[mythtv-users] Consequences of drive failure

[Mike Perkins]

On 03/12/12 00:23, tortise wrote:
> On 3/12/2012 11:22 a.m., Nick Rout wrote:
>> Actually in the end I had to use dd or dd_rescue to make an image and
>> then copied them off the image (the drive was quite badly broken as a
>> result of stupid me ignoring the first sign of trouble - never again!)
>
> What was the first sign of trouble?  In my experience lost sectors may be a few
> lost randomnly or a sign that more are looming but whether more are looming and
> how quickly is the next batch going to occur seems pure speculation.  How
> quickly the number change seems a relevant consideration, if the number is
> climbing.... = alarm bells?!
>
> Also the temp the drive is running at also seems relevant to me.  A recent case
> change for me resulted in temp drops from ~ 52 degrees down to ~ 32 degrees.
> Got to be good for my drives, surely?!
>
> One of my 2TB drives has suddenly got 516 bad sectors, should I replace it
> asap?  (!)  I am considering how I copy it to a 3TB drive as a replacement...
>
In my case we were watching a program that froze for ~3 secs and then jumped 
maybe 30 secs forward. Repeatedly. Our setup does *not* experience any kind of 
freezing or jumps normally.

The next time I played this program it got to where we had abandoned it 
previously and then froze. I had to hard-poweroff the front end.

So, checked the back end dmesg and there are errors against a drive... fun when 
you have four in the box and /none/ of the messages use the same labels to refer 
to the drives!

I eventually found the correct drive and pulled the SATA and power cables; it 
then hung waiting for fsck until I connected keyboard and screen, rebooted and 
edited the duff drive out of /etc/fstab.

-- 

Mike Perkins