C H A P T E R  4

Known Issues

This chapter contains the latest supplementary information for the preceding chapters in this guide. Specific Change Request (CR) identification numbers are provided for service personnel, when necessary. This chapter contains the following topics:

 


Ship Kit Issues

This section describes the known ship kit issue.

Some or All CDs Are Not Included in the HBA Ship Kit

Workaround: You can obtain the latest drivers and software at: http://support.intel.com/support/go/sunraid.htm


BIOS Utility Issues

The following are known issues with the BIOS RAID Configuration utility:

Creating an Array With the BIOS Utility Changes the BIOS Boot Order

Workaround: After creating the array, check the BIOS settings to verify the correct boot order and make changes as necessary. For more information, see Best Practices For Controlling the Boot Order of Logical Drives.

Hot-Plug Functionality Does Not Work in the BIOS Utility

Workaround: Hot-plugging of enclosures is not supported in the BIOS RAID Configuration utility. Hot-plugging of SAS/SATA hard disk drives (HDDs) is supported only within hard disk enclosures and only under the conditions specified in Understanding Hot-Plug Limitations and Conditions Within the BIOS RAID Configuration Utility.


Performance Issues

This section contains the known performance issues:

ZFS Forces a Flush of the NVRAM on the HBA When Completing Synchronous Writes, Which Impacts Performance

Workaround: As of Solaris 10 8/07 (s10u4), you can prevent ZFS from issuing SYNCHRONIZE CACHE commands to the NVRAM on the HBA by defining a ZFS global setting in the Solaris /etc/system file. This setting improves ZFS performance and is appropriate for Solaris 10 8/07. However, this setting will likely not be required in subsequent releases of the Solaris OS. The setting must only be used if all devices managed by ZFS are managed with non-volatile caches.

To define the ZFS global setting, do the following:

1. On the system in which the HBA is installed, add the following line to the Solaris /etc/system file:

 


set zfs:zfs_nocacheflush=1



Note - This global setting affects all ZFS file systems on the system in which the HBA is installed. Keep in mind that you must not define this setting if ZFS is managing any disks with volatile caching, as the setting can put data at risk on those disks.


2. Reboot the system.

For more information about how to reboot the system, see your system documentation.

The System Freezes When a Sun Storage 32GB SLC SATA Solid State Drive (SSD) Is Configured In the HBA

CR 6806467

Issue: This occurs because the HBA has firmware prior to version 16732 installed on it and the Sun Storage 32GB SLC SATA SSD has firmware version 8626 or earlier installed on it.

Workaround: Do the following:

1. Upgrade the HBA firmware to version 16732, at minimum.

You can obtain the latest HBA firmware at: http://support.intel.com/support/go/sunraid.htm

2. Power cycle the HBA host system.

3. Upgrade the SSD firmware to version 8850, at minimum.

You can obtain the latest SSD firmware at: http://www.sunsolve.sun.com

4. Power cycle the SSD.

A Space Usage Error Message Is Displayed When Trying to Expand an Existing Volume

CR 6871696

Issue: When attempting to grow or expand a volume on a device that has enough storage space to do so, the following error message might be displayed:

The specified operation failed because there was not enough space on the specified device.

This error message occurs if the write cache setting is unkown on some of the drives used by the HBA.

Workaround: Initialize the drive and then grow the volume onto that initialized drive. To avoid encountering this error message, initialize all ready drives before the drives are used by the HBA.

Cannot Access the HBA From the GUI Nor Access a LUN From the Host

CR 6820225

Workaround: Reboot the host system on which the HBA resides.


JBOD Issues

The following are known issues with JBODs:

Difficulty With Detecting Disks in a JBOD

Workaround: If you have difficulty detecting disks in a given JBOD, do the following:

1. Make sure the JBOD has been power cycled to clear affiliations.

For more information about affiliations, see Connecting a JBOD With SATA Disks.

2. If the JBOD has not been power cycled, disconnect the JBOD, power cycle it, and reconnect the JBOD.

3. If you are still not seeing disks, disconnect the JBOD and fully bring up the host system with the JBOD cable unplugged.

This ensures that no disks are connected to the card.

4. Start the Sun StorageTek RAID Manager GUI and, from the GUI, confirm the disks are not connected to the card.

5. Reconnect the cable and verify in the GUI that the disks are displayed and reconnected to the card.

Keep in mind that the more disks on a given channel, the longer the disks will take to be displayed in the GUI. It may take several minutes.

During System Boot Time, JBOD Affiliations Cause HBA Inoperability and System Panic

CR 6723287

Issue: If you connect a JBOD with SATA disks to an HBA, and the JBOD has affiliations, the HBA might become inoperable and the Solaris system might panic during boot time. This occurs when the system has an old firmware version.

Workaround: Do either of the following:

After Upgrading Firmware on a JBOD, the JBOD Is No Longer Detected by the GUI

CR 6792854

Issue: If you incorrectly move cables connected to the JBOD during the JBOD firmware upgrade process, the GUI can no longer detect the JBOD.

Workaround: Before upgrading firmware on a JBOD, review Best Practices For Switching Cables and Making New Connections and Best Practices For Cabling to Disk Enclosures. If you have already moved cables from the JBOD during the firmware upgrade, see Difficulty With Detecting Disks in a JBOD.

The Solaris System Panics After Attaching a JBOD to the HBA

CR 6818045

Issue: This occurs if a JBOD that is already attached to the HBA has incoming IO requests running while you are trying to attach the new JBOD.

Workaround: Quiesce IO on the HBA prior to attaching the new JBOD to the HBA.