Hardware problem on a Drive [Predictive failure]

[Georgios76]

My server is a Intel S5520UR.

hard disk : 2x 146GB Seagate ST9146802SS (RAID 1)

I have the ESXi 4.0

I have the following warning now in my configuration : "DRIVE 1 on controller 0 Fw: 0003PREDECTIVE FAILURE".

please you can see below a snap shot of my configuration. Is-it possible that a drive is "down" ? Does I have to replace a drive in my Raid1 ?

I have 4 virtual machines on this server : the 2 first virtual machines works fine, the 2 other are very very slow now. They take 15 mins to load..normally a virtual machine took max 2 minuntes to load!

Is the slowness coming from this problem?

For information : on my "Chassis" of the server, all the LED are green, I did not have any "orange" LED".

thank you very much for your help.

[a_p_]

Welcome to the Community,

usually the drive is going to die soon if this error pops up. This error occurs if too many errors within a certain amount of time occurred. So make sure you replace the disk as soon as possible.

André

[Georgios76]

thank you so much for your answer.

I have some others questions regarding the problem I have with the VM and the Server,

VM:

• The two VM of the 4 which have not been stopped are running fine
• The two other VM of the 4 which have been stopped, they are very very very slow, and when they are powered on, they are block for 10/15 mins at 95%, is it normal?

Also on my INTEL Server, all my led are in green, the hard disk led too, normally if had a hard disk problem, they should be in orange, no?

Does I have to stop the server to replace the Hotplug SAS Hard disk? It will take a lot of time to rebuild the disk?

many thanks,

george

[a_p_]

The two other VM of the 4 which have been stopped, they are very  very very slow, and when they are powered on, they are block for 10/15  mins at 95%, is it normal?

I can't answer this question for sue, because I'm not familiar on how the Intel RAID works. However I guess that due to the unhealthy HDD still being active, there might be access issues which are responsible for the slowness.

Also  on my INTEL Server, all my led are in green, the hard disk led too,  normally if had a hard disk problem, they should be in orange, no?

That's the problem with a "predictive" failure. The HDD indicator only switches from green to amber if the HDD completely failed. If you can't identify the HDD, you may need to check whether Intel provides tools to identify the disks (i.e. switch the HDD light to amber or let it flash)

Does I have to stop the server to replace the Hotplug SAS Hard disk? It will take a lot of time to rebuild the disk?

With hotplug HDD's it's usually recommended to replace them with the system powered on, to ensure the controller will rebuild the new disk from the existing one and not the other way around. When you replace the disk, wait for a minute or two after unplugging the old disk and before inserting the new one, to ensure the controller can recognize that a disk is missing. Anyway, please consult your documentation about how to replace HDDs on your system! How long it takes to rebuild the disk also depends on the controller and the disk speed. With the 10k SAS disk I'd guess it will take less than an hour.

André

[Georgios76]

Thank you very much André for your answers, It will help me a lot, I will take a look to buy a new disk, the same as I have.

[Mourad_MADMIN]

Hey,

Im working on an old HP server, and the spare Disk is in "Predictive failure" state!, and i can not buy another one!
what tricks procedues i can follow to backup maximum my data
otherwise, is it possible to reconfigure the RAID to not lose my VMs

Best Regards !