Isolate System Application Faults - Seagate Exos CORVAULT Hardware Installation And Maintenance Manual

controller module (CM), cable, connector, or switch. It also could be a failure involving more than one of these
components.
Use this section to gather common installation hardware fault information and isolate the fault.
CAUTION When you suspect a disk drive or connection is the fault, halt all input and output operations to the
drive group or groups from all hosts as a data protection precaution. Make sure your regularly scheduled backup is good
and that you conduct another backup at the time of failure as an additional data protection precaution.
To isolate hardware and connectivity faults associated with the storage enclosure:
1. Complete the following actions to gather fault information:
a. Examine host application or operating system logs for fault identification.
b. Examine system logs from both controller modules for errors, such as log and status output for fault identification.
c. Use system alerts to help narrow the fault to an area of the storage enclosure.
d. Visually inspect the front panel LEDs. A fault LED indicates there is a hardware issue and provides a general
location: the main bay, the auxiliary bay, or the rear panel area.
e. Access the identified location, then look for an amber fault LED to help identify a faulty module. The SAS
expander module fault LED faces the sidewall, so look carefully along the sidewall for faults in the main bay.
f. If you cannot find a fault in any of the listed areas, the fault most likely lies outside the storage enclosure.
2. Isolate the fault to one of the components in the storage enclosure. There can be a significant number of
components interacting to create a viable data path, each potentially creating an issue. For example, if a host-side
data error occurs, it could be hardware, such as the controller module or a faulty cable, or it could be the data host.
3. Proceed to the next section and take the recommended corrective action. This may mean obtaining a replacement
FRU of the same type before replacing the faulty module.

Isolate system application faults

System application faults generally fall into two categories: disk drive and array faults or controller module cache faults.
In the case of a faulty CM or a power loss, the cache memory flushes to nonvolatile memory. During the process of
writing the cache to nonvolatile memory, only those components used in the flush process receive super-capacitor
power while the Cache LED momentarily flashes. The cache memory flush typically takes 60 seconds for each 1 Gb of
cache. After the flush, the system uses the remaining super-capacitor power to refresh cache memory.
Chapter 6 Hardware installation and configuration issues
59