IBM SAN768B-2, Руководство пользователя

Страница: 197 / 228

Appendix D. Tuning the system
Some sensitive applications may experience performance impact if the bit-error-rate
is high enough even though it still adheres to Fibre Channel specifications. The
probability of performance impact due to this condition is extremely low. In these
cases tuning may be required to optimize your application performance.
Tuning your system in the manner described in this section is only required if one
of the following blades installed in your chassis and are experiencing issues on
those blades:
v
FC 8 Gbps 64-port switch blade (FC 3864)
v
FC 8 Gbps Encryption blade (FC 3895)
v
FC 8 Gbps FCIP Extension Blade (FC 3890/FC 3891)
If your system only contains FC 16-Gbps port blades (FC 3648, FC 3632, or FC
3664) or Enhanced FC 8-Gbps port blades (FC 3649 or FC3633) then the steps
outlined in this section are
not applicable. If you are experiencing issues in your
system with only these blades then try other Troubleshooting recommendations or
call IBM Support for additional help.
Identifying whether you need to tune your system
To identify whether you need to tune your system begin by reviewing the chassis
error log for C2-1010 or C3-1010 events. The Cx-1010 error messages triggers when
a certain number of errors are detected on a port over a certain period of time.
This counter includes CRC and other bit errors for internal ports.
The Cx-1010 event message has the following format:
Message <timestamp>, [Cx-1010], <sequence number>, Slot <CP Slot#>, CRITICAL,
<system name>, S<slot number>, C<chip index>: Internal monitoring has identified
suspect hardware, blade may need to be reset or replaced: fault1:<fault1_cnt>,
fault2:<fault2_cnt> th2:0x<threshold_used>.
An example of finding a C2-1010 message using the
errdump
command on an IBM
SAN768B-2 is shown below:
errdump –a | grep C2-1010
2013/03/05-01:01:01:307873: [C2-1010], 2/0, Slot6, CRITICAL,,, S3,C1:
Internal monitoring has identified suspect hardware, blade may need to
be reset or replaced: fau1:0x10, fau2:0x02 th2:0x0
Identify the port blade that is having an issue in the error message by noting the
S<slot number>. In the above example, the S<slot number> is S3, meaning that the
port blade in slot 3 potentially has the issue.
Note: In a SAN768B-2 chassis slots 1-4 and 9-12 are port blades while slots 5 and 8
are core blades. In a SAN384B-2 chassis slots 1-2 and slots 7-8 are port blades
while slots 3 and 6 are core blades. The S<slot number> value may indicate an
error being detected either on a port blade or core blade.
If the error found is a C3-1010 message that is originating from a slot that has a FC
16 Gbps port blade (FC 3648 or FC 3632) or Enhanced FC 8 Gbps port blade (FC
© Copyright IBM Corp. 2011, 2014
169
|
|
|