RAID 5 w/ IBM 4Mx controller problem

[penguinroundup]

Hi, odd question here...

we have an IBM xSeries 342 w/ a ServeRAID 4Mx controller (RAID 5). Yesterday, one of the hdd's went bad, and it automatically started recreating data on the hot-spare (so far, so good, right?) BTW, this server has Win2K Server running on it.

well, after the controller had finished the rebuild, we got a bunch of error messages saying that "the volume E: is corrupt. Please run chkdsk" - so we run chkdsk, which takes over 2 hours, and afterwards, more than half the files are corrupted on the volume. We wound up doing a restore of the entire volume today.

Any ideas why or how flipping over to the hot spare would cause this much damage? We've had hdd's go bad in other (identical) servers, and not had one problem.

thanks for any information.

 

[xmsre]

Ask IBM.

http://www-1.ibm.com/support/docvie...g1MIGR-44091&loc=en_US&cs=utf-8&cc=us&lang=en

Oops. I imagine the filesystem didn't look to healthy after the order of the stripes gets rearrainged.

 

[w3x]

It is possibly that the HDD you used as a hotspare (or one of the other HDDs) had a bad stripe on it therefore, when your system did the rebuild the bad information was passed to all the HDDs in the system.