Best Practice guide for x4540 running Solaris OS

Asset ID:	1-79-1563657.1
Update Date:	2018-05-23
Keywords:

Solution Type Predictive Self-Healing Sure

Solution 1563657.1 : Best Practice guide for x4540 running Solaris OS

Applies to:

Sun Fire X4540 Server - Version Not Applicable and later
Information in this document applies to any platform.

Purpose

To discuss this information further with Oracle experts and industry peers, we encourage you to review, join or start a discussion in the My Oracle Support Community - Sun x86 Systems

Best practice guide for x4540 running Solaris OS

Details

Best practice guide for x4540 running Solaris O/S

1. Make sure the system BIOS and Service Processor (SP) firmware is up-to-date.
2. Make sure the LSI firmware is up-to-date.
       Related: In certain conditions, the LSI 1068E firmware may hang [ID 1545687.1]
3. Make sure the firmware for each HDD is up-to-date.
       Related: Certain Disk Drives May Experience Data Miscompare or Drive Initialization Issues [ID 1020312.1]
       Related: Certain 500.0GB Disk Drive in Sun Fire X4500 and X4540 Systems May Experience Data Miscompare Issues [ID 1020313.1]
   Related: non-public FABs 1022189.1, 1499574.1, and 1512680.1
       Related: Sun x4500 and x4540 Disk Models and Firmware Revisions (Doc ID 1561202.1)
       Related: Seagate Firmware Update Requirement (Doc ID 1640003.1)
4. Make sure the SUNWhd package is installed.
5. Keep the OS patched and up-to-date with recommended patches.
6. Monitor for HDD faults.

Note: This document makes reference to the Oracle Hardware Installation Assistant (OHIA). The OHIA is also known as HIA (Hardware Installation Assistant) and it is also known as the legacy SIA (Sun Installation Assistant).

1. Make sure the system BIOS and Service Processor (SP) firmware is up-to-date.

Note:
The most recent BIOS for x4540 is 0ABNF040
The most recent SP version for x4540 is 3.0.10.13c r83015

To check the current BIOS version from Solaris OS, use the smbios command:

# smbios | grep ABNF
Version String: 0ABNF040

You can also view the system BIOS version after a power cycle, during POST, as shown here.

CLI screenshot:

BIOScliPOST

GUI screenshot:

BIOSguiPOST

To check the current SP version, connect to the SP via CLI and use the version command:

-> version
SP firmware 3.0.10.13.c
SP firmware build number: 83015
SP firmware date: Thu Aug 8 22:07:14 CST 2013
SP filesystem version: 0.1.23

Another method to check both the SP firmware version and the BIOS version is from the SP using the GUI.
Connect to the SP via a supported browser and look on the 'System Information / Overview' tab as shown here:

When upgrading the system BIOS and SP firmware, the process will update them both together, as a set. Here are two possible methods to update them:

Method #1

Use the the .pkg file located in Patch 17617719: X4540 SW 2.3.3 - ILOM_AND_BIOS:

   ILOM-3_0_10_13_c_r83015-Sun_Fire_X4540.pkg

Instructions for this method exist for both GUI (Web Interface) and CLI scenarios:

Chapter 11 "Updating ILOM Firmware" of the Oracle Integrated Lights Out Manager (ILOM) 3.0 Web Interface Procedures Guide (820-6411-12) here:
    http://docs.oracle.com/cd/E19469-01/820-6411-12/firmware_web.html#50634852_pgfId-1000552
or

Chapter 11 "Updating ILOM Firmware" of the Oracle Integrated Lights Out Manager (ILOM) 3.0 CLI Procedures Guide (820-6412-12) here:
    http://docs.oracle.com/cd/E19469-01/820-6412-12/firmware_cli.html#50561098_pgfId-1014696

Method #2:

Use the SW 2.3.2 OHIA (Oracle Hardware Install Assistant)

Download Patch 16044285: X4540 SW 2.3.2 - HIA 2.4.10.5

Instructions for this method are documented in the section titled "How to Upgrade the System BIOS and ILOM Firmware" in the Sun Installation Assistant 2.3 through 2.4 User's Guide for x64 Servers here:
http://docs.oracle.com/cd/E19593-01/821-0694/p26.html#scrolltoc

Note: There was no OHIA released for SW 2.3.3. You must use Method #1 to get the latest BIOS and SP firmware.

2. Make sure the LSI firmware is up-to-date.

Note:
The most recent LSI firmware for x4540 is LSI BIOS v6.26.00 FW 1.27.92.

To check version LSI version on all six (0-5) controllers, here are four possible methods:

Method #1

Use raidctl

# raidctl -l 0 1 2 3 4 5
Controller Type Version
----------------------------------------------------------------
c0 LSI_1068E 1.27.92.00
Controller Type Version
----------------------------------------------------------------
c1 LSI_1068E 1.27.92.00
Controller Type Version
----------------------------------------------------------------
c2 LSI_1068E 1.27.92.00
Controller Type Version
----------------------------------------------------------------
c3 LSI_1068E 1.27.92.00
Controller Type Version
----------------------------------------------------------------
c4 LSI_1068E 1.27.92.00
Controller Type Version
----------------------------------------------------------------
c5 LSI_1068E 1.27.92.00

Method #2

Look in /var/adm/messages file:

# grep Firmware /var/adm/messages* | grep mpt[0-5]
Jun 06 06:44:01 gl-x64-thor     mpt0 Firmware version v1.1b.5c.0 (IT)
Jun 06 06:44:20 gl-x64-thor     mpt1 Firmware version v1.1b.5c.0 (IT)
Jun 06 06:44:28 gl-x64-thor     mpt3 Firmware version v1.1b.5c.0 (IT)
Jun 06 06:44:28 gl-x64-thor     mpt2 Firmware version v1.1b.5c.0 (IT)
Jun 06 06:44:28 gl-x64-thor     mpt4 Firmware version v1.1b.5c.0 (IT)
Jun 06 06:44:37 gl-x64-thor     mpt5 Firmware version v1.1b.5c.0 (IT)

Note:
Convert hex to decimal to confirm the versions match: v1.1b.5c.0 = v1.27.92.0

Method #3a

Look in the LSI config utility during POST (GUI)

Method #3b

Look in the LSI config utility during POST (CLI)

Method #4

Use the fwupdate command.

Note: The fwupdate command in one of the commands included with the Oracle Hardware Management Pack (OHMP)
It is beyond the scope of this document to discuss installation or use of OHMP, but if you have it already installed you can use it to determine firmware revision on all six controllers.

# fwupdate list controller

==================================================
CONTROLLER
==================================================
ID        Manufacturer   Model     Product Name              FW Version          BIOS Version        XML Support
----------------------------------------------------------------------------------------------------------------
c0        LSI Logic      0x0058    SAS1068E                  01.27.92.00-IT      6.26.00.00          N/A
c1        LSI Logic      0x0058    SAS1068E                  01.27.92.00-IT      6.26.00.00          N/A
c2        LSI Logic      0x0058    SAS1068E                  01.27.92.00-IT      6.26.00.00          N/A
c3        LSI Logic      0x0058    SAS1068E                  01.27.92.00-IT      6.26.00.00          N/A
c4        LSI Logic      0x0058    SAS1068E                  01.27.92.00-IT      6.26.00.00          N/A
c5        LSI Logic      0x0058    SAS1068E                  01.27.92.00-IT      6.26.00.00          N/A

To update the LSI firmware, the only supported method is to use the Oracle Hardware Install Assistant (OHIA) iso image:

Oracle_HIA-2_4_10_5-Sun_Fire_X4540.iso

That iso file is located in Patch 16044285: X4540 SW 2.3.2 - HIA 2.4.10.5.

Boot from the ISO contained in the patch and follow the onscreen instructions. You may want to refer to this doc, it has details and screenshots of the process. See "How to Upgrade HBA Firmware" in the OHIA documentation, here:

Sun Installation Assistant 2.3 through 2.4 User's Guide for x64 Servers
http://docs.oracle.com/cd/E19593-01/821-0694/p28.html#scrolltoc

#3 Make sure the firmware for each HDD is up-to-date

Note: Replacement disks from logistics may not be on the most recent version. You should check the disk firmware whenever a disk is replaced. On x4540 (and x4500 as well) it is highly recommended that all HDDs are on the most recently available firmware.

To check the firmware on the disks, refer to this MOS doc: "Sun x4500 and x4540 Disk Models and Firmware Revisions" 1561202.1

#4 Make sure the SUNWhd package is installed.

Note: For more information about SUNWhd, see MOS document 1565521.1 "Sun Fire X4500/X4540 Hard Disk Suite (SUNWhd) and hdtool"

Generally speaking, there are two main reasons to have the SUNWhd package installed.

Reason #1

The Solaris storage device have different logical names. They can be identified by cXtY, sd# or by their PCI path. The SUNWhd package includes the hd command. Using hd and hd -i you can determine the slot location of any disk.

For example, using the following hd and hd -i outputs, we can tell the following:
   disk c4t5 is in slot #37,
   disk sd22 maps to c0t6 which is in slot #6, and
   disk /pci@3c,0/pci10de,377@a/pci1000,1000@0/sd@7,0 maps to sd15 (which also maps to c3t7) all of which are in slot #31

# hd

platform = Sun Fire X4540

   [snip]

---------------------SunFireX4540-------Rear----------------------------
3:    7:   11:   15:   19:   23:   27:   31:   35:   39:   43:   47:
c0t3 c0t7 c1t3 c1t7 c2t3 c2t7 c3t3 c3t7 c4t3 c4t7 c5t3 c5t7
^++   ^++   ^++   ^++   ^++   ^++   ^++   ^++   ^++   ^++   ^++   ^++
2:    6:   10:   14:   18:   22:   26:   30:   34:   38:   42:   46:
c0t2 c0t6 c1t2 c1t6 c2t2 c2t6 c3t2 c3t6 c4t2 c4t6 c5t2 c5t6
^++   ^++   ^++   ^++   ^++   ^++   ^++   ^++   ^++   ^++   ^++   ^++
1:    5:    9:   13:   17:   21:   25:   29:   33:   37:   41:   45:
c0t1 c0t5 c1t1 c1t5 c2t1 c2t5 c3t1 c3t5 c4t1 c4t5 c5t1 c5t5
^b+   ^++   ^b+   ^++   ^++   ^++   ^++   ^++   ^++   ^++   ^++   ^++
0:    4:    8:   12:   16:   20:   24:   28:   32:   36:   40:   44:
c0t0 c0t4 c1t0 c1t4 c2t0 c2t4 c3t0 c3t4 c4t0 c4t4 c5t0 c5t4
^b+   ^++   ^b+   ^++   ^++   ^++   ^++   ^++   ^++   ^++   ^++   ^++
-------*-----------*-SunFireX4540---*---Front----*---------*--------

# hd -i
c0t0d0 sd16 /pci@0,0/pci10de,377@a/pci1000,1000@0/sd@0,0
c0t3d0 sd19 /pci@0,0/pci10de,377@a/pci1000,1000@0/sd@3,0
c0t7d0 sd23 /pci@0,0/pci10de,377@a/pci1000,1000@0/sd@7,0
c1t7d0 sd7 /pci@0,0/pci10de,375@b/pci1000,1000@0/sd@7,0
c1t3d0 sd3 /pci@0,0/pci10de,375@b/pci1000,1000@0/sd@3,0
c3t6d0 sd14 /pci@3c,0/pci10de,377@a/pci1000,1000@0/sd@6,0
c2t3d0 sd43 /pci@0,0/pci10de,376@f/pci1000,1000@0/sd@3,0
c4t3d0 sd27 /pci@3c,0/pci10de,375@b/pci1000,1000@0/sd@3,0
c2t7d0 sd47 /pci@0,0/pci10de,376@f/pci1000,1000@0/sd@7,0
c4t7d0 sd31 /pci@3c,0/pci10de,375@b/pci1000,1000@0/sd@7,0
c1t2d0 sd2 /pci@0,0/pci10de,375@b/pci1000,1000@0/sd@2,0
c3t7d0 sd15 /pci@3c,0/pci10de,377@a/pci1000,1000@0/sd@7,0
c3t3d0 sd11 /pci@3c,0/pci10de,377@a/pci1000,1000@0/sd@3,0
c0t6d0 sd22 /pci@0,0/pci10de,377@a/pci1000,1000@0/sd@6,0
c5t7d0 sd39 /pci@3c,0/pci10de,376@f/pci1000,1000@0/sd@7,0
c1t6d0 sd6 /pci@0,0/pci10de,375@b/pci1000,1000@0/sd@6,0
c5t2d0 sd34 /pci@3c,0/pci10de,376@f/pci1000,1000@0/sd@2,0
c2t2d0 sd42 /pci@0,0/pci10de,376@f/pci1000,1000@0/sd@2,0
c0t1d0 sd17 /pci@0,0/pci10de,377@a/pci1000,1000@0/sd@1,0
c0t2d0 sd18 /pci@0,0/pci10de,377@a/pci1000,1000@0/sd@2,0
c0t4d0 sd20 /pci@0,0/pci10de,377@a/pci1000,1000@0/sd@4,0
c0t5d0 sd21 /pci@0,0/pci10de,377@a/pci1000,1000@0/sd@5,0
c1t0d0 sd0 /pci@0,0/pci10de,375@b/pci1000,1000@0/sd@0,0
c1t1d0 sd1 /pci@0,0/pci10de,375@b/pci1000,1000@0/sd@1,0
c1t4d0 sd4 /pci@0,0/pci10de,375@b/pci1000,1000@0/sd@4,0
c1t5d0 sd5 /pci@0,0/pci10de,375@b/pci1000,1000@0/sd@5,0
c2t0d0 sd40 /pci@0,0/pci10de,376@f/pci1000,1000@0/sd@0,0
c2t1d0 sd41 /pci@0,0/pci10de,376@f/pci1000,1000@0/sd@1,0
c2t4d0 sd44 /pci@0,0/pci10de,376@f/pci1000,1000@0/sd@4,0
c2t5d0 sd45 /pci@0,0/pci10de,376@f/pci1000,1000@0/sd@5,0
c2t6d0 sd46 /pci@0,0/pci10de,376@f/pci1000,1000@0/sd@6,0
c5t0d0 sd32 /pci@3c,0/pci10de,376@f/pci1000,1000@0/sd@0,0
c5t1d0 sd33 /pci@3c,0/pci10de,376@f/pci1000,1000@0/sd@1,0
c5t3d0 sd35 /pci@3c,0/pci10de,376@f/pci1000,1000@0/sd@3,0
c5t4d0 sd36 /pci@3c,0/pci10de,376@f/pci1000,1000@0/sd@4,0
c5t5d0 sd37 /pci@3c,0/pci10de,376@f/pci1000,1000@0/sd@5,0
c5t6d0 sd38 /pci@3c,0/pci10de,376@f/pci1000,1000@0/sd@6,0
c3t0d0 sd8 /pci@3c,0/pci10de,377@a/pci1000,1000@0/sd@0,0
c3t1d0 sd9 /pci@3c,0/pci10de,377@a/pci1000,1000@0/sd@1,0
c3t2d0 sd10 /pci@3c,0/pci10de,377@a/pci1000,1000@0/sd@2,0
c3t4d0 sd12 /pci@3c,0/pci10de,377@a/pci1000,1000@0/sd@4,0
c3t5d0 sd13 /pci@3c,0/pci10de,377@a/pci1000,1000@0/sd@5,0
c4t0d0 sd24 /pci@3c,0/pci10de,375@b/pci1000,1000@0/sd@0,0
c4t1d0 sd25 /pci@3c,0/pci10de,375@b/pci1000,1000@0/sd@1,0
c4t2d0 sd26 /pci@3c,0/pci10de,375@b/pci1000,1000@0/sd@2,0
c4t4d0 sd28 /pci@3c,0/pci10de,375@b/pci1000,1000@0/sd@4,0
c4t5d0 sd29 /pci@3c,0/pci10de,375@b/pci1000,1000@0/sd@5,0
c4t6d0 sd30 /pci@3c,0/pci10de,375@b/pci1000,1000@0/sd@6,0

Note: The disk mapping for each system can be different depending on how the OS was installed. Do not assume that the disk layout from one system is identical to another.

Reason #2

When the SUNWhd package is installed, helpful information is included in explorer data files for Oracle support teams. This data is collected in ..\disks\hd directory of the explorer output.

5. Keep the OS patched and up-to-date with recommended patches.

Check MOS 'Recommended Patch Advisor' to determine the Recommended Patchset for the version of Solaris you have installed.

6. Monitor for HDD faults.

    FIGURE C-1 Close Up of Controls and Indicators.
    TABLE C-2 #5 Top fault LED - Drive or fan fault.
    FIGURE C-6 Disk Drive and Fan Tray LEDs. Fault (Amber) Service action is required.
    X4500 - Diagnose Disk Faults
    X4540 - Diagnose Disk Faults

Note: Unrecovered read error "Error level: Fatal" is not an indication the disk itself has suffered a fatal error. Reference Doc ID 1477074.1

End of Document.

References

<NOTE:1565521.1> - Sun Fire X4500/X4540 Hard Disk Suite (SUNWhd) and hdtool
<NOTE:1561202.1> - Sun x4500 and x4540 Disk Models and Firmware Revisions

Attachments

This solution has no attachment