[lug] best way to check a new hd that might be going bad

[George Sexton]

It's time consuming, but badblocks -wv works pretty well. Of course it's 
destructive...

On 1/10/12 12:55 PM, David L. Willson wrote:
> Many times for me, repeatedly dd'ing zeroes and pseudo-random bits onto a drive has cured or killed a questionable drive.
>
> David L. Willson
> Trainer, Engineer, Enthusiast
> RHCE MCT MCSE Network+ A+ Linux+ LPIC-1 NovellCLA UbuntuCP
> tel://720.333.LANS
> Freedom is better when you earn it. Learn Linux.
>
> ----- Original Message -----
>>
>> On 1/6/12 11:53 AM, karl horlen wrote:
>>> per my other recent post.
>>>
>>>
>>> i just bought two 1T drives and installed them in a software raid 1
>>> mirror in my server.  i'm already noticing a very high pitched
>>> whine that wasn't there with the older smaller drives.  it's
>>> doesn't sound super serious right now and appears to come and go
>>> but my guess is it's a harbinger of things to come.
>>>
>>>
>>> i imagine i can remove one of the drives at a time from the array,
>>> pull the drive from the bay (supermicro type .. not sure if hot
>>> plug or not) and reboot to determine which drive is producing the
>>> noise.
>>>
>>> outside of that, are there any cmds / procedures the audience might
>>> recommend for determining the physical health of my drives?  like
>>> i said they're brand new and newly formatted.
>> You can use the SMART management tools to monitor the physical health
>> of
>> the drives. smartd and smartctl are the tools to use. They also
>> support
>> some testing capabilities while online.
>>
>>> i would prefer to replace the drive if necessary before i restore
>>> my old server from backups.  or at a minimum, at least know what
>>> to expect going forward so i can prepare for it.
>>>
>>> if one drive goes down, since the entire system is living on a sw
>>> raid (two raid devices md0 and md1), i guess it's really a
>>> no-brainer to just slap a new drive in, partition correctly and
>>> let the array recover.  so perhaps, i just have to let the disk do
>>> it's thing and proceed through warranty if real errors start
>>> showing up.  it would save me from having to dban the drive though
>>> once it's configured.
>>>
>>> _______________________________________________
>>> Web Page:  http://lug.boulder.co.us
>>> Mailing List: http://lists.lug.boulder.co.us/mailman/listinfo/lug
>>> Join us on IRC: irc.hackingsociety.org port=6667
>>> channel=#hackingsociety
>>>
>> --
>> George Sexton
>> MH Software, Inc.
>> 303 438-9585
>> http://www.mhsoftware.com/
>>
>> _______________________________________________
>> Web Page:  http://lug.boulder.co.us
>> Mailing List: http://lists.lug.boulder.co.us/mailman/listinfo/lug
>> Join us on IRC: irc.hackingsociety.org port=6667
>> channel=#hackingsociety
>>
> _______________________________________________
> Web Page:  http://lug.boulder.co.us
> Mailing List: http://lists.lug.boulder.co.us/mailman/listinfo/lug
> Join us on IRC: irc.hackingsociety.org port=6667 channel=#hackingsociety
>

-- 
George Sexton
MH Software, Inc.
303 438-9585
http://www.mhsoftware.com/