Filesystems Raid, ReiserFS and hours of fun

[Nicholas Leippe]

On Tuesday 15 March 2005 08:05 am, Jeff Davies wrote:
> I believe so I would have to double check.
> If we did could you tell me the affects?

Well, from what you described, as long as you did a clean shutdown, I wouldn't 
think it to have been part of the problem this time.  However, with reiserfs 
(and probably other journalled filesystems) it is extremely dangerous, since 
proper operation of the journal depends on known, physical commit states (ie, 
when the drive acks a write, the fs expects it to be physically written).  
W/write caching, it is possible for a hardware or even a software failure to 
lose the cached data before it has been committed to the platter.  So, I 
would recommend disabling it.  This doesn't help your current problem--I've 
digressed.

What kernel are you running?  The only threads I've seen recently are
a) some reports of reiserfs corruption with 2.6.8.? kernel
b) some reports of reiserfsck corruption with IDE-TCQ enabled in 2.6 kernels.



-- 
Respectfully,

Nicholas Leippe
Sales Team Automation, LLC
1335 West 1650 North, Suite C
Springville, UT  84663 +1 801.853.4090
http://www.salesteamautomation.com