LSI drivers for LSI 1064E RAID for Centos 6

samylines · 08-22-2015, 03:54 AM

Hello,

We have our server configured with 2 x 1TB SATA drives and hardware RAID1.

Recently one of the drives went bad and had to be replaced. Our NOC technician suggested we install "LSI drivers for LSI 1064E RAID" so that we monitor the RAID health for future purpose.

We are using Centos 6.7 (64 bit) on our server.

I have tried searching for "LSI drivers for LSI 1064E RAID" for Centos 6 but no luck. Can anyone please guide me from where can I download the drivers as suggested by our NOC engineer, so that we can monitor its health status in future.

Thank you for your help in advance.

k64 · 08-22-2015, 04:14 AM

You can Download from this link

IBM and LSI Basic or Integrated RAID SAS controller software matrix

After installitions please update your Kernel and the system

Regards

wroom · 08-24-2015, 03:44 PM

Quote:

Originally Posted by samylines

...Our NOC technician suggested we install "LSI drivers for LSI 1064E RAID" so that we monitor the RAID health for future purpose.

We are using Centos 6.7 (64 bit) on our server...

?
The LSI 1064E is one of the most compatible raid controller there is.
All Linux distributions i come in contact with have the mptsas drivers already.

And at least OpenSUSE always installs with the mpt-status and all.

Maybe you have the mpt-status command already?
Try "mpt-status -p" and it will probe for the HBA.

If you have a hardware raid setup with your LSI 1064E, it will tell you what ID your controller responds to. Like this:

Code:

# mpt-status -p
Checking for SCSI ID:0
Checking for SCSI ID:1
Checking for SCSI ID:2
Found SCSI id=2, use ''mpt-status -i 2`` to get more information.

Then, if you also got id=2, you can use the command "mpt-status -i 2" to get the following response:

Code:

# mpt-status -i 2
ioc0 vol_id 2 type IM, 2 phy, 278 GB, state OPTIMAL, flags ENABLED
ioc0 phy 1 scsi_id 4 SEAGATE  ST330055SSUN300G 0791, 279 GB, state ONLINE, flags NONE
ioc0 phy 0 scsi_id 3 SEAGATE  ST330055SSUN300G 0791, 279 GB, state ONLINE, flags NONE

Ah! How nice! My raid mirror is OPTIMAL.

Also, instead of probing for the controller, you can do this:

Code:

# cat /proc/scsi/mptsas/2
ioc0: LSISAS1064E B3, FwRev=01170400h, Ports=1, MaxQ=286

But, if you use the LSI 1064E without setting up a raid on the disks, it will literally "go native".
Running "mpt-status -p" will give the following answer:

Code:

# mpt-status 
open /dev/mptctl: No such file or directory
  Try: mknod /dev/mptctl c 10 220
Make sure mptctl is loaded into the kernel

Then your SAS or SATA disks will be natively connected like any SAS/SATA disks on any non-raid controller.

You will be able to issue "smartctl -a /dev/sda" commands to watch your SAS or SATA disks wellness.
But since you have not set up any raid sets on the controller, it will not show up like an mptsas device.

However, if you have set up a hardware raid set on the LSI 1064E controller, it can be monitored using the mpt-status command.
Also, looking at the disks smart status can be done like the following:

First find out where the disks and the controllers are:

Code:

# cat /proc/scsi/scsi
Attached devices:
Host: scsi0 Channel: 00 Id: 00 Lun: 00
  Vendor: TEAC     Model: DV-28E-N         Rev: P.6A
  Type:   CD-ROM                           ANSI  SCSI revision: 05
Host: scsi2 Channel: 00 Id: 00 Lun: 00
  Vendor: SEAGATE  Model: ST330055SSUN300G Rev: 0791
  Type:   Direct-Access                    ANSI  SCSI revision: 05
Host: scsi2 Channel: 00 Id: 01 Lun: 00
  Vendor: SEAGATE  Model: ST330055SSUN300G Rev: 0791
  Type:   Direct-Access                    ANSI  SCSI revision: 05
Host: scsi2 Channel: 01 Id: 02 Lun: 00
  Vendor: LSILOGIC Model: Logical Volume   Rev: 3000
  Type:   Direct-Access                    ANSI  SCSI revision: 02

Ok. SCSI device 0, aka /dev/sg0, is a TEAC CD/DVD reader on SCSI bus 0 with bus ID 0 and LUN 0.
Devices /dev/sg1 and /dev/sg2 are my SAS disks.
And SCSI device /dev/sg3 is my LSI 1064E HBA/RAID controller.

So, "smartctl -a /dev/sg1" will give me all the smart status on the first disk, and "smartctl -a /dev/sg2" will give me all the smart status on the second disk. And i can run a smart long selftest on the first drive using "smartctl -t long /dev/sg1".
Marvellous!

But what about "smartctl -a /dev/sg3"?

Code:

# smartctl -a /dev/sg3
smartctl 6.0 2012-10-10 r3643 [x86_64-linux-3.7.10-1.16-desktop] (SUSE RPM)
Copyright (C) 2002-12, Bruce Allen, Christian Franke, www.smartmontools.org

Vendor:               LSILOGIC
Product:              Logical Volume  
Revision:             3000
User Capacity:        298 999 349 248 bytes [298 GB]
Logical block size:   512 bytes
Logical Unit id:      0x600508e0000000009ea7d217ceea7302
Device type:          disk
Local Time is:        Mon Aug 24 22:32:25 2015 CEST
Device does not support SMART

Error Counter logging not supported
Device does not support Self Test logging

bummer. But of course the controller itself does not have smart data. Only the disks.
But at least it told me it constitutes a "LSILOGIC Logical Volume".

The same result is given by the command: "smartctl -a /dev/sda", (where my raid1 is mapped by BIOS to /dev/sda).

You can of course install the megaraid cli to be able to both monitor and manage the configuration of the LSI 1064E raid controller.

But you don't need the cli to monitor the raid and/or perform smart operations on the disks.