IBM ServeRAID-8, ServeRAID-8i, ServeRAID-8s, ServeRAID-8k, ServeRAID-8k-I Best Practices And Maintenance Information
ServeRAID-8 Series
Best Practices and Maintenance
Information
ServeRAID-8 Series
Best Practices and Maintenance
Information
Note: Before using this information and the product it supports, read the general information in “Notices,” on page 45.
Second Edition (September 2011)
© Copyright IBM Corporation 2008, 2011.
US Government Users Restricted Rights – Use, duplication or disclosure restricted by GSA ADP Schedule Contract
with IBM Corp.
Contents
Chapter 1. ServeRAID-8 series best practices and maintenance information 1
Notices and statements in this document ................1
Obtaining ServeRAID updates ....................1
Initializing previously used good hard disk drives .............2
Synchronizing logical drives after upgrading firmware from build 8263 or earlier 2
Upgrading to ServeRAID Manager 9.0 within Microsoft Windows .......3
Synchronization options when you create a new array and the impact on
performance .........................3
Physical drive write-cache policies changed with firmware build 15407 and later 4
Defunct hard disk drives might return to normal operation after a system restart 5
The Automatic Failover feature and how it affects replaced drives.......6
Performance is degraded with multiple ongoing background tasks ......6
Replacing and rebuilding a drive in a PFA state .............7
PHY settings and AMSU updates ..................9
Periodic ServeRAID maintenance updates and evaluation .........9
Zero-mode flash update overview ..................11
Zero-mode flash for ServeRAID-8i (optional controller) ..........11
Zero-mode flash for ServeRAID-8s (optional controller) ..........13
Zero-mode flash for ServeRAID-8k and ServeRAID-8k-l (BladeCenter Storage
Expansion Unit 3) .......................14
Zero-mode flash for ServeRAID-8k and ServeRAID-8k-l (System x3650)....16
Zero-mode flash for ServeRAID-8k and ServeRAID-8k-l (System x3550)....18
Zero-mode flash for ServeRAID-8k and ServeRAID-8k-l (System x3500)....20
The ServeRAID-8k and ServeRAID-8k-l RAID enabler modules have a very low
failure rate ..........................23
Correct mounting of the ServeRAID-8k RAID enabler battery ........24
Physical drives are online, but the logical drive is offline ..........24
Valid RAID 5EE state transitions after a disk fails ............25
System hang issues with ServeRAID-8i on enterprise servers........26
After upgrading firmware, the server crashes every 8-10 days .......26
Evaluating hard disk drive problems .................27
Logical drive states and descriptions .................30
Using the force online feature ...................33
Recover Data option ......................33
Force-online operation .....................36
Displaying an Offline array ...................39
Invalid process ........................39
Displaying an Offline array in POST ................40
Revived array operations ....................41
Offline array operations .....................42
Chapter 2. Getting help and technical assistance ...........43
Before you call .........................43
Using the documentation .....................43
Getting help and information from the World Wide Web ..........43
Software service and support ...................44
Hardware service and support ...................44
IBM Taiwan product service ....................44
Appendix. Notices .......................45
Trademarks ..........................45
Important notes .........................46
Documentation format ......................47
© Copyright IBM Corp. 2008, 2011 iii
Index ............................49
iv ServeRAID-8 Series: Best Practices and Maintenance Information
Chapter 1. ServeRAID-8 series best practices and maintenance information
This document provides best practices and maintenance information for the IBM
ServeRAID-8 series controllers.
Notices and statements in this document
The following notices and statements are used in this document:
v Note: These notices provide important tips, guidance, or advice.
v Important: These notices provide information or advice that might help you avoid
inconvenient or problem situations.
v Attention: These notices indicate potential damage to programs, devices, or
data. An attention notice is placed just before the instruction or situation in which
damage might occur.
Obtaining ServeRAID updates
For updated versions of the ServeRAID software, go to http://www-304.ibm.com/
systems/support/supportsite.wss/docdisplay?lndocid=SERV-RAID
&brandind=5000008 or complete the following steps.
Note: Changes are made periodically to the IBM Web site. Procedures for locating
firmware and documentation might vary slightly from what is described in this
document.
1. Go to http://www.ibm.com/systems/support/.
2. Under Product support, click System x.
3. Under Popular links, click Software and device drivers.
4. Under Related downloads, click ServeRAID to display the matrix of
downloadable ServeRAID files.
®
Note: If you download ServeRAID software, you must download and install all
ServeRAID software at the same time to make sure that all levels of the
software are compatible. The ServeRAID software includes the following
types of software:
v BIOS and firmware code
v Device drivers
v ServeRAID Manager program
v Command-line programs
If you do not have access to the World Wide Web, contact your place of purchase,
your IBM reseller, or your IBM marketing representative for updated CDs.
© Copyright IBM Corp. 2008, 2011 1
Initializing previously used good hard disk drives
A previously used good hard disk drive is defined in this document as a drive that
was previously a member of an array that was attached to a ServeRAID-8 series
controller. The drive is being reused within a new RAID configuration. Before you
can add a previously used good hard disk drive to a new array, either as a
replacement drive or to expand an array, you must first initialize the drive to remove
existing configuration information. Existing configuration information can cause the
ServeRAID controller to behave differently than expected and, in some rare cases,
can result in data loss.
You can initialize a hard disk drive by using the Array Configuration Utility (ACU)
(accessible by pressing Ctrl+A when you are prompted at system startup), the IBM
ServeRAID Support CD version 8 or 9, or the ServeRAID Manager program (from
within the operating system), or by using the arcconf task command.
For more information, see the following RETAIN tips.
A rebuild does not start after replacing a defunct drive - IBM ServeRAID
http://www-304.ibm.com/systems/support/supportsite.wss/
docdisplay?brandind=5000008&lndocid=MIGR-5074510
(Type 5074510 in the Search field at http://www.ibm.com.)
Lost configuration when drive added - ServeRAID 8k, 8k-l
http://www-304.ibm.com/systems/support/supportsite.wss/
docdisplay?brandind=5000008&lndocid=MIGR-5073723
(Type 5073723 in the Search field at http://www.ibm.com.)
Synchronizing logical drives after upgrading firmware from build 8263 or earlier
ServeRAID firmware builds 7777 through 8263 use a legacy method to track
bad-stripe unit errors. Starting with firmware build 8264 and later, the process to
manage bad-stripe units is improved with a bad-stripe table. The firmware upgrade
process does not import existing bad-stripe information from the old method to the
new, and the legacy method is perceived by the new code as disk errors when the
affected stripe units are read. In most cases, these types of errors are corrected
automatically without any user intervention. However, in rare cases, if a large
number of these errors are reported by a single drive, the drive might be marked
Defunct prematurely.
To avoid this situation, start a synchronization on each logical drive. The
synchronization process scrubs the physical drives, corrects the errors as they are
detected, and when applicable, creates an equally equivalent bad-stripe table entry.
Note that the ServeRAID-8 series controllers are designed to correct or self-heal
bad-stripe table entries when the next successful write occurs to that stripe unit.
You can start synchronizations by using the IBM ServeRAID Support CD, the
ServeRAID Manager installable application (from within the operating system), or
the arcconf task command.
2 ServeRAID-8 Series: Best Practices and Maintenance Information
Note: The ServeRAID-8 series controllers can run multiple synchronizations, at the
expense of an incremental performance decrease. You can modify the
priority of synchronizations from high to low by using the ServeRAID
Manager application or by using the arcconf setpriority command to help
balance incoming user workload to synchronization workload.
Upgrading to ServeRAID Manager 9.0 within Microsoft Windows
ServeRAID Manager 9.0 changed a fundamental method that is used to manage
attached storage. In earlier versions 8.00 to 8.40 two components are always
needed to communicate with the storage subsystem. These two components are
the operating-system ARCSAS device driver and the AACMGT.SYS filter driver.
With the ServeRAID Manager 9.0 application, only the latest version of the
operating-system device driver is needed, because the AACMGT filter driver
functions are added to the new device drivers.
On a new installation of ServeRAID Manager 9.0, the filter driver is not installed,
and a restart is not required; however, the management function works only if the
latest ARCSAS driver, which is version 5.2.0.12913, is installed. If an earlier version
of the driver is installed, the ServeRAID Manager application reports a driver
update is required message.
If you are upgrading an existing installation of ServeRAID Manager 8.x to version
9.0, the upgrade installation preserves the existing AACMGT filter driver. The new
software uses the existing filter driver to obtain the storage-management data as
needed. This is the only way that ServeRAID Manager 9.0 can manage a system
with an earlier version of the device driver. If the device driver is ever upgraded to
version 5.2.0.12913, ServeRAID Manager 9.0 uses the native storage-management
functions of the new driver.
If it is necessary to upgrade an existing installation of an earlier version of
ServeRAID Manager 8.x, you must first upgrade to ServeRAID Manager 8.40 and
then upgrade to 9.0 to incorporate the broadest support for the ServeRAID 8-series
controllers and the fixes in the later code.
For more information, see the following RETAIN tip.
ServeRAID Manager reports “SAS driver update required”
http://www-304.ibm.com/systems/support/supportsite.wss/
docdisplay?brandind=5000008&lndocid=MIGR-5073797
(Type 5073797 in the Search field at http://www.ibm.com.)
Synchronization options when you create a new array and the impact on performance
When you create a new redundant array, three methods are available to
synchronize the data on the new array: QuickInit, AutoSync, and Clear. The initial
synchronization on a new array creates the redundant stripe units that are used to
recover the array if a disk has failed.
The QuickInit option has the least impact on system performance because the
synchronization is done as new write operations are made to the array; however,
the array becomes available to you immediately. The array shows a status of
Impacted until the synchronization is completed. During idle time, the array slowly
synchronizes in the background. QuickInit might take days to several weeks to be
Chapter 1. ServeRAID-8 series best practices and maintenance information 3
completed and change the array status to Optimal. You can make the new array
Optimal faster by synchronizing the logical drive.
The AutoSync option has minimal impact on performance because an active
synchronization is started after the array is created. The new array immediately
becomes available; however, the synchronization defaults to a high priority. When
the synchronization task is set to High, the task can take as much as 33% of the
controller resources. After the array is created, you can modify the priority from High
to Medium to Low by using the IBM ServeRAID Support CD, the ServeRAID
Manager application, or the arcconf setpriority command. Modifying the priority to
low reduces the resources that are used by the synchronization to about 16% but
increases the time that it takes to be completed. The array becomes Optimal after
the synchronization is completed.
Use the Clear option to perform an immediate and exclusive synchronization on the
new array. The synchronization might take several hours to finish; the array is
unavailable until the synchronization is completed. When the synchronization is
completed, the array is Optimal, and the new logical drive becomes available.
For more information, see the following RETAIN tip.
Rebuild and sync priorities on ServeRAID-8x controllers
http://www-304.ibm.com/systems/support/supportsite.wss/
docdisplay?brandind=5000008&lndocid=MIGR-5070858
(Type 5070858 in the Search field at http://www.ibm.com.)
Physical drive write-cache policies changed with firmware build 15407 and later
With ServeRAID firmware build 15407 and later, managing hard disk drive
write-cache policies on the ServeRAID-8 series controllers is improved to make this
a global controller setting so that all hard disk drives that are attached to the
subsystem conform to a common write-cache policy. You can set the write-cache
policy differently for SAS and SATA drives.
You can set the hard disk drives write-cache setting to either write-through (disabled
or Off) or write-back (enabled or On). In earlier firmware versions, the hard disk
drive write-cache policy is changed on an individual device basis. Different devices
can have different cache states.
The cache policy changes offer four new options: SAS = On or Off and SATA = On
or Off. The controller default is SAS = Off and SATA = On. When you upgrade the
firmware to build 15407 and later from a version earlier than 15407, the controller
applies the default settings to the storage devices.
Avoid using more than one cache mode. The new firmware options help to make
sure that all drives conform to a common write-cache configuration. The write-cache
mode of a hard disk drive supports one of two general policies on a RAID controller.
When you disable (Off) the write-cache, the RAID controller firmware can protect
the uncommitted write operations at the logical drive level, which prevents the loss
of data if there is an unexpected power outage. When the write-cache setting is
enabled (On), there is a moderate performance gain at the risk of losing data if
there is an unexpected power loss. With earlier versions of firmware, it is possible
to configure multiple write-cache modes on array member devices, which
undermines both policies.
4 ServeRAID-8 Series: Best Practices and Maintenance Information
Attention: Data loss can occur when the write-cache policy is enabled on the
physical drives, because the ServeRAID controller cannot manage write operations
that are owned by the individual hard disk drive and are not committed to the drive
media. The battery-backed memory cache on the ServeRAID controller can protect
the write operation only at a logical drive level. Therefore, set the hard disk drive
write-cache policy to Off or Disabled. Turning the write-cache policy Off on all drives
ensures the lowest risk to data loss if there is an unexpected power outage.
You can modify the write-cache policy settings from the Controller Configuration
section, which you can access by pressing Ctrl+A, and which is available from the
IBM ServeRAID Support CD, the ServeRAID Manager application, and the arcconf
setcache command.
Note: The IBM default hard disk drive write-cache state for SAS drives is Disabled
or Off. For SATA drives, the IBM default hard disk drive write-cache policy is
Enabled or On.
For more information, see the following RETAIN tip.
ServeRAID changes hard drive write-cache mode unexpectedly
http://www-304.ibm.com/systems/support/supportsite.wss/
docdisplay?brandind=5000008&lndocid=MIGR-5070673
(Type 5070673 in the Search field at http://www.ibm.com.)
Defunct hard disk drives might return to normal operation after a system restart
A hard disk drive that is marked Defunct during normal operation on a ServeRAID-8
series controller is treated as a replaced drive when the host server is restarted.
Defunct is a physical-drive state in which the ServeRAID controller cannot
communicate correctly with the drive.
Visual indicators, such as the amber LEDs, are also reset and turn off.
During a server restart process, the ServeRAID-8 series controller executes normal
startup sequences for each drive that is attached to the system, assuming that each
drive is good. If a drive that is marked Defunct spins up, the controller takes
user-defined actions as if it were replaced in the system. If a drive that is marked
Defunct does not spin up, the disk might remain in a Defunct state, or appear
missing from the controller scan or in the ServeRAID Manager program. In this
situation, the amber LED that is lit still turns off.
If a hard disk drive returns to normal operation, the drive might not be bad. The
ServeRAID controller marks a drive Defunct if any component in the subsystem
prevents normal communication to the drive. This includes issues with the hard disk
drive firmware, cables, backplanes, drive connectors, and tray interposers. The
ServeRAID controller cannot determine which component has contributed to the
offline drive.
A strong indicator that the drive might be good is that the hard disk drive does
return to normal operation after a system restart. If this occurs, perform a
subsystem health check to make sure that all known issues are avoided by applying
known fixes. For more information, see “Periodic ServeRAID maintenance updates
and evaluation” on page 9.
Chapter 1. ServeRAID-8 series best practices and maintenance information 5
For more information, see the following RETAIN tip.
Hard Disk Drives rebuild after going defunct and a system restart
http://www-304.ibm.com/systems/support/supportsite.wss/
docdisplay?brandind=5000008&lndocid=MIGR-5072365
(Type 5072365 in the Search field at http://www.ibm.com.)
The Automatic Failover feature and how it affects replaced drives
The Automatic Failover feature of the Adaptec RAID Configuration Utility controls
the rebuilding of a logical drive when a defunct drive is replaced. Defunct is a
physical-drive state in which the ServeRAID controller cannot communicate
correctly with the drive.
When Automatic Failover is enabled (the default setting), the ServeRAID controller
automatically starts a rebuild operation when a hard disk drive is installed to replace
a defunct drive and no other hot-spare options are defined. Hot spares always have
priority over the Automatic Failover feature.
When Automatic Failover is disabled and a new (never used) hard disk drive is
installed to replace a defunct drive, the new drive status becomes Ready. Then, you
must manually configure the drive for operation. The drive can be used in a new
array or defined as a hot spare. If an array is critical and the drive is defined as a
hot spare, a rebuild operation starts for the drive, if its size qualifies the drive as a
replacement disk.
Attention: When Automatic Failover is disabled, do not replace a defunct hard
disk drive with a hard disk drive that has existing ServeRAID configuration
information (metadata). The ServeRAID controller tries to use the existing
information on the drive to recover the array. The newly installed drive might appear
as part of an incomplete array, or, if the drive is a simple volume or one part of a
mirror, the array might come online. If the newly installed drive is part of a mirror, it
might come online in a Critical state. In some cases, this causes unpredictable
behavior and data corruption.
Important: If Automatic Failover is disabled, before you install a hard disk drive
that has existing ServeRAID configuration information, you must first
initialize the drive to remove the existing configuration information. For
more information, see “Initializing previously used good hard disk
drives” on page 2.
To change the Automatic Failover feature setting, press Ctrl+A when you are
prompted at system startup. From the Adaptec RAID Configuration Utility menu,
select SerialSelect Utility; then, select Controller Configuration. You can also
use the arcconf failover command to change the Automatic Failover feature
setting, if you are using firmware version build 15407 or later.
Performance is degraded with multiple ongoing background tasks
The ServeRAID-8 series controllers are designed to run multiple background tasks
at the same time. Background tasks are rebuilds, synchronizations, and copy back
operations. There can be a significant performance impact, to the point of making
the controller unusable, if too many operations are started and you do not balance
the total number of tasks and the priority of those tasks.
6 ServeRAID-8 Series: Best Practices and Maintenance Information
Synchronizing spanned arrays is one of the main sources of an unexpected high
impact to performance. Spanned arrays are made up of multiple subarrays or legs.
When you start a synchronization on a logical drive, it actually starts one
synchronization per leg. If a RAID 10 is made up of eight subarrays, synchronizing
the logical drive results in a synchronization of each leg, or eight synchronizations.
An active synchronization defaults to a high priority, which takes as much as 33% of
the controller resources. Adjusting the priority to low reduces the resources that are
used by the synchronization to approximately 16% but increases the time it takes to
be completed.
If two synchronizations are active with a high priority, over 50% of the controller
resources are used by these operations. The controller firmware moderately
throttles synchronization operations; however, the high priority setting remains
aggressive. Additional background tasks will continue to reduce resources that are
available to user workload without manual intervention.
Considering the number of drives and array spans within a ServeRAID
configuration, you can manage tasks by controlling the total number of background
tasks that are running concurrently and by lowering the priority of the operations.
On systems that are servicing clients in a business capacity, do not run more than
two background tasks concurrently, and lower the task priority to free resources for
business workloads. You can stop synchronization tasks, if necessary, and restart
them after other operations are completed.
The tools that are available from a graphical interface to modify tasks are on the
IBM ServeRAID Support CD (you must start to the CD) and the
operating-system-installable version of the ServeRAID Manager application. From
the command line, you can use the arcconf task command to start and stop
operations, and you can use the arcconf setpriority command to change the
priority.
Note: Firmware builds 11849 and earlier do not preserve modifications that you
make to task priorities between system resets. After a system reset is
completed, the tasks switch to high priority. Firmware builds 15407 and later
preserve the modifications that you make to task priorities before a system
reset.
For more information, see the following RETAIN tip.
Rebuild and sync priorities on ServeRAID-8 series SAS controllers
http://www-304.ibm.com/systems/support/supportsite.wss/
docdisplay?brandind=5000008&lndocid=MIGR-5070858
(Type 5070858 in the Search field at http://www.ibm.com.)
Replacing and rebuilding a drive in a PFA state
IBM SAS and SATA hard disk drives support Predictive Failure Analysis (PFA) or
SMART alerts. This specification is designed into the hard disk drives to internally
monitor and diagnose a likely failure within the near future. If a disk issues a PFA
event, the ServeRAID controller captures and forwards the event locally to the
ServeRAID Management console and agent. ServeRAID Manager Agent software
supports industry-standard Simple Network Management Protocol (SNMP) for alert
automation. If a PFA alert occurs on an IBM hard disk drive, the hard disk drive is
eligible for replacement under the warranty terms and conditions.
Chapter 1. ServeRAID-8 series best practices and maintenance information 7
Depending on whether the RAID configuration is redundant, recovering an array
when a member hard disk drive has an active PFA alert requires different actions.
Make sure that there is a recent backup of the data before you proceed.
If the affected array is a redundant logical drive (RAID 1, 1E, 5, 6, 5EE, 10, 1E0,
50, or 60), the controller protects and keeps the data available to you while the
logical drive is critical. The following procedure provides the steps to replace the
PFA drive and to regenerate the Critical logical drive back to an online or Okay
state. If a qualified hot-spare drive is configured, ServeRAID automatically uses the
hot-spare drive to regenerate logical drives to an Optimal state.
If the affected array is a nonredundant logical drive (a simple volume, RAID 0, or
RAID 00), all data on the logical drive is lost when the disk eventually fails. To
recover the logical drive, you must delete the nonredundant array and recreate it;
then, you can restore the data from a recent backup.
To replace a PFA drive, complete the following steps:
1. Make sure that you back up the data.
2. Identify the slot location of the PFA drive. (Use the ServeRAID Manager
application, the IBM ServeRAID Support CD, or the arcconf identify
command.)
3. Mark the drive Defunct and wait for the amber LED to light. (Use the ServeRAID
Manager application, the IBM ServeRAID Support CD, or the arcconf setstate
command.)
4. Without removing the drive completely, gently remove the physical drive from
the connector, using the handle of the tray.
5. Wait 45 seconds to allow the hard disk drive motor to completely stop spinning.
6. Remove the defunct drive from the slot and insert the replacement hard disk
drive.
7. For redundant arrays, go to step 8. For nonredundant arrays, recreate the array
and restore the data from a recent backup.
8. Observe the drive for normal device startup behavior and LED activity. Within
approximately one minute, the ServeRAID controller and the ServeRAID
Manager application detect the hot-swap event. Depending on the configuration,
the following actions occurs:
v The replacement hard disk drive spins up.
v The green drive activity LED begins flashing and might eventually turn off or
flash rapidly, depending on the configuration.
v If no hot-spare drive is defined, ServeRAID Manager automatically initiates a
rebuild operation to the replacement disk (the green LED flashes rapidly and
the amber LED flashes).
v If a hot-spare drive is defined and copy back is enabled, ServeRAID Manager
automatically performs a rebuild operation back to the replaced drive, after
the hot-spare drive rebuild operation is completed.
v The steady amber drive LED turns off after a transition to a Hot-Spare,
Ready, or Online state is completed. Otherwise, the amber LED is flashing
during a rebuild or copy back operation
9. Observe the drive for normal device startup behavior and LED activity:
v The replacement hard disk drive spins up.
v The activity LED (green drive LED) begins flashing and might eventually turn
off or flash rapidly, depending on the configuration.
8 ServeRAID-8 Series: Best Practices and Maintenance Information
v The solid amber drive LED turns off after a transition to a Hot-Spare, Ready,
or Online state is complete. Otherwise, it flashes during a rebuild or copy
back operation.
Note: You can see the drive state transitions in real time by using ServeRAID
Manager.
(Optional) You can use these same instructions for an offline hard disk drive
replacement by starting to the IBM ServeRAID Support CD.
PHY settings and AMSU updates
The term PHY equates to the port or channel on SAS and SATA controllers and
devices. PHY ports on a SAS controller have settings that tune the signal for
correct communication across the channel. SAS devices use one group of settings,
and SATA devices use another. Each device type uses separate values and can be
optimized individually.
The tool that is used to program these settings is the Adaptec Manufacturing SAS
Update (AMSU) utility. PHY settings are not firmware updates. Over time, IBM has
made improvements to the signal quality through continued testing and as new
devices are released. AMSU updates can improve communication to disk devices
but is not necessary in all situations.
In most situations, the original PHY settings operate normally; however, in some
cases, updating to the latest AMSU settings is necessary. The symptoms that are
generated by previous levels of PHY settings often have multiple origins; therefore,
it is very important to apply all other updates and fixes before you consider the
AMSU updates.
The following symptoms are observed in IBM lab testing:
v A server experiences repeated hard disk drive failures in the same slot location.
v A drive intermittently becomes missing from the configuration.
v The ServeRAID Manager RaidEvtA.log file indicates sense data 0b/4b/04
“informational” events.
Apply the following fixes before you update the AMSU settings:
v Apply the latest ServeRAID-8k or ServeRAID-8k-l firmware and device drivers.
v Apply the latest hard disk drive firmware.
v Apply the latest backplane update (if one is available for the system).
v Review and apply host system updates that are applicable.
If the symptoms continue, obtain the current AMSU update package and apply the
updates. For more information about a complete ServeRAID maintenance review,
see “Periodic ServeRAID maintenance updates and evaluation.”
Periodic ServeRAID maintenance updates and evaluation
On a periodic basis, the server and the ServeRAID subsystem require maintenance
and service to remain optimal. To help avoid known issues before they might affect
the server, complete the following steps:
1. Make sure that you back up the data.
2. Update the ServeRAID device driver.
Chapter 1. ServeRAID-8 series best practices and maintenance information 9
3. Update the ServeRAID controller firmware. If this results in a failed flash
update or a blinkLED event, run a zero-mode flash update to recover the
controller.
4. Update the ServeRAID Manager application, monitoring agents, or IBM
Systems Director components, as applicable.
5. Update the command-line tools.
6. Update the hard disk drive firmware.
7. Update the backplane firmware, as applicable to the system.
8. For external SAS/SATA storage, update the firmware of any attached IBM
System Storage
®
EXP3000.
9. Evaluate the host system updates that are available and make sure that the
current microcode is installed. (Sometimes these updates can dramatically
improve ServeRAID runtime operations.)
10. Make sure that Data Scrubbing is enabled.
11. Make sure that Automatic Failover is enabled.
12. Run a synchronization on each logical drive (lower the task priority as needed
to minimize the production impact).
13. Evaluate the current AMSU settings for the servers that are listed in Table 1.
Note: You can check the AMSU settings by using the following arcconf
getlogs command:
arcconf getlogs n uart > filename.txt
Search this log for the following string:
mfg connector name=
The string is followed by a value for the AMSU settings, for example:
mfg connector name=j1sprucewood-v2
Table 1. AMSU settings for servers
Server AMSU version Release date Changes
IBM System x3500 J1SPRUCEWOOD-V3 August 2008 Improved SATA PHY
settings
J1SPRUCEWOOD-V2 July 2007 Improved SAS PHY
settings
J1SPRUCEWOOD Original Original version
IBM System x3550 J1DEFIANT-V3 August 2008 Improved SATA PHY
settings
J1DEFIANT-V2 May 2008 Improved SATA PHY
settings
J1RIGDE LITE-V4 July 2007 Improved SAS PHY
settings
J1SPRUCEWOOD Original Original version
IBM System x3650 J1VALIANT-V2 July 2007 Improved SAS PHY
settings
J1SPRUCEWOOD Original Original version
IBM System x3655 VJ1THUNDER-V2 July 2007 Improved SAS PHY
settings
J1SPRUCEWOOD Original Original version
10 ServeRAID-8 Series: Best Practices and Maintenance Information
14. (Optional) Evaluate whether the Copy Back option is necessary in the
configuration and set it accordingly.
15. (Optional) Inspect the SAS cables for the correct connection to the backplanes
and system boards.
16. (Optional) Review the configuration and evaluate the need to define a
hot-spare drive.
17. (Optional) Run the operating-system level “file system” health check tools.
18. (Optional) Run the native application-based data integrity tools to make sure
that the data is good.
19. Note the date and time of completion.
20. Observe and evaluate the ServeRAID logs to verify the correct operation.
Note: A healthy log contains informational events but minimal warning and
critical events. Repeated warning or critical events after a complete
health check might require additional troubleshooting.
Zero-mode flash update overview
A zero-mode flash update is a recovery procedure for ServeRAID controllers that
you can perform when a flash update fails or otherwise appears to have corrupted
firmware. Some of the following symptoms might appear if a ServeRAID controller
flash update has failed:
v The ServeRAID controller is not seen during POST after a recent flash update.
v The flash update program returns an error code or failed message.
v The system loses power during a flash update.
v A kernel panic event is displayed during the ServeRAID controller POST.
v A blinkLED message is displayed during the ServeRAID controller POST.
The following sections contain information about zero-mode flash updates for the
following controllers:
v ServeRAID-8i
v ServeRAID-8s
v ServeRAID-8k and 8k-l in the following systems:
– IBM BladeCenter Storage Expansion Unit 3
– System x3500 server
– System x3550 server
– System x3650 server
Zero-mode flash for ServeRAID-8i (optional controller)
To perform a zero-mode flash update for a ServeRAID-8i SAS PCI-X Controller,
complete the following steps:
1. Obtain a 2-pin jumper connector.
2. Download and create the ServeRAID-8i flash diskettes from the IBM
ServeRAID Support CD version 9.0 (containing build 15407 or later) or
download the individual diskettes (two total) from the ServeRAID Software
Matrix Web site; go to http://www-304.ibm.com/systems/support/
supportsite.wss/docdisplay?lndocid=SERV-RAID&brandind=5000008 or
complete the following steps.
Chapter 1. ServeRAID-8 series best practices and maintenance information 11
Note: Changes are made periodically to the IBM Web site. Procedures for
locating firmware and documentation might vary slightly from what is
described in this document.
a. Go to http://www.ibm.com/systems/support/.
b. Under Product support, click System x.
c. Under Popular links, click Software and device drivers.
d. Under Related downloads, click ServeRAID to display the matrix of
downloadable ServeRAID files.
Detailed instructions for building these diskettes are provided in the readme
file.
3. Turn off the server and remove the cover.
4. Determine whether the flash jumper is accessible. If it is not, remove the
ServeRAID-8i controller.
5. Locate the flash jumper, which is labeled on the top of the controller as shown
in Figure 1.
Figure 1. ServeRAID-8i zero-mode flash jumper
6. Install the jumper connector across the two pins that are shown in Figure 1.
7. If you removed the controller in step 4, reinstall it.
8. Insert the ServeRAID-8i firmware disk 1 into the diskette drive.
9. Turn on the server and observe the server POST process. If the zero-mode
jumper is installed correctly, the ServeRAID-8i controller does not display a
POST banner.
10. As the server starts to diskette 1, the following message is displayed:
This program will update the firmware on all IBM ServeRAID-8i controllers
in the system. Do you want to continue. [Y,N]?
11. Type y to respond to the prompt, and then follow the instructions on the screen
to complete the flash update.
12. After the flash update is completed, turn off the server.
13. Remove the zero-mode jumper from the flash pins (if necessary, remove and
reinstall the ServeRAID-8i controller).
12 ServeRAID-8 Series: Best Practices and Maintenance Information
Loading...