Parity Errors

[lboregard]

i've just ran a parity sync that came up with errors.

attached is the syslog.

can you please help me interpret the log ?

from what i can perceive, seems like disk0 is going bad ...

pls advise

[Joe L.]

Your parity disk (disk0) is reporting a massive number of media errors.  (un-readable sectors)

Your first step should be to get a SMART report on the drive to learn its view on what is happening.
Type:
smartctl -d ata -a /dev/sdh


We are specifically looking for lines involving "sectors pending re-allocation" and re-allocated sectors, but post the entire smart report output.

Odds are very good you will need to RMA the drive, but the smart report will confirm this.

Joe L.

[lboregard]

thanks Joe L.

here's the smartctl report ...

[Joe L.]

The disk has already re-allocated 218 sectors and has 36 more pending re-allocation.

I'd consider the disk as a prime candidate for an RMA.  Replace it as soon as practical.

Joe L.

  5 Reallocated_Sector_Ct   0x0033   096   096   005    Pre-fail  Always       -       218
196 Reallocated_Event_Count 0x0032   095   095   000    Old_age   Always       -       268
197 Current_Pending_Sector  0x0022   100   100   000    Old_age   Always       -       36

[BRiT]

 5 Reallocated_Sector_Ct   0x0033   096   096   005    Pre-fail  Always       -       218
196 Reallocated_Event_Count 0x0032   095   095   000    Old_age   Always       -       268
197 Current_Pending_Sector  0x0022   100   100   000    Old_age   Always       -       36

You have some issues to work out with reallocated sectors and pending sector reallocation.

[lboregard]

...
I'd consider the disk as a prime candidate for an RMA.  Replace it as soon as practical.
...

got it ... will do

[lboregard]

i'm set to replace the parity drive ... just to be sure, is the following the correct procedure ?

Stop the array
power down
remove the failed parity drive
insert new parity drive in the same slot
Power up.
Go to the devices page, assign the new parity drive to the parity slot
Log in via telnet, or log in at the system console and type
initconfig
Refresh the main web-interface page in your browser. All the indicators on the data drives should turn blue, the parity drive will probably be orange.
Press "Start" to begin a full parity calculation of the newly initialized drive configuration.  All your data on the other drives will still be there.

(this taken from a previuos post by Joe L.)

[Joe L.]

i'm set to replace the parity drive ... just to be sure, is the following the correct procedure ?

Stop the array
power down
remove the failed parity drive
insert new parity drive in the same slot
Power up.
Go to the devices page, assign the new parity drive to the parity slot
Log in via telnet, or log in at the system console and type
initconfig
Refresh the main web-interface page in your browser. All the indicators on the data drives should turn blue, the parity drive will probably be orange.
Press "Start" to begin a full parity calculation of the newly initialized drive configuration.  All your data on the other drives will still be there.

(this taken from a previuos post by Joe L.)

No, that is not the procedure.

Stop the array
Power down
Replace the parity drive
Power up
The array will notice you changed the parity drive and wait for you to press "Start"  (You may have to check the checkbox under start to enable it)
Press Start

Wait for the parity drive to be rebuilt. 
When the re-build of parity is complete press the "Check" button.

The initial calculation of parity writes to the parity drive.  The subsequent "Check" reads it to be certain it was written to properly.

No need to go to the devices page unless it is not recognized as the replacement (it would not be if you used a different disk controller port)
No need to log in via telnet to type "initconfig"  That is only if you are removing a drive from the array and not replacing it, or setting a new disk configuration.   You are replacing an existing parity drive, and that is not a new configuration.

[lboregard]

again, thanks a lot !!!