[SOLVED] MegaRAID - How to identify bad drive

[jeffshead]

I have an IBM 5015 which has been crossed flashed to LSI 9260-8i. It is installed in a re-purposed PC that is now being used as an esxi server in a home lab. I have four SATA drives configured in RAID10.

I encounter disc latency issues and the following errors in MSM:

110    [Information, 0]    2016-09-28, 22:10:24    Controller ID:  0 Event From : 192.168.0.100   Corrected medium error during recovery:   PD   Port 0 - 3:0:2      Location   0x562db028    127
113    [Information, 0]    2016-09-28, 22:10:23    Controller ID:  0 Event From : 192.168.0.100   Unexpected sense:   PD 
    =   Port 0 - 3:0:2Unrecovered read error,   CDB   =    0x28 0x00 0x56 0x2d 0xb0 0x00 0x00 0x00 0xcf 0x00    ,   Sense   =    0xf0 0x00 0x03 0x56 0x2d 0xb0 0x28 0x0a 0x00 0x00 0x00 0x00 0x11 0x00 0x00 0x00 0x00 0x00    126
113    [Information, 0]    2016-09-28, 22:10:14    Controller ID:  0 Event From : 192.168.0.100   Unexpected sense:   PD 
    =   Port 0 - 3:0:2Unrecovered read error,   CDB   =    0x28 0x00 0x56 0x2d 0xb0 0x00 0x00 0x00 0xcf 0x00    ,   Sense   =    0xf0 0x00 0x03 0x56 0x2d 0xb0 0x09 0x0a 0x00 0x00 0x00 0x00 0x11 0x00 0x00 0x00 0x00 0x00    125
110    [Information, 0]    2016-09-28, 18:05:25    Controller ID:  0 Event From : 192.168.0.100   Corrected medium error during recovery:   PD   Port 0 - 3:0:2      Location   0x562e374c    124
113    [Information, 0]    2016-09-28, 18:05:24    Controller ID:  0 Event From : 192.168.0.100   Unexpected sense:   PD 
    =   Port 0 - 3:0:2Unrecovered read error,   CDB   =    0x28 0x00 0x56 0x2e 0x37 0x3f 0x00 0x00 0xc1 0x00    ,   Sense   =    0xf0 0x00 0x03 0x56 0x2e 0x37 0x4c 0x0a 0x00 0x00 0x00 0x00 0x11 0x00 0x00 0x00 0x00 0x00    123
113    [Information, 0]    2016-09-28, 18:05:19    Controller ID:  0 Event From : 192.168.0.100   Unexpected sense:   PD 
    =   Port 0 - 3:0:2Unrecovered read error,   CDB   =    0x28 0x00 0x56 0x2e 0x37 0x3f 0x00 0x00 0xc1 0x00    ,   Sense   =    0xf0 0x00 0x03 0x56 0x2e 0x37 0x4b 0x0a 0x00 0x00 0x00 0x00 0x11 0x00 0x00 0x00 0x00 0x00    122
110    [Information, 0]    2016-09-28, 18:05:15    Controller ID:  0 Event From : 192.168.0.100   Corrected medium error during recovery:   PD   Port 0 - 3:0:2      Location   0x562e3729    121
113    [Information, 0]    2016-09-28, 18:05:14    Controller ID:  0 Event From : 192.168.0.100   Unexpected sense:   PD 
    =   Port 0 - 3:0:2Unrecovered read error,   CDB   =    0x28 0x00 0x56 0x2e 0x37 0x00 0x00 0x00 0x3f 0x00    ,   Sense   =    0xf0 0x00 0x03 0x56 0x2e 0x37 0x29 0x0a 0x00 0x00 0x00 0x00 0x11 0x00 0x00 0x00 0x00 0x00    120
113    [Information, 0]    2016-09-28, 18:05:09    Controller ID:  0 Event From : 192.168.0.100   Unexpected sense:   PD 
    =   Port 0 - 3:0:2Unrecovered read error,   CDB   =    0x28 0x00 0x56 0x2e 0x37 0x00 0x00 0x00 0x3f 0x00    ,   Sense   =    0xf0 0x00 0x03 0x56 0x2e 0x37 0x29 0x0a 0x00 0x00 0x00 0x00 0x11 0x00 0x00 0x00 0x00 0x00    119
113    [Information, 0]    2016-09-28, 18:04:49    Controller ID:  0 Event From : 192.168.0.100   Unexpected sense:   PD 
    =   Port 0 - 3:0:2Unrecovered read error,   CDB   =    0x28 0x00 0x56 0x2d 0xc3 0x00 0x00 0x00 0xdf 0x00    ,   Sense   =    0xf0 0x00 0x03 0x56 0x2d 0xc3 0x20 0x0a 0x00 0x00 0x00 0x00 0x11 0x00 0x00 0x00 0x00 0x00    118

I'm assuming one of the four discs is failing??? How do I identify the bad one without shutting the server down, pulling each drive, attaching each drive to another machine and running a drive scanning utility?

 

[andrewbedia]

Hard Disk Sentinel

 

[jeffshead]

Hard Disk Sentinel

Please elaborate. How do I scan the drives with Hard Disk Sentinel? It's an esxi box. Do I install Hard Disk Sentinel on one of the VM's? I don't think that will work since none of the VM's access the RAID array directly.

 

[andrewbedia]

Oh, I thought you were on Windows. There's a way to read smart data on linux. You're looking for smartctl in the smartmontools package on red hat/centos. I don't have ESXI in front of me, otherwise I'd tell you how to get it on there.

EDIT: Now I see the esxi text at the top of the MSM window.

 

[azev]

If you check the far right on msm window it should give you some statistic about the drive; and as far as I can tell it is accurate as it increases as hdsentinel does.

 

[jeffshead]

If you check the far right on msm window it should give you some statistic about the drive; and as far as I can tell it is accurate as it increases as hdsentinel does.

Thanks for the info. Luckily, my cables are labeled so they do match the slot number listed in MSM.

I also learned MSM lists the serial number of the drives as well. That's how I actually identified the drive.