![]() | Oracle System Handbook - ISO 7.0 May 2018 Internal/Partner Edition | ||
|
|
![]() |
|||||||||||||||||||||||||||||||||||||
Solution Type Problem Resolution Sure Solution 1011273.1 : Sun Blade[TM] 1000 Fault Manager FRU mislabeling
PreviouslyPublishedAs 215475 Applies to:Sun Blade 1000 Workstation - Version Not Applicable and laterAll Platforms SymptomsFault diagnosis on Sun Blade[TM] 1000 systems may have mislabeled FRU(s). Fault diagnosis produces a list of components suspected as faulty. The output of fmdump -v -u <case-id> describes suspected FRUs on lines beginning with 'rsrc:' or 'FRU:'. Follow the instructions in the resolution section below to determine if any FRUs are mislabeled in diagnoses of your system. CauseThis problem is due to a software defect on Sun Blade 1000 systems. SolutionA fix for this defect, Bug ID 6444795, is under development. Meanwhile, administrators should consult the following table to understand what FRUs are actually suspected to be faulty.
For example, consider this fmdump -v output from a Sun-Blade-1000 where diagnosis has discovered a faulty network adapter: # fmdump -v TIME UUID SUNW-MSG-ID Jul 10 13:48:48.4691 4a926ba5-d3d9-e036-e158-958d7eedc6fa PCI-8000-8S 33% fault.io.pci.bus-linkerr Problem in: hc:///motherboard=0/hostbridge=0/pcibus=1/pcidev=3/pcifn=0/pcibus=2 Affects: dev:////pci@8,700000/pci@3 FRU: hc:///component=PCI 3 67% fault.io.pci.device-interr Problem in: hc:///motherboard=0/hostbridge=0/pcibus=1/pcidev=3/pcifn=0/pcibus=2/pcidev=0/pcifn=0 Affects: dev:////pci@8,700000/pci@3/network@0 FRU: hc:///component=PCI 3 One of two faults is suspected here. According to this output, the FRU needing replacement in either case is the board in the slot marked PCI 3. Because of defect 6444795, we have to consult the table above to know that the faulty board is actually in the slot marked PCI 2. If a faulty component's FRU is labeled in the initial fmdump -v as PCI 2 or MB, the administrator should examine the resource (rsrc) or 'Problem in:' output associated with the component, and consult the table below.
In this partial fmdump -v output from a Sun Blade 1000, diagnosis has determined multiple components are suspected faulty, and a FRU component of MB or PCI 2 has been indicated. Due to defect 6444795, we must consult the table above to determine the actual location of these FRUS: 9% fault.io.pci.device-invreq Problem in: hc:///motherboard=0/hostbridge=0/pcibus=1/pcidev=5/pcifn=0 Affects: dev:////pci@8,700000/ebus@5 FRU: hc:///component=MB 9% fault.io.pci.device-invreq Problem in: hc:///motherboard=0/hostbridge=0/pcibus=1/pcidev=1/pcifn=0/pcibus=1/pcidev=4/pcifn=0 Affects: dev:////pci@8,700000/pci@1/scsi@4 FRU: hc:///component=PCI 2 9% fault.io.pci.device-invreq Problem in: hc:///motherboard=0/hostbridge=0/pcibus=1/pcidev=1/pcifn=0/pcibus=1/pcidev=5/pcifn=0 Affects: dev:////pci@8,700000/pci@1/scsi@5 FRU: hc:///component=MB For the first fault, the 'Problem in:' string matches none of patterns 1-3, meaning the FRU associated with the fault is indeed the motherboard. For the second and third faults, the 'Problem in:' strings match table pattern 3. This implies that the FRU associated with these faults is in the system's PCI 4 slot.
Attachments This solution has no attachment |
|||||||||||||||||||||||||||||||||||||
|