Using Cisco RFGW-10 MIBs
This chapter describes the objects and MIBs that are needed to use Simple Network Management Protocol (SNMP) requests to perform the following tasks on a Cisco RFGW-10.
Tips and Guidelines
When using SNMP to manage the Cisco RFGW-10, be aware of the following points.
IF-MIB Caching
In this release, the Cisco RFGW-10 implements a cache to allow continuous polling of the ifTable interface counters, without creating spikes in the CPU usage. An SNMP request for these counters returns the values that were last stored in the counter cache memory, instead of returning the current run-time value of these counters. This improves performance, because it means the Cisco IOS or IOS-XE software does not have to poll each line card to obtain these counters when an SNMP request is made.
The ifTable counter cache is updated approximately every 10 seconds, which means that if you read the ifTable interface counters more quickly than every 10 seconds, the SNMP request might not return new values. The run-time counters do continue to increment, however, to account for the actual traffic occurring on the interfaces, and another SNMP request in 10 seconds does show the new values.
SNMP-Based and CLI-Based Counters
The SNMP specifications do not allow most SNMP-based counters to be cleared, except at system initialization. Instead, during normal operations the counters continue incrementing until they reach their maximum value, at which point they wrap around to zero and continue incrementing again.
This behavior requires the following considerations when managing the RFGW-10 using SNMP commands:
- 32-bit counters—A 32-bit counter wraps around to zero after reaching approximately 4.2 billion. On a busy RFGW-10, this means that byte and packet counters could wrap around after only a few days. To ensure that you are maintaining the correct counts for packets and other objects, regularly poll the desired counters and always save the previous values. Subtract the previous value from the current value, and if the difference between the two counters becomes negative, it indicates that the counters have wrapped.
To accurately total the counters over a period of several weeks or months, you might also need to keep track of the number of times that the counter wraps during this time period. You should poll the counters often enough so that they do not wrap around to zero more than once without being detected.
Tip Some SNMPv3 MIBs are beginning to include 64-bit counters, as well as 32-bit counters, for many of the same objects. If given a choice, use the 64-bit counters, because they typically will not wrap around to zero for months or years, if ever.
- Counting from a specified event or time period—SNMP-based counters begin incrementing from zero when the RFGW-10 is powered on, and continue incrementing until they wrap. To track the number of packets or other objects from a particular event, you must save the value of the counters at the time of the event. Then when you want to obtain a new packet count, compare the current value of the counters with the saved value.
- Comparison with command-line interface (CLI) values—Many show commands have a corresponding clear command that resets the counters to zero. The clear command, however, affects only the counters that are displayed by the CLI, not the SNMP-based counters. In addition, many CLI-based counters automatically reset whenever a certain function, such as resetting an interface, is performed. This means that the counters displayed using CLI commands are not usually the same as the counters displayed by SNMP commands. Be aware of these differences when comparing the CLI-based and SNMP-based counters.
Route Processor Redundancy (RPR) Modules on the Cisco RFGW-10
On a Cisco RFGW-10 running Cisco IOS Release 12.2(44)SQ, SNMP configuration commands and CLI commands are not synchronized to the standby Supervisor. Hence, the active configuration is lost when a switchover occurs and the standby Supervisor is active. When the RFGW-10 switches back to the original Supervisor, the original configuration is restored.
Use CLI commands for critical configurations and save the configuration to the startup-config to ensure that the critical configuration is active during any switchovers.
Obtaining Basic Information About the RFGW-10
Basic information about the Cisco RFGW-10 can be obtained from objects in the following MIBs:
OLD-CISCO-CHASSIS-MIB
The following object in the OLD-CISCO-CHASSIS-MIB provides a convenient location to store the chassis serial number for the RFGW-10, so that it can be easily retrieved when calling Cisco Technical Support:
- chassisId—Provides the serial number or ID number for the chassis, as defined by the snmp-server chassis-id command, which is typically used to identify the service contract and levels of service that you have purchased from Cisco Technical Support. This object defaults to the empty string, so you must use the snmp-server chassis-id command to set the value of this object before you can retrieve it.
SNMPv2-MIB
The following objects in the SNMPv2-MIB provide basic information about the RFGW-10, its software, and other run-time information:
- sysDescr—Provides an overall description of the RFGW-10, including its model number and the version of Cisco IOS or IOS-XE software that it is running. For example:
- sysObjectID—Provides the specific model number, as it is defined in the CISCO-PRODUCTS-MIB. For example:
- sysUpTime—Provides the time, in hundredths of a second, since the RFGW-10 was last initialized. For example:
- sysContact—Provides the name, phone number, or other identifying information for the person or department responsible for this RFGW-10, as it was entered using the snmp-server contact command. For example:
- sysLocation—Provides a description of the RFGW-10’s location, as it was entered using the snmp-server location command. For example:
ENTITY-MIB
The following objects in the ENTITY-MIB provide basic information about the RFGW-10 hardware:
- entPhysicalDescr—Provides a description of each hardware component in the RFGW-10. For example, the following is a typical description for the Cisco RFGW-10 chassis:
- entPhysicalHardwareRev—Provides the hardware revision of each component, if present and supported for that particular component. For example:
- entPhysicalSerialNum—Provides the serial number for each component, if present and supported for that particular component. For example:
- entPhysicalModelName—Provides the model name for each component, if present and supported for that particular component. For example:
Note
Also see the next section for more information about the ENTITY-MIB and how to use it.
Managing Physical Components
The Cisco RFGW-10 supports a number of MIBs for the management of the physical components. These MIBs provide the following functions:
- Organizes the physical entities in the chassis into a containment tree that describes the relationship of each entity to all other entities
- Monitors and configures the status of field-replaceable units (FRUs)
- Maps physical ports to their respective interfaces
- Provides asset information for asset tagging
- Provides firmware and software information for chassis components
See the following sections for a description of each MIB, as well as instructions on how to use the MIBs to track the components in the RFGW-10:
Tip To retrieve the chassis serial number for the RFGW-10, retrieve the chassisId object from the OLD-CISCO-CHASSIS-MIB. This object defaults to the empty string, so you must use the snmp-server chassis-id command to set the value of this object before you can retrieve it.
ENTITY-MIB
The Cisco RFGW-10 uses the ENTITY-MIB, which is defined as the standard RFC 2737, to manage its physical components, which are known as entities. An entity could be a card, a port on a card, a major subsystem on a card, a slot in the chassis, a field-replaceable unit (FRU), or any other equipment that is installed in the RFGW-10.
The ENTITY-MIB defines a set of objects that uniquely identify each entity in the RFGW-10, using a hierarchical containment tree that shows how each entity relates to each other. Other MIBs can then use the objects defined by the ENTITY-MIB to provide additional information about each entity.
The following are the most important objects in the ENTITY-MIB for the management of physical entities on the RFGW-10:
- entPhysicalTable—Describes each physical component (entity) in the RFGW-10. The table contains a row entry for the top-most entity (the chassis) and then for each entity in the chassis. Each entry provides the name and description of the entry, its type and vendor, and a description of how the entity fits into the containment tree.
- entPhysicalIndex—Uniquely identifies each entry. This value is guaranteed to be unique across all equipment in this chassis and across all MIBs, allowing you to correlate the data from several MIBs for any particular entity.
- entAliasMappingTable—Maps each physical port’s entPhysicalIndex value to the corresponding ifIndex value in the ifTable in the IF-MIB. This provides a quick way of identifying a particular port with a particular interface.
- entPhysicalContainsTable—For each physical entity, lists the entPhysicalIndex value for any child objects of the entity. This provides an easy way of creating the container tree for the RFGW-10, which shows the relationship between physical entities in the chassis.
Typically, the container tree is organized as follows:
– The chassis is the topmost level and contains the processor card and chassis slots.
– Chassis slots contain the individual line cards and I/O controller (if installed).
Performing Inventory Management
The ENTITY-MIB provides all of the information needed to collect an inventory of the physical components in the RFGW-10. The following procedure illustrates one way this can be done, using a RFGW-10. In this example, the RFGW-10 contains the following cards:
Slot 1: RFGW-10 Supervisor V-10GE, 2x10GE(X2) and 4x1GE(SFP)
Slot 2: RFGW-10 Supervisor V-10GE, 2x10GE(X2) and 4x1GE(SFP)
Slot 3: RFGW-10 Universal Downstram EQAM Card, 12 RF ports, 48 QAMs
Slot 13: RFGW-10 Timing, Communication, and Control Card
Slot 14: RFGW-10 Timing, Communication, and Control Card
To collect and organize the information in the ENTITY-MIB, use the following procedure.
Step 1 Collect the list of physical entities by displaying all of the entPhysicalDescr objects. For example:
Step 2 Obtain additional information about each entPhysicalDescr object by collecting the entPhysicalVendorType, entPhysicalName, and entPhysicalClass objects. Use the index value to match the objects with their corresponding entPhysicalDescr object. Table A-1 shows typical descriptions for the objects used in this example.
|
|
|
|
|
|
|---|---|---|---|---|
Supervisor V-10GE with 2 10GE X2 ports, and 4 1000BaseX SFP ports |
||||
Supervisor V-10GE with 2 10GE X2 ports, and 4 1000BaseX SFP ports |
||||
Step 3 To create the containment tree for the RFGW-10, collect the EntPhysicalContainedIn object for each entPhysicalDescr object. The value in EntPhysicalContainedIn is the index number for the parent (or “container”) for the corresponding entPhysicalDescr device.
Table A-2 shows the parent container for the entPhysicalDescr objects being used in this example:
Step 4 (Optional)If a parent object contains multiple children that are the same type of object, such as a RFGW-10 that contains multiple line card slots (Chassis Slots), use the entPhysicalParentRelPos objects to organize the child objects into their proper order. The entPhysicalParentRelPos objects contain an integer that shows the sequential order of the child objects. This integer typically starts incrementing from 0, so that it matches the actual numbering of the physical objects (slot 0 has an entPhysicalParentRelPos value of 0, slot 1 has an entPhysicalParentRelPos value of 1, and so forth).
Note If entPhysicalParentRelPos contains –1, then the object does not have an identifiable relationship with the other objects.
Table A-3 shows how the entPhysicalDescr objects that refer to chassis slots can be put into their physical order by using their entPhysicalParentRelPos values. For example, entPhysicalDescr.4 has an entPhysicalParentRelPos value of 3, which indicates that this slot is slot 3 in the RFGW-10 chassis.
|
|
|
|
|
|
|---|---|---|---|---|
Step 5 (Optional)To map a physical interface to its ifIndex, which is defined in IF-MIB and used in other MIBs to uniquely identify a logical interface, use the entAliasMappingIdentifier object.
For example, the following shows the entAliasMappingIdentifier values for the RFGW-10 used in this example. In this example, entPhysicalDescr.3019 (which Table A-1 identifies as the Qam3/1.1 interface) maps to an ifIndex value of 199.
Generating SNMP Traps
This section describes how to configure the Cisco RFGW-10 to generate SNMP traps when certain events or conditions occur on the RFGW-10. To use SNMP commands to configure the RFGW-10 to generate SNMP traps, you must define at least one target host to receive the traps, using the following procedure:
Tip You can also use the command-line interface (CLI) to enable and configure the generation of traps on the RFGW-10. For information on using the CLI, see the “Enabling Notifications” section.
Step 1 Create an entry in the snmpTargetAddrTable, which is defined in SNMP-TARGET-MIB, for each host that is to receive traps. Each entry contains the following objects:
- snmpTargetAddrName—Unique string, up to 32 characters long, that identifies this host.
- snmpTargetAddrTDomain—The TCP/IP transport service to be used when delivering traps to this host, typically snmpUDPDomain.
- snmpTargetAddrTAddress—The transport address for the host, typically a six-octet value that is composed of the host’s four-byte IP address followed by the two-byte UDP port number to which the traps should be sent.
- snmpTargetAddrTimeout—Maximum period of time, in hundredths of a second, that the Cisco RFGW-10 waits for a response from the host (if any). The default is 1500 (15 seconds).
- snmpTargetAddrRetryCount—Default number of times that the Cisco RFGW-10 resends a trap if a response is not received within the timeout period. The default value is 3 retries.
- snmpTargetAddrTagList—List of tags (defined below) that should be associated with this particular target host. If a host’s tag value matches an snmpNotifyTag value, the host receives the types of notifications that are defined by the corresponding snmpNotifyType.
- snmpTargetAddrParams—Arbitrary string, up to 32 characters long, that identifies an entry in the snmpTargetParamsTable, which defines the parameters to be used in generating traps.
- snmpTargetAddrStorageType—Type of storage to be used for this row entry: volatile(2), nonVolatile(3), permanent(4), or readOnly(5). The default is nonVolatile(4).
- snmpTargetAddrRowStatus—Must be set to createAndGo(4) or createAndWait(5) to create this row entry. This object must be set only after all of the other entries in the row have been set.
Step 2 Create an entry in the snmpTargetParamsTable, which is defined in SNMP-TARGET-MIB, to define the SNMP parameters that the RFGW-10 should use when generating SNMP notifications. Each entry contains the following objects:
- snmpTargetParamsName—Unique string, up to 32 characters long, that defines this particular entry. This string is also used in the snmpTargetAddrParams to define the parameters to be used when sending traps to any particular host.
- snmpTargetParamsMPModel—Version of SNMP to be used in sending this trap: 0=SNMPv1, 1=SNMPv2c, and 3=SNMPv3.
- snmpTargetParamsSecurityModel—Version of SNMP security to be used in sending traps: 0=SNMPv1, 1=SNMPv2c, and 3=SNMPv3.
- snmpTargetParamsSecurityName—String, up to 32 characters long, to be used in identifying the Cisco RFGW-10 when sending traps.
- snmpTargetParamsSecurityLevel—Type of security to be used when sending traps: noAuthNoPriv(1), authNoPriv(2), and authPriv(3).
- snmpTargetParamsStorageType—Type of storage to be used for this row entry: volatile(2), nonVolatile(3), permanent(4), or readOnly(5). The default is nonVolatile(4).
- snmpTargetParamsRowStatus—Must be set to createAndGo(4) or createAndWait(5) to create this row entry. This object must be set only after all of the other entries in the row have been set.
Step 3 Create an entry in the snmpNotifyTable, which is defined in the SNMP-NOTIFICATION-MIB. Each row in this table contains the following objects, which define a set of host targets that are to receive traps:
- snmpNotifyName—Unique string, up to 32 characters, that identifies this particular row entry.
- snmpNotifyTag—Arbitrary string, up to 255 characters, that identifies the set of hosts to receive traps. This tag value is matched against the snmpTargetAddrTagList object to determine which hosts should receive which traps.
- snmpNotifyType—Defines the type of trap to be set: trap(1) or inform(2). The default is trap(1).
- snmpNotifyStorageType—Type of storage to be used for this row entry: volatile(2), nonVolatile(3), permanent(4), or readOnly(5). The default is nonVolatile(4).
- snmpNotifyRowStatus—Must be set to createAndGo(4) or createAndWait(5) to create this row entry. This object must be set only after all of the other entries in the row have been set.
Step 4 Optionally create rows in the snmpNotifyFilterProfileTable and snmpNotifyFilterTable, which are defined in the SNMP-NOTIFICATION-MIB. These tables create notification filters that limit the types of notifications that the RFGW-10 sends to particular hosts.
Step 5 Optionally enable traps and notifications to be sent. Most other MIBs include their own objects of NOTIFICATION-TYPE that enable or disable feature-specific traps. These notification objects also define the varbinds that are sent with each trap, which contain the specific information about the event that occurred.
A number of notifications and traps can also be enabled using CLI commands. Table A-4 lists some of the most common traps, how they can be enabled through the CLI, and the situations that generate these traps.
Identifying Cisco Unique Device Identifiers
In order to use UDI retrieval, the Cisco product in use must be UDI-enabled. A UDI-enabled Cisco product supports five required Entity MIB objects. The five Entity MIB v2 (RFC-2737) objects are as follows:
Although the show inventory command may be available, using that command on devices that are not UDI-enabled produces no output.
Before using the UDI Retrieval feature, you should understand the following concepts:
Unique Device Identifier Overview-Each identifiable product is an entity, as defined by the Entity MIB (RFC-2737) and its supporting documents. Some entities, such as a chassis, will have subentities like slots. An Ethernet switch might be a member of a superentity such as a stack. Most Cisco entities that are orderable products will leave the factory with an assigned UDI.
The UDI information is printed on a label that is affixed to the physical hardware device, and it is also stored electronically on the device in order to facilitate remote retrieval.
A UDI consists of the following elements:
- Product identifier (PID) The PID is the name by which the product can be ordered; it has been historically called the Product Name or Part Number. This is the identifier that one would use to order an exact replacement part.
- Version identifier (VID) The VID is the version of the product. Whenever a product has been revised, the VID will be incremented. The VID is incremented according to a rigorous process derived from Telcordia GR-209-CORE, an industry guideline that governs product change notices.
- Serial number (SN) The SN is the vendor-unique serialization of the product. Each manufactured product will carry a unique serial number assigned at the factory, which cannot be changed in the field. This is the means by which to identify an individual, specific instance of a product.
Feedback