Table Of Contents
Frequently Asked Questions
System Configuration and Operation
PAC FAQs
General PAC Issues
Troubleshooting Toolset
Traffic Capture Toolset
Planning and Reporting Toolset
Web Client Connectivity Issues
Report Generation and Display Issues
Understanding Reports
Frequently Asked Questions
This chapter covers Frequently Asked Questions (FAQs) that may arise when running the PAC software. For FAQs that address PAM issues, see the Cisco WAN Access Performance Management System System Administration Guide, 2.0.
This chapter contains the following topics:
•
"System Configuration and Operation"
•"PAC FAQs" section.
Note Please consult this section before calling your vendor's technical support.
System Configuration and Operation
Q. What must I do so my Cisco WAPMS technical support personnel can access my PAM to troubleshoot problems?
A. Three things:
•connect a phone line to the PAM's modem port
•provide the technical support personnel with the phone number
•launch the PC Anywhere application (which is resident on the PAM). To set PC Anywhere to wait for a call, click the Be a host button, then double-click the modem icon.
When the technical support personnel are finished with the troubleshooting task, close
PC Anywhere and disconnect the phone line.
Q. If I have an enhancement to the Cisco WAPMS application I believe would be beneficial, where do I send the requests?
A. Customers are encouraged to submit product enhancements. Please send all feature requests to your Cisco Systems sales representative or systems engineer. The Cisco Systems Product Management Team will review your request and consult the parties necessary to determine whether to include your feature request in an upcoming release.
Q. When my system elements (ASEs, PAMs, and clients) are located in different time zones, which time zone should I set each ASE to?
A. ASEs, PAMs, and clients work together to assure the statistics collected are stored with the proper time reference. There is no need to set clocks on the ASEs. However, each PAM and client should be set to its own local time zone so data you display on the PAM or client is relative to your viewing location.
The client's local time determines the time displayed for all events (in the Event Processor window), conversations, traffic captures, troubleshooting data, and client-displayed report data.
Some tasks—data collection, database maintenance, and scheduled report group generation—which you schedule from the client, actually occur on the PAM and their results are displayed in log files on the PAM. Consequently, enter the PAM's time in the client's corresponding window when scheduling these tasks.
PAC FAQs
This section contains frequently asked questions relating specifically to the PAC.
General PAC Issues
Q. When I start the PAC, the Performance Monitoring, Troubleshooting, Traffic Capture, and Conversations Toolset buttons are grayed out. The user ID I'm using has access to these options, but they are unselectable. Why?
A. The first time you run the PAC on a workstation or the first time you run a new version of the PAC, you must choose a network for these PAC Toolsets to view. Click Network Selector at the bottom of the window, then choose a network in the window that appears. This enables the Toolsets when the network element selected has the corresponding capability.
See "Using the PAC," for further details.
Q. When starting the PAC, the error message RPC Program Mismatch appears. What does this mean?
A. The system running your PAC is using a version that is incompatible with the software version of your PAM. Upgrade all clients to the same version as your PAM as part of your upgrade procedure. See the Cisco WAN Access Performance Management System Installation and Configuration Guide, 2.0.
Q. What do I do when my PAC communication with a PAM goes down and the Retry option in the window does not recover?
A. Exit and restart the PAC to reconnect to the PAM.
Q. Why does no data appear when I attempt functions involving real-time communication with the ASE? (Real-time functions include: the File menu options ASE Inventory, Update ASE Software, and Collect Data Now; the Network Configuration function ASE Setup; Performance Monitoring; Troubleshooting [real-time mode] and Troubleshooting current status and current statistics; Traffic Capture; and Conversations.)
A. Real-time communication to ASEs requires IP access from your workstation to the ASEs. You may have a firewall between your workstation and the ASEs preventing this communication.
Q. How does Cisco WAPMS report times when an ASE, PAC, and PAM in a system are all in different time zones?
A. All dates and times are relative to the date and time settings of each PAC. This gives all events and data local significance to the user viewing the information. For example, when an ASE in Dallas reboots at 6:00 a.m. CST, then a PAC in New York would show the ASE reboot at 7:00 a.m. EST while a PAC in Seattle would show the reboot happened at 4:00 a.m. PST.
Q. How do I save the ASE inventory results as an Excel file?
A. After performing an ASE inventory, you have the option to save the information as a comma-delimited file. You can then open the file in Excel, which initiates the text import wizard. Select Comma as the delimiter to display the information in spreadsheet format.
Network Configuration Toolset
Q. What are my options in configuring IP Transport ASEs from the PAC?
A. From the PAC, you can set up IP Transport ASE access lines and end-to-end circuits on an IP-switched network.
When you configure IP Transport ASEs, you can configure up to ten protocols (and three additional custom protocols), up to 32 subnet ranges per site, and the committed access rate (CAR). You can also view the class of service settings, and modify the policing threshold for the ASE. You configure the ASE's Frame Relay or HDLC link layer settings and its physical layer settings in the standard way. See "Configuring IP Transport Settings" section.
IP Transport ASE model 49 supports up to 155 IP circuits. You need to configure each circuit's far end point; single ended IP circuits are not supported. You have options for viewing circuit subnets, and for removing old circuits when the ASE's circuit is full. See "Configuring IP Circuits" section.
You can also upgrade a Frame Relay or HDLC model 49 to an IP Transport ASE. See "Updating ASE Software" section.
Q. Why does the automatic detection of my ASE during ASE Setup not detect the committed information rate (CIR)?
A. Annex-D link management interface (LMI) status messages do not include information regarding CIR as part of the full status response—only Frame Relay Forum status messages include CIR information. Furthermore, depending on how your WAN service provider has configured the switch, the switch may not disclose the CIR value. Consequently, when your WAN service provider is using Annex-D or is not setting the switch option to disclose CIR, Cisco WAPMS cannot automatically detect CIR.
Q. After clicking ASE Setup in the Network Configuration Toolset, the PAC application times out when attempting to contact the ASE.
A. Verify the following:
•check the correct IP address is configured for the ASE in the Network Configuration window. From the Network Configuration window, choose the access line, then click Edit.
•determine if a compatible version of ASE software is installed on the target ASE. The ASE software version must be at the PAC/PAM version or earlier. To check the ASE software version from the Network Configuration window, choose the access line, click Edit, then click ASE Setup. To check the PAC version, select Help > About Cisco WAPMS.
•the SNMP community strings in the ASE and PAC/PAM are identical. On the PAC, choose Setup > Domain Options. At the ASE's Administrator prompt, enter community.
•check IP connectivity to the ASE by pinging the IP address assigned to the ASE. Enter ping at the ASE's Admin prompt.
Q. Why do some ASE network probes auto-detect T1 channels and others do not?
A. A significant amount of traffic must be present across all T1 channels for a probe to auto detect the channels. When the line is idle or very under utilized, the ASE cannot detect the channels because it looks specifically at the data to derive channel information.
Event Processor Toolset
Q. Can I forward events to my third-party management program?
A. Yes. You can forward ASE events from the PAM to a third party management program such as HP OpenView. See "Encapsulation Protocols," for instructions on forwarding events.
Q. Why are events displayed with the wrong date and time?
A. The date, time, and time zone on either the PAC or the PAM are not correctly set for their respective locations. Each PAC and PAM should be set to its own local time zone. The date and time displayed in the Event Processor is relative to the PAC's time setting.
Q. Why do I see events with dates in the future in the Event Processor?
A. Future events are generated by ASEs that are running older versions of Cisco WAPMS and which have not been rebooted for eight months. Upgrade the ASE, PAC, and PAM to the most current software release to prevent future events from being sent.
Q. Many events in the Event Processor are tagged with the message Incorrect SNMP community name contacting ASE. What does this mean?
A. Another management system is sending traps to the ASE. You can use the LOG command while logged into the ASE to determine which devices are contacting it.
Q. What does it mean when an event message indicates: Performance data collection poll failure, 4 out of 65 sections could not be updated?
A. When only some of the sections can be updated, one of the following may be the cause:
•inactive and/or invalid DLCIs
•software mismatch between the PAC/PAM and the ASE
•misconfigured PVCs.
When none of the sections can be updated, then the circuit or router is down, or there is some other communications failure between the ASE and the PAM. When inband communication is being used, the management PVC could be the cause of failure.
Q. Can you configure Statistical Threshold Events for multiple ASEs instead of one ASE at a time?
A. Yes. You can create a template of the threshold events using PAM Manager, then save the template to a multiple ASEs. See the "Using the Templates Tool" section on page 4-12 in the Cisco WAN Access Performance Management System System Administration Guide, 2.0 for instructions on how to create templates.
Troubleshooting Toolset
Q. When the PAM is inactive at its scheduled time for ASE data collection, does the PAM receive that ASE data the next time it becomes active, or is this data lost?
A. Whenever the PAM performs ASE data collection, the PAM automatically retrieves all data from the ASE it does not already have in the short-term database. At any given time, the ASE has data from the two most recent days.
Q. What does it mean in the Troubleshooting Summary window when part of the graph is mud, yet it is surrounded on either side by what appears to be valid, plotted information?
A. This indicates the ASE was not operational during the time when the graph was mud, or data was not retrieved from the ASE before the ASE's day window expired.
Q. Under what circumstances can the Cisco WAPMS database miss network performance data, and how can I prevent this from occurring?
A. The database does not receive network performance data when data is not downloaded from the ASE for 48 hours or when database maintenance is not executed for two weeks. To prevent these situations, leave the PAM running 24 hours a day.
Q. Why can I display 2-hour and 2-day history data but I cannot display real-time data?
A. 2-hour and 2-day history statistics are transmitted through TFTP between the ASE and the PAM during data collection. When TFTP communication is interrupted during this data collection process, the PAC does not display these statistics.
You can test TFTP connectivity from the PAM using the IPTool utility (located at c:\pam\iptool.exe), which is installed with the PAM software. Run this utility and issue a TFTP Get request to the ASE. When a response is not received, TFTP communication is not being transmitted. Check the appropriate routers access list.
Q. When viewing real-time traffic, I can see the bandwidth being used bursting up to the port speed. However, when viewing 2-hour or 2-day history, the line does not show any bursts. Why?
A. Real-time statistics average utilization data over 5 second time intervals; 2-hour and 2-day history statistics average the data over 15 minute time intervals, which may result in a relatively smooth utilization pattern. You can view a more detailed display of circuit burst events by using a Burst view that uses 1-second time intervals to calculate precise line utilization.
Q. When I am using the Troubleshooting Toolset and click on the PVC button, the following error message is generated: Cannot communicate with ASE.
A. This message indicates an inactive or invalid DLCI exists on the line. To verify which DLCI is inactive or invalid, use the Circuit Details function. Delete the appropriate DLCIs from Network Configuration to stop receiving the error message.
Q. When viewing a particular PVC view from within the Troubleshooting Toolset, the following message is generated: Cannot contact the far side ASE. Why?
A. This typically is the result of an incorrect DLCI association and incorrectly mapped PVC. Verify the PVC mapping information you configured is correct. Do this by performing a Traffic Capture on each end of the circuit and validating the DLCI numbers used at each end.
Q. Why is my round-trip-delay button grayed out?
A. An ASE must be present on both ends of the circuit for round-trip-delay to be enabled. In addition, ASE probes that are not "inline"—that is, any ASE probe using a y-cable connection—are incapable of performing round-trip-delay tests.
Q. Why do I receive the error message: Unmatched Frame Traffic when using the round-trip-delay function?
A. This error message indicates one or more of the following are present:
•one or more PVC are mapped incorrectly. Verify within the Network Configuration Toolset that all PVCs are mapped correctly.
•there is little data going across the circuit (less then 5% utilization)
•traffic utilization is one sided, meaning one side has low utilization and the other side has a high utilization.
Q. Why do I get the message: Unable to Start Connectivity Test when using the Verify Connectivity button?
A. This message is displayed when Inband or SLA communication is not enabled on the ASE. The message also appears when the ASE is a non-inline ASE probe (that is, it is connected by a y-cable connection) and, therefore, is incapable of performing a connectivity test.
Q. When viewing the Protocols view within the Troubleshooting Toolset, why are all of the protocols displayed as "Other"?
A. "Other" indicates Cisco WAPMS has detected a VC or PVC on the line, but could not determine its type. Cisco WAPMS fails to correctly decode the protocol information for several reasons:
•the incorrect encapsulation protocol has been configured in the ASEs Frame Relay settings. In most cases the default setting Frame Relay (Auto) allows Cisco WAPMS to automatically detect the appropriate encapsulation being used. However, cases can occur where the system does not correctly identify the correct setting and you must set the encapsulation protocol manually.
•compression or encryption has been enabled on the router interface. Cisco WAPMS does not have the ability to decode past the Frame Relay header on circuits using these technologies.
•traffic on the circuit does not include the defined protocols. You can perform a Traffic Capture to verify the data.
•certain Token Ring bridges are known to display similar results as a result of "bit swapping" the bridge may be performing through a proprietary algorithm.
Q. I just changed the protocols my ASE monitors, but when I switch back to the Troubleshooting Toolset, I continue to see the old protocols.
A. Real-time views of the new protocols should be available after the ASE rolls over to the next 15-minute sampling time "bucket" and the window refreshes. Also, data collected before the protocol change does not reflect the revised settings because the data was collected with the old protocol configuration.
Q. When viewing PVC statistics, why is the number of far-end offered frames less than the number of delivered frames?
A. Offered frames are the total number of frames offered below the CIR rate. The number of delivered frames may exceed the number of offered frames when "bursted" frames get through.
Note It is important to set the Excess Burst Allowed parameter to get accurate statistics. This parameter can be found within ASE setup.
Traffic Capture Toolset
Q. How do I capture only LMI packets on my Frame Relay circuit?
A. Depending on the type of LMI being used, the LMI packets transmit over DLCI 0 or DLCI 1023. Create a filter for each DLCI when you do not know which is used. Because a full status message is sent every minute, capture a minimum of 12 frames to make certain you captured a full status request and reply. See "Traffic Capture," for step by step instructions on how to perform traffic captures.
Q. Can I save the information obtained from a traffic capture to a file so I can view the information later?
A. Yes. After performing the capture you can choose File > Save as from the menu and save the information. You have the option to save the file in traffic capture format (use the extension *.pkt), which allows you to open and view the file in the Traffic Capture Toolset. You also can save the file in CSV format (use the extension *.txt), which allows you to import the data into a database or spreadsheet.
Q. Can I filter for all hosts on an IP subnet?
A. Yes. Use an asterisk (*) as a wildcard character in place of any part of the IP source or destination address in the IP Advanced Filter Setup window. For example, to filter on all hosts on the class C 204.217.134.0 network, enter 204.217.134.*.
Q. Can I filter using nonstandard IP subnet masks?
A. Yes. See the bit-level edit box for the IP address and use asterisk (*) to replace any bit in the binary address.
Planning and Reporting Toolset
Web Client Connectivity Issues
Q. When accessing Planning and Reporting with my browser, I get the error message: Server is not responding. Why?
A. You receive this message when a firewall exists between the Web Client workstation and the PAM. The firewall must be configured properly to allow communication between the Web Client and the PAM. See "Supported Standards," for the ports and protocols used by Cisco WAPMS.
Q. When launching the Planning and Reporting Toolset, the browser hangs and displays a blank white window and hourglass cursor. Why?
A. If this occurs, check the following:
•the WWW service must be running on the PAM. Ensure the World Wide Web Publishing Service status is set to start within Settings > Control Panel > Services.
•the WWW service account's user ID and password must match in both User Manager For Domains and Internet Service Manager. During most installations, the WWW Service Anonymous login user ID is set to the IUSR<xxxxx> (where <xxxxx> is the NETBIOS name of the PC) and the password is randomly generated. You can find this user ID in the NT User Manager for Domains Toolset and the Internet Service Manager. The accounts must match in both places. Because the passwords are not visible (noted by *****), there is no way to verify a match. When the user IDs match in both locations but you believe the password could have been changed, you can set them manually through the corresponding Toolsets.
Report Generation and Display Issues
Q. When I choose Planning and Reporting, it does not start—the window remains blank.
A. When the PAC is running on a Windows NT system, the PAC may not have been installed by someone using a user ID with Windows NT administrator privileges. Consequently, the Java modules required by Planning and Reporting were not installed correctly. When this is the problem, reinstall the PAC using a user ID with Windows NT administrator privileges.
See the Cisco WAN Access Performance Management System Installation and Configuration Guide, 2.0 for step-by-step guidance through the installation process.
Q. When attempting to view SLA reports, the following message appears: No Data Available for all of the SLA reports. Why?
A. SLA reports are available only for ASE models capable of inband communication. SLA must be enabled on the ASE for SLA report information to populate.
Q. When creating End-To-End VC Service Level Verification reports, the data-delivery-ratio and round-trip-delay columns contain the value Undefined. Why?
A. To generate data-delivery-ratio and round-trip-delay data, the ASEs on each end of the PVC must have SLA enabled. You also must enable the SLA option on each ASE.
Q. Other users cannot view reports I have created. Why?
A. Report groups and scheduled reports can only been viewed through the user ID under which they were created. Each user must define his or her own report groups or, when security is less important, you can define one or two common user IDs for which you can create reports.
Q. In some of my reports, such as the Frame Relay Service Level Reports, the CIR on some sites are reporting as zero, while other sites are reporting a non-zero CIR value. Why?
A. Cisco WAPMS attempts to auto-configure the CIR value of a line, based on the CIR value set in the LMI full status message. Some types of LMI, such as Annex D, do not identify the CIR value; however, you can manually configure it within Cisco WAPMS. To do this, open the Network Configuration window and edit the ASE Setup of the appropriate ASE.
Note When the CIR is changed, data collection and database maintenance must run before the change is reflected in Cisco WAPMS.
Understanding Reports
Q. How is the Most Active VCs report calculated?
A. The Most Active VCs report is the average maximum throughput for the specified time period. In other words, a report over a ten day period would provide a ranking of the average of each day's maximum throughput over ten days.
Q. When viewing Most Over Utilized Channel reports, why does it occasionally display the line as being overutilized when, in fact, it is only used for brief periods of time?
A. The Most Over Utilized Access Channel reports (Tx and Rx) do not include the time the utilization is 0 to 10 percent of configured bandwidth. The report only takes into account the time the line is being utilized, and does not include circuit "idle" time.
Q. How is the Frame Delivery Ratio (FDR) computed?
A. FDR is the ratio of the number of frames delivered on a circuit by the network to the adjusted number of frames offered to the network on that circuit. It is measured on a per-PVC basis. The data is stored in the ASE as a far-end "frames offered" count and a near-end "frames delivered" count. The counts required to compute the FDR are accumulated by the ASEs on the ends of the PVCs by communicating with each other. The statistic is used for "non-burst" offered frame counts only.
For each monitored DLCI, each ASE maintains a counter of:
•delivered frames—the total number of frames delivered by the network
•offered frames—the total number of frames offered below Bc + Be (that is, below the excess burst rate).
When the console presents the DDR data for a circuit, it computes it from the aggregate of counts from both ends of the PVC using the following formula:
FDR = [DeliveredCount1 + DeliveredCount2] รท [OfferedCount1 + OfferedCount2]
