Cable Modem Troubleshooting Tips

Cable Modem diagnostics and signal levels:

• Signal levels: a caution  New 2002 Sep 7 
• Upstream and Downstream channels  New 2002 Sep 7 
• Symptoms of poor signal levels: what to expect  New 2002 Sep 7 
• Cable modem spontaneously re-booting: other causes
• Desirable signal levels: for acceptable service  New 2002 Sep 7 
• Advanced diagnostics: for the techie-minded  New 2002 Sep 7 
• How to discover your cable modem signal levels  New 2002 Sep 7 
• Inability to connect to cable modem diagnostic address: what to check
• DocsDiag diagnostics: typical report  New 2002 Sep 7 
• Surfboard diagnostics: how to read  New 2002 Sep 7 
• Surfboard event log: what's important and what's not
• 3Com HomeConnect (Tailfin) diagnostics: how to read  New 2002 Sep 7 
• 3Com HomeConnect (Tailfin) event log: how to read  New 2002 Sep 7 
• 3Com CMX diagnostics: how to read
• Terayon TeraJet diagnostics: how to read
• ntl:home 100/120 (Ambit) diagnostics: how to read
• Pace digital TV set-top-box cable modem diagnostics

Signal levels: a caution

So also with signal levels: they are a diagnostic aid, but you wouldn't request a visit from an ISP's service technician on the strength of a suspect signal level alone: there would need also to be service problems of the sort that might be caused by poor signal conditions.

Upstream and Downstream channels

Data communication between your cable modem and the ISP's network is split into two:

• The Downstream channel carries data from the ISP's head end (UBR) to your cable modem;
• The Upstream channel carries data from your cable modem to the ISP's head end (UBR).

These two channels are on very different frequencies on the cable, and each has their own signal level and noise level.

The downstream channel looks just like a normal digital TV channel, and occupies a standard TV channel slot in the frequency spectrum, between 91 to 857 MHz (DOCSIS) or between 112 to 858 MHz (EuroDOCSIS). The downstream channel broadcasts downstream data for every cable modem in the area served by the UBR. Cable modems discover the downstream channel by searching for a data-carrying digital TV channel when they boot up, or by remembering what frequency was used last time.

Downstream frequencies known to be in use are:

For every downstream channel, there might be up to six upstream channels on which cable modems send data to the UBR. These channels are normally allocated to frequencies below 42 MHz (DOCSIS) or below 65 MHz (EuroDOCSIS). Your cable modem discovers the upstream frequencies by listening on the downstream channel for Upstream Channel Descriptor (UCD) packets. The UBR's service area might be physically segmented so that not all upstream channels are available on all segments. The cable modem will choose (or be allocated to) one of the available upstream channels at your particular location.

Symptoms of poor signal levels

If cable signal levels become unacceptably poor, the LEDs on the cable modem will start indicating problems:

• 3Com CMX: Cable Modem Status goes out or flashes.
• 3Com HomeConnect (Tailfin): Link Status goes out or flashes.
• Surfboard: the Receive light goes out or flashes for downstream problems; the Send light goes out or flashes for upstream problems.
• Terayon TeraJet: the Cable light goes out or flashes.
• ntl:home (Ambit): the SYNC light goes out or flashes for downstream problems; the RDY light goes out or flashes for upstream problems.
• Pace digital TV set top boxes: the middle light goes out or flashes.

If the above occur, an immediate call to your ISP's technical support is justified. There is no point trying to get the PC working while the above conditions persist.

Alternatively, in conditions of poor signal levels, the cable modem might spontaneously reboot. Other possible causes of spontaneous reboots are given below.

When signal levels are poor but not poor enough to provoke the above symptoms, you might observe degraded throughput, or packet loss, or poor latency, or failure to acquire or renew a DHCP lease. But poor signal levels are not the only cause of these symptoms, so some diagnosis is required before you blame the signal levels.

Cable modem spontaneously re-booting

If the cable modem is spontaneously re-booting, and there is no evidence of signal-related problems, then you should check on electrical conditions near to the cable modem. Possible causes of these problems are:

• The cable modem itself is faulty.
• Mains wiring to the cable modem is faulty or intermittent.
• Mains spikes arising from e.g. vacuum cleaners, power tools, heating thermostats, refrigerator thermostats.
• Electromagnetic Interference (EMI), e.g. mobile phone, local radio transmitter.

There are plenty of reports of mobile phones disrupting cable modems if kept too close.

Desirable signal levels

For diagnostic purposes, the cable modem can measure and report the Downstream Received Power and the Downstream Signal-to-Noise Ratio (SNR). The SNR is important because if there is too much noise on the cable, the data cannot be decoded correctly, even with downstream power levels within acceptable limits. If the SNR is good enough so that the cable modem is working correctly, the exact power level, even if slightly outside normal ranges, does not matter.

Downstream SNR: should be 30 dB or higher: the higher the better. As the SNR decreases below 30 dB, performance will steadily decrease, and errors will increase. The cable modem might stop working properly if the SNR drops below 23.5 dB.

Downstream Received Power: the DOCSIS specification requires cable modems to function correctly with downstream power levels in the range -15 dBmV to +15 dBmV: power readings at or close to those extremes are likely to be unacceptable. However, when connected to a real-life CATV network, a cable modem might be functional over a narrower range than this. Accordingly, cable ISPs will specify an even narrower target range when commissioning their network: this target range will differ according to ISP.

Provided the Downstream SNR is acceptable, the downstream power should be satisfactory if your cable modem reports a figure either within these specified ranges or close to them.

If a cable modem reports a downstream receive power of exactly 0.0 dB, this means that measurement of downstream power is disabled, and no information is available. In this case, you will need to judge downstream quality by SNR alone.

Upstream Transmit Power: the cable modem's Upstream Transmit Power (or return power) is set by commands from the UBR so that the UBR hears the same signal level from all cable modems on that upstream channel. Because of different cable losses for each cable modem in the area, each cable modem will tend to settle onto a different upstream transmit power level. The better the return path is, the lower the upstream transmit power will be.

The upstream transmit power will lie within the range +8 to +58 dBmV, with many ISPs specifying a target commissioning level below +55 dBmV. Values in the forties are the most common. Many cable modems are unable to transmit any more powerfully than +58 dBmV. One cannot tell how far this is below the figure that the UBR would need to see a strong enough signal at its end to maintain satisfactory performance, so a figure as high as +58 dBmV is normally a sign of an unacceptable return path. If other problem symptoms are also present, an upstream transmit power of +58 dBmV would constitute valid supporting evidence for requesting technical support from a cable ISP.

The upstream SNR can be sensed only at the UBR, so cannot be discovered by end-users.

The external cable infrastructure is affected by weather conditions, so it is normal to see power levels fluctuating slowly with temperature. If your cable signal conditions are marginal, the CM might stop working in certain weather conditions, and recover in others.

However, rapidly fluctuating power levels might be a sign of a failing amplifier in the ISP's network, or a bad cable connection, and should normally be investigated.

Advanced diagnostics

Cable modems maintain counters of various error conditions, some of which can be more definite indicators of signal problems than the power and SNR measurements. Only a few models of cable modem report these counters in the diagnostic web pages available to end-users, but the advanced diagnostics can usually be read by the DocsDiag program (unless the ISP has denied the use of SNMP to the end-user), using its -v verbose keyword.

All the error counters are cumulative counts of errors since the last cable modem restart. Therefore, some non-zero counts might relate to short-lived problems which cured themselves long ago. What matters is whether the counters are increasing now. To sense this, you should take the advanced diagnostics twice, and look at the differences in the counters.

The downstream data transmitted from the UBR is split into codewords (usually between 16 and 256 bytes). Each codeword is packaged with extra Forward Error Correction (FEC) data bits, which enables the cable modem to reconstruct the original codeword if no more than a few bits of it are in error when received. The algorithm used for generating the FEC bits and correcting the received codeword is known as Reed-Solomon. Here is an extract from a verbose DocsDiag report:

System up time                       = 2 days 00h 17m 32.48s
SigQu: received without error        = 3451610121 codewords
SigQu: correctable errors            = 2560 codewords
SigQu: uncorrectable errors          = 0 codewords

and the same report 23 minutes later:

System up time                       = 2 days 00h 41m 08.63s
SigQu: received without error        = 3496202395 codewords
SigQu: correctable errors            = 2658 codewords
SigQu: uncorrectable errors          = 0 codewords

The received without error count is the number of codewords received on the downstream channel (whether destined for this cable modem or not) with no errors detected by the FEC system. This counter increases very rapidly, and exhausts the capacity of a 32-bit integer after only a few hours, after which it starts from zero again. It can however be safely used when taking differences between two successive readings a few minutes apart. In the above example, in 23 minutes, the CM has received 3496202395 - 3451610121 = 44592274 perfect codewords.

The correctable errors count is the number of codewords which were detected to be damaged but could be repaired by the FEC system. In a perfect world, this would be zero. In the above example, in 23 minutes, the CM has detected 2658 - 2560 = 98 faulty codewords which could be corrected by FEC. This low rate of error (roughly 2 per million) can be tolerated.

The uncorrectable errors count is the number of codewords which were detected to be damaged beyond possibility of repair by the FEC system. If this count is increasing, then downstream signal conditions are unacceptable, no matter how good the SNR and power figures look. In the above example, the current rate of uncorrectable codewords is zero, which is acceptable. Indeed, no uncorrectable errors have been seen in the last 2 days.

If the T3 time-out counter is incrementing, in the absence of any downstream codeword errors, this is an indicator of upstream problems.

How to discover your cable modem signal levels

Many models of cable modem provide diagnostic information on web pages which can be read at address http://192.168.100.1/. Details specific to certain models will be discussed below. The following models are known not to provide diagnostics on web pages:

• 3Com CMX (try DocsDiag instead)
• Terayon TeraJet (try DocsDiag instead)
• Pace digital TV set top boxes (view the User Information TV screen instead)

If you are unable to browse the above address, see the article Inability to connect to cable modem diagnostic address below.

For brands of cable modem which do not have diagnostics on web pages, see DocsDiag for reading DOCSIS diagnostics.

Inability to connect to cable modem diagnostic address

If you cannot connect to the diagnostic address 192.168.100.1 given above, check the following:

• if you have set up an explicit web proxy, ensure that you have included the cable modem address in the proxy exclusion list. See How to sidestep the transparent web caches for an example of setting an exclusion, or turn off web proxying for the duration.
• if you have just swapped computers on the cable modem, see Swapping computers on the cable modem.
• if you are using a NAT router between your PC and the cable modem, ensure that the router's own configuration will route data for 192.168.100.1 out of the WAN port towards the cable modem.
• For many models of cable modem (and for the Surfboard or Tailfin when they are not online to the cable network), you must temporarily manually configure your computer to have an IP address in the same sub-net as the cable modem diagnostic address. For instance, if your cable modem address is 192.168.100.1, you would configure your IP address to be 192.168.100.2. Remember to return your configuration to "automatic" DHCP working afterwards before you try going online to the network.

DocsDiag diagnostics

A typical default DocsDiag report looks like this:

General Instrument SB3100 Cable Modem: Hardware version: 1; OS: VxWorks 5.3.1; Software version: 3.2.9p

System up time                       = 2 days 04h 56m 22.63s
Ethernet multiple-collision errs.1   = 16
Downstream channel ID                = 3
Downstream channel frequency         = 402750000 Hz
Downstream received signal power     = 0.0 dBmV (or not supported)
Upstream channel ID                  = 2
Upstream channel frequency           = 40783269 Hz
QoS max upstream bandwidth           = 128000 bps
QoS max downstream bandwidth         = 600000 bps
SigQu: Signal to Noise Ratio         = 34.7 dB
Cable modem status                   = Registration complete
Upstream transmit signal power       = 55.0 dBmV
Date and Time                        = 2002-09-07,19:42:57.0+00:00
Configuration filename               = mota4100-silver.cm

To include the advanced diagnostics, use the keyword -v. To print the event log, use the keyword -log.

Surfboard diagnostics

Send your web browser to http://192.168.100.1/ and click the Signal tab, or go directly to http://192.168.100.1/signal.html. You will see a page similar to:

The power levels are highlit above. The downstream power level and SNR will be valid only if the downstream frequency is shown to be Locked. The upstream power level will be valid only if the upstream frequency is shown to be Ranged.

Surfboard event log

Surfboard users can view the cable modem event log by sending a web browser to http://192.168.100.1/logs.html or by clicking on the Logs tab in the diagnostic pages. The log is useful when diagnosing sporadic problems, and can reveal the reason for the cable modem having previously gone offline. Not every line in the log represents an error, even if it is flagged as an error. Interpreting the log is not easy.

The log is presented in reverse chronological order, so the most recent log entry is at the top. Each entry in the log is timestamped in the format YYMMDDhhmmss in GMT. The timestamp is replaced by asterisks if the cable modem has not yet acquired the time from the UBR.

Here's a not entirely abnormal re-boot sequence, which should be read from the bottom up:

010404174626 7-Information F502.1  Bridge Forwarding Enabled.
010404174626 7-Information F502.3  Bridge Learning Enabled.
010404174626 7-Information B518.0  Baseline Privacy is skipped
010404174626 7-Information I500.0  Registration Completed
010404174626 7-Information I0.0    REG-RSP Registration Response
010404174626 7-Information I0.0    REG-REQ Registration Request
010404174626 7-Information D509.0  Retrieved TFTP Config silver.cm SUCCESS
010404174626 7-Information D507.0  Retrieved Time....... SUCCESS
************ 7-Information D511.0  Retrieved DHCP .......... SUCCESS
************ 5-Warning     D520.2  DHCP Attempt# 1 BkOff: 4s Tot DSC:1 OFF:1 REQ:1 ACK:1
************ 7-Information D0.0    DHCP CM Net Configuration download and Time of Day
************ 7-Information T500.0  Acquired Upstream .......... SUCCESS
************ 8-Debug       T503.1  Acquire US with status OK, powerLevel 48, tempSid 16
************ 8-Debug       T505.0  Acquired Upstream with status OK
************ 7-Information T501.0  Acquired Downstream (403000000 Hz)........ SUCCESS
************ 8-Debug       T509.0  Acquired DS with status OK, DS Freq 403000000, US Id 3
************ 7-Information H501.18 HFC: UCD Symbol Rate change.
************ 7-Information H501.17 HFC: UCD Minislot size change.
************ 8-Debug       T509.0  Acquired DS with status NO FEC lock, DS Freq 143000000, US Id 0
  [snipped many similar lines]
************ 8-Debug       T509.0  Acquired DS with status NO FEC lock, DS Freq 403000000, US Id 0
************ 4-Error       M506.0  Invalid scan limits. Scan will use entire range.
************ 7-Information I510.0  *** BOOTING *** SB3100-3.2.9-SCM-NOSHELL

Note the name of the configuration file flagged above on the D509.0 line. See Cable Modem configuration file for more about this file.

In some cases, after (i.e. above) the F502.1, the cable modem will go on to attempt to load a firmware update by TFTP. Errors flagged during this process should not be a source of concern for the end-user. Here is an example:

010404145522 4-Error       X501.35 Unit Update Failed.
010404145522 7-Information F502.1  Bridge Forwarding Enabled.
010404145522 4-Error       X501.43 Unit Update Failed. Access Violation returned from TFTP server.
010404145519 7-Information F502.2  Bridge Forwarding Disabled.
010404145519 7-Information X500.0  Attempting Unit Update

The X501.43 error above should not be happening, but will not be affecting data service other than delaying the end of the re-boot sequence (by 3 seconds in this case). Here is another case with a longer delay of 2 minutes at the end of the boot sequence:

010409205631 4-Error       X501.35 Unit Update Failed.
010409205631 7-Information F502.1  Bridge Forwarding Enabled.
010409205631 4-Error       X501.38 Unit Update Error. TFTP transfer timed out after 16 read requests.
010409205425 7-Information F502.2  Bridge Forwarding Disabled.
010409205425 7-Information X500.0  Attempting Unit Update

In the next example, the X501.28 is not an error at all: it is simply that the firmware update offered by the network is not intended for this model of cable modem, and has therefore been correctly declined:

010419191646 4-Error       X501.35 Unit Update Failed.
010419191646 7-Information F502.1  Bridge Forwarding Enabled.
010419191645 4-Error       X501.28 Unit Update - Not updating. This image is not for this platform.
010419191645 7-Information F502.2  Bridge Forwarding Disabled.
010419191645 7-Information X500.0  Attempting Unit Update

Log entries of events affecting the data service are flagged Critical, Alert, or Emergency, for instance:

010331125025 7-Information H501.6  HFC: TRC SYNC Recovery OK 
010331125024 3-Critical    H501.4  HFC: LOST TRC SYNC- trying to recover

The above shows a transient downstream signal problem, which was recovered from one second later. Other possible serious service-affecting errors are:

************ 3-Critical    H501.9  HFC: T1 Timer Expired
************ 3-Critical    H501.7  HFC: T2 Timer Expired
************ 2-Alert       T507.0  Received Async Error Range Failed
************ 3-Critical    H501.8  HFC: T4 Timer Expired
************ 3-Critical    H501.16 HFC: FEC LOCK recovery failed

The above, if repeated frequently, all indicate serious signal problems on the CATV system, which would need attention from your local cable operator. In severe cases you might also see:

************ 7-Information H501.2  HFC: Shutting Downstream Down
************ 8-Debug       T503.1  Acquire US with status NO UCD for US, powerLevel 0, tempSid 0

With the H501.2, the cable modem is announcing that previous errors are so severe that it is now giving up. Following (i.e. above) the H501.2 event, the entire re-boot sequence will be executed, and this will appear as a substantial loss of service, with much flashing of lights.

3Com HomeConnect (Tailfin) diagnostics

Send your web browser to http://149.112.50.65/ for early firmware versions, or http://192.168.100.1/ for recent firmware versions. Firmware can be automatically updated by your cable operator down the cable, so by the time you read this, most Tailfins will have been updated to use the new address. Click on the Session Status link. The downstream and upstream power levels are displayed. To view the downstream SNR, see the event log below.

3Com HomeConnect (Tailfin) event log

Send your web browser to http://149.112.50.65/ for early firmware versions, or http://192.168.100.1/ for recent firmware versions. Firmware can be automatically updated by your cable operator down the cable, so by the time you read this, most Tailfins will have been updated to use the new address. Click the link to the log page, which is implemented with an embedded Java applet. A typical log output might look like this, reading from the top downwards:

2 0:0:2.0 1\1\1970 +0:0 ROOT 1 Cable Interface MAC Addr:    00:01:03:4B:0C:6F
2 0:0:2.0 1\1\1970 +0:0 ROOT 1 Ethernet Interface MAC Addr: 00:01:03:4B:0C:70
2 0:0:2.0 1\1\1970 +0:0 ROOT 1 USB Interface MAC Addr:      00:01:03:4B:0C:71
2 0:0:2.0 1\1\1970 +0:0 ROOT 1 Serial Number: HHP24B0C6F 
2 0:0:2.0 1\1\1970 +0:0 ROOT 1 Initializing Audio...
2 0:0:2.0 1\1\1970 +0:0 ROOT 1 Sound is disable.
2 0:0:2.0 1\1\1970 +0:0 ROOT 1 Default volume is 3.
2 0:0:2.0 1\1\1970 +0:0 ROOT 1 Entering Normal Operation Mode!
2 0:0:2.0 1\1\1970 +0:0 ROOT 1 ********************************************************
 	   	CM RESET REASON ID       
2 0:0:2.0 1\1\1970 +0:0 ROOT 1 ********************************************************
3 0:0:3.0 1\1\1970 +0:0 cmmr 1    ***** CMM ***** Acquiring Freq 0609000000 Hz
4 0:0:4.0 1\1\1970 +0:0 bcmr 1 Downstream Information
4 0:0:4.0 1\1\1970 +0:0 bcmr 1 Tuner Freq    : 609.0000 MHz
4 0:0:4.0 1\1\1970 +0:0 bcmr 1 CurFreqErr    : 000000
4 0:0:4.0 1\1\1970 +0:0 bcmr 1 Symbol Rate   : 5.056941 MHz
4 0:0:4.0 1\1\1970 +0:0 bcmr 1 Lock Threshold: 20.50 dB
4 0:0:4.0 1\1\1970 +0:0 bcmr 1 SNR Value     : 33.46 dB
4 0:0:4.0 1\1\1970 +0:0 bcmr 1 QAM mode      : 64
4 0:0:4.0 1\1\1970 +0:0 bcmr 1 Interleave Depth : 32,4,32
4 0:0:4.0 1\1\1970 +0:0 bcmr 1 QAM and FEC locked
4 0:0:4.0 1\1\1970 +0:0 bcmr 1 Uncorrectable Errors: 0 
4 0:0:4.0 1\1\1970 +0:0 bcmr 1 Downstream Power Level:   3.1 
4 0:0:4.0 1\1\1970 +0:0 bcmr 1 Acquiring Time Sync
5 0:0:5.0 1\1\1970 +0:0 cmmr 1 Upstream symbol rate:  2.560 MHz
5 0:0:5.0 1\1\1970 +0:0 cmmr 1 Upstream center frequency: 37.200 MHz
6 0:0:6.0 1\1\1970 +0:0 bcmr 1 Course sync check is valid
6 0:0:6.0 1\1\1970 +0:0 bcmr 1 Fine sync check is valid
6 0:0:6.0 1\1\1970 +0:0 cmmr 1    CMM Primary SID updated: 0 
6 0:0:6.0 1\1\1970 +0:0 cmmr 1 Prog UsChanId: 04 ChgCnt: 1D
6 0:0:6.0 1\1\1970 +0:0 cmmr 1  Setting power level to 27.0
6 0:0:6.0 1\1\1970 +0:0 cmmr 1 Range Request sent at 27.0 power level
6 0:0:6.0 1\1\1970 +0:0 cmmr 1 mcns: power level adjust (1/4 dB) = 016
6 0:0:6.0 1\1\1970 +0:0 cmmr 1  Setting power level to 31.0
6 0:0:6.0 1\1\1970 +0:0 cmmr 1 mcns: timing adjust = 0x09f8
6 0:0:6.0 1\1\1970 +0:0 cmmr 1    CMM Primary SID updated: 1db 
6 0:0:6.0 1\1\1970 +0:0 cmmr 1    CMM Unicast SID updated: 1db in US Queue idx: 0 
6 0:0:6.0 1\1\1970 +0:0 cmmr 1    ***** CMM ***** Start Unicast Ranging TS=00000AE9
7 0:0:7.0 1\1\1970 +0:0 cmmr 1    CMM Unicast SID updated: 1db in US Queue idx: 0 
7 0:0:7.0 1\1\1970 +0:0 cmmr 1    ***** CMM ***** Ranging Complete, MslotCnt: 005F1066
7 0:0:7.0 1\1\1970 +0:0 dhcp 1 Sending DHCP message
8 0:0:8.0 1\1\1970 +0:0 netp 1 Sending DHCP message
8 0:0:8.0 1\1\1970 +0:0 netp 1 Sent DHCPREQUEST and entering requesting state !
8 0:0:8.0 1\1\1970 +0:0 netp 1 Configuration file acquired from DHCP ACK pdu
8 0:0:8.0 1\1\1970 +0:0 netp 1 DHCP server provided 1 gateway router(s)
8 0:0:8.0 1\1\1970 +0:0 netp 1 Successfully added default route 
8 0:0:8.0 1\1\1970 +0:0 netp 1  to Cable
8 0:0:8.0 1\1\1970 +0:0 netp 1 Successfully added static route to gateway
8 0:0:8.0 1\1\1970 +0:0 netp 1 DHCP -- DHCP process complete(DHCPACK received) 0xff
8 0:0:8.0 1\1\1970 +0:0 netp 1 	TimeOffset			= 0
8 0:0:8.0 1\1\1970 +0:0 netp 1 	IP address lease time 		= 345600 seconds
8 0:0:8.0 1\1\1970 +0:0 netp 1 DHCP state => Address acquisition successful
8 0:0:8.0 1\1\1970 +0:0 netp 1 Adding IP address to Cable port successful !
8 22:51:57.0 7\22\2002 +0:0 cmmr 1    ***** CMM ***** Config File Processing Complete: MslotCnt: 005F1FB1
9 22:51:58.0 7\22\2002 +0:0 cmmr 1    CMM Primary SID updated: 1db 
9 22:51:58.0 7\22\2002 +0:0 cmmr 1    CMM Unicast SID updated: 1db in US Queue idx: 0 
11 22:52:0.0 7\22\2002 +0:0 cmmr 1 MCNS Upstream Statistics
11 22:52:0.0 7\22\2002 +0:0 cmmr 1 Ref Freq:      25.000 MHz
11 22:52:0.0 7\22\2002 +0:0 cmmr 1 Tx Freq :      37.200 MHz
11 22:52:0.0 7\22\2002 +0:0 cmmr 1 Symbol Rate:    2.560 MHz
11 22:52:0.0 7\22\2002 +0:0 cmmr 1 Tx QAM mode:   QPSK
11 22:52:0.0 7\22\2002 +0:0 cmmr 1 Tx Channel ID: 0x04
11 22:52:0.0 7\22\2002 +0:0 cmmr 1    ***** CMM ***** MAC Registration Complete, MslotCnt: 00600C8F
11 22:52:0.0 7\22\2002 +0:0 cmmr 1 ********** START FORWARDING DONE| **********
12 22:52:1.0 7\22\2002 +0:0 fdsp 1 sound_play_file connect22.wav volume = 3
12 22:52:1.0 7\22\2002 +0:0 fdsp 1 Successfully added route to CPE

The columns are: seconds elapsed since boot, time of day, date, timezone offset, task name, trace level, and message. Signal-related information is highlit above.

3Com CMX diagnostics

The 3Com CMX does not have a diagnostics web page, but does have a serial communication port on which it produces a log. To read this log, connect your PC's serial port to the CMX's serial port with a null modem cable (or laplink cable). Launch a terminal emulator such as Hyperterminal, and set the serial communication parameters to 38400 bps, 8 data bits, No parity, 1 stop bit. (You might need to experiment to find the correct flow-control settings to avoid data over-run and a corrupt log view). Power the CMX off and on again, and watch the log stream by as it configures itself. The information in the log includes the signal levels, as described above. The log is produced only during the boot process, and when the CMX is recovering from errors.

For a potential solution for reading diagnostics while in service, see DocsDiag.

Terayon TeraJet diagnostics

There are no diagnostics available to the end-user via a web browser. For a potential solution by other means, see DocsDiag. The default settings of DocsDiag do not work with the Terayon: use the Recipe for use of DocsDiag with the ISP-side address of the cable modem given in the documentation.

ntl:home 100/120 (Ambit) diagnostics

Send your web browser to http://192.168.100.1/. When challenged, enter userid root and password root. Not all these diagnostic screens will be present on all models.

The first diagnostic screen looks like this:

The most important signal diagnostics above are Downstream SNR and Upstream Transmit Power Level. For acceptable ranges, see the discussion above in the Surfboard section.

The above is normal for an NTL cable connection. If the cable modem has failed to connect properly to the cable network, fewer of the tasks will be registered as Done. The first one not showing as Done will be the task that is failing. It is not a problem for BPI to be showing as not enabled.

This shows more technical information on the downstream channel, including the Downstream Receive Power Level. For acceptable ranges, see the Surfboard discussion above.

This shows more technical information on the upstream channel. There is nothing here that is affected by signal quality. Note that the Upstream Symbol Rate is shown as 1280 ksym/sec. With QPSK modulation on the upstream giving 2 bits per symbol, this equates to 2560 kbits/sec. With the cable modem having an upstream rate of 128 kbits/sec (see below), this means that one upstream channel can support only 20 cable modems without congestion if those modems are all sending as fast as they can. The screen given by Upstream Burst Descriptor is too technical to be of interest to end users.

In Operation Configuration, the items of interest are Maximum Downstream Data Rate and Maximum Upstream Data Rate: these are the cable modem rate caps. The settings shown above are typical of the NTL 600kbps service for stand-alone cable modems.

This screen is refreshed every second, and shows the number of packets sent and received on each interface.

The Event Log screen is refreshed every 10 seconds. It is presented in reverse chronological order of Last Time, so the most recent event is at the top. Unlike the Surfboard log, events with the same IDs are combined into a single entry, with a date range between the first and most recent incidences, and a repeat count. Note that the event with ID 95 will show the name of the cable modem configuration file in use.

For troubleshooting information on the ntl:100 cable modem, see http://www.ntlworld.com/help/manual/100Manual.htm.

Return to Index.