<div dir="ltr"><div dir="ltr">Hej Erik,<br></div><br><div class="gmail_quote"><br><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
If I understand things correctly this is about what HW-raid-controllers do (kick out disk if some smart-value is considered bad).<br>
Of course I can myself do some scripting that let smartd kick out disks, but still... why can't I found example of this -- and why isn't it considered as a part of setting up a mdadm raid?<br></blockquote><div><br></div><div><div>Actually this is the job of mdadm to kick a failed drive out of an
array. There are mechanisms which do that. A failed drive then has the
flag "F" (failed) in the output of /proc/mdstats .</div><div>Smartmontools are for monitoring only, they don't manage a raid array.</div></div><br></div></div>