<div dir="ltr"><div class="gmail_extra"><div class="gmail_quote">On Wed, Jul 23, 2014 at 10:50 AM, Dan Wilga <span dir="ltr"><<a href="mailto:mythtv-users2@dwilga-linux1.amherst.edu" target="_blank">mythtv-users2@dwilga-linux1.amherst.edu</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div class="">On 7/23/14, 10:20 AM, Ian Evans wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
I've never looked at raid before. Here's one question: How long do you have to replace a dead drive? My drive appears to have died a few days into a two week vacation. IF the rest of the drives were still healthy, would you be able to toss a new drive in at that point and still be okay?<br>
</blockquote></div>
As long as only one drive dies in a RAID 5, you can take as long as you want. But if a second drive happens to die in the meantime, you're sunk. If you go to RAID 6, you're only out of luck if three drives die at once.<br>
<br>
After spending way too much time a few years ago recovering data from a failed drive, I've since gone to RAID 5 and now 6. Is it overkill? Perhaps. But this way I can have a drive fail, send the bad unit back for RMA and get the replacement--all while still having a one-drive safety net.<br>
<br>
Two other things I've found very helpful are smartmontools and raid-check. Between them, they help to point out a drive that is likely to fail before it becomes critical.<div class="HOEnZb"><div class="h5"><br><br></div>
</div></blockquote><div><br></div><div>Are you using software raid or a raid controller? <br></div></div><br></div></div>