RFC 3202
Definitions of Managed Objects for Frame Relay Service Level Definitions
Network Working Group R. Steinberger Request for Comments: 3202 Paradyne Networks Category: Standards Track O. Nicklass RAD Data Communications Ltd. January 2002 Definitions of Managed Objects for Frame Relay Service Level Definitions Status of this Memo This document specifies an Internet standards track protocol for the Internet community, and requests discussion and suggestions for improvements. Please refer to the current edition of the "Internet Official Protocol Standards" (STD 1) for the standardization state and status of this protocol. Distribution of this memo is unlimited. Copyright Notice Copyright (C) The Internet Society (2002). All Rights Reserved.Abstract
This memo defines an extension of the Management Information Base (MIB) for use with network management protocols in TCP/IP-based internets. In particular, it defines objects for managing the Frame Relay Service Level Definitions.Table of Contents
1. The SNMP Management Framework ............................... 2 2. Conventions ................................................. 3 3. Overview .................................................... 3 3.1. Frame Relay Service Level Definitions ..................... 4 3.2. Terminology ............................................... 5 3.3. Network Model ............................................. 5 3.4. Reference Points .......................................... 6 3.5. Measurement Methodology ................................... 8 3.6. Theory of Operation ....................................... 9 3.6.1. Capabilities Discovery .................................. 9 3.6.2. Determining Reference Points for Row Creation ........... 10 3.6.2.1. Graphical Examples of Reference Points ................ 11 3.6.2.1.1. Edge-to-Edge Interface Reference Point Example ...... 12 3.6.2.1.2. Edge-to-Edge Egress Queue Reference Point Example ... 13 3.6.2.1.3. End-to-End Using Reference Point Example ............ 14 3.6.3. Creation Process ........................................ 15 3.6.4. Destruction Process ..................................... 15
3.6.4.1. Manual Row Destruction ................................ 15 3.6.4.2. Automatic Row Destruction ............................. 16 3.6.5. Modification Process .................................... 16 3.6.6. Collection Process ...................................... 16 3.6.6.1. Remote Polling ........................................ 16 3.6.6.2. Sampling .............................................. 17 3.6.6.3. User History .......................................... 17 3.6.7. Use of MIB Module in Calculation of Service Level Definitions .................................................... 17 3.6.8. Delay ................................................... 20 3.6.9. Frame Delivery Ratio .................................... 20 3.6.10. Data Delivery Ratio .................................... 21 3.6.11. Service Availability ................................... 21 4. Relation to Other MIB Modules ............................... 22 5. Structure of the MIB Module ................................. 23 5.1. frsldPvcCtrlTable ......................................... 23 5.2. frsldSmplCtrlTable ........................................ 23 5.3. frsldPvcDataTable ......................................... 23 5.4. frsldPvcSampleTable ....................................... 24 5.5. frsldCapabilities ......................................... 24 6. Persistence of Data ......................................... 24 7. Object Definitions .......................................... 24 8. Acknowledgments ............................................. 61 9. References .................................................. 61 10. Security Considerations .................................... 63 11. Authors' Addresses ......................................... 63 12. Full Copyright Statement ................................... 641. The SNMP Management Framework
The SNMP Management Framework presently consists of five major components: o An overall architecture, described in RFC 2571 [1]. o Mechanisms for describing and naming objects and events for the purpose of management. The first version of this Structure of Management Information (SMI) is called SMIv1 and described in STD 16, RFC 1155 [2], STD 16, RFC 1212 [3] and RFC 1215 [4]. The second version, called SMIv2, is described in STD 58, RFC 2578 [5], RFC 2579 [6] and RFC 2580 [7]. o Message protocols for transferring management information. The first version of the SNMP message protocol is called SNMPv1 and described in STD 15, RFC 1157 [8]. A second version of the SNMP message protocol, which is not an Internet standards track protocol, is called SNMPv2c and described in RFC 1901 [9] and RFC
1906 [10]. The third version of the message protocol is called
SNMPv3 and described in RFC 1906 [10], RFC 2572 [11] and RFC 2574
[12].
o Protocol operations for accessing management information. The
first set of protocol operations and associated PDU formats is
described in STD 15, RFC 1157 [8]. A second set of protocol
operations and associated PDU formats is described in RFC 1905
[13].
o A set of fundamental applications described in RFC 2573 [14] and
the view-based access control mechanism described in RFC 2575
[15].
A more detailed introduction to the current SNMP Management Framework
can be found in RFC 2570 [16].
Managed objects are accessed via a virtual information store, termed
the Management Information Base or MIB. Objects in the MIB are
defined using the mechanisms defined in the SMI.
This memo specifies a MIB module that is compliant to the SMIv2. A
MIB conforming to the SMIv1 can be produced through the appropriate
translations. The resulting translated MIB must be semantically
equivalent, except where objects or events are omitted because no
translation is possible (use of Counter64). Some machine readable
information in SMIv2 will be converted into textual descriptions in
SMIv1 during the translation process. However, this loss of machine
readable information is not considered to change the semantics of the
MIB.
2. Conventions
The keywords MUST, MUST NOT, REQUIRED, SHALL, SHALL NOT, SHOULD,
SHOULD NOT, RECOMMENDED, NOT RECOMMENDED, MAY, and OPTIONAL, when
they appear in this document, are to be interpreted as described in
RFC 2119 [22].
3. Overview
This MIB module addresses the items required to manage the Frame
Relay Forum's Implementation Agreement for Service Level Definitions
(FRF.13 [17]). At present, this applies to these values of the
ifType variable in the Internet-standard MIB:
o frameRelay (32)
o frameRelayService (44)
This section provides an overview and background of how to use this MIB module.3.1. Frame Relay Service Level Definitions
The frame relay service level definitions address specific characteristics of a frame relay service that can be used to facilitate the following tasks: o Evaluation of frame relay service providers, offerings or products. o Measurement of Quality of Service. o Enforcement of Service Level Agreements. o Planning or describing a frame relay network. The following parameters are defined in FRF.13 [17] as a sufficient set of values to accomplish the tasks previously stated. o Delay - The amount of time elapsed, in microseconds, from the time a frame exits the source to the time it reaches the destination. NOTE: FRF.13 [17] defines this value in terms of milliseconds. o Frame Delivery Ratio - The ratio of the number of frames delivered to the destination versus the number of frames sent by the source. This ratio can be further divided by inspecting either only the frames within the CIR or only the frames in excess of the CIR. o Data Delivery Ratio - The ratio of the amount of data delivered to the destination versus the amount of data sent by the source. This ratio can be further divided by inspecting either only the data within the CIR or only the data in excess of the CIR. o Service Availability - The amount of time the frame relay service was not available. There are three types of availability statistics defined in FRF.13 [17]: Mean Time to Repair, Virtual Connection Availability, and Mean Time Between Service Outages. The later two require information about the scheduled outage time. It is assumed that scheduled outage time information will be maintained by the network management software, so it is not included in the MIB module. Consult FRF.13 [17] for more details.
3.2. Terminology
o CIR - The Committed Information Rate (CIR) is the subscriber data rate (expressed in bits/second) that the network commits to deliver under normal network conditions [18]. o DLCI - Data Link Connection Identifier [18]. o Logical Port - This term is used to model the Frame Relay "interface" on a device [18]. o NNI - Network to Network Interface [18]. o Permanent Virtual Connection (PVC) - A virtual connection that has its end-points and bearer capabilities defined at subscription time [18]. o Reference Point (RP) - The point of reference within the network model at which the calculations or data collection takes place. o UNI - User to Network Interface [18].3.3. Network Model
The basic model, as illustrated in figure 1 below, contains two frame relay DTE endpoints connected to a network cloud via a frame relay UNI interface. The network cloud can contain zero or more internal frame relay NNI connections that interconnect multiple networks. The calculations and data collection can be performed at any reference point within the network. +-------------+ +-------------+ | Frame Relay | | Frame Relay | | DTE Device | | DTE Device | +------+------+ +------+------+ | | UNI UNI Connection Connection | | +------+------+ NNI +-------------+ NNI +------+------+ | Network A +------------+ Network B +------------+ Network C | +-------------+ Connection +-------------+ Connection +-------------+ Figure 1 Frame Relay Network Reference Model
3.4. Reference Points
The collection and calculations of the service level definitions apply to two reference points within the network. These two points are the locations where the frames are referenced in the collection of the service level specific information. The reference points used in the MIB module are shown in figure 2 below. For completeness, the module also allows for proprietary reference points which MAY exist anywhere in the network that is not a previously defined reference point. The meaning of the proprietary reference points is insignificant unless defined by the device manufacturer.
+---------------------------+ |+-----------+ +-----------+| || | |Measurement|| ||Frame Relay---Engine --(Source RP)----+ ||DTE | |(If Exists)|| | |+-----------+ +-----------+| | +---------------------------+ | Frame Relay Source | +------------------------------------------+ | Frame Relay Network | +----------------------------------+ | | +------------------------------+ | | | | +---------+ +---------+ | | | | | | | | Traffic | | | +--(Ingress RP)--- L1 / L2 --- Policing| | | | | | Control | | Engine | | | | | +---------+ +----|----+ | | | | | | | | | (Traffic Policing RP)| | | +------------------|-----------+ | | Ingress Node | | | | | | +-----------|-----------+ | | | Intermediate Nodes | | | +-----------|-----------+ | | | | | Egress Node | | | +--------------|-----------+ | | | (Egress Queue Input RP) | | | | | | | | | +-------+------+ | | | | | Egress Queue | | | | | +-------+------+ | | | | | | | | | (Egress Queue Output RP) | | | +--------------|-----------+ | +--------------------|-------------+ Frame Relay Destination | +---------------------------+ +-----------+ |+-----------+ +-----------+| | || | |Measurement|| | ||Frame Relay---Engine --(Destination RP)--+ ||DTE | |(If Exists)|| |+-----------+ +-----------+| +---------------------------+ Figure 2 Reference Points (FRF.13 [17])
The MIB variables frsldPvcCtrlTransmitRP and frsldPvcCtrlReceiveRP allow the user to view and configure the reference points at which the calculations occur. These variables are specific to the device on which they are located. Frame relay devices act as both frame sources and frame destinations. The definitions in this MIB module apply to the interaction of a pair of devices on the network path. The same device can potentially use different reference points for calculation and collection of the statistics based on whether the referenced frame is sent or received by the device. When the device is acting as a frame source, the value of frsldPvcCtrlTransmitRP reflects the reference point used for all source calculations pertaining to the specified PVC. When the device is acting as a frame destination, the value of frsldPvcCtrlReceiveRP reflects the reference point used for all destination calculations pertaining to the specified PVC. For example, FRF.13 [17] defines an Edge-to-Edge Egress Queue measurement domain as a domain in which measurement is performed between an Ingress Reference Point and an Egress Queue Input Reference Point. For this domain between a source device and a destination device, the value of frsldPvcCtrlTransmitRP for the source device would be set to ingTxLocalRP(2) and the value of frsldPvcCtrlReceiveRP for the destination device would be set to eqiRxLocalRP(4). While it is usually the case that the reference points would be equivalent on the remote device when monitoring frames going in the opposite direction, there is no requirement for them to be so. It can be seen from the above example that a total of four reference points are required in order to collect information for both directions of traffic flow. The reference points represent the transmit and receive directions at both ends of a PVC. If a device has knowledge of the information from the remote device, it is possible to collect the statistics from a single device. This is not always the case. In most instances, two devices will need to be monitored to capture a complete description of the service level on a PVC. The reference points a single device is capable of monitoring are contained in the frsldRPCaps object.3.5. Measurement Methodology
This document neither recommends nor suggests a method of implementation. This is left to the device manufacturer and should be independent of the data that is actually collected. Periodic collection of this data can be performed through either polling of the data table, use of the sample tables or use of the user history group of RFC 2021 [19].
3.6. Theory of Operation
The following sections describe how to use this MIB module. They include row handling, data collection and data calculation. The recommendations here in are suggestions as to implementation and do not infer that they are the only method that can be used to perform such operations.3.6.1. Capabilities Discovery
Three objects are provided specifically to aid the network manager in discovering the capabilities of the device with respect to this MIB module. o frsldPvcCtrlWriteCaps This object reports the write capabilities of the PVC Control Table. Use this object to determine which objects can be modified. This need only be referenced if row creation or modification is to be performed. o frsldSmplCtrlWriteCaps This object reports the write capabilities of the Sample Control Table. Use this object to determine which objects can be modified. The group need only be referenced if the sample tables will be used to collect historical information. o frsldRPCaps This object reports the reference points at which the device is capable of collecting information. This object needs to be referenced if row creation is to be performed in the PVC Control Table. Devices can only create rows containing supported reference points. These objects do not imply that there is no need for an Agent Capabilities macro for devices that do not fully support every object in this MIB module. They are provided specifically to aid in the ensured network management operations of this MIB module with respect to row creation and modification. An additional four objects are provided to report and control memory the utilization of this MIB module. These objects are frsldMaxPvcCtrls, frsldNumPvcCtrls, frsldMaxSmplCtrls are frsldNumSmplCtrls. Together, they allow a manager to control the
amount of memory allocated for specific utilization by this MIB module. This is done by setting the maximum allowed allocation of controls.3.6.2. Determining Reference Points for Row Creation
The performance of a PVC is monitored by evaluating the uni- directional flow of frames from an ingress point to an egress point. Reference points describe where each of the two measurements are made. Monitoring both of the uni-directional flows that make-up the PVC frame traffic requires a total of four reference points as shown in Figures 3 through 5. A monitoring point that evaluates traffic is restricted to counting frames that pass the reference points hosted locally on the monitoring point. Thus, if the monitoring point is near the ingress point of the flow, it will count the frames entering into the frame relay network. The complete picture of frame loss for the uni-directional flow requires information from the downstream reference point located at another (remote) monitoring point. The local monitoring point MAY be implemented in such way that the information from the downstream monitoring point is moved to the local monitoring point using implementation-specific mechanisms. In this case all information required to calculate frame loss becomes available from the local measurement point. The local measurement point agent is capable of reporting all the objects in the FrsldPvcDataEntry row - the counts for offered frames entering the network and delivered frames exiting the network. Alternatively, the local monitoring point MAY be restricted to counts of frames observed on the local device only. In this case, the objects of the FrsldPvcDataEntry row reporting what happened on the remote device are not available. The following list shows the possible valid reference points for an FRF.13 SLA from the source reference point to the destination reference point in both directions. o Local Information Only Local Device: srcLocalRP, desLocalRP Remote Device: srcLocalRP, desLocalRP o Remote Information Only Local Device: srcRemoteRP, desRemoteRP Remote Device: srcRemoteRP, desRemoteRP
o Mixed Two Device Model 1 (Local Device Always Transmitter)
Local Device: srcLocalRP, desRemoteRP
Remote Device: srcLocalRP, desRemoteRP
o Mixed Two Device Model 2 (Local Device Always Receiver)
Local Device: srcRemoteRP, desLocalRP
Remote Device: srcRemoteRP, desLocalRP
o Mixed One Device Model 1 (Directional Rows)
First Row: srcRemoteRP, desLocalRP (Receiver Row)
Second Row: srcLocalRP, desRemoteRP (Sender Row)
o Mixed One Device Model 2 (Device Based Rows)
First Row: srcLocalRP, desLocalRP (Local Row)
Second Row: srcRemoteRP, desRemoteRP (Remote Row)
Each of the above combinations is valid and provides the same
information.
The following steps are recommended to find which reference points
need to be configured:
1) Locate both of the devices at either end of the PVC to be
monitored.
2) Determine the capabilities by referencing the frsldRPCaps object
of each device.
3) Locate the best combination of the two devices such that the
necessary reference points are all represented.
4) If any one of the necessary reference points does not exist in the
combination of the two devices, it is not possible to monitor the
FRF.13 defined SLA between the two reference point on the PVC.
3.6.2.1. Graphical Examples of Reference Points
FRF.13 [17] defines three specific combinations of reference points:
Edge-to-Edge Interface, Edge-to-Edge Egress Queue and End-to-End.
Examples of valid reference points that may be used for each of these
are discussed in the sections below.
It is often the case that a device knows as a minimum either only local information or both local and remote information. Because these are two common examples, each will be illustrated below.3.6.2.1.1. Edge-to-Edge Interface Reference Point Example
Device 1 Device 2 +-------------+ +-------------+ | Ingress | | Egress | | +-----+ | | +-----+ | |(A)| | | Traffic Flow | | |(B)| -->-->-- -->-->-->-->-->-->-->-->-->-->-->- -->-->--> | | | | From Device 1 to 2 | | | | | +-----+ | | +-----+ | | | | | | Egress | | Ingress | | +-----+ | | +-----+ | |(D)| | | Traffic Flow | | |(C)| <--<--<- -<--<--<--<--<--<--<--<--<--<--<-- --<--<-- | | | | From Device 2 to 1 | | | | | +-----+ | | +-----+ | +-------------+ +-------------+ where (A), (B), (C) and (D) are reference points Figure 3 For devices with only local knowledge, one row is required on each device as follows: (A) frsldPvcCtrlTransmitRP for Device 1 = ingTxLocalRP(2) (B) frsldPvcCtlrReceiveRP for Device 2 = eqoRxLocalRP(5) (C) frsldPvcCtrlTransmitRP for Device 2 = ingTxLocalRP(2) (D) frsldPvcCtlrReceiveRP for Device 1 = eqoRxLocalRP(5) In which a single row is created on Device 1 containing reference points (A) and (D), and a single row is created on Device 2 containing reference points (C) and (B). For devices with both local and remote knowledge, the two rows can exist in any combination on either device. For this example, the transmitting devices will be responsible for information regarding the flow for which they are the origin. Only one row is required per device for this example.
(A) frsldPvcCtrlTransmitRP for Device 1 = ingTxLocalRP(2) (B) frsldPvcCtlrReceiveRP for Device 1 = eqoRxRemoteRP(11) (C) frsldPvcCtrlTransmitRP for Device 2 = ingTxLocalRP(2) (D) frsldPvcCtlrReceiveRP for Device 2 = eqoRxRemoteRP(11)3.6.2.1.2. Edge-to-Edge Egress Queue Reference Point Example
Device 1 Device 2 +-------------+ +-------------+ | Ingress | | Egress | | +-----+ | | +-----+ | |(A)| | | Traffic Flow |(B)| | | -->-->-- -->-->-->-->-->-->-->-->-->-->-->- -->-->--> | | | | From Device 1 to 2 | | | | | +-----+ | | +-----+ | | | | | | Egress | | Ingress | | +-----+ | | +-----+ | | | |(D)| Traffic Flow | | |(C)| <--<--<- -<--<--<--<--<--<--<--<--<--<--<-- --<--<-- | | | | From Device 2 to 1 | | | | | +-----+ | | +-----+ | +-------------+ +-------------+ where (A), (B), (C) and (D) are reference points Figure 4 For devices with only local knowledge, one row is required on each device as follows: (A) frsldPvcCtrlTransmitRP for Device 1 = ingTxLocalRP(2) (B) frsldPvcCtlrReceiveRP for Device 2 = eqiRxLocalRP(4) (C) frsldPvcCtrlTransmitRP for Device 2 = ingTxLocalRP(2) (D) frsldPvcCtlrReceiveRP for Device 1 = eqiRxLocalRP(4) In which a single row is created on Device 1 containing reference points (A) and (D), and a single row is created on Device 2 containing reference points (C) and (B).
For devices with both local and remote knowledge, the two rows can exist in any combination on either device. For this example, the transmitting devices will be responsible for information regarding the flow for which they are the origin. Only one row is required per device for this example. (A) frsldPvcCtrlTransmitRP for Device 1 = ingTxLocalRP(2) (B) frsldPvcCtlrReceiveRP for Device 1 = eqiRxRemoteRP(10) (C) frsldPvcCtrlTransmitRP for Device 2 = ingTxLocalRP(2) (D) frsldPvcCtlrReceiveRP for Device 2 = eqiRxRemoteRP(10)3.6.2.1.3. End-to-End Using Reference Point Example
Device 1 Device 2 +-------------+ +-------------+ | Source | | Destination | | +-----+ | | +-----+ | |(A)| | | Traffic Flow | | |(B)| -->-->-- -->-->-->-->-->-->-->-->-->-->-->- -->-->--> | | | | From Device 1 to 2 | | | | | +-----+ | | +-----+ | | | | | | Destination | | Source | | +-----+ | | +-----+ | |(D)| | | Traffic Flow | | |(C)| <--<--<- -<--<--<--<--<--<--<--<--<--<--<-- --<--<-- | | | | From Device 2 to 1 | | | | | +-----+ | | +-----+ | +-------------+ +-------------+ where (A), (B), (C) and (D) are reference points Figure 5 For devices with only local knowledge, one row is required on each device as follows: (A) frsldPvcCtrlTransmitRP for Device 1 = srcLocalRP(1) (B) frsldPvcCtlrReceiveRP for Device 2 = desLocalRP(1) (C) frsldPvcCtrlTransmitRP for Device 2 = srcLocalRP(1) (D) frsldPvcCtlrReceiveRP for Device 1 = desLocalRP(1)
In which a single row is created on Device 1 containing reference points (A) and (D), and a single row is created on Device 2 containing reference points (C) and (B). For devices with both local and remote knowledge, the two rows can exist in any combination on either device. For this example, the transmitting devices will be responsible for information regarding the flow for which they are the origin. Only one row is required per device for this example. (A) frsldPvcCtrlTransmitRP for Device 1 = srcLocalRP(1) (B) frsldPvcCtlrReceiveRP for Device 1 = desRemoteRP(7) (C) frsldPvcCtrlTransmitRP for Device 2 = srcLocalRP(1) (D) frsldPvcCtlrReceiveRP for Device 2 = desRemoteRP(7)3.6.3. Creation Process
In some cases, devices will automatically populate the rows of PVC Control Table and potentially the Sample Control Table. However, in many cases, it may be necessary for a network manager to manually create rows. Manual creation of rows requires the following steps: 1) Ensure the PVC exists between the two devices. 2) Determine the necessary reference points for row creation. 3) Create the row(s) in each device as needed. 4) Create the row(s) in the sample control tables if desired.3.6.4. Destruction Process
3.6.4.1. Manual Row Destruction
Manual row destruction is straight forward. Any row can be destroyed and the resources allocated to it are freed by setting the value of its status object (either frsldPvcCtrlStatus or frsldSmplCtrlStatus) to destroy(6). It should be noted that when frsldPvcCtrlStatus is set to destroy(6) all associated sample control, sample and data table rows will also be destroyed. Similarly, when frsldSmplCtrlStatus is set to destroy(6) all sample rows will also be
destroyed. The frsldPvcCtrlPurge objects do not apply to manual row destruction. If the row is set to destroy(6) manually, the rows are destroyed as part of the set.3.6.4.2. Automatic Row Destruction
Rows is the tables may be destroyed automatically based on the existence of the DLCI on which they rely. This behavior is controlled by the frsldPvcCtrlPurge and frsldPvcCtrlDeleteOnPurge objects. When a DLCI no longer exists in the device, the data in the tables has no relation to anything known on the network. However, there may be some need to keep the historic information active for a short period after the destruction or removal of a DLCI. If the basis for the row no longer exists, the row will be destroyed at the end of the purge interval that is controlled by frsldPvcCtrlPurge. The effects of automatic row destruction are the same as manual row destruction.3.6.5. Modification Process
All read-create items in this MIB module can be modified at any time if they are fully supported. Write access is not required. To simplify the use of the MIB frsldPvcCtrlWriteCaps and frsldSmplCtrlWriteCaps state which of the read-create variables can actually be written on a particular device.3.6.6. Collection Process
3.6.6.1. Remote Polling
This MIB module supports data collection through remote polling of the free running counters in the PVC Data Table. Remote polling is a common method used to capture real-time statistics. A remote management station polls the device to collect the desired information. It is recommended all statistics for a single PVC be collected in a single PDU. The following objects are designed around the concept of real-time polling: o frsldPvcDataMissedPolls o frsldPvcDataFrDeliveredC o frsldPvcDataFrDeliveredE o frsldPvcDataFrOfferedC o frsldPvcDataFrOfferedE o frsldPvcDataDataDeliveredC o frsldPvcDataDataDeliveredE
o frsldPvcDataDataOfferedC o frsldPvcDataDataOfferedE o frsldPvcDataHCFrDeliveredC o frsldPvcDataHCFrDeliveredE o frsldPvcDataHCFrOfferedC o frsldPvcDataHCFrOfferedE o frsldPvcDataHCDataDeliveredC o frsldPvcDataHCDataDeliveredE o frsldPvcDataHCDataOfferedC o frsldPvcDataHCDataOfferedE o frsldPvcDataUnavailableTime o frsldPvcDataUnavailables3.6.6.2. Sampling
The sample tables provide the ability to historically sample data without requiring the additional overhead of polling. At key periods, a network management station can collect the samples needed. This method allows the manager to perform the collection of data at times that will least affect the active network traffic. The sample data can be collected using a series of SNMP getNext or getBulk operations. The value of frsldPvcSmplIdx increments with each new collection bucket. This allows the managers to skip information that has already been collected. However, care should be taken in that the value can roll over after a long period of time. The start and end times of a collection period allow the manager to know what the actual period of collection was. It is possible for there to be discontinuities in the sample table, so both start and end should be referenced.3.6.6.3. User History
User history, as defined in RFC 2021 [19], is an alternative mechanism that can be used to get the same benefits as the sample table by using the objects provided for real-time polling. Some devices MAY have the ability to use user history and opt not to support the sample tables. If this is the case, the information from the data table can be used to define a group of user history objects.3.6.7. Use of MIB Module in Calculation of Service Level Definitions
The objects in this MIB module can be used to calculate the statistics defined in FRF.13 [17]. The description below describes the calculations for one direction of the data flow, i.e., data sent from local transmitter to a remote receiver. A complete set of bidirectional information would require calculations based on both
directions. For the purposes of this description, the reference
points used SHOULD consistently represent data that is sent by one
device and received by the other.
A complete evaluation requires the combination of two uni-directional
flows. It is possible for a management station to combine all of the
calculated information into one conceptual row. Doing this requires
that each of the metrics are collected for both flow directions and
grouped by direction If the information is split between two
devices, the management station must know which two devices to
communicate with for the collection of all information. The grouping
of information SHOULD be from ingress to egress in each flow
direction.
The calculations below use the following terminology:
o DelayAvg
The average delay on the PVC. This is represented within the
MIB module by frsldPvcSmplDelayAvg.
o FrDeliveredC
The number of frames received by the receiving device through
the receive reference point that were delivered within CIR.
This is represented within the MIB module by one of
frsldPvcDataFrDeliveredC, frsldPvcDataHCFrDeliveredC,
frsldPvcSmplFrDeliveredC, or frsldPvcSmplHCFrDeliveredC.
o FrDeliveredE
The number of frames received by the receiving device through
the receive reference point that were delivered in excess of
CIR. This is represented within the MIB module by one of
frsldPvcDataFrDeliveredE, frsldPvcDataHCFrDeliveredE,
frsldPvcSmplFrDeliveredE, or frsldPvcSmplHCFrDeliveredE.
o FrOfferedC
The number of frames offered by the transmitting device through
the transmit reference point that were sent within CIR. This
is represented within the MIB module by one of
frsldPvcDataFrOfferedC, frsldPvcDataHCFrOfferedC,
frsldPvcSmplFrOfferedC, or frsldPvcSmplHCFrOfferedC.
o FrOfferedE
The number of frames offered by the transmitting device through
the transmit reference point that were sent in excess of CIR.
This is represented within the MIB module by one of
frsldPvcDataFrOfferedE, frsldPvcDataHCFrOfferedE,
frsldPvcSmplFrOfferedE, or frsldPvcSmplHCFrOfferedE.
o DataDeliveredC
The number of octets received by the receiving device through
the receive reference point that were delivered within CIR.
This is represented within the MIB module by one of
frsldPvcDataDataDeliveredC, frsldPvcDataHCDataDeliveredC,
frsldPvcSmplDataDeliveredC, or frsldPvcSmplHCDataDeliveredC.
o DataDeliveredE
The number of octets received by the receiving device through
the receive reference point that were delivered in excess of
CIR. This is represented within the MIB module by one of
frsldPvcDataDataDeliveredE, frsldPvcDataHCDataDeliveredE,
frsldPvcSmplDataDeliveredE, or frsldPvcSmplHCDataDeliveredE.
o DataOfferedC
The number of octets offered by the transmitting device through
the transmit reference point that were sent within CIR. This
is represented within the MIB module by one of
frsldPvcDataDataOfferedC, frsldPvcDataHCDataOfferedC,
frsldPvcSmplDataOfferedC, or frsldPvcSmplHCDataOfferedC.
o DataOfferedE
The number of octets offered by the transmitting device through
the transmit reference point that were sent in excess of CIR.
This is represented within the MIB module by one of
frsldPvcDataDataOfferedE, frsldPvcDataHCDataOfferedE,
frsldPvcSmplDataOfferedE, or frsldPvcSmplHCDataOfferedE.
o UnavailableTime
The amount of time the PVC was not available during the
interval of interest. This is represented within the MIB
module by either frsldPvcDataUnavailableTime or
frsldPvcSmplUnavailableTime.
o Unavailables
The number of times the PVC was declared to be unavailable
during the interval of interest. This is represented within
the MIB module by either frsldPvcDataUnavailables or
frsldPvcSmplUnavailables.
3.6.8. Delay
The frame transfer delay is defined as the amount of time elapsed, in
microseconds, from the time a frame exits the source to the time it
reaches the destination. The average delay can be found using the
MIB variable described in DelayAvg above. The delay may be
calculated as either round trip or one way, and this information is
held in the frsldPvcCtrlDelayType MIB variable. If the delay be
calculated as round trip, the value of DelayAvg represents the
average of the total delays of the round trips. In this case, the
manager SHOULD divide the value returned by the agent by two to
obtain the frame transfer delay. In the case that
frsldPvcCtrlDelayType is oneWay, the value of DelayAvg represents the
average of the frame transfer delays and SHOULD be used as is.
3.6.9. Frame Delivery Ratio
The frame delivery ratio is defined as the total number of frames
delivered to the destination divided by the frames offered by the
source. The destination values can be obtained using FrDeliveredC
and FrDeliveredE. The source values can be obtained using FrOfferedC
and FrOfferedE.
FrDeliveredC + FrDeliveredE
Frame Delivery Ratio = ---------------------------
FrOfferedC + FrOfferedE
FrDeliveredC
Committed Frame Delivery Ratio = ------------
FrOfferedC
FrDeliveredE
Excess Frame Delivery Ratio = ------------
FrOfferedE
3.6.10. Data Delivery Ratio
The data delivery ratio is defined as the total amount of data delivered to the destination divided by the data offered by the source. The destination values can be obtained using DataDeliveredC and DataDeliveredE. The source values can be obtained using DataOfferedC and DataOfferedE. DataDeliveredC + DataDeliveredE Data Delivery Ratio = ------------------------------- DataOfferedC + DataOfferedE DataDeliveredC Committed Data Delivery Ratio = -------------- DataOfferedC DataDeliveredE Excess Data Delivery Ratio = -------------- DataOfferedE3.6.11. Service Availability
Some forms of service availability measurement defined in FRF.13 [17] require knowledge of the amount of time the network is allowed to be unavailable during the period of measurement. This is called the excluded outage time and will be represented in the measurements below as ExcludedTime. It is assumed that the management software will maintain this information in that it often relates to specific times and dates that many devices are not capable of maintaining. Further, it may change based on a moving maintenance window that the device cannot track well. Mean Time to Repair (FRMTTR) = 0 if Unavailables is 0. UnavailableTime Otherwise, FRMTTR = --------------- Unavailables Virtual Connection Availability (FRVCA) = 0 if IntervalTime equals ExcludedTime. IntervalTime - ExcludedTime - UnavailableTime Otherwise, FRVCA = --------------------------------------------- *100 IntervalTime - ExcludedTime Mean Time Between Service Outages (FRMTBSO) = 0 if Unavailables is 0.
Otherwise, FRMTBSO = IntervalTime - ExcludedTime - UnavailableTime
---------------------------------------------
Unavailables
4. Relation to Other MIB Modules
There is no explicit relation to any other frame relay MIB module nor
are any required to implement this MIB module. However, there is a
need for knowledge of ifIndexes and some understanding of DLCIs. The
ifIndex information can be found in the IF-MIB [21] which is
required. The DLCI information can be found in either the Frame
Relay DTE MIB (RFC 2115) [20] or the Frame Relay Network Services MIB
(RFC 2954) [18]; however, neither is required.
Upon setting of frsldPvcCtrlStatus in the frsldPvcCtrlTable to
active(1) the system can be in one of the following three states:
(1) The respective DLCI is known and is active. This corresponds to
a state in which frPVCEndptRowStatus is active(1) and
frPVCEndptRcvdSigStatus is either active(2) or none(4) for the
Frame Relay Network Services MIB (RFC 2954) [18]. For the Frame
Relay DTE MIB, the same state is shown by frCircuitRowStatus of
active(1) and frCircuitState of active(2).
(2) The respective DLCI has not been created. This corresponds to a
state in which the row with either frPVCEndptDLCIIndex or
frCircuitDlci equal to the respective DLCI does not exist in
either the frPVCEndptTable or the frCircuitTable respectively.
(3) The respective DLCI has just been removed. This corresponds to a
state in which either frPVCEndptRowStatus is no longer active(1)
or frPVCEndptRcvdSigStatus is no longer active(2) or none(4) for
the Frame Relay Network Services MIB (RFC 2954) [18]. For the
Frame Relay DTE MIB, the same state is shown when either
frCircuitRowStatus is no longer active(1) or frCircuitState is no
longer active(2).
For the first case, the row in the frsldPvcDataTable will be filled.
If frsldSmplCtrlStatus in the frsldSmplCtrlTable for the respective
DLCI is also `active' the frsldPvcSampleTable will be filled as well.
For the second case, the respective rows will not be added to any of
the data or sample tables and frsldPvcCtrlStatus SHOULD report
notReady(3).
For the third case, frsldPvcCtrlDeleteOnPurge should direct the behavior of the system. If all tables are purged, this case will be equivalent to the second case above. Otherwise, frsldPvcCtrlStatus SHOULD remain active(1).5. Structure of the MIB Module
The FRSLD-MIB consists of the following components: o frsldPvcCtrlTable o frsldSmplCtrlTable o frsldPvcDataTable o frsldPvcSampleTable o frsldCapabilities Refer to the compliance statement defined within for a definition of what objects MUST be implemented.5.1. frsldPvcCtrlTable
The frsldPvcCtrlTable is the central control table for operations of the Frame Relay Service Level Definitions MIB. It provides variables to control the parameters required to calculate the objects in the other tables. A row in this table MUST exist in order for a row to exist in any other table in this MIB module.5.2. frsldSmplCtrlTable
This is an optional table to allow control of sampling of the data in the data table.5.3. frsldPvcDataTable
This table contains the calculated data. It relies on configuration from the control table.
5.4. frsldPvcSampleTable
This table contains samples of the delivery and availability information from the data table as well as delay information calculated over the sample period. It relies on configuration from both the control table and the sample control table.5.5. frsldCapabilities
This is a group of objects that define write capabilities of the read-create objects in the tables above.6. Persistence of Data
The data in frsldPvcCtrlTable and frsldSmplCtrlTable SHOULD persist through power cycles. Note, however, that the symantics of readiness for the rows still applies. This means that it is possible for a row to be reprovisioned as notReady(3) if the underlying DLCI does not persist. The data collected in the other tables SHOULD NOT persist through power cycles in that the reference TimeStamp is no longer valid.
(page 24 continued on part 2)
