Advanced T1 Troubleshooting

READ THIS FIRST

*** WARNING ***

This article directs you to enter configuration mode in order to execute the commands. When performing these steps, session logging should be turned on to record commands entered, and extra care should be taken to verify the commands are being entered properly. Failure to do so could result in potentially critical information loss, up to and including loss of all device configuration.

*** WARNING ***

Looping a circuit from remote equipment will cause connection to be lost unless an alternate means of connection is available. If connection is lost and there is no alternate means of connecting, you will be unable to remove the loop without physical access to the device.

*** WARNING ***

*** CAUTION ***

Looping a circuit will cause loss of connectivity for that connection. Ensure that intrusive testing is authorized and the connection is not in use.

*** NOTE ***

This knowledge base article assumes that you have already checked the circuit status as outlined in the Troubleshooting T1/DS3 Circuits knowledge base article.

If a loop already exists on the interface you are going to be working with, you will be unable to loop it again. This loop must be cleared before continuing. If you are unable to clear the loop, it is likely that the carrier has placed a loop on the circuit within their equipment. You will need to engage them to clear the loop before continuing.
Applying or removing a loop on a circuit generates a small amount of errors on the interface and controller. Make sure to note error counts prior to applying a loop and after clearing it.

Extended Ping Testing

Cisco networking devices provide a mechanism to automate the sending of many ping packets in sequence. The code below illustrates the menu used to specify extended ping options. This example specifies 100 successive pings. However, when testing the components on your serial line, you should specify a much larger number, such as 1000 pings. Also increase the datagram size to a larger number, such as 1500.

A Repeat count of 10000 will take approximately 15 minutes on core equipment, which will fill a single interval for T1/T3 controller statistics.

Running the test

*** NOTE ***

If a serial interface is part of a bonded multilink, any other serial interfaces that are part of the multilink will need to be looped as well. In this instance, interface error counts should be recorded for all involved interfaces.
Encapsulation must be set to High-Link Data Control (HDLC) in order for traffic to pass through the loop.

Verify encapsulation type using the show interface [interface name] command:

Router#show interface serial1/0/1:0

Serial1/0/1:0 is down, line protocol is down

Hardware is PA-MC-T3

MTU 1500 bytes, BW 1536 Kbit, DLY 20000 usec,

reliability 255/255, txload 1/255, rxload 1/255

Encapsulation HDLC, crc 16, loopback not set

If encapsulation is anything other than HDLC, reconfigure interface to use HDLC encapsulation.

Make sure to annotate current encapsulation setting prior to changing, as it will need to be reconfigured to the correct value once testing is completed.

config t

interface serial 1/0/1:0

no encapsulation

encapsulation hdlc

Ctrl-Z

Loop circuit as outlined in Loop a T1/T3 Circuit

Clear interface counters

Note interface counters and controller statistics.

Send ping pattern

Repeat as needed

Clear loop

Clear counters

Example Ping Test

Router#ping

Protocol [ip]:

Target IP address: 10.10.1.5                               <--- Target IP

Repeat count [5]: 100                                      <--- Ping count specification

Datagram size [100]: 1500

Timeout in seconds [2]:

Extended commands [n]: y                                   <--- Extended commands selected option

Source address or interface:

Type of service [0]:

Set DF bit in IP header? [no]:

Validate reply data? [no]:

Data pattern [0xABCD]: 0xffff                              <--- Data pattern specification

Loose, Strict, Record, Timestamp, Verbose[none]:

Sweep range of sizes [n]:

Type escape sequence to abort.

Sending 100, 1500-byte ICMP Echos to 10.10.1.5, timeout is 2 seconds:

Packet has data pattern 0xFFFF

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

Success rate is 100 percent (100/100), round-trip min/avg/max = 20/22/24 ms

Data Sequence Types

0x0000 - all zeroes
0xFFFF - all ones
0x5555 - alternating zeros and ones
0x4040 - one with seven zeros (1 in 7)
0x8080 - one with seven zeros (1 in 7)
0x1000 - one with 15 zeros
0x1111 - one after every 8 bits (tests B8ZS vs AMI)
0x4242 - one followed by 4 consecutive zeros (2 in 8)

All Zeros

0000 0000 0000 0000

This pattern is composed of only zeros and must be encoded with either B8ZS or ZBTSI zero suppression techniques before transmission into the network. The pattern is effective in finding equipment mis-optioned for AMI. This is the same as the 0x0000 Cisco extended ping test (with out header and footer data).

All Ones

1111 1111 1111 1111

This pattern is composed of only ones. This pattern causes the repeater to consume the maximum amount of power. If the DC baseline current is properly regulated, then the repeater will have no trouble transmitting long ones sequences. In some networks, an unframed pattern of all ones is an Alarm Indicating Signal (AIS) (blue alarm or an all ones condition).

This is the same as the 0xFFFF Cisco extended ping test (without header and footer data).

Zero-One

0101 0101 0101 0101

The alternating zeros and ones pattern is used to test the regeneration circuitry of the repeaters on the circuit. If the repeaters are in spec and properly calibrated then they will be able to quickly alternate from +3 volts to the baseline, to -3 volts, to the baseline, and so on.

This is the same as the 0x5555 Cisco extended ping test (without header and footer data).

Quasi-Random

Pseudo-random sequence based on a 20-bit shifting register. Generates every combination of 20-bit words, repeats every 1,048,575 bits, and suppresses consecutive zeros to no more than 14. Contains high density sequences, low density sequences, and sequences that change from low density to high density and vice versa. This pattern is preferred by voice techs since it most closely matches voice conversations on a digital circuit. In the past, when voice dominated services provided by the telephone company, this was fine. However, data streams stress the circuit beyond traditional voice conversations, thus additional, and more stressful patterns must be run to complete a proper test.

3 in 24

0100100 00000000 00000100

Pattern contains the longest string of consecutive zeros (15), with the lowest ones density (12.5%). A framed 3 in 24 pattern could generate a yellow alarm on circuits using D4 framing. This is dependent on the alignment of ones bits to the frame.

1 in 7

0100 0000 0100 0000

Only a single one in an 8-bit repeating sequence. This is the similar to the Cisco extended 0x8080 test (with out header and footer data) except that the one shows up in the first most significant bit on the Cisco extended test, instead of the second most significant bit on the BERT test. A closer test would be 0x4040. This would be exactly the same except for the header and footer information introduced by TCP/IP and frame relay encoding.

The framed version of this pattern may cause a D4 yellow alarm for framed circuits depending on alignment of one bits to frames. One bits are strategically aligned with frame bits to avoid D4 yellow alarm.

2 in 8

0100 0010 0100 0010 0100 0010

Pattern contains a maximum of 4 consecutive zeros. The framed version of 2 in 8 will not invoke B8ZS sequence because eight consecutive zeros are required to cause B8ZS substitution. May cause D4 yellow alarm for framed circuits depending on alignment of the one bits to frame. One bits are strategically aligned with frame bits to avoid D4 yellow alarm. Pattern is effective in finding equipment mis-optioned for B8ZS.

This is the same as the 0x4242 Cisco extended ping test (with out header and footer data).

55 Octet

55 Octet Pattern contains rapid sequence changes from low density to high density. Pattern can only be used unframed to comply with ones density specifications.

Bit Error Rate Tests (BERT)

Words here

Hard/Physical Loopback

Building a Loopback Plug/Cable

Words

Multiplexer Loopback

Words

Reference

Extended Ping Testing

http://ipsn6tap.blogspot.com/2009/04/extended-ping-test-pattern-poor-mans.html

http://www.cisco.com/en/US/docs/internetworking/troubleshooting/guide/tr1915.html

Page updated

Google Sites

Report abuse