FAILED SMART self-check - offlist

[m.roth at 5-cent.us]

Paul,

   Oh, good: you read his emails the same way I did, that he really didn't
understand what RAID 1 is, or how the system handles hardware. I was
afraid I was missing something, and didn't intend to insult him. I
really do try to make sure of my aim before I start coming down on
someone....

       mark

Paul Tader wrote:
> On 7/25/12 7:40 AM, Robert Canary wrote:
>> Sorry for the confusion.
>>
>> Two drives sda and sdb are mirrored with RAID 1
>> Getting errors on sda.
>> I removed the sda, and moved sdb into the sda position
>>
>> However, I am still getting the same errors on the sda.
>>
>> Robert Canary
>> OCDirect Electrical-Datacomm
>> (866) 594-0786 Fax
>> (270) 955-0362 Voice
>>
>> ----- Original Message -----
>>> On 07/25/12 07:18, Robert Canary wrote:
>>>> I have RHEL machine here that becomes so unresponsive the mail
>>>> server
>>>> it is hosting practically stops. It is currently running a mirrored
>>>> drive between sda and sdb. I have been getting the following errors
>>>> in the log files, so I moved the mirrored drive to sda and disabled
>>>> the (old) sda. Currently the system seems to be stable again,
>>>> however, I am still getting these errors.
>>>
>>> I'm sorry, but I don't understand what you did. You mention two
>>> drives,
>>> and then "the (old) sda". ?
>>>
>>> Next, is this RAID 1?
>>>>
>>>> Any ideas where to go to fix it. Could it be bad sectors where
>>>> transferred to the mirrored drive? Running smartmontools-5.3 I
>>>
>>> No. Bad sectors are *physical* errors on the platter.
>>> <snip>
>>> 	mark
>>>
>>> --
>>>    When was the pleistocene?
>>>    Around 1.8 million years ago, til about 10K years ago.
>>>    What era are we in now?
>>>    Trouble.
>>>
>>> --
>>> redhat-list mailing list
>>> unsubscribe mailto:redhat-list-request at redhat.com?subject=unsubscribe
>>> https://www.redhat.com/mailman/listinfo/redhat-list
>>>
>>
>
> Sorry, if you mean you physically move the sdb drive to the physical
> location sda was in?  Or in other words swapped hardware around?
>
> mark is correct.  This drive is failing and needs to be replaced.  In
> short you should run the mdadm tool to remove the drive from the
> mirrored set (if it's not already), physically replace the drive,
> partition and add the new drive into the mirror with the mdadm utility.
>
> As far as the server being slow, it could be that the OS is trying to
> read/write data from this failing drive and is slow because it's
> encountering I/O errors.  Run the command "dmesg".  I'm betting it
> filled with I/O errors.
>
> Hope this helps.
>
> Paul
>
> --
> redhat-list mailing list
> unsubscribe mailto:redhat-list-request at redhat.com?subject=unsubscribe
> https://www.redhat.com/mailman/listinfo/redhat-list
>