Unsolved
This post is more than 5 years old
4 Posts
0
2710
April 20th, 2016 13:00
NX3100 internal RAID1 OS disk exchange
Hi all,
We have this NX3100 + PV MD1200 configured as NFS server running Debian. OS is installed on 2x 2.5" 136GB SAS internal disks configured in RAID1. One of the OS disk is having media errors and we would like to exchange this disk before it fails.
Enclosure Device ID: 32
Slot Number: 13
Drive's position: DiskGroup: 0, Span: 0, Arm: 1
Enclosure position: N/A
Device Id: 13
WWN:
Sequence Number: 2
Media Error Count: 9
Other Error Count: 1
Predictive Failure Count: 0
Last Predictive Failure Event Seq Number: 0
PD Type: SAS
System is out of warranty/support. I ordered new SAS disk from dell and I got 300GB SAS. Since it is internal disk, following the manual system needs to be shutdown to exchange the disk. I found youtube video "Dell PowerEdge R510 Internal Hard Drive Replacement". In this video its mentioned to be very careful when exchanging internal disk because system is off and on boot raid controller can mirror new empty disk to old OS disk destroying the OS. Usually when I exchange hot swap disk I put it in offline mode and prepare to remove with megasascli.
I plan to make full OS disk image before exchange. I will also do the same thing like with hotswap disks, putting failing disk in to offline mode, prepared to remove.
Did someon have same problem? How did you do the exchange?
Best,
Drazen
0 events found
tommo666
3 Apprentice
•
1.2K Posts
0
April 21st, 2016 15:00
The disk holder has label on the top indicating the disk positions, disk 12 is on the top and disk 13 below that. It is always a good idea to make a backup before changing storage components.
djalsovec
4 Posts
0
April 29th, 2016 05:00
I created full /sda disk backup with clonezilla before disk exchange. I didn't use megacli to put disk in offline mode, mak missing and prepare to removal, as I usually do with hotswap disks. I did it from raid controller bios.
After disk exchange, new disk was recognized but rebuild didn't start automatically, disk was ready, but radi1 disk was missing.
I used megacli to solve the problem.
megacli -Pdgetmissing -a0
Adapter 0 - Missing Physical drives
No. Array Row Size Expected
0 0 1 139392 MB
megacli -PdReplaceMissing -PhysDrv [32:13] -Array0 -row1 -a0
Adapter: 0: Missing PD at Array 0, Row 1 is replaced.
megacli -PDRbld -Start -PhysDrv [32:13] -a0
Started rebuild progress on device(Encl-32 Slot-13)