Flashback substitution events, Software raid and health monitoring – HP IO Accelerator for BladeSystem c-Class User Manual
Page 55
Monitoring IO Accelerator health 55
...
Media status: Healthy; Reserves: 100.00%, warn at 10.00%; Data: 99.12%
Lifetime data volumes:
...Physical bytes written: 6,423,563,326,064
...Physical bytes read: 5,509,006,756,312
The following Health Status messages are produced by the fio-status utility:
•
Healthy
•
Low metadata
•
Read-only
•
Reduced-write
•
Unknown
HP IO Accelerator Management Tool
In the Device Report tab, look for the Reserve Space percentage in the right column. The higher the
percentage, the healthier the drive is likely to be.
SNMP
For details on how to configure SNMP health indicators, see "Setting up SNMP for Windows operating
systems (on page
)."
Flashback substitution events
The IO Accelerator is equipped with Flashback substitution capability. This capability allows for predictive
NAND flash failures to occur and keep the IO Accelerator functioning by substituting out bad locations or
even a whole NAND chip. The output of the fio-status –a command will show if any flashback
substitution events have occurred.
The following is an example of the output after entering the fio-status –a command. The numbers
represent the bank (0-3) and chip number that was substituted.
First flashback event -> No action to take as infant mortality condition on NAND
chip was caught per design.
Second flashback event –> Run fio-bugreport. Return the card (if under warranty)
when you see two or more flashback substitution events.
•
Flashback active on (bank a/chip m) and (bank b/chip n) (multi-bank failure)
•
Flashback active on (bank a/chip m) and (bank b/chip m) (chip failure)
On the second flashback event, include the output from running the fio-bugreport command when you
return the card.
Software RAID and health monitoring
Software RAID stacks are typically designed to detect and mitigate the failure modes of traditional storage
media. The IO Accelerator attempts to fail as gracefully as possible, and its new failure mechanisms are
compatible with existing software RAID stacks. A drive in write-reduced mode participating in a write-heavy
workload is evicted from a RAID group for failure to receive data at a sufficient rate. A drive in read-only