the show line command, see the "Interpreting
show line Output" section earlier in this chapter.
Step 2 If the line speed is not configured to the speed you
desire, you must reconfigure the line. Use the speed line
configuration command to set the line speed on the
access server or router line. Set the value to the highest
speed in common between the modem and the access
server or router port.
NOTE: If for some reason you cannot use flow control,
limit the line speed to 9600 bps. Faster speeds are likely
to result in lost data.
Step 3 Issue the show line EXEC command again and confirm
that the line speed is set to the desired value.
Step 4 When you are certain that the access server or router line
is configured for the desired speed, initiate a reverse
Telnet session to the modem via that line. For more
information, see the section "Initiating a Reverse Telnet
Session to a Modem."
Step 5 Issue a modem command string that includes the lock
DTE speed command for your modem. See your modem
documentation for exact configuration command syntax.
NOTE: The lock DTE speed command, which might
also be referred to as port rate adjust or buffered mode,
is often related to the way in which the modem handles
error correction. This command varies widely between
modems.
Locking the modem speed ensures that the modem
always communicates with the Cisco access server or
router at the speed configured on the Cisco auxiliary
port. If this command is not used, the modem will revert
to the speed of the data link (the telephone line) instead
of communicating at the speed configured on the access
server.
--------------------------------------------------------------------------------------------------------
High Rate of Data Loss Over Modem Connection
Symptom: Remote sessions over a modem connection experience a high
rate of data loss. Table
3-7 shows possible causes and suggests actions when there is a high
rate of data loss over a modem connection.
Table 3-7 Modem: High Rate of Data Loss Over
Modem Connection
---------------------------------------------------------------------------------------------------------------------
Possible Causes Suggested Actions
---------------------------------------------------------------------------------------------------------------------
Error correction is not configured on the modem. Step 1 Make certain the modem is configured for error
correction. For the exact syntax of the command, see
your modem documentation.
Flow control is not enabled, is enabled only on one Step 1 Display detailed information about the auxiliary line
device (either DTE or DCE), or is misconfigured. using the show line aux-line-number EXEC
command.
In the Capabilities field (see Figure 3-13), look for the
following:
Capabilities: Hardware Flowcontrol In,
Hardware Flowcontrol Out...
If there is no mention of hardware flow control in this
field, hardware flow control is not enabled on the line.
Cisco recommends hardware flow control for access
server-to-modem connections. For an explanation of
the output from the show line command, see the
"Interpreting show line Output" section earlier in this
chapter.
Step 2 Configure hardware flow control on the line using the
flowcontrol hardware line configuration command.
NOTE: If for some reason you cannot use flow control,
limit the line speed to 9600 bps. Faster speeds are
likely to result in lost data.
Step 3 After enabling hardware flow control on the access
server or router line, initiate a reverse Telnet session to
the modem via that line. For more information, see the
section "Initiating a Reverse Telnet Session to a
Modem."
Step 4 Issue a modem command string that includes the
RTS/CTS Flow command for your modem. This
command ensures that the modem is using the same
method of flow control (that is, hardware flow control)
as the Cisco access server or router. See your modem
documentation for exact configuration command
syntax. Figure 3-11 shows the hardware flow control
command string for a Hayes-compatible modem.
---------------------------------------------------------------------------------------------------------------------
Modem Does Not Disconnect Properly
Symptom: Modem does not disconnect properly. Connection to modem
does not terminate when quit command is entered. Table
3-8 describes possible causes and suggests actions for a modem that
does not disconnect properly.
Table 3-8 Modem: Modem Not Disconnecting Properly
------------------------------------------------------------------------------------------------------------------
Possible Causes Suggested Actions
------------------------------------------------------------------------------------------------------------------
Modem is not sensing DTR. Step 1 Enter the Hangup DTR modem command string. This command
tells the modem to drop carrier when the DTR signal is no longer
being received. On a Hayes-compatible modem the &D3 string
is commonly used, as shown in Figure 3-11. For the exact syntax
of this command, see the documentation for your modem.
Modem control is not configured on the Step 1 See Table 3-4 for instructions on configuring modem control on
router or access server (modem control a router or access server port.
on auxiliary ports is only available in
Software Release 9.21 and later).
------------------------------------------------------------------------------------------------------------------
Remote Dial-In Client Receives No EXEC Prompt
Symptom: Remote dial-in client opens a session and appears to be
connected, but the user does not receive an EXEC prompt (for example, a
Username or Router> prompt). Table
3-9 describes possible causes and suggests actions for a remote dial-in
client that is not receiving an EXEC prompt.
Table 3-9 Modem: Remote Dial-In Client Is
Not Receiving an EXEC Prompt
----------------------------------------------------------------------------------------------------------------
Possible Causes Suggested Actions
----------------------------------------------------------------------------------------------------------------
Autoselect is enabled on the line. Step 1 Attempt to access EXEC mode by issuing a carriage
return.
Line is configured with the no exec command. Step 1 Use the show line line-number EXEC command to view
the status of the appropriate line.
Check the Capabilities field to see if it says "EXEC
suppressed." If this is the case, the no exec line
configuration command is enabled.
Step 2 Configure the exec line configuration command on the
line to allow EXEC sessions to be initiated.
Flow control is not enabled, is enabled only on Step 1 For information on configuring flow control, see
one device (either DTE or DCE), or is Table 3-7.
misconfigured.
Modem speed setting is not locked. Step 1 For information on setting the speed of your access
server or modem, see Table 3-6.
----------------------------------------------------------------------------------------------------------------
Remote Dial-In Interrupts Existing Sessions
Symptom: Remote dial-in session interrupts an already existing session
initiated by another user. Table
3-10 describes possible causes and suggests actions for remote dial-in
sessions interrupting existing sessions.
Table 3-10 Modem: Remote Dial-In Interrupts
Existing Sessions
---------------------------------------------------------------------------------------------------------------------
Possible Causes Suggested Actions
---------------------------------------------------------------------------------------------------------------------
Modem configured for DCD always high. Step 1 The modem should be reconfigured to have DCD high only
on carrier detect (CD). This is usually done with the &C1
modem command string (see Figure 3-11), but check your
modem documentation for the exact syntax for your modem.
Step 2 You might have to configure the access server line to which
the modem is connected with the no exec line configuration
command. Clear the line with the clear line privileged
EXEC command, initiate a reverse Telnet session with the
modem, and reconfigure the modem so that DCD is high
only on CD.
Step 3 End the Telnet session by entering disconnect and
reconfigure the access server line with the exec line
configuration command.
Modem control is not configured on the Step 1 See Table 3-4 for instructions on configuring modem control
router or access server (modem control on on a router or access server port.
auxiliary ports is only available in Software
Release 9.21 and later).
Incorrect cabling configuration Step 1 See Table 3-4 for information on the recommended cabling
configuration.
---------------------------------------------------------------------------------------------------------------------
Copyright 1988-1995
© Cisco Systems Inc.
