Downloads |
Table Of Contents
Response Time Reporter Enhancements
Configure Response Time Reporter Control Message Authentication
Configure Optional Probe Characteristics
Capture Statistics and Collect Error Information
Monitor the Response Time Reporter Feature
Perform Normative Analysis for SNA LU2
Perform Troubleshooting for IP/ICMP
Configure a Trigger for Connection Loss
distributions-of-statistics-kept
show rtr collection-statistics
show rtr distributions-statistics
statistics-distribution-interval
Response Time Reporter Enhancements
Feature Summary
The Response Time Reporter (RTR) Enhancement feature allows you to monitor network performance, network resources, and applications by measuring response times and availability.
Use the Response Time Reporter Enhancement feature to troubleshoot problems by checking the time delays between devices and the time delays on the path from the source device to the destination device at the protocol level.
You can also use this feature to send any combination of SNMP traps and SNA Alerts/Resolutions when one of the following has occurred: a user-configured threshold is exceeded, a connection is lost and reestablished, or when a timeout occurs. Thresholds can also be used to trigger additional collection of time delay statistics.
You can use this feature to perform preliminary problem analysis by scheduling the response time reporter and collecting the results as history and accumulated statistics. You can then use the statistics to model and predict future network topologies.
Note
The terms probe and collector are used interchangeably in this document.
Benefits
The Response Time Reporter Enhancement feature extends IP support and allows you to measure various types of IP traffic, such as UDP and TCP. Specifically, the enhancements enable you to perform the following tasks:
•
•
•
•
•
•
•
•
Platforms
This feature is supported on the following platforms:
•
•
•
•
•
•
•
•
•
•
•
•
•
•
Supported MIBs and RFCs
The response time reporter supports the Cisco Round Trip Time Monitor (RTTMON) MIB and the following MIB enhancements:
•
•
•
•
For descriptions of supported MIBs and how to use MIBs, see Cisco's MIB website on CCO at http://www.cisco.com/public/sw-center/netmgmt/cmtk/mibs.shtml.
No RFCs are supported by this feature.
Configuration Tasks
To configure the response time reporter feature, complete the tasks in the following sections. Refer to the "Command Reference" section for detailed syntax description of the commands used in these tasks. Configuring the probe and scheduling the probe are required tasks; the remaining tasks are optional.
•
•
•
•
See the "Configuration Examples" section for example configurations.
Enable the Responder
The responder is an enhancement that extends response time reporter support to new types of probes, such as the UDP echo responder and TCP connection probes. The RTR responder code must exist on target routers to support probes to non-native services such as the UDP echo and TCP connection probes. If UDP and TCP services (ports) are chosen for a probe to which a router does not normally respond, the RTR responder must be enabled to respond to RTR probe packets. If services that are already provided by the target router (such as Telnet or HTTP) are chosen, the RTR responder does not need to be enabled. For non-Cisco devices, the RTR responder can not be configured and RTR can probe only services native to those devices.
To enable the responder, use the following command in global configuration mode:
Configure Response Time Reporter Control Message Authentication
RTR uses a control message protocol to communicate with the Cisco routers that are the target of RTR probe operations. For security reasons, users have the option to enable authentication on the RTR Control Protocol. The authentication is provided using MD5 authentication. This authentication requires key definition on the source and target RTR routers. The existing Cisco IOS software commands are used to define the keys. Refer to the Configuration Fundamentals Configuration Guide for more information on these commands. The rtr key-chain command notifies RTR that it should use a specific key for authentication.
To configure the RTR authentication, use the following command in global configuration mode:
rtr key-chain name
Configures RTR authentication.
Configure the Probe
Response time and availability information is collected by probes (devices specifically placed in a network to collect data about the network) that you configure on the router. You must configure the probe type before you can configure any of the other characteristics.
The tasks in this section describe how to:
Define an echo Probe
You can measure response time from source to target by defining an echo probe. To define an echo probe, use the following commands starting in global configuration mode:
1
rtr number
Enter RTR configuration mode.
2
type echo protocol type type-target
Defines an echo probe.
Define a pathEcho Probe
You can measure time from hop to hop by defining a pathEcho probe. To define a pathEcho probe, use the following commands starting in global configuration mode:
1
rtr number
Enter RTR configuration mode.
2
type pathEcho protocol type type-target
Defines a pathEcho probe.
Note
Define a tcpConnect Probe
You can measure TCP connection time from source to target by defining a tcpConnect probe. To define a tcpConnect probe, use the following commands starting in global configuration mode:
1
rtr number
Enter RTR configuration mode.
2
type tcpConnect dest-ipaddr {name | ip addr} dest-port port number [source-ipaddr {name | ip addr} source-port port number] [control {enable | disable}]
Defines a tcpConnect probe.
Define a udpEcho Probe
You can measure UDP response time from source to target by defining a udpEcho probe. To define a udpEcho probe, use the following commands starting in global configuration mode:
1
rtr number
Enter RTR configuration mode.
2
type udpEcho dest-ipaddr {name | ip addr} dest-port port number [source-ipaddr {name | ip addr} source-port port number] [control {enable | disable}]
Defines a udpEcho probe.
Configure Optional Probe Characteristics
To configure optional characteristics, use one or more of the following commands in response time reporter configuration mode:
frequency seconds
Sets the frequency for RTR probe operation.
lsr-path {name | ip addr} [name | ip addr] ...
Specifies the path on which to measure the ICMP echo response time.
owner text
Configures the SNMP owner of the probe.
request-data-size bytes
Sets the protocol data size in the payload of the probe's request packet.
response-data-size bytes
Sets the protocol data size in the payload of the probe's response packet.
tag text
Logically links probes together in a group.
timeout milliseconds
Sets the amount of time the probe waits for a response from its request packet.
threshold milliseconds
Sets the rising threshold (hysteresis) that generates a reaction event and stores history information for the probe.
tos number
Defines the IP ToS byte for request packets. (Only valid on the echo probe in this release.)
Checks each probe response for corruption.
Capture Statistics and Collect Error Information
The main purpose of the probe is to capture statistics and collect error information. By default, the following information is captured and collected:
•
•
•
•
•
•
•
A statistical distribution of response times can be thought of as a set of buckets that holds the results of a probe. Each bucket holds the completion count that falls into that specific time interval. To modify the time intervals use the statistics-distribution-interval command. To modify the number of buckets use the distributions-of-statistics-kept command. For example, if the statistics-distribution-interval is 20 ms and the distributions-of-statistics-kept is 3 (buckets a, b and c) and 3 round-trip time (RTT) operations are performed with response times of 10 ms, 15 ms, and 30 ms, then the completion count for the buckets is 2 for a, 1 for b, and 0 for c.
In most situations, you do not need to modify the time intervals or number of buckets (also referred to as size). You should change the size only when distribution adjustments are needed (for example, when performing statistical modeling of your network).
To control how many and what type of statistics are collected on the router, use the following optional commands in response time reporter configuration mode:
statistics-distribution-interval milliseconds
Sets the time interval for each statistical distribution.
Sets the number of buckets or statistical distributions kept during the probe's lifetime. Size is the number of buckets that contain data counts for their intervals.
Collects pathEcho statistical distributions per hop per path. Size specifies the number of hops for which statistics are collected per path for each probe.
Collects statistical distributions for multiple paths. Size specifies the number of paths for which statistical distribution buckets are maintained per hour for each probe.
hours-of-statistics-kept hours
Sets the number of hours for which statistics are maintained for the probe.
Note
Collect History
The RTR can collect data samples for a given probe; these samples are called history data. By default, history data is not collected. When history collection is enabled, RTR collects the last n data points. The number of data points are configured using the buckets-of-history-kept command.
Additionally, when collecting history, RTR introduces the concept of lives. A life is defined as the operational lifetime of a probe. When a probe is stopped and restarted, data is kept in new life entries (if the number of entries is 2 or less). If the number of entries is more than 2, the oldest entry is overwritten by the new entry.
Note
To control how much and what type of history is collected on the router, use the following commands in response time reporter configuration mode. The first command is required; the remaining commands are optional.
1
samples-of-history-kept samples
For a pathEcho probe, sets the number of hops in a path. For all other probes, RTR sets the number of samples to 1.
2
For a pathEcho probe, sets the number of paths to store. For all other probes, sets the number (size) of data points to be kept.
3
lives-of-history-kept lives
Enables history collection and sets the number of lives maintained in the history table for the probe.
4
filter-for-history {none | all | overthreshold | failures}
Defines the type of information kept in the history table for the probe. This is a required command to enable history. All, overthreshold, or failures must be specified for history to work.
To disable history collection, use the default value (0) for the lives-of-history-kept command rather than the filter-for-history none command. The lives-of-history-kept command disables history collection before the probe's operation is attempted, and the filter-for-history command with the none keyword checks for history inclusion after the probe's operation attempt is made.
Set Reaction Conditions
You can configure the probe to send threshold notifications and use those notifications to trigger additional collection of time delay statistics. You can also configure the probe to send notifications when the probe loses connection, reestablishes connections, times out, and first succeeds after a timeout.
To configure the probe's reaction conditions, use the following optional commands in global configuration mode:
1
rtr reaction-configuration number [connection-loss-enable] [timeout-enable] [threshold-falling milliseconds] [threshold-type option] [action-type option]
Configures certain actions (for example, checking for connection losses or timeouts) to occur based on events controlled by the RTR.
2
rtr reaction-trigger number target-number
Defines the target probe to make the transition from a pending state to an active state when one of the trigger action-type options is defined for the probe.
Schedule the Probe
After you have configured the probe, you must schedule the probe to begin capturing statistics and collecting error information. When scheduling a probe, it can start immediately or start at a certain month, day, or hour. The pending keyword is used when setting the probe to start at a later time. The pending keyword is also an internal state of the probe visible through SNMP. The pending state is also used when a probe is a reaction probe waiting to be triggered.
To schedule an RTR probe, use the following command in global configuration mode:
rtr schedule number [life seconds] [start-time {pending | now | hh:mm [month day | day month]}] [ageout seconds]
Schedules the probe by configuring the time parameters.
Note
If the probe is in a pending state (the default), you can define the conditions under which the probe makes the transition from pending to active with the rtr reaction-trigger command. When the probe is in an active state it immediately begins collecting information.
Reset the Probe
To shut down the RTR (stop all probes and clear the RTR configuration), use the following command in global configuration mode:
CautionUse the rtr reset command only in extreme situations such as the incorrect configuration of a number of probes. The rtr reset command reconfigures the router to its startup configuration
In addition to stopping all probes and clearing the RTR configuration information, the rtr reset command returns the RTR feature to the startup condition. This command does not reread the configuration stored in NVRAM. You must retype the RTR's configuration or perform a config memory command.
Monitor the Response Time Reporter Feature
To display information about the status and configuration of the RTR feature, use the following commands in EXEC mode. You can display information in a tabular or full format. Tabular format displays information in a column reducing the number of screens required to display the information. Full format displays all information using identifiers next to each displayed value.
show rtr application [tabular | full]
Displays global information about the RTR feature.
Displays authentication information.
show rtr collection-statistics [number] [tabular | full]
Displays error totals collected for all probes or a specified probe.
show rtr configuration [number] [tabular | full]
Displays configuration values including all defaults for all probes or a specified probe.
show rtr distributions-statistics [number] [tabular | full]
Displays statistical distribution information (captured response times) for all probes or a specified probe.
show rtr history [number] [tabular | full]
Displays history collected for all probes or a specified probe.
show rtr operational-state [number] [tabular | full]
Displays the operational state of all probes or a specified probe.
show rtr reaction-trigger [number] [tabular | full]
Displays the reaction trigger information for all probes or a specified probe.
Displays responder information.
show rtr totals-statistics [number] [tabular | full]
Displays the total statistic values (accumulation of error counts and completions) for all probes or a specified probe.
Configuration Examples
The following sections contain examples of setting up probes on the router to monitor network performance and send notifications:
•
•
•
Configure a tcpConnect Probe
The example in shows a tcpConnect probe configured from Router B to the Telnet port (TCP port 23) of IP Host 1 (IP address 10.0.0.1).
Figure 1 Configuring a tcpConnect Probe
Configuration for Router B
RouterB(config)# rtr 5RouterB(config-rtr)# type tcpConn dest-ipaddr 10.0.0.1 dest-port 23 control disableRouterB(config-rtr)# exitRouterB(config)# rtr schedule 5 start nowIn the example the control protocol for the probe is disabled. RTR collector uses the RTR control protocol to notify the RTR responder on the responder router to enable the target port temporarily. This action allows the responder to respond to the probe packet. In this case, because the target is not a router and a well known TCP port is used, there is no need to send the control message.
Configure a udpEcho Probe
The example in shows a udpEcho probe configured from Router B to UDP port 888 on Router A (IP address 20.0.0.1).
Figure 2
Configuring a udpEcho Probe
Note
Configuration for Router A
RouterA(config)# key chain rtr-keyRouterA(config-keychain)# key 1RouterA(config-keychain-key)# key-string secreteRouterA(config-keychain-key)# exitRouterA(config-keychain)# exitRouterA(config)# rtr key-chain rtr-keyRouterA(config)# rtr responderIn the configuration for Router B we create a keychain called "rtr-key." The rtr key-chain command enables RTR MD5 authentication on the control messages.
Configuration for Router B
RouterB(config)# key chain rtr-keyRouterB(config-keychain)# key 1RouterB(config-keychain-key)# key-string secreteRouterB(config-keychain-key)# exitRouterB(config-keychain)# exitRouterB(config)# rtr key-chain rtr-keyRouterB(config)# rtr 6RouterB(config-rtr)# type udpEcho dest-ipaddr 20.0.0.1 dest-port 888 control enableRouterB(config-rtr)# exitRouterB(config)# rtr schedule 6 start nowPerform Normative Analysis for SNA LU2
The example in Figure 3 shows probe 1 configured from Router A to Host 2, and Probe 2 is configured from Router B to Host 2. This configuration allows normative analysis of the network to determine a baseline from which triggers (and general reactions) are configured. Also, two SNA PUs must be configured: CWBC0A and CWBC0B. For information on configuring PUs, see the dspu host or the sna host commands in the Bridging and IBM Networking Command Reference.
Figure 3 Configure Probes for Normative Analysis—SNA LU2
Configuration for Router A
RouterA(config)# rtr 1RouterA(config-rtr)# type echo protocol snaLU2EchoAppl CWBC0ARouterA(config-rtr)# exitRouterA(config)# rtr schedule 1 start-time nowRouterA(config)# exitConfiguration for Router B
RouterB(config)# rtr 2RouterB(config-rtr)# type echo protocol snaLU2EchoAppl CWBC0BRouterB(config-rtr)# exitRouterB(config)# rtr schedule 1 start-time nowRouterB(config)# exitConfiguration Files for Router A and Router B
After you save the configurations for Router A and Router B (using the copy running-config startup-config command), information is stored in the configuration files. kept commands are added automatically to the configuration file because they differ depending on the type you specify for the probe. The following information is stored:
!Router A Configuration File! Router A's PU Configurationsna host CWBC0A xid-snd 05dcc00a rmac 4001.3745.1088 rsap 4 lsap 12 focalpointrtr 1type echo protocol snaLU2EchoAppl CWBC0Apaths-of-statistics-kept 1hops-of-statistics-kept 1samples-of-history-kept 1rtr schedule 1 start-time now!Router B Configuration File!Router B's PU Configuration from the Configuration File:sna host CWBC0B xid-snd 05dcc00b rmac 4001.3745.1088 rsap 4 lsap 12 focalpointrtr 2type echo protocol snaLU2EchoAppl CWBC0Bpaths-of-statistics-kept 1hops-of-statistics-kept 1samples-of-history-kept 1rtr schedule 2 start-time nowPerform Troubleshooting for IP/ICMP
The example in Figure 4 shows that Probe 3 is configured from Router B to Router A to perform network troubleshooting and identify network problems that configure triggers and general reactions.
Figure 4 Configure a Probe for Troubleshooting—IP/ICMP
This example sets up a pathEcho (with history) pending entry from Router B to Router A via IP/ICMP. It attempts to execute 3 times in 25 seconds (first attempt starts at 0 seconds) and keeps those 3 times with 3 buckets. The entry can be started 5 times before wrapping over stored history (lives-of-history-kept = 5). Because this configuration keeps history, it uses more RAM on the router.
Configuration for Router B
RouterB(config)# rtr 3RouterB(config-rtr)# type pathEcho protocol ipIcmpEcho RouterARouterB(config-rtr)# frequency 10RouterB(config-rtr)# lives-of-history-kept 5RouterB(config-rtr)# buckets-of-history-kept 3RouterB(config-rtr)# filter-for-history allRouterB(config-rtr)# exitRouterB(config)# rtr schedule 3 life 25RouterB(config)# exitConfiguration File for Router B
After you save the configuration (using the copy running-config startup-config command) the information is stored in the configuration file. Note the addition of commands in the configuration file. They are automatically included because they differ depending on the type you specify for the probe. The following information is stored:
rtr 3type pathEcho protocol ipIcmpEcho 172.28.161.21frequency 10response-data-size 1lives-of-history-kept 5buckets-of-history-kept 3filter-for-history allrtr schedule 3 life 25 start-time pendingConfigure a Trigger for Connection Loss
Figure 5 shows Probes 1, 2, and 3 in the network. This example shows how to configure a trigger if Probe 2 encounters a connection loss from Router B to Host 2. If a connection loss occurs between Router B and Host 2, a trap is issued, an SNA NMVT Alert is issued, and the Probe 3 state is changed to active.
Figure 5 Configure a Trigger for Connection Loss
Router B Configuration
RouterB(config)# rtr reaction-configuration 2 connection-loss-enableaction-type trapNmvtAndTriggerRouterB(config)# rtr reaction-trigger 2 3
Note
Command Reference
The following commands are new or modified in Cisco IOS Release 12.0(3)T.
•
The following commands have not been modified, but are necessary to configure RTR:
•
•
•
•
•
•
Note
buckets-of-history-kept
To set the number of history buckets that are kept during the RTR probe's lifetime, use the buckets-of-history-kept response time reporter configuration command. Use the no form of this command to return to the default value.
buckets-of-history-kept size
no buckets-of-history-keptSyntax Description
Default
50 buckets
Command Mode
Response time reporter configuration
Usage Guidelines
This command first appeared in Cisco IOS Release 11.2.
An RTR probe collects history and captures statistics. By default, history is not collected. When a problem arises where history is useful (for example, a large number of timeouts are occurring), you can configure the lives-of-history-kept command to collect history. You can optionally adjust the buckets-of-history-kept, filter-for-history, and samples-of-history-kept commands.
RTR keeps the last n buckets of data.
Note
If history is collected, each bucket contains one or more history entries from the probe. When the probe type is pathEcho, an entry is created for each hop along the path that the probe takes to reach its destination. The type of entry stored in the history table is controlled by the filter-for-history command. The total number of entries stored in the history table is controlled by the combination of samples-of-history-kept, buckets-of-history-kept, and lives-of-history-kept commands.
Each time the probe starts an RTR operation, a new bucket is created until the number of history buckets matches the specified size or the probe's lifetime expires. History buckets do not wrap. The probe's lifetime is defined by the rtr schedule command. The probe starts an RTR operation based on the seconds specified by the frequency command.
Example
In the following example, probe 1 is configured to keep 25 history buckets during the probe's lifetime:
rtr 1type echo protocol ipIcmpEcho 172.16.161.21buckets-of-history-kept 25lives-of-history-kept 1Related Commands
filter-for-history
lives-of-history-kept
rtr
rtr schedule
samples-of-history-keptdistributions-of-statistics-kept
To set the number of statistic distributions kept per hop during the RTR probe's lifetime, use the distributions-of-statistics-kept response time reporter configuration command. Use the no form of this command to return to the default value.
distributions-of-statistics-kept size
no distributions-of-statistics-keptSyntax Description
Default
1 distribution
Command Mode
Response time reporter configuration
Usage Guidelines
This command first appeared in Cisco IOS Release 11.2.
In most situations, you do not need to change the statistic distribution size for the RTR. Only change the size when distributions are needed (for example, when performing statistical modeling of your network).
Note
When the number of distributions reaches the specified size, no further distribution information is stored.
Example
In the following example, the distribution is set to 5 and the distribution interval is set to 10 ms. In this configuration the first distribution contains statistics from 0 to 9 ms, the second distribution contains statistics from 10 to 19 ms, the third distribution contains statistics from 20 to 29 ms, the fourth distribution contains statistics from 30 to 39 ms, and the fifth distribution contains statistics from 40 ms to infinity.
rtr 1type echo protocol ipIcmpEcho 172.16.161.21distributions-of-statistics-kept 5statistics-distribution-interval 10Related Commands
hops-of-statistics-kept
hours-of-statistics-kept
paths-of-statistics-kept
rtr
statistics-distribution-intervalfilter-for-history
To define the type of information kept in the history table for the RTR probe, use the filter-for-history response time reporter configuration command. Use the no form of this command to return to the default value.
filter-for-history {none | all | overThreshold | failures}
no filter-for-history {none | all | overThreshold | failures}Syntax Description
Default
No history is kept.
Command Mode
Response time reporter configuration
Usage Guidelines
This command first appeared in Cisco IOS Release 11.2.
Use the filter-for-history command to control what is stored in the history table for the RTR. To control how much history is saved in the history table, use the lives-of-history-kept, buckets-of-history-kept, and the samples-of-history-kept commands.
A probe collects history and captures statistics. By default, history is not collected. When a problem arises where history is useful (for example, a large number of timeouts are occurring), you can configure the lives-of-history-kept command to collect history.
Note
Example
In the following example, only probe packets that fail are kept in the history table:
rtr 1type echo protocol ipIcmpEcho 172.16.161.21lives-of-history-kept 1filter-for-history failuresRelated Commands
buckets-of-history-kept
lives-of-history-kept
rtr
samples-of-history-keptfrequency
To set the rate at which the RTR probe starts a response time operation, use the frequency response time reporter configuration command. Use the no form of this command to return to the default value.
frequency second
no frequencySyntax Description
Default
60 seconds
Command Mode
Response time reporter configuration
Usage Guidelines
This command first appeared in Cisco IOS Release 11.2.
CautionFor normal operation, do not set the frequency value to less than 60 seconds for the following reasons: it is not needed when keeping statistics (the default), and it can slow down the WAN because of the potential overhead that numerous probes can cause.
If the probe takes longer to execute the current RTR operation than the specified frequency value, a statistics counter called busy is incremented in lieu of starting a second operation.
The value specified for the frequency command cannot be less than the value specified for the timeout command.
Example
In the following example, the probe is configured to execute an RTR operation every 90 seconds:
rtr 1type echo protocol ipIcmpEcho 172.16.1.176frequency 90Related Commands
hops-of-statistics-kept
To set the number of hops for which statistics are maintained per path for the RTR probe, use the hops-of-statistics-kept response time reporter configuration command. Use the no form of this command to return to the default value.
hops-of-statistics-kept size
no hops-of-statistics-keptSyntax Description
size
Number of hops for which statistics are maintained per path. The default is 16 hops for type pathEcho and 1 hop for all other types.
Default
16 hops for type pathEcho
1 hop for all other types
Command Mode
Response time reporter configuration
Usage Guidelines
This command first appeared in Cisco IOS Release 11.2.
One hop is the passage of a timed packet from this router to another network device. The other network device is a device along the path to the destination (including the destination) when the probe type is pathEcho, or is just the destination when the type is echo.
When the number of hops reaches the specified size, no further hop information is stored.
Example
In the following example, the statistics for probe 2 are maintained for only 10 hops:
rtr 2type pathecho protocol ipIcmpEcho 172.16.1.177hops-of-statistics-kept 10Related Commands
distributions-of-statistics-kept
hours-of-statistics-kept
paths-of-statistics-kept
rtr
statistics-distribution-intervalhours-of-statistics-kept
To set the number of hours for which statistics are maintained for the RTR probe, use the hours-of-statistics-kept response time reporter configuration command. Use the no form of this command to return to the default value.
hours-of-statistics-kept hours
no hours-of-statistics-keptSyntax Description
Default
2 hours
Command Mode
Response time reporter configuration
Usage Guidelines
This command first appeared in Cisco IOS Release 11.2.
When the number of hours exceeds the specified value, the statistics table wraps (the oldest information is replaced by newer information).
Example
In the following example, the statistics for probe 2 are maintained for 3 hours:
rtr 2type pathecho protocol ipIcmpEcho 172.16.1.177hours-of-statistics-kept 3Related Commands
distributions-of-statistics-kept
hops-of-statistics-kept
paths-of-statistics-kept
rtr
statistics-distribution-intervallives-of-history-kept
To set the number of lives maintained in the history table for the RTR probe, use the lives-of-history-kept response time reporter configuration command. Use the no form of this command to return to the default value.
lives-of-history-kept lives
no lives-of-history-keptSyntax Description
Default
0 lives
Command Mode
Response time reporter configuration
Usage Guidelines
This command first appeared in Cisco IOS Release 11.2.
The default value means that history is not collected for the probe. To disable history collection, use the default value for the lives-of-history-kept command rather than the filter-for-history none command. The lives-of-history-kept command disables history collection before the probe's operation is attempted, and the filter-for-history command checks for history inclusion after the probe makes an operation attempt.
When the number of lives exceeds the specified value, the history table wraps (the oldest information is replaced by newer information).
When a probe makes a transition from pending to active, a life starts. When a probe's life ends, the probe makes a transition from active to pending.
Example
In the following example, the history for probe 1 is maintained for 5 lives:
rtr 1type echo protocol ipIcmpEcho 172.16.1.176lives-of-history-kept 5Related Commands
buckets-of-history-kept
filter-for-history
rtr
samples-of-history-keptlsr-path
To define a loose source routing (LSR) path for IP echo probe, use the lsr-path response time reporter configuration command. Use the no form of this command to remove the definition.
lsr-path {name | ip addr} [{name | ip addr}] ...
no lsr-pathSyntax Description
Default
LSR path is disabled.
Command Mode
Response time reporter configuration
Usage Guidelines
This command first appeared in Cisco IOS Release 12.0(3)T.
The maximum number of hops available is 8 when an LSR path is configured.
Example
In the following example, LSR is defined for the echo probe with IP address 172.16.1.176:
rtr 1type echo protocol ipIcmpEcho 172.16.1.176lsr-path 172.18.4.149 172.18.26.155Related Commands
owner
To configure the SNMP owner of the RTR probe, use the owner response time reporter configuration command. Use the no form of this command to return to the default value.
owner text
no ownerSyntax Description
Default
No owner is specified.
Command Mode
Response time reporter configuration
Usage Guidelines
This command first appeared in Cisco IOS Release 11.2.
The owner name contains one or more of the following: ASCII form of the network management station's transport address, network management station name (the domain name), and network management personnel's name, location, or phone number. In some cases, the agent is the owner of the probe. In these cases, the name can begin with "agent."
Example
In the following example, the owner is set for probe 1:
rtr 1type echo protocol ipIcmpEcho 172.16.1.176owner 172.16.1.189 cwb.cisco.com John Doe RTP 555-1212Related Commands
paths-of-statistics-kept
To set the number of paths for which statistics are maintained per hour for the RTR probe, use the paths-of-statistics-kept response time reporter configuration command. Use the no form of this command to return to the default value.
paths-of-statistics-kept size
no paths-of-statistics-keptSyntax Description
size
Number of paths for which statistics are maintained per hour. The default is 5 paths for type pathEcho and 1 path for all other types.
Default
5 paths for type pathEcho
1 path for all other types
Command Mode
Response time reporter configuration
Usage Guidelines
This command first appeared in Cisco IOS Release 11.2.
A path is the route the probe's request packet takes through the network to get to its destination. The probe may take a different path to reach its destination for each RTR operation.
When the number of paths reaches the specified size, no further path information is stored.
Example
In the following example, the statistics for probe 2 are maintained for only 3 paths:
rtr 2type pathEcho protocol ipIcmpEcho 172.16.1.177paths-of-statistics-kept 3Related Commands
distributions-of-statistics-kept
hops-of-statistics-kept
hours-of-statistics-kept
rtr
statistics-distribution-intervalrequest-data-size
To set the protocol data size in the payload of the RTR probe's request packet, use the request-data-size response time reporter configuration command. Use the no form of this command to return to the default value.
request-data-size byte
no request-data-sizeSyntax Description
Default
1 byte for ICMP echo probe
4 bytes for udpEcho probe
Command Mode
Response time reporter configuration
Usage Guidelines
This command first appeared in Cisco IOS Release 11.2.
When the protocol name has the suffix "appl," the packet uses both a request and respond data size (see the response-data-size command), and the data size is 12 bytes smaller than the normal payload size (this 12-byte data size is the ARR Header used to control send and data response sizes).
Example
In the following example, the request packet size for probe 3 is set to 40 bytes:
rtr 3type echo protocol snalu0echoappl cwbc0arequest-data-size 40Related Commands
response-data-size
To set the protocol data size in the payload of the RTR probe's response packet, use the response-data-size response time reporter configuration command. Use the no form of this command to return to the default value.
response-data-size byte
no response-data-sizeSyntax Description
Default
For "appl" protocols, 0 bytes
For all others, the same value as specified in the request-data-size command
Command Mode
Response time reporter configuration
Usage Guidelines
This command first appeared in Cisco IOS Release 11.2.
+The response-data-size command is only applicable for "appl" protocols defined with the type command (for example, snalu0echoappl). When the protocol ends in "appl," the response data size is 12 bytes smaller than normal payload size.
Example
In the following example, the response packet size for probe 3 is set to 1440 bytes:
rtr 3type echo protocol snalu0echoappl cwbc0aresponse-data-size 1440Related Commands
rtr
To configure an RTR probe, use the rtr global configuration command. Use the no form of this command to remove all configuration information for a probe including the probe's schedule, reaction configuration, and reaction triggers.
rtr number
no rtr numberSyntax Description
Default
None
Command Mode
Global configuration
Usage Guidelines
This command first appeared in Cisco IOS Release 11.2.
A probe is used for the purpose of collecting response time information.
RTR allows a maximum of 500 probes.
The RTR feature allows customers to monitor the performance of their network, network resources, and applications by measuring response times and availability. With this feature, you can perform troubleshooting, problem notification, and preliminary problem analysis. For more information, refer to the "Monitoring the Router and Network" chapter in the Configuration Fundamentals Configuration Guide and the Cisco Round-Trip Time Monitor (RTTMON) MIB.
The rtr command places you in response time reporter configuration mode.
Note
Use the following response time reporter configuration commands to configure the probe's characteristics:
•
•
•
After you configure a probe, you must schedule it. For information on scheduling a probe, refer to the rtr schedule command. You can also optionally set reaction triggers for the probe. For information on reaction triggers, refer to the rtr reaction-configuration and rtr reaction-trigger commands.
To display the probe's current configuration settings, use the show rtr configuration EXEC command.
Example
In the following example, probe 1 is configured to perform end-to-end response time operations using an SNA LU Type 0 connection with the host name cwbc0a. Only the type command is required; all others are optional.
rtr 1type echo protocol snalu0echoappl cwbc0arequest-data-size 40response-data-size 1440
Note
Related Commands
rtr reaction-configuration
rtr reaction-trigger
rtr reset
rtr schedulertr key-chain
To enable RTR control message authentication and specify an MD5 key chain, use the rtr key-chain global configuration command. Use the no form of this command to remove control message authentication.
rtr key-chain name
no rtr key-chainSyntax Description
Default
None
Command Mode
Global configuration
Usage Guidelines
This command first appeared in Cisco IOS Release 12.0(3)T.
The authentication configuration on the RTR collector and responder must be the same. Both sides must configure the same key chain or both sides must not use authentication.
Example
In the following example, the RTR control message uses MD5 authentication, and the key chain name is rtr:
rtr key-chain rtrRelated Commands
rtr reaction-configuration
To configure certain actions to occur based on events under the control of the RTR, use the rtr reaction-configuration global configuration command. Use the no form of this command to return to the probe's default values.
rtr reaction-configuration number [connection-loss-enable] [timeout-enable]
[threshold-falling milliseconds] [threshold-type option] [action-type option]
no rtr reaction-configuration numberSyntax Description
Default
No reactions are generated.
Command Mode
Global configuration
Usage Guidelines
This command first appeared in Cisco IOS Release 11.2.
Triggers are used for diagnostics purposes and are not used in normal operation.
Use triggers to determine where delays are happening in the network when excessive delays are seen on an end-to-end basis.
The reaction applies only to attempts to the target (that is, attempts to any hops along the path in pathEcho do not generate reactions).
To specify trap as an action-type option, you must configure the snmp-server community, snmp-server enable, and snmp-server host commands. Refer to the Configuration Fundamentals Configuration Guide for more information on these commands.
Examples
In the following example, probe 19 sends an SNMP trap when there is an over threshold or falling threshold violation:
rtr reaction-configuration 19 threshold-type immediate action-type trapOnlyshows that an alert (rising trap) is issued immediately when the response time exceeds the rising threshold and a resolution (falling trap) is issued immediately when the response time drops below the falling threshold.
Figure 6 Rising and Falling Thresholds
In the following example, traps are sent when probe 1 times out.
snmp-server enable traps rtrsnmp-server host 172.69.1.129 traps public rtrrtr 1type echo protocol ipIcmpEcho 172.69.1.129exitrtr reaction-configuration 1 timeout-enable action-type trapOnlyRelated Commands
rtr
rtr reaction-trigger
threshold
timeoutrtr reaction-trigger
To define a second RTR probe to make the transition from a pending state to an active state when one of the trigger action-type options are defined with the rtr reaction-configuration command, use the rtr reaction-trigger command. Use the no form of this command to remove the trigger combination.
rtr reaction-trigger number target-number
no rtr reaction-trigger numberSyntax Description
Default
None
Command Mode
Global configuration
Usage Guidelines
This command first appeared in Cisco IOS Release 11.2.
Triggers are used for diagnostics purposes and are not used in normal router operation.
The target probe must be scheduled to start using the rtr schedule command with the start-time pending option configured.
Example
In the following example, the state of probe 1 is changed from pending to active when probe 2's action-type occurs:
rtr reaction-trigger 2 1Related Commands
rtr
rtr reaction-configuration
rtr schedulertr reset
To shut down the RTR (stop all probes and clear the RTR configuration), use the rtr reset global configuration command.
rtr reset
Syntax Description
This command has no arguments or keywords.
Default
None
Command Mode
Global configuration
Usage Guidelines
This command first appeared in Cisco IOS Release 11.2.
CautionUse the rtr reset command only in extreme situations (for example, if the incorrect number of probes is configured.
The rtr reset command stops all probes, clears RTR configuration information, and returns the RTR feature to the startup condition. This command does not reread the RTR configuration stored in the startup configuration file in NVRAM. You must retype the configuration or perform a config memory command.
Example
The following example resets the RTR feature:
rtr resetRelated Commands
rtr responder
To enable the RTR responder, use the rtr responder global configuration command. Use the no form of this command to disable the responder.
rtr responder
no rtr responderSyntax Description
This command has no arguments or keywords.
Default
None
Command Mode
Global configuration
Usage Guidelines
This command first appeared in Cisco IOS Release 12.0(3)T.
Example
The following example enables the RTR responder:
rtr responderRelated Commands
rtr schedule
To configure the time parameters for an RTR probe, use the rtr schedule global configuration command. Use the no form of this command to stop the probe and restart it with the default parameters (pending).
rtr schedule number [life seconds] [start-time {pending | now | hh:mm [month day |
day month]}] [ageout seconds]
no rtr schedule numberSyntax Description
Default
Pending state (the probe is started but not actively collecting information).
Command Mode
Global configuration
Usage Guidelines
This command first appeared in Cisco IOS Release 11.2.
After you schedule the probe with the rtr schedule command, you cannot change the probe's configuration with the rtr global configuration command. To change the probe's configuration, use the no form of the rtr command and reenter the configuration information.
If the probe is in a pending state, you can define the conditions under which the probe makes the transition from pending to active with the rtr reaction-trigger and rtr reaction-configuration commands. When the probe is in an active state, it immediately begins collecting information.
The following time line shows the probe's age-out process:
W----------------------X----------------------Y----------------------ZWhere:
•
•
•
•
The ageout starts counting down at W and Y, is suspended between X and Y, and is reset to its configured size at Y.
It is possible for the probe to age out before it executes (Z can occur before X). To ensure that the probe does not age out, the difference between the probe's configuration time and start time (X and W) must be less than the ageout seconds.
Note
Example
In the following example, probe 25 begins collecting data at 3:00 p.m. on April 5. This probe ages out after 12 hours of inactivity, which can be before it starts or after it has finished with its life. When this probe ages out, all configuration information for the probe is removed (the configuration information is no longer in the running configuration file in RAM.
rtr schedule 25 life 43200 start-time 15:00 apr 5 ageout 43200Related Commands
rtr
rtr reaction-configuration
rtr reaction-triggersamples-of-history-kept
To set the number of entries kept in the history table for each bucket in the RTR probe, use the samples-of-history-kept response time reporter configuration command. Use the no form of this command to return to the default value.
samples-of-history-kept samples
no samples-of-history-keptSyntax Description
samples
Number of entries for each bucket kept in the history table. The default is 16 entries for type pathEcho and 1 entry for all other probes.
Default
16 entries for type pathEcho
1 entry for all other probes
Command Mode
Response time reporter configuration
Usage Guidelines
This command first appeared in Cisco IOS Release 11.2.
Use the samples-of-history-kept command to control how many entries are saved in the history table. To control the type of information that gets saved in the history table, use the filter-for-history command. To set how many buckets are created in the history table, use the buckets-of-history-kept command.
A probe collects history and captures statistics. By default, history is not collected. When a problem arises where history is useful (for example, a large number of timeouts are occurring), you can configure the lives-of-history-kept command to collect history.
Note
Example
In the following example, 10 entries are kept in the history table for each of the of probe's 3 lives:
rtr 1type pathecho protocol ipIcmpEcho 172.16.1.176lives-of-history-kept 3samples-of-history-kept 10Related Commands
buckets-of-history-kept
filter-for-history
lives-of-history-kept
rtr
show rtr historyshow rtr application
Use the show rtr application EXEC command to display global information about the RTR feature.
show rtr application [tabular | full]
Syntax Description
Default
All information is displayed.
Command Mode
EXEC
Usage Guidelines
This command first appeared in Cisco IOS Release 11.2.
Use the show rtr application command to display information such as supported operation types and supported protocols.
Sample Display
The following is sample output from the show rtr application command in full format:
Router# show rtr application fullResponse Time ReporterVersion: 1.0.0 Initial Round Trip Time MIBMax Packet Data Size (ARR and Data): 16384Time of Last Change in Whole RTR: *16:49:53.000 UTC Thu May 16 1996System Max Number of Entries: 20Supported Operation TypesType of Operation to Perform: echoType of Operation to Perform: pathEchoSupported ProtocolsProtocol Type: ipIcmpEchoProtocol Type: snaRUEchoProtocol Type: snaLU0EchoApplProtocol Type: snaLU2EchoApplRelated Commands
show rtr authentication
Use the show rtr authentication EXEC command to display RTR authentication information.
show rtr authentication
Syntax Description
This command has no arguments or keywords.
Command Mode
EXEC
Usage Guidelines
This command first appeared in Cisco IOS Release 12.0(3)T.
Use the show rtr authentication command to display information such as supported operation types and supported protocols.
Sample Display
The following is sample output from the show rtr application command:
Router# show rtr authenticationRTR control message uses MD5 authentication, key chain name is: rtrRelated Commands
show rtr collection-statistics
Use the show rtr collection-statistics EXEC command to display statistical information for all RTR probes or a specified probe.
show rtr collection-statistics [number] [tabular | full]
Syntax Description
Default
All information is displayed.
Command Mode
EXEC
Usage Guidelines
This command first appeared in Cisco IOS Release 11.2.
Use the show rtr collection-statistics command to display information such as the number of failed operations and the failure reason. You can also use the show rtr distribution-statistics and show rtr totals-statistics commands to display additional statistical information.
Sample Display
The following is sample output from the show rtr collection-statistics command in full format:
Router# show rtr collection-statistics 1 fullCollected StatisticsEntry Number: 1Start Time Index: *17:15:41.000 UTC Thu May 16 1996Path Index: 1Hop in Path Index: 1Number of Failed Operations due to a Disconnect: 0Number of Failed Operations due to a Timeout: 0Number of Failed Operations due to a Busy: 0Number of Failed Operations due to a No Connection: 0Number of Failed Operations due to an Internal Error: 0Number of Failed Operations due to a Sequence Error: 0Number of Failed Operations due to a Verify Error: 0Target Address: 172.16.1.176Related Commands
show rtr configuration
show rtr distributions-statistics
show rtr totals-statisticsshow rtr configuration
Use the show rtr configuration EXEC command to display configuration values including all defaults for all RTR probes or the specified probe.
show rtr configuration [number] [tabular | full]
Syntax Description
Default
All information is displayed.
Command Mode
EXEC
Usage Guidelines
This command first appeared in Cisco IOS Release 11.2.
Sample Display
The following is sample output from the show rtr configuration command in full format:
Router# show rtr configuration 1 fullComplete Configuration Table (includes defaults)Entry Number: 1Owner:"Sample Owner"Tag:"Sample Group"Type of Operation to Perform:echoReaction and History Threshold (milliseconds):5000Operation Frequency (seconds):60Operation Timeout (milliseconds):5000Verify Data:FALSEStatus of Entry (SNMP RowStatus):activeProtocol Type:ipIcmpEchoTarget Address:1.0.0.1Source Address:0.0.0.0Target Port:0Source Port:0Request Size (ARR data portion):28Response Size (ARR data portion):1Control Packets:enabledLoose Source Routing:disabledLSR Path:Type of Service Parameters:0x0Life (seconds):3600Next Scheduled Start Time:Pending TriggerEntry Ageout:neverConnection Loss Reaction Enabled:FALSETimeout Reaction Enabled:FALSEThreshold Reaction Type:neverThreshold Falling (milliseconds):3000Threshold Count:5Threshold Count2:5Reaction Type:noneNumber of Statistic Hours kept:2Number of Statistic Paths kept:1Number of Statistic Hops kept:1Number of Statistic Distribution Buckets kept:1Statistic Distribution Interval (milliseconds):20Number of History Lives kept:0Number of History Buckets kept:15Number of History Samples kept:1History Filter Type:noneRelated Commands
show rtr application
show rtr collection-statistics
show rtr distributions-statistics
show rtr history
show rtr operational-state
show rtr reaction-trigger
show rtr totals-statisticsshow rtr distributions-statistics
Use the show rtr distributions-statistics EXEC command to display statistic distribution information (captured response times) for all RTR probes or a specified probe.
show rtr distributions-statistics [number] [tabular | full]
Syntax Description
Default
Information is displayed in tabular format.
Command Mode
EXEC
Usage Guidelines
This command first appeared in Cisco IOS Release 11.2.
The distributions statistics include the following information:
•
•
•
•
You can also use the show rtr collection-statistics and show rtr totals-statistics commands to display additional statistical information.
Sample Display
The following is sample output from the show rtr distributions-statistics command in tabular format:
Router# show rtr distributions-statistics tabularCaptured StatisticsMultiple Lines per EntryLine 1Entry = Entry NumberStartT = Start Time of Entry (hundredths of seconds)Pth = Path IndexHop = Hop in Path IndexDst = Time Distribution IndexComps = Operations CompletedOvrTh = Operations Completed Over ThresholdsSumCmp = Sum of Completion Times (milliseconds)Line 2SumCmp2L = Sum of Completion Times Squared Low 32 Bits (milliseconds)SumCmp2H = Sum of Completion Times Squared High 32 Bits (milliseconds)TMax = Completion Time Maximum (milliseconds)TMin = Completion Time Minimum (milliseconds)Entry StartT Pth Hop Dst Comps OvrTh SumCmpSumCmp2L SumCmp2H TMax TMin1 17417068 1 1 1 2 0 1288192 0 64 64Related Commands
show rtr collection-statistics
show rtr configuration
show rtr totals-statisticsshow rtr history
Use the show rtr history EXEC command to display history collected for all RTR probes or a specified probe.
show rtr history [number] [tabular | full]
Syntax Description
Default
Information displayed in tabular format.
Command Mode
EXEC
Usage Guidelines
This command first appeared in Cisco IOS Release 11.2.
The response return codes are listed in .
Table 1 Response Return Codes
1
Okay
2
Disconnected
3
Over threshold
4
Timeout
5
Busy
6
Not connected
7
Dropped
8
Sequence error
9
Verify error
10
Application specific
Sample Display
The following is sample output from the show rtr history command in tabular format:
Router# show rtr history tabularPoint by point HistoryMultiple Lines per EntryLine 1Entry = Entry NumberLifeI = Life IndexBucketI = Bucket IndexSampleI = Sample IndexSampleT = Sample Start TimeCompT = Completion Time (milliseconds)Sense = Response Return CodeLine 2 has the Target AddressEntry LifeI BucketI SampleI SampleT CompT Sense2 1 1 1 17436548 16 1AB 45 A0 162 1 2 1 17436551 4 1AC 12 7 292 1 2 2 17436551 1 1AC 12 5 222 1 2 3 17436552 4 1AB 45 A7 222 1 2 4 17436552 4 1AB 45 A0 16Related Commands
show rtr operational-state
Use the show rtr operational-state EXEC command to display the operational state of all RTR probes or a specified probe.
show rtr operational-state [number] [tabular | full]
Syntax Description
Default
Full information is displayed.
Command Mode
EXEC
Usage Guidelines
This command first appeared in Cisco IOS Release 11.2.
Use the show rtr operational-state command to determine whether a connection loss, timeout, or over threshold occurred; how much life the probe has left; whether the probe is active; and the completion time. This command also displays the results of the latest operation attempt.
Sample Display
The following is sample output from the show rtr operational-state command in full format:
Router# show rtr operational-state 1 fullCurrent Operational StateEntry Number: 1Modification Time: *17:15:41.000 UTC Thu May 16 1996Diagnostics Text:Last Time this Entry was Reset: NeverNumber of Octets in use by this Entry: 2438Connection Loss Occurred: FALSETimeout Occurred: FALSEOver Thresholds Occurred: FALSENumber of Operations Attempted: 6Current Seconds Left in Life: 3336Operational State of Entry: activeLatest Completion Time (milliseconds): 60Latest Operation Return Code: okLatest Operation Start Time: *17:19:41.000 UTC Thu May 16 1996Latest Target Address: 172.16.1.176Related Commands
show rtr reaction-trigger
Use the show rtr reaction-trigger EXEC command to display the reaction trigger information for all RTR probes or a specified probe.
show rtr reaction-trigger [number] [tabular | full]
Syntax Description
Default
Full information is displayed.
Command Mode
EXEC
Usage Guidelines
This command first appeared in Cisco IOS Release 11.2.
Use the show rtr reaction-trigger command to display the configuration status and operational state of target probes that are triggered as defined with the rtr reaction-configuration command.
Sample Display
The following is sample output from the show rtr reaction-trigger command in full format:
Router# show rtr reaction-trigger 1 fullReaction TableEntry Number: 1Target Entry Number: 2Status of Entry (SNMP RowStatus): activeOperational State: pendingRelated Commands
show rtr responder
Use the show rtr responder EXEC command to display RTR responder information.
show rtr responder
Syntax Description
This command has no arguments or keywords.
Command Mode
EXEC
Usage Guidelines
This command first appeared in Cisco IOS Release 12.0(3)T.
Use the show rtr responder command to display information about recent sources of RTR control messages, such as who has sent recent control messages and who has sent invalid control messages.
Sample Display
The following is sample output from the show rtr responder command:
Router# show rtr responderRTR Responder is: EnabledNumber of control message received: 19 Number of errors: 1Recent sources:4.0.0.1 [19:11:49.035 UTC Sat Dec 2 1995]4.0.0.1 [19:10:49.023 UTC Sat Dec 2 1995]4.0.0.1 [19:09:48.707 UTC Sat Dec 2 1995]4.0.0.1 [19:08:48.687 UTC Sat Dec 2 1995]4.0.0.1 [19:07:48.671 UTC Sat Dec 2 1995]Recent error sources:4.0.0.1 [19:10:49.023 UTC Sat Dec 2 1995] RTT_AUTH_FAILRelated Commands
show rtr totals-statistics
Use the show rtr totals-statistics EXEC command to display the total statistical values (accumulation of error counts and completions) for all RTR probes or a specified probe.
show rtr totals-statistics [number] [tabular | full]
Syntax Description
Default
Full information is displayed.
Command Mode
EXEC
Usage Guidelines
This command first appeared in Cisco IOS Release 11.2.
The total statistics consist of the following items:
•
•
•
•
You can also use the show rtr distributions-statistics and show rtr collection-statistics commands to display additional statistical information.
Sample Display
The following is sample output from the show rtr totals-statistics command in full format:
Router# show rtr totals-statistics fullStatistic TotalsEntry Number: 1Start Time Index: *17:15:41.000 UTC Thu May 16 1996Age of Statistics Entry (hundredths of seconds): 48252Number of Initiations: 10Related Commands
show rtr collection-statistics
show rtr configuration
show rtr distributions-statisticsstatistics-distribution-interval
To set the time interval for each statistics distribution kept for the RTR, use the statistics-distribution-interval response time reporter configuration command. Use the no form of this command to return to the default value.
statistics-distribution-interval milliseconds
no statistics-distribution-intervalSyntax Description
milliseconds
Number of milliseconds used for each statistics distribution kept. The default is 20 ms.
Default
20 ms
Command Mode
Response time reporter configuration
Usage Guidelines
This command first appeared in Cisco IOS Release 11.2.
In most situations, you do not need to change the statistical distribution interval or size. Change the interval or size only when distributions are needed, (for example, when performing statistical modeling of your network). To set the statistical distributions size, use the distributions-of-statistics-kept response time reporter configuration command.
Example
In the following example, the distribution is set to 5 and the distribution interval is set to 10 ms. The first distribution will contain statistics from 0 to 9 ms, the second distribution will contain statistics from 10 to 19 ms, the third distribution will contain statistics from 20 to 29 ms, the fourth distribution will contain statistics from 30 to 39 ms, and the fifth distribution will contain statistics from 40 ms to infinity.
rtr 1type echo protocol ipIcmpEcho 172.28.161.21distribution-of-statistics-kept 5statistics-distribution-interval 10Related Commands
distributions-of-statistics-kept
hops-of-statistics-kept
hours-of-statistics-kept
paths-of-statistics-kept
rtrtag
To create a user-specified identifier for an RTR probe, use the tag RTR configuration command. It is normally used to logically link probes in a group. Use the no form of this command to remove a tag from a probe.
tag text
no tagSyntax Description
Default
None
Command Mode
Response time reporter configuration
Usage Guidelines
This command first appeared in Cisco IOS Release 11.2.
Tags can be used to support automation (for example, by using the same tag for two different probes on two different routers echoing the same target).
Example
In the following example, probe 1 is tagged with the label bluebell:
rtr 1type echo protocol ipIcmpEcho 172.16.1.176tag bluebellRelated Commands
threshold
To set the rising threshold (hysteresis) that generates a reaction event and stores history information for the RTR probe, use the threshold response time reporter configuration command. Use the no form of this command to return to the default value.
threshold millisecond
no thresholdSyntax Description
millisecond
Number of milliseconds required for a rising threshold to be declared. The default value is 5000 ms.
Default
5000 ms
Command Mode
Response time reporter configuration
Usage Guidelines
This command first appeared in Cisco IOS Release 11.2.
The value specified for the threshold command must not exceed the value specified for the timeout command.
The threshold value is used by the rtr reaction-configuration and filter-for-history commands.
Example
In the following example, the threshold for probe 1 is set to 2500 ms:
rtr 1type echo protocol ipIcmpEcho 172.16.1.176threshold 2500Related Commands
filter-for-history
rtr
rtr reaction-configurationtimeout
To set the amount of time the RTR probe waits for a response from its request packet, use the timeout response time reporter configuration command. Use the no form of this command to return to the default value.
timeout millisecond
no timeoutSyntax Description
millisecond
Number of milliseconds the probe waits to receive a response from its request packet. The default is 5000 ms.
Default
5000 ms
Command Mode
Response time reporter configuration
Usage Guidelines
This command first appeared in Cisco IOS Release 11.2.
Use the timeout command to set how long the probe waits to receive a response, and use the frequency command to set the rate at which the probe starts an RTR operation.
The value specified for the timeout command cannot be greater than the value specified for the frequency command.
Example
In the following example, the timeout is set for 2500 ms:
rtr 1type echo protocol ipIcmpEcho 172.16.1.176timeout 2500Related Commands
tos
To define a type of service byte in the IP header of an RTR probe, use the tos response time reporter configuration command. Use the no form of this command to return to the default value.
tos number
no tosSyntax Description
Default
The default type of service value is 0.
Command Mode
Response time reporter configuration
Usage Guidelines
This command first appeared in Cisco IOS Release 12.0(3)T.
Example
In the following example, probe 1 is configured as an echo probe using the IP/ICMP Echo protocol and the destination IP address 172.16.1.175. The ToS value is set to 0x80:
rtr 1type echo protocol ipIcmpEcho 172.16.1.176tos 0x80Related Commands
type echo
To configure an echo probe, use the type echo response time reporter configuration command. You must configure the probe's type before you can configure any of the other characteristics of the probe. Use the no form of this command to remove the type configuration for the probe.
type echo protocol type type-target
no type echo protocol type type-targetSyntax Description
Default
None
Command Mode
Response time reporter configuration
Usage Guidelines
This command first appeared in Cisco IOS Release 11.2.
Support of echo to a protocol and pathEcho to a protocol is dependent on the protocol type and implementation. In general most protocols support echo and few protocols support pathEcho.
Note
Example
In the following example, probe 10 is created and configured as echo using the IP/ICMP Echo protocol and the destination IP address 172.16.1.175:
rtr 10type echo protocol ipIcmpEcho 172.16.1.175Related Commands
type pathEcho
To configure a pathEcho probe, use the type pathEcho response time reporter configuration command. You must configure the probe's type before you can configure any of the other characteristics of the probe. Use the no form of this command to remove the type configuration for the probe.
type pathEcho protocol type type-target
no type pathEcho protocol type type-targetSyntax Description
Default
None
Command Mode
Response time reporter configuration
Usage Guidelines
This command first appeared in Cisco IOS Release 11.2.
Support of echo to a protocol and pathEcho to a protocol is dependent on the protocol type and implementation. In general most protocols support echo and few protocols support pathEcho.
Note
Example
In the following example, probe 10 is created and configured as pathEcho using the IP/ICMP Echo protocol and the destination IP address 172.16.1.175:
rtr 10type pathEcho protocol ipIcmpEcho 172.16.1.175Related Commands
type tcpConnect
To define a tcpConnect probe use the type tcpConnect response time reporter configuration command. Use the no form of this command to remove the type configuration for the probe.
type tcpConnect dest-ipaddr {name | ip addr} dest-port port number [source-ipaddr {name |
ip addr} source-port port number] [control {enable | disable}]
no type tcpConnect dest-ipaddr {name | ip addr} dest-port port numberSyntax Description
Default
The control protocol is enabled. Prior to sending a probe packet to the responder, the RTR collector sends a control message to the responder to enable the destination port.
Command Mode
Response time reporter configuration
Usage Guidelines
This command first appeared in Cisco IOS Release 12.0(3)T.
You must configure the probe's type before you can configure any of the other characteristics of the probe.
Example
In the following example, probe 11 is created and configured as a tcpConnect probe using the destination IP address 172.16.1.175 and the destination port 2400:
rtr 11type tcpConnect dest-ipaddr 172.16.1.175 dest-port 2400Related Commands
type udpEcho
To define a udpEcho probe use the type udpEcho response time reporter configuration command. Use the no form of this command to remove the type configuration for the probe.
type udpEcho dest-ipaddr {name | ip addr} dest-port port number [source-ipaddr {name |
ip addr} source-port port number] [control {enable | disable}]
no type udpEcho dest-ipaddr {name | ip addr} dest-port port numberSyntax Description
Default
The control protocol is enabled. Prior to sending a probe packet to the responder, the RTR collector sends a control message to the responder to enable the destination port.
Command Mode
Response time reporter configuration
Usage Guidelines
This command first appeared in Cisco IOS Release 12.0(3)T.
You must configure the probe's type before you can configure any of the other characteristics of the probe.
The source IP address and port number are optional. If they are not specified, RTR selects the IP address nearest to the target and an available UDP port.
Example
In the following example, probe 12 is created and configured as udpEcho probe using the destination IP address 172.16.1.175 and destination port 2400:
rtr 12type udpEcho dest-ipaddr 172.16.1.175 dest-port 2400Related Commands
verify-data
To cause the RTR probe to check each response for corruption, use the verify-data response time reporter configuration command. Use the no form of this command to return to the default value.
verify-data
no verify-dataSyntax Description
This command has no arguments or keywords.
Default
Disabled
Command Mode
Response time reporter configuration
Usage Guidelines
This command first appeared in Cisco IOS Release 11.2.
Only use the verify-data command when corruption may be a problem in your network.
CautionDo not enable this feature during normal operation because it causes unnecessary overhead. Use the verify-data command only if you suspect there is a problem in your network.
Example
In the following example, probe 5 is configured to verify the data for each response:
rtr 5type echo protocol ipIcmpEcho 172.16.1.174request-data-size 2verify-dataRelated Commands
