<div dir="ltr"><div class="gmail_quote">On Wed, Aug 20, 2008 at 11:19 AM, r2d2 <span dir="ltr"><<a href="mailto:r2d2z4j6y8@gmail.com">r2d2z4j6y8@gmail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="PADDING-LEFT: 1ex; MARGIN: 0px 0px 0px 0.8ex; BORDER-LEFT: #ccc 1px solid">
<div dir="ltr">
<div class="Ih2E3d">> Almost definitely a hardware error on the disk.<br></div>Bad news.
<div class="Ih2E3d"><br><br>> If you type "dmesg" do you see anything about IO errors, DMA resets, or timeouts?<br></div>I didn't notice anything about IO errors or other errors. As I mentioned, there was "Corrupt dmap page" in /var/log/messages before the filesystem was first damaged.
<div class="Ih2E3d"><br><br>> Depending on how important your recordings are, I'd suggest buying a new disk and copying<br>> them off one at a time then you'll find out which is on the broken bits of disk.<br>
</div>I will consider buying a new disk. It would be quite difficult to identify the broken bits as the filesystem resides on a logical volume built over 2 software RAID arrays (raid1, 2 disks and raid5, 4 disks). I know this is an awful mess. TV shows and movies are not so important, but I am disappointed : the RAID was inefficient to prevent this kind of problem.</div>
</blockquote>
<div> </div>
<div>Well, to be fair, if you are on RAID 5, it won't prevent filesystem errors, but it should allow you to replace the underlying damaged disk without additional data loss as it will rebuild the disk using the parity information. I personally don't like RAID/LVM type systems on Myth with the introduction of storage groups because I fear filesystem errors wiping out an entire volume more than a single disk failure taking a % of my data with it. But that's the risk I'm willing to take. </div>
<div> </div>
<div>Put another drive in it's place and let RAID do it's work</div>
<div> </div>
<div>Kevin</div></div></div>