Drive 1 on adapter 0 is predicting a future

[Patricia5959]

We have a Dell Power edge 1750 windows server 2003, Recently whenever we reboot the server, we receive this message:

Drive 1 on adapter 0 is predicting a future failure qualifier 00h, Press any key to continue.

After pressing any key, the server boots normal without issues. We are concerned with the message. How do I find out if our hard drive is really failing? And what is causing the failure.

Thanks Patricia

 

[technome]

The hard drive and the adapter maintain information concerning physical disk errors, speed of reads /writes, read/write errors, temperature, error on the interface bus etc. The threshold of critical errors over period of time has been exceeded, thus the warning. Basically this disk is likely to fail, could an hour or 5 years.
In raid 1, the message is not as critical as with a parity raid. With a parity raid, if another disk fails, and a that disk is replaced, a rebuild takes place. During a rebuild all disk surfaces are checked for errors, if the original disk with predictive failure alerts should go offline or fail during this time your array is toast.
If you have a resident program monitoring the array such as Dell's OpenManage or Lsi's Power Console, you should have numerous events as to the predictive failure, and some details.
I would order a new drive as a replacement, not a refurb or recert if possible.

........................................
Chernobyl disaster..a must see pictorial

http://www.kiddofspeed.com/default.htm

 

[Patricia5959]

Thanks for the information. I checked into the log and found several items. Since, this server not being used for 2 years the two errors were ESM log near capacity size. And a temp reaching minimum by 1 degree 'c'. Immediate after the the temp warning the system made the correction and posted a system ok.

I ran a test on Drive 1 and it failed the SMART test.

We are looking for a new hard drive but the old drive was a 73 gb Hard drive U320 SCSI 10k for a Dell poweredge 1750, aren't popular. I'm trying to find an upgrade that is new that is compatible.

Thanks for the direction towards location the problems
Patricia

 

[technome]

Generally Dell send of 15k drives as replacements, had no issues with 15k drives as such. If you use third party suppliers, try to get new verses refurb or recerts. If it has a 5 year warranty, it is new, anything less is a pull/recert/refurb.

That was minimum, right, not maximum. Minimum is not much of an issue unless condensation can take place <grin>

Good luck


........................................
Chernobyl disaster..a must see pictorial

http://www.kiddofspeed.com/default.htm

 

[Patricia5959]

Ok we have are replacement in hand. I know this is a really bad question but I'm going to ask it because we don't know. On the system hardware the Power Edge server have numbers "O" "1" and "2" imprinted over each hard drive slot to install the physical hardware.

On my partition software the drives start with "1".

So which hard drive is failing. Hard drive in the "O" slot or hard drive in the "1" slot.

We are all hoping it's not the main hard drive where the windows server software sits.

thanks for the help.

 

[technome]

If you have three drives, likely you have a raid 5. In raid 5, data is evenly divided over all the disks in the array, if a disk fails, the data on the failed disk, needed to keep the server going is supplied from the parity information residing on the remaining disks (very simplified) so basically you have lost no data. Should you lose another disk before the present failed disk is replaced, only then will data be lost.

Go into the raid bios setup, find the failed drive, you have a choice to "blink" the drive (causes the disks LED to blink). Blink it, the drive which blinks the LED approx every second, is the dead drive. "Un-blink" it when you are sure,( its a toggle feature).
Believe you can "blink" the drive from the Dell OpenManage software from within windows also.
Tis good to be cautious, as you would lose your array if the incorrect drive is "pulled".
The failed drive can be replaced with the server booted into Windows or if the server is powered off.


........................................
Chernobyl disaster..a must see pictorial

http://www.kiddofspeed.com/default.htm