IPPM-REPORTING-MIB: View SNMP OID List / Download MIB

This memo defines a portion of the Management Information Base (MIB) for use with network management protocols in TCP/IP-based internets. In particular, it specifies the objects used for managing the results of the IPPM metrics measurements, alarms and reporting of measurement results.

A notification sent because the value of the measure is under the high threshold value and greater than the low threshold value. The notification contains the instances of the ippmHistoryValue object that exceeded the threshold. The notification contains the instances of the ippmHistoryTimestamp identifying the time the event occurred. ippmHistoryPathToResults is a link to the file name, which contains detailed results corresponding to this event.

A notification sent when the size of the history of a metric of a aggregated measure exceeds ippmAggrMeasureHistorySize. The agent will then manage the reports according to the policy described in ippmAggrMeasureResultsMgmt.

A notification sent when the size of the history of a metric of a network measure exceeded ippmNetMeasureHistorySize. Then the agent manages the records according to the policy described in ippmNetMeasureResultsMgmt.

The current time of the system running the IPPM REPORTING MIB SNMP agent. When the agent is running in proxy mode, it is the current time of the proxy agent. When the agent is located in the probe, it is the current time of the probe agent.

ippmSystemSynchronizationType describes the mechanism used to synchronize the system running the IPPM REPORTING MIB SNMP agent. Other(0) The synchronization process must be defined in the ippmSystemSynchonizationDescription. Ntp(1) The system is synchronized using the network time protocol. The NTP synchronization must be described in the ippmSystemSynchonizationDescription. Gps(2) The system is synchronized using the GPS clocks. Cdma(3) The system is synchronized using the CDMA clocks. Enumeration: 'ntp': 1, 'cdma': 3, 'other': 0, 'gps': 2.

The description of the synchronization process of the system running the IPPM REPORTING MIB SNMP agent.

ippmSystemClockResolution provides the precision of the clock used for the measures . The unit is the nanosecond. For example, the clock on an old Unix host might advance only once every 10 msec, and thus have a resolution of 10 msec. So its resolution is 10000000 nanoseconds and the value of ippmSystemClockResolution is 10000000.

This object describes the status of the system running the IPPM REPORTING MIB SNMP agent. It does not describe end point measurement status: unknown(0) means the service is unknown. up(1) means the service is operational and available for general use. down(2) means the agent is not available for use. Enumeration: 'down': 2, 'unknown': 0, 'up': 1.

ippmSystemAggregatedMetrics lists the aggregated metrics that are performed in the SNMP agent instead of in the point of measure.

This table registers the event related to the synchronization of the points of measure. Each event is described in an ippmSynchronizationEntry. ippmSynchronizationTable is mandatory. ippmSynchronizationTable content is read only.

An entry describes a modification of the synchronization status.

An index that identifies the synchronization events in chronological order. 65535 is an arbitrary size. It is not recommended to keep permanently a history of 65535 events.

The time when the synchronization event occurs.

The stratum level of the clock computed when the synchronization event occurs.

The new time resolution computed after the synchronization event occurred.

This table is the list of measurement end points available in the measurement system. Proxy mode: It is the list of the measurement end points of the set of probes for which the IPPM agent provides an SNMP interface. IPPM MIB implemented in a probe: It is the list of the measurement end points of the probe. The ippmPointOfMeasureTable content is read only. This implies that the measurement software handles the table internally ippmPointOfMeasureTable is mandatory.

An entry may be the management address of some middleware in charge of the management of a set of probes. It may the management address of a probe that contains several line cards. An entry describes the capability of a point of measure. ippmPointOfMeasureMetrics lists the metrics handles by the point of measure.

A local index that identifies an entry in the point of measure table.

The address type associated with the management address.

The management address on the point of measure

Defines the address type of the measurement interface of the point of measure.

Specifies the address of the measurement interface for the point of measure.

ippmPointOfMeasureMetrics lists the metrics this point of measure implements.

This table is mandatory. It describes the current implementation. Each IPPM standardized metric must be described in the table. ippmMetricsTable content is read only.

An entry describes the static capabilities of a metric implementation.

ippmMetricsIndex defines an unambiguous index for each standardized metric. It identifies a metric, and as such its value is the value of the node of the metric in the IPPM registry. This value is the same in any implementation of the IPPM REPORTING MIB. Example: In the IPPM-METRICS-REGISTRY, onewayPacketLossAverage is registered as the node 14 of ippmMetricsRegistry.metrics.rfc. Consequently the index of the metric onewayPacketLossAverage in the IppmMetricsTable will always be '14'

Indicates the metric type, whether it is network or aggregated. Enumeration: 'network': 0, 'aggregated': 1.

The unit used in the current entity for the results of the measurement of this metric. Enumeration: 'millisecond': 2, 'noUnit': 0, 'nanosecond': 4, 'packet': 6, 'second': 1, 'microsecond': 3, 'kilobyte': 8, 'percentage': 5, 'byte': 7, 'megabyte': 9.

A textual description of the metric implementation following the exact name of this metric in the registry. For example: oneWayDelay: OWD Metric .

A management entity wishing to access or aggregate remote Ippm measurements in an agent must previously be registered in the ippmOwnersTable. This table is read-create and contains at least the owner 'monitor'.

The description of the resources granted to an SNMP application. For example, an instance of ippmOwnersOwner with an IppmOwnerString 'acme', which represents the 14th owner created in ippmOwnersTable would be named ippmOwnersEntryOwner.14.

The owner described by this entry.

Defines the metrics granted to an owner for which measurements can be performed.

The maximum number of records that this owner may have in the history table and in the report table.

The IP address type of the management entity corresponding to this owner. InetAddressType is restricted to ipv4(1),ipv6(2)and dns(16).

The IP address of the management entity corresponding to this owner. For example, the IP address of the management console used to send SNMP requests.

The email address of the management entity corresponding to this owner.

The status of this table entry. Once this status is set to active, the corresponding entry in the table may not be modified.

The table containing the measurement results.

An ippmHistoryEntry entry is one of the results of a measure identified by ippmHistoryMeasureOwner, ippmHistoryMeasureIndex, ippmHistoryMetricIndex and ippmHistorySequence. In the index : + ippmHistoryMeasureOwner identifies the owner of the measure; + ippmHistoryMeasureIndex identifies the measure in the owner namespace; + ippmHistoryMetricIndex identifies the metric measured by the measure. The metric is described in the corresponding entry of the ippmMetricsTable; + ippmHistorySequence is the sequence number of the measurement result for an entry in the history table.

The owner of the measure that produced this result. The measure is either an ippmNetMeasure or an ippmAggrMeasure.

The owner index of the measure that produced this result. The measure is either an entry of the ippmNetMeasureTable or of the ippmAggrMeasureTable.

ippmHistoryMetricIndex identifies the metric measured by the measure. The metric is described in the corresponding entry of the ippmMetricsTable.

ippmHistorySequence is the sequence number of the measurement results for a metric. Network metrics: It's the sequence number of a measurement packet. Typically, it identifies the order of the packet in the stream of packets sent by the source. Aggregated metrics: It is the sequence order of the aggregation computed.

The timestamp when the measurement occurred.

The observed value of the measurement.

It is typically an URL describing the file location where bulk results are logged.

An entry is a measurement that performs network measures and provides results. It performs several metric measurements per packet exchange. Each step of a measure produces a singleton result per metric. The time of the measurement and the value of the metric are saved in the ippmHistoryTable.

The IppmNetMeasureTable is mandatory, and its content is read only. It means that the measurement software handles the table internally. The setup of the network measure is not permitted through the IPPM REPORTING MIB. As an example, OWAP provides a setup protocol to setup and tear down networks measures. The ippmNetMeasureMetrics is set to a list of metrics to be computed from the same raw packet exchange. Each step of measurement delivers a singleton per metric. Results are timestamped and saved in the ippmHistoryTable. One may create aggregated measures by using the results of network measures.

The owner of the network measure.

The owner index of the network measure.

The name of the metric instance(s) as defined in ippmNetMeasureMetrics. It illustrates the specificity of the metric(s) and includes the metric(s) and the PacketType. Example: IP-TCP-HTTP-One-way-Delay: free text

ippmNetMeasureMetrics defines the metrics to compute within this measure. Only network metrics of the same type are allowed in this field (e.g. poisson-based metrics and periodic-based metrics are incompatibles, while one-way delay and packet loss are generally processed simultaneously: a very bad delay is potentially a very good packet loss). Results are saved in the ippmHistoryTable. Results of a metric are identified using an index of type IppmMetricsRegistryIndex. Example: Given a multi-metrics measure of One-way-Delay(6) and One-way- Packet-Loss(12). The value of the field ippmNetMeasureMetrics is '0001000001000000'b, '1040'B. Results are logged in the ippmHistoryTable where One-way-Delay singletons have a value of ippmMetricsIndex of 6 while One-way-Packet-Loss singletons have a value of ippmMetricsIndex of 12.

Specifies the time at which the measurement begins.

Specifies the unit of the measurement period.

Gives the period used to collect singletons from the point of measures. This value is used as the cycle period in the report.

Specifies the measurement duration unit.

Specifies the measurement duration.

Specifies the maximum number of results saved for each metric of this measure. Overflow condition will be managed by the object ippmNetMeasureResultsMgmt.

This object defines whether this row (and the measure controlled by this row) is restarted automatically, manually, or discarded upon failure, or reboot of the measurement system: 'auto' The measure is restarted automatically. 'manual' The measure has to be restarted manually. 'discarded' The measure and it results are discarded. Enumeration: 'discarded': 3, 'auto': 1, 'manual': 2.

Action to take when the log is full. The measurement system owner may choose to either wrap, in which case the agent writes over existing records. The user may choose to suspend writing to the log in the event that he wishes to archive the data. The resume action causes the agent to begin to write in the log, and assumes the data has been cleared. This object indicates the way the measurement results are managed when the owner quota has been exceeded: 'wrap' continue the measurement and erase the older entries in the history. 'suspend' stop the measure and keep the results in the history. Enumeration: 'wrap': 1, 'suspend': 2.

Defines the type P of the source address of the packets sent by the measure.

Specifies the address of the source of the measure. It is represented as a list of parameters corresponding to those of the PROTOCOL IDENTIFIER set in ippmNetMeasureSrcPacketType.

Defines the type P of the destination address of the packets sent by the measure.

Specifies the address of the destination of the measure. It is represented as a list of parameters corresponding to those of the PROTOCOL IDENTIFIER set in ippmNetMeasureDstPacketType.

The transmit mode used to send the packets: 'other' The rule used to send the packets is unknown. 'periodic' Packets are sent periodically at ippmNetMeasureTxPacketRate rate. 'poisson' Packets are sent using a Poisson law, the median is the value of ippmNetMeasureDevOrIntBurstSize, the deviation is ippmNetMeasureMedOrBurstSize. 'multiburst' Packets are sent bursty at ippmNetMeasureTxPacketRate. The size of the burst is made of ippmNetMeasureMedOrBurstSize packets. Between 2 consecutive bursts, transmission stops during the time needed to send ippmNetMeasureDevOrIntBurstSize. Enumeration: 'multiburst': 3, 'other': 0, 'poisson': 2, 'periodic': 1.

The packet rate unit used to send the packets.

The rate the packets are sent.

Multi-burst mode: This field represents the burst size in number of packets. Poisson mode: This field indicates the number of packets sent, on average, during each period corresponding to the median. The median is therefore MedOrBurstSize*TxPacketRateUnit/TxPacketRate. Example: If TxPacketRateUnit/TxPacketRate is 100 packets/second, and if MedOrBurstSize, the number of packets sent during the period corresponding to the median is 50 packets, then the median equals 50*1/100 = 1/2 seconds.

Multi-burst mode: This field indicates the gap between 2 bursts, in number of packets. Example: If TxPacketRateUnit/TxPacketRate is 100 packets/second, and DevOrIntBurstSize equals 50 packets, then the gap between 2 bursts is equal to 50*1/100, or 1/2 second. Poisson mode: This field indicates the typical difference between the packets of the period corresponding to the median.

Specifies the delay after which the packet is considered lost by the sink.

Specifies the size of the packets counted at the IP network layer in regards to the PacketType definition. Example: For a PacketType 'ip ipip4' the L3 size will be the size of the packet at ipip4 level.

The pattern used to fill the payload of the packet.

Reports the current number of packets received since the beginning of the measure. This parameters is useful to monitor the measure and it is needed to compute statistics.

The time when the last aggregation was computed.

Reports the operational status of the network measure. Enumeration: 'unknown': 0, 'running': 1, 'stopped': 2.

An aggregated measure summarizes the results of previous network or aggregated measures. The results are saved in the ippmHistoryTable. Each step of the calculation for the measure produces a singleton result per metric.

Typically, the configuration operation creates and sets the value of the fields of a new ippmAggrMeasureEntry. ippmAggrMeasureOwner and ippmAggrMeasureIndex identify the instance created. The field ippmAggrMeasureMetrics identifies the metric to compute. As such its ippmMetricsType should be 'aggregated'. The measure aggregates the results of a measure identified by ippmAggrMeasureHistoryOwner, ippmAggrMeasureHistoryIndex and ippmAggrMeasureHistoryMetric. The measure to aggregate belongs to ippmNetMeasureTable or ippmAggrMeasureTable. The aggregation starts at ippmAggrMeasureBeginTime and ends after ippmAggrMeasureDuration. An aggregated result is computed for each ippmMeasureCollectionRate tick and saved in the ippmHistoryTable.

The owner who has configured this entry.

The index of the aggregated measure. The value is managed by the owner.

The name of the instance of the metric. It illustrates the specificity of the metric and includes the metric and the typeP. example: IP-port-HTTP-connectivity: free text.

ippmAggrMeasureMetrics defines the metrics to compute within this aggregated measure. Only aggregated metrics of the same type are allowed in this field (e.g. Measurement of minimum, average and maximum metrics are generally processed simultaneously on the same network measure). Results are saved in the ippmHistoryTable. Results of a metric are identified using an index of type IppmMetricsRegistryIndex. Example: Given a multi-aggregation of One-way-Delay-Median(9) and One-way-Delay-Minimum(10). The value of the field ippmAggrMeasureMetrics is '0000011000000000'b, '0600'B. Results are logged in the ippmHistoryTable where One-way-Delay-Median singletons have a value of ippmMetricsIndex of 9 while One-way-Delay-Minimum singletons have a value of ippmMetricsIndex of 10. NOTE WELL: It is not recommended to use the multi aggregation capability in conjunction with the filter feature.

The owner of the measure to summarize.

The owner index of the measure to summarize.

The metric of the measure to summarize.

ippmAggrMeasureFilter defines the kind of filter to apply on a result to determine if the result is stored or not. The parameters of the filter are ippmAggrMeasureLowThreshold and ippmAggrMeasureHighThreshold. Thresholds have the same unit as the metric value. In the following examples we consider an aggregated measure. Its low threshold is set to 80.its high threshold is set to 100. The aggregation produced a flow of 12 aggregated results {40 30 60 85 140 130 190 95 50 90 30 20}. If the filter is set to 'logInBandValue' then the results 85, 95, 90 will be stored. If the filter is set to 'logOutBandValue' then the results 40 30 60 140 130 190 50 30 20 will be stored. If the filter is set to 'logAboveValue' then the results 140 130 190 will be stored. If the filter is set to 'logBelowValue' then the results 40 30 60 50 30 20 will be stored. If the filter is set to 'logUpAndDownValue' then the results 40, 140, 50 will be stored.

An event is generated when the result of the measure of the metric is lower that the value of ippmAggrMeasureLowThreshold. The threshold has the same unit as the metric. The metric unit is recorded in the object ippmMetricsUnit of this metric entry in the ippmMetricsTable.

An event is generated when the result of the measure of the metric exceeds the value of ippmAggrMeasureHighThreshold. The threshold has the same unit as the metric. The metric unit is recorded in the object ippmMetricsUnit of this metric entry in the ippmMetricsTable.

Specifies the time at which the aggregated measure starts.

Specifies the unit of the aggregated measure period.

Specifies the amount of time between 2 measurement action intervals. The action is specific to the semantic of the measure. Network metrics: The ippmNetMeasureClockPattern transforms the flow of periodical instants as a flow of unpredictable instants of measurement packet emission. As the source and the sink share the definition of the clock of the measure, and as the sending timestamp is part of the measurement packet, the sink has the information to verify that the stream of packets generated by the source respects the clock law. Aggregated metrics: They are performed periodically on a sequence of results of other measures. The period corresponds to the interval between two successive computations of the metric. The value of ippmHistoryTimestamp result of a aggregated metric computed corresponds to the value of the ippmHistoryTimestamp of the last metric result of the sequence used to compute the aggregated metric.

Specifies the unit of the measure duration.

Specifies the duration of the measure.

Specifies the maximum number of results saved for each metric of this measure. Overflow condition will be managed by the object ippmAggrMeasureResultsMgmt.

This object defines whether this row and the measure controlled by this row are kept in volatile storage and lost upon reboot or if this row is backed up by non-volatile or permanent storage. Possible values are: other(1), volatile(2), nonVolatile(3), permanent(4), readOnly(5).

This object displays the way the history of the aggregated measure is managed. 'wrap' continue the measure and erase the older entries in the history. 'suspend' stop the measure and keep the results in the history. Enumeration: 'wrap': 1, 'suspend': 2.

This object controls the activity of the aggregated measure. 'start' The aggregated measure is started. 'stop' The aggregated measure is stopped. Enumeration: 'start': 0, 'stop': 1.

A fast report is required in order to verify quickly that a measure is running well. 'fast report' feature is active if ippmAggrMeasureFastReport is not null and points to a notification. A fast report consists of sending by email to the owner of the measure, a table of the results of all the metrics computed by this aggregated measure. The owner email address is read from the ippmOwnersTable. ippmAggrMeasureFastReport identifies the notification which defines the header of the report. The results part of the report is made of a column of results per metrics. Results are separated using commas. To avoid disaster, an aggregated measure using a fast report must have a cycle of aggregation greater than or equal to 1 second and should not sent more than an email every 5 minutes and should not sent more than 12 emails.

The time when the last aggregated measure was computed.

Reports the operational status of the aggregated measure. Enumeration: 'unknown': 0, 'running': 1, 'stopped': 2.

Reports the current number of packets used to calculate the aggregation since the start of the measure. This parameters is useful to monitor the measure and it is needed to compute statistics.

The status of this entry. Once the entry status is set to active, the associate entry cannot be modified.

The compliance statement for SNMP entities which implement the IPPM MIB as a proxy in interdomain. The implementation of the VACM control is mandatory.

The compliance statement for SNMP entities which implement the IPPM MIB as a proxy.

The compliance statement for SNMP entities which implement the IPPM MIB in a probe.

The IPPM System Group

The IPPM Network Measure Group

The IPPM History Group

The IPPM AggregatedMeasure Group

The IPPM Owners Group

The IPPM Notification Group