Reply to thread

Message

<blockquote data-quote="kluisi" data-source="post: 291944" data-attributes="member: 17548">I thought there were a couple of problems with RAID 5 implementations. The one I was thinking of was the Write Hole:<a href="http://www.raid-recovery-guide.com/raid5-write-hole.aspx" target="_blank">&quot;Write hole&quot; phenomenon in RAID5, RAID6, RAID1, and other arrays.</a>When I searched for it, I also found this information which I was not aware of before, but it seems to be a new issue that is popping up with the increasing size of disks. It seems that if one drive in a RAID 5 array fails and is replaced, the error rate for huge disks is still low enough that when it tries to read all of the parity information off of the other disks in the array, it is likely that one of the reads for the parity bits will fail and cause your entire rebuild of the array to fail with it:<a href="http://www.zdnet.com/blog/storage/why-raid-5-stops-working-in-2009/162" target="_blank">Why RAID 5 stops working in 2009 | ZDNet</a>Yikes.</blockquote>

[QUOTE="kluisi, post: 291944, member: 17548"] I thought there were a couple of problems with RAID 5 implementations. The one I was thinking of was the Write Hole: [url=http://www.raid-recovery-guide.com/raid5-write-hole.aspx]"Write hole" phenomenon in RAID5, RAID6, RAID1, and other arrays.[/url] When I searched for it, I also found this information which I was not aware of before, but it seems to be a new issue that is popping up with the increasing size of disks. It seems that if one drive in a RAID 5 array fails and is replaced, the error rate for huge disks is still low enough that when it tries to read all of the parity information off of the other disks in the array, it is likely that one of the reads for the parity bits will fail and cause your entire rebuild of the array to fail with it: [url=http://www.zdnet.com/blog/storage/why-raid-5-stops-working-in-2009/162]Why RAID 5 stops working in 2009 | ZDNet[/url] Yikes. [/QUOTE]

Verification