Error/Fault Data for Dial and ISDNLike the discussion of performance management, this section on error and fault management looks at MIBs and CLI commands. In addition, as mentioned in previous chapters, the use of syslog is an important part of any fault management plan. In the case of remote access servers, the use of both syslog and the radius logs can add an important aspect to the data that is gathered and analyzed for remote access. In this section, the following will be discussed:
MIB Variables for Monitoring Connection ErrorsThe cmLineStatisticsTable provides essential data on modem connection errors. As explained earlier in the section "MIBS for Modem Status Monitoring," cmDisconnectReason is one of the most important variables (see Table 17-4 for the different results). In fault management, these variables help you identify the exact type of disconnect. For fault management, the cmDisconnectReason variable would be mapped against the different result categories just as in performance management. The table and scripts used for performance can be reused here. In addition, the entries in cmLineStatisticsTable could indicate the number of connection completions, failures, and other data needed to monitor the modem faults. As mentioned earlier, this data is very useful for troubleshooting the modem connections. CLI Commands for Monitoring Connection StatisticsSee the output from show modem call-stat (see Example 17-2). For fault management, you'd use the command show modem call-stat to look specifically for disconnect reasons. CLI Commands for Monitoring Modem FirmwareAs mentioned in the first section of this chapter, one of the fundamental parts of provisioning a Network Access Server (NAS) is inspecting and upgrading modem firmware; the modem firmware is not accessible through SNMP. Although keeping track of the firmware is not usually considered part of fault management with remote access networks, it is an essential part. During the fault identification process, the type of firmware is one of the first things that is checked. Finally, because faults are isolated on servers, the need for understanding the modem firmware becomes essential. Often, the upgrade of modem firmware is one of the first requirements. Example 17-8 shows output from the show modem version command from from the as5800. Examples from other platforms are not included because of their similarity. Example 17-8 Using show modem version on an as5800 platform. as5800#show mod ver Modem Range Module Firmware Rev 1/3/00 1/3/11 0 2.4.1.0 1/3/12 1/3/23 1 2.6.1.0 . . . 1/5/12 1/5/23 1 Unknown 1/5/24 1/5/35 2 2.6.1.0 1/5/36 1/5/47 3 Unknown 1/5/48 1/5/59 4 2.6.1.0 - - 11 - Modem board HW version info: Modem Range: 1/3/00 1/3/11 Modem Module: 0 Manufacture Cookie Info: EEPROM Type 0x0101, EEPROM Version 0x01, Board ID 0xDC, Board Hardware Version 1.0, Item Number 73-2988-2, Board Revision A48, Serial Number 09820408, PLD/ISP Version 255.255, Manufacture Date 3-Aug-1998. Example 17-9 displays information for modem firmware. This information includes modem firmware version, boot code version, controller and DSP code version (56 KB modem modules only), modem board hardware version, and carrier card information. This particular example displays information for 56 KB modem cards installed in a Cisco AS5xoo. If the version number is displayed as 0.0.0, verify that out-of-band status polling is functioning. Example 17-9 Using show modem version to obtain firmware information. router# show modem version Modem module Firmware Boot DSP MdmA NumberB RevC RevD RevE 0/0 0 3.1(21) 3.0(4) 1.1(0)/1.1(0) . . . 0/12 1 2.2(8) 1.0(5) . . . 0/23 1 2.2(8) 1.0(5) 2/0 0 3.1(21) 3.0(4) 1.1(0)/1.1(0) . . . 2/23 1 3.1(21) 3.0(4) 1.1(0)/1.1(0) Modem board HW version info: Slot 0: Carrier card: hw version= 8, number_of_ports= 24, max_modules= 2, max_oob_ports= 2 Modem Module 0: number_of_modems= 12, option_bits= 1, rev_num= 02.00, vendor_model_number= 02, vendor_banner= Microcom MNP10 K56 Modem Modem Module 1: number_of_modems= 12, option_bits= 1, rev_num= 03.00, vendor_model_number= 01, vendor_banner= Microcom MNP10 V34 Modem Example 17-10 displays modem version information for V.110 modem cards for unmanaged modems. Example 17-10 Obtaining modem version information with show modem version. router# show modem version Modem module Firmware Boot Mdm A NumberB RevC RevD 0/0 0 Unmanaged Unmanaged . . . 2/23 1 Unmanaged Unmanaged Modem board HW version info: Slot 0: F Carrier card: G hw version= 3 H, number_of_ports= 12I, max_modules= 1J, max_oob_ports= 1 K Modem Module 0: L number_of_modems= 12M, option_bits= 1N, rev_num= 03.01O, vendor_model_number= 01P, vendor_banner= V.110 Terminal AdaptorQ Highlighted field descriptions from Examples 17-9 and 17-10 are as follows:
Modem board HW version info:
The CLI command show modem version provides you with the data necessary to begin troubleshooting a firmware problem. It is the first place that you can look to when the Cisco Technical Assistance Center (TAC) asks for the firmware version on the NAS or more particularly on the modem with which you are having a problem. Table 17-9 summarizes the information collected from the show modem ver, show version, and show CPU CLI commands. It is intended as an example of how an administrator could organize the information for use. The table's software inventory enables you to look at the access server's IOS version and the different modem versions as well.
MIB Variables for Measuring ISDN/PRI ServiceFrom CISCO-POP-MGMT-MIB, cpmDS0UsageTable is the table that provides some of the data needed to monitor the ISDN/PRI Service. This MIB is implemented only on the AS5200/5300/5800/c3620/c7200 images for 11.3(5)T and greater releases. The primary variables that are useful from the CISCO-POP-MGMT-MIB for ISDN/PRI monitoring are listed as follows with the descriptions from that MIB:
The preceding variables provide the basis for managing the PRI for the NAS. These variables enable you to measure the usage of the PRI and can be helpful for troubleshooting problems. One cautionary note on the use of this table is that it is not populated for all Cisco devices and is available only in later releases of IOS. MIB Variables for Monitoring ISDN and Modem ConnectionsLooking at the CISCO-POP-MGMT-MIB, the cpmCallFailure provides objects that are optimizations so as not to retrieve and count tables. The variables in this section supplement the existing call-termination conditions from the CISCO-CALL-HISTORY-MIB. These variables address the both ISDN and Modem terminations. Again, this MIB is only implemented on a few of the Cisco platform (AS5200/5300/5800/c3620/c7200) images. Also, it is only available in the later IOS images (11.3(5)T and greater releases). The primary variables that are useful from the CISCO-POP-MGMT-MIB for ISDN and modem monitoring are listed as follows with the descriptions from that MIB:
The information from the preceding variables can be organized into a table such as Table 17-10. The table provides a starting point for identifying and isolating problems with resources on an NAS.
CLI Commands for ISDN/PRI ServiceThe show isdn service command is used for PRI interfaces only. Also, if the channel were 0, then the state would be valid; in all other cases, the state is ignored. This command provides different information from the cpmDS0UsageTable discussed in the section "MIB Variables for Monitoring ISDN/PRI Service." Although gathering data through the show command is usually preferred because of the reduction of CPU overhead, in this case the cpmDS0UsageTable variables provide a good deal of additional information that is not available through the CLI. Example 17-11 shows sample output from show isdn service. Example 17-11 Obtaining ISDN/PRI service information with show isdn service. as5800#show isdn service PRI Channel Statistics: ISDN Se1/0/0:15, Channel (1-31)A Activated dsl 0 State (0=Idle 1=Propose 2=Busy 3=Reserved 4=Restart 5=Maint)B 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 3 Channel (1-31) Service (0=Inservice 1=Maint 2=Outofservice)C 0 0 0 0 0 0 0 0 0 0 2 2 2 2 2 2 2 0 0 0 0 0 0 0 0 0 0 0 0 0 2 rtp-isdn>show isdn service PRI Channel Statistics: ISDN Se1/0:23, Channel (1-31)A Activated dsl 0 State (0=Idle 1=Propose 2=Busy 3=Reserved 4=Restart 5=Maint)B 2 2 2 2 2 2 2 2 2 2 2 0 2 2 2 2 2 2 2 2 2 2 2 3 3 3 3 3 3 3 3 Channel (1-31) Service (0=Inservice 1=Maint 2=Outofservice)C 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 2 2 2 2 2 2 2 The highlighted items in Example 17-11 are as follows:
The numbers given in state and channel provide an excellent way to look at the status of the PRI interface. For instance, in Example 17-11, you can readily see which are busy and that the "reserve" states are actually out of service. Knowing that this is output from a T1 rather than an E1 makes this piece of information important. Once again, mapping these to a table format and possibly graphing the states/channels over the servers can be a valuable source of information of the fault status of the PRI lines. MIB Variables for Monitoring Modem and ISDN Trap StatesFrom CISCO-ISDN-MIB, DemandNbrCallInformation is a trap/inform sent to the manager whenever a successful call clears, or a failed call attempt is determined to have ultimately failed. Syslog Messages for Modem and ISDN Trap InformationA number of syslog messages are useful for Dial and ISDN fault management and apply directly to the MIB objects and CLI commands previously discussed. They are collected in Table 17-11.
|