HP StorageWorks 2/64 HP StorageWorks Fabric OS 5.x MIB Reference Guide (AA-RVH - Page 31
Understanding SNMP traps - storageworks director
View all HP StorageWorks 2/64 manuals
Add to My Manuals
Save this manual to your list of manuals |
Page 31 highlights
Once loaded, the MAX-ACCESS provides access levels between the agent and management station. The access levels are as follows: • Not-accessible: You cannot read or write to this variable. • Read-create: Specifies a tabular object that can be read, modified, or created as a new row in a table. • Read-only - Public: You can only monitor information. • Read-write - Private: You can read or modify this variable. Understanding SNMP traps An unsolicited message that comes to the management station from the SNMP agent on the device is called a trap. HP StorageWorks switches send traps out on UDP port 162 only, therefore, an SNMP management application must be running on UDP port 162. In order to receive traps, the management station IP address and severity level must be configured on the switch. Up to six trap recipients can be configured using Web Tools or the snmpConfig command. You can define a different message severity level for each of the recipients, so that some recipients receive all trap messages and others receive only the most critical. There are two main MIB trap choices: • FA MIB trap - Associated with the FA MIB (FA-MIB), this MIB manages SAN switches and devices from any company that complies with FA specifications. • HP-specific MIB trap - Associated with the HP-specific MIB (SW-MIB), manages HP StorageWorks switches only. The functionality of these MIBs have some overlap. If you enable both SW-MIB and FA-MIB traps, you might receive duplicate messages for switch events that trigger the trap. You can also use these additional MIBs and their associated traps: high-availability MIB (HA-MIB); FICON-MIB; and SW-EXTTRA. You can use the snmpConfig command to disable the FA-MIB, HA-MIB, FICON-MIB, and SW_EXTTRA; but neither the SW-MIB or the FE-MIB can be disabled. An event trap (swEventTrap, connUnitEventTrap, or swFabricWatchTrap) is basically an error message (errShow output) that is SNMP formatted and delivered. FA traps Consider enabling this trap if you want to use SNMP to monitor multiple connectivity units including HP StorageWorks switches. The switchStatusPolicySet command determines the FA-TRAP switch status related outputs: • connUnitStatusChange • connUnitSensorStatusChange • connUnitPortStatusChange The MIB-II system description swEventTrapLevel determines the output for the connUnitEventTrap. Events in the Error Log of a severity at or above the configured threshold will generate SNMP traps. The FibreAlliance Trap (FA-TRAP) can be configured to send traps using the snmpConfig command. See the HP StorageWorks Fabric OS 5.x command reference guide for more information on this command. HA traps Consider enabling these traps to monitor Field Replaceable Unit (FRU) status and Control Processor (CP) status when you have an HP StorageWorks director in your environment: • fruStatusChanged: This trap is generated by a FRU status change, such as a switch reboot or adding or removing a FRU component. • cpStatusChanged: This trap is generated by a change in the status of a CP, including a reboot, or firmware download. • fruHistoryTrap: This trap is generated when a FRU is added or removed. The high-availability trap (HA-TRAP) can be configured to send traps using the snmpConfig command. See the HP StorageWorks Fabric OS 5.x command reference guide for more information on this command. Fabric OS 5.x MIB reference guide 31