XFS corruption on 3ware RAID6-volume

Justin Piszcz jpiszcz at lucidpixels.com
Wed Feb 23 09:42:17 CST 2011



On Wed, 23 Feb 2011, Emmanuel Florac wrote:

> Le Wed, 23 Feb 2011 10:29:15 -0500 (EST)
> Justin Piszcz <jpiszcz at lucidpixels.com> écrivait:
>
>>> /c0 Firmware Version = FE9X 4.10.00.007
>> The latest is:
>>
>> 9.5.1-9650-Upgrade.zip
>> 9.5.2-9650-Upgrade.zip
>> 9.5.3-9650-Upgrade.zip
>> 9650SE_9690SA_firmware_beta_fw4.10.00.016.zip
>> 9650SE_9690SA_firmware_beta_fw_4.10.00.019.zip <- latest
>
> Hum well, this is beta, I'd rather stick to released firmware :) His
> firmware is actually the latest stable version for the 9650.

Yes, but it specifically addresses issue with resets:


019:
  SCR FIRM03219 Pchip reset interrupt not handled
  SCR FIRM03220 Put parity segment to free state completely when deallocate unused parity segment

016:
SCR 2196:  Unexpected controller soft resets
Fixed an issue with regards to  deferral of write and read commands to help eliminate unexpected soft resets.

SCR 2217: Incomplete unit after multiple power failures when a RAID 1 unit is initializing after a rebuild.
This issue is fixed in this firmware version.

SCR 819: Controller may assert when cache is disabled in a RAID5/RAID6 configuration under heavy I/O load.
This issue is fixed in this firmware version.

SCR 2214: Performance drops during SES polling.
Fixed the issue where performance drops when a Storage Enclosure Processor (SEP) takes up to 700-800 msec to respond to SES polling.

Justin.


More information about the xfs mailing list