IBM Flex System p270 Compute Node Installation And Service Manual page 475

v Follow the suggested actions in the order in which they are listed in the Action column until the problem is
solved.
v See Chapter 7, "Parts listing for IBM Flex System p270 Compute Node," on page 95 to determine which
components are CRUs and which components are FRUs.
v If an action step is preceded by Trained service technician only, that step must be performed only by a trained
service technician.
Isolation
procedure or
symbolic failing
item Symptom
FSPSP10 The failing item that
follows this
procedure in the
failing item list is not
valid or is missing
for this system's
configuration.
FSPSP11 The service processor
has detected an error
in the system unit.
The DIMM that was
FSPSP12
in the failing item
list did not correct
the memory error.
FSPSP14 The service processor
cannot establish
communication with
the compute node
firmware. The
compute node
firmware will
continue to run the
system and
partitions while it
attempts to recover
the communications.
Compute node
firmware recovery
actions will continue
for approximately 30
- 40 minutes.
Action
2. See "Removing a DIMM" on page 44 for location information and the
removal procedure.
3. Install new memory DIMMs, as described in "Installing a DIMM" on
page 47.
1. If there is only one failing item after this FSPSP10 isolation procedure
in the failing item list:
a. Verify that the failing item is installed, connected, and seated
properly.
b. If the failing item is seated properly and the problem persists,
replace the failing item.
c. If the failing item is missing, add the item.
2. If multiple failing items are listed, they have identical serial numbers.
Remove all but one of the failing items.
1. Perform a firmware update. See "Updating the firmware" on page 16.
2. If the problem persists, replace the system-board and chassis assembly,
as described in "Replacing the system-board and chassis assembly" on
page 83.
Replace the system-board and chassis assembly, as described in
"Replacing the system-board and chassis assembly" on page 83.
1. View the event log in the management module to locate the SRC and
the time that the event was logged. See "Error logs" on page 441.
If progress codes are being displayed, the compute node firmware was
able to reset the service processor and solve the problem.
2. Record the time the log was created or when you first noticed this
SRC.
3. If progress codes are not being displayed, examine the management
module event log to see if an A7006995 SRC has been displayed.
If an A7006995 SRC has been displayed, the compute node is
powering off partitions and attempting a compute node dump. Follow
the action in the A7006995 SRC description if the partitions do not
power down as requested.
4. If an A7006995 SRC has not been displayed, has the A1xx SRC
remained for more than 40 minutes?
If so, the compute node firmware could not begin powering down the
partitions. Contact your next level of support to assist in attempting to
power down any remaining partitions and forcing a compute node
dump. Collect the dump for support and power off and power on the
compute node.
5. If an A1xx SRC has not remained more than 40 minutes, contact your
next level of support.
Chapter 8. Troubleshooting
463