User`s guide

System Troubleshooting and Diagnostics
5.2 Product Fault Management and Symptom-Directed Diagnosis
Example 5–7 (Cont.) Error Log Entry Indicating Q-Bus Error
BPCR ECC80024
.
.
.
CESR 80000200
CP2 IO ERROR
ERROR SUMMARY
.
.
.
DSER 00000080
Q-22 BUS NXM
.
.
.
CIOEAR2 00001468
cp2 IO error address = 20001468
NDAL commander id (cp2 transac) = 0(X)
.
.
.
IPCR0 00000020
LOCAL MEMORY EXTERNAL ACCESS ENABLED
ANAL/ERR/OUT=QBUS QBUS.ZPD
5.2.8 Interpreting DMA Host Transaction Faults Using
ANALYZE/ERROR
Some kernel errors may result in two or more entries being logged. If the
SHAC DSSI adapters or the SGEC Ethernet controller or other CDAL device
(residing on the processor module) encounter host main memory uncorrectable
ECC errors, main memory NXMs or CDAL parity errors or timeouts, more
than one entry results. Usually there will be one Polled Error entry logged by
the host, and one or more Device Attention and other assorted entries logged
by the device drivers.
In these cases the processor module or one of the four memory modules are
the most likely cause of the errors. Therefore, it is essential to analyze Polled
Error entries, since a polled entry usually represents the source of the error
versus other entries, which are simply aftereffects of the original error.
Example 5–8 provides an abbreviated error log for a polled error. Example 5–9
provides an example of a device attention entry.
5–28 System Troubleshooting and Diagnostics