ProLiant Servers (ML,DL,SL)
1753599 Members
6609 Online
108796 Solutions
New Discussion

Re: Unable to replace failed disk in ML350G5

 
Alex90
Visitor

Unable to replace failed disk in ML350G5

Hi!

I have RAID1+0 with four disks in it and now one of them is degraded (750 GB, bay 1). Unfortunately it's not possible to install the same disk as degraded one. So I've tried to replace the disk with another one with higher capacity (1TB). This disk must be compatible with e200i controller because it was part of another array in the past. But it doesn't start rebuilding, it doesn't recognize the new disk either when installed in first bay.

Array disks are seated in 1-4 bays. The fifth bay is taken by that disk I've tried to install. I'm also attaching ADU log.

Spoiler

So the questions are:

1) Why the rebuild process doesn't start?

2) What could be done to repair array without stopping the server (through ACU maybe)?

Thank you in advance!

4 REPLIES 4
Jimmy Vance
HPE Pro

Re: Unable to replace failed disk in ML350G5

Drives of larger capacity can be used to replace a failed drive. The replacement drive must be put in the same drive bay as the failed drive before a rebuild will start.  The only time a drive in a different bay can be used as a replacment is when the drive is configured as an online spare.

No support by private messages. Please ask the forum! 
Alex90
Visitor

Re: Unable to replace failed disk in ML350G5

Thank you for reply!

When I insert a new drive in first bay (where failed drive resided) the orange indcator goes out for about ten seconds then lights up again without any changes in ACU. ADU report attached above describes the situation where first bay is taken by 2TB disk which was functional in another RAID (RAID was destroyed due to urgent need of a new HDD). The second attempt was done with 1TB disk (also functional). But ACU doesn't even see the replacement!

Screenshot attached shows the physical disks layout. The first bay is occupied by functional 2TB disk.

1) What kind of problem could prevent the auto rebuild?

2) Any way to add a new spare to the damaged array?

2) What else could be done to rebuild the array?

Thank you in advance for any help!

Jimmy Vance
HPE Pro

Re: Unable to replace failed disk in ML350G5

The first thing you need to do is make sure you have a good backup of the data.   Have you ran the Array Diagnostics Utility?  ACU isn't the same.  What is the unused disk in bay 5?  Was the drive you are trying to install in bay 1 removed from another array? If so was the array deleted using ACU on the other system before moving the drive?  THe array information is written on each disk and if the replacement disk has conflicting array data it can cause problems

No support by private messages. Please ask the forum! 
Alex90
Visitor

Re: Unable to replace failed disk in ML350G5

Mr, Vance, thank you for you help and attention!

I took all the precautions and solutions you gave me. I also ran ADU and attached report above. 

Neveretheless, the problem has gone away after server reboot. During post the bay 1 green led lit and finally rebuild process was started with the new disk installed. The one thing I still don't understand is why the reboot was nessesary, why the hot replacement with the clear disk wasn't enough.