This README file contains the latest information about the IBM TotalStorage™ DS300 and DS400 Storage Server controller firmware version 7.00 Refer to the IBM TotalStorage™ Support Web Site or CD for the IBM ServeRAID™ Manager Installation and Support Guides for controller firmware download instructions. For other related publications (refer to Related Publications in the installation and support guide) and the latest information and updated README file. Last Update: 10/19/2005 Products Supported: ------------------------------------------------------ | | Machine Type | Model | |-------------|--------------|-------------------------| | DS300 | 1701 | 1RL ,1RS, 1ES, 2RD ,2ES | |-------------|--------------|-------------------------| | DS400 | 1700 | 1RS, 1ES, 2RD ,2ES | ------------------------------------------------------ CONTENTS -------- 1.0 Overview 2.0 Change History 3.0 Installation and Setup Instructions 4.0 Configuration Information 5.0 Unattended Mode 6.0 Web Sites and Support Phone Number 7.0 Trademarks and Notices 8.0 Disclaimer ======================================================================= 1.0 Overview -------------- This IBM TotalStorage DS300 and DS400 release includes the IBM version 7.00 Storage Server Controller Firmware for upgrading the following IBM machine types - 1700/1701-1RL,1RS,1ES,2RD,2ES. The IBM ServeRAID Manager version 8.20.00 Installation and Support Guides are available on IBM's Support Web Site as a downloadable Portable Document Format (PDF file). Refer to the IBM Support Web Site for the latest Firmware files for the IBM DS300 & DS400 Storage Subsystems. http://www.ibm.com/servers/storage/support/disk/ ======================================================================= 1.1 Limitations --------------------- 1. The Management Station Service currently only runs on a Windows,IA32 bit platform. 2. The DS400 does not coexist on the same SAN, using the DS4000 HBA device driver for DS400, as the DS4000 (formally FAStT) Storage Subsystem. 3. The multipath DSM software for the DS4000 HBA FC2-133 must be installed, in conjunction with the corresponding DS4000 HBA device driver for DS400, in order to provide multipathing support. 4. The default password for the controller is passw0rd. 5. On the DS400, the loopback diagnostic function does not function correctly. 6. A Raid 10 array that is in a rebuild state may change to the synchronization /rebuild state on a failover. This is OK. If the synchronization does not complete before the problem is fixed and a failback ensues, the array will go back to the rebuild state on the original owning controller. 7. Event log time stamp maybe off by one hour from the controller time due to daylight savings. 8. After a reboot, the data scrubbing rate will maintain the last selected state of High, Medium or Low. The only exception is if the last state was Off, then the rate default to Low. 9. Sometimes the preferred owner of some LUNs may change after a reboot or a move. The preferred owner can be reset by doing a move of the LUN back to the other controller. 10. If the host system boots up prior to the storage subsystem completing its power up cycle, LUNs assigned to that host may not appear. Rescan for disks after the storage subsystem has powered up fully. 11. On occasion, when using ServeRAID Manager to create Arrays and/or Logical Drives, the creation may result in failure of all or part of the array or logical drive creation. Try creating the array again. 12. The maximum number of Basic arrays that can be created is 24. 13. In the CLI, if a command does not complete successfully, retry the command. 14. If your storage subsystem identifies arrays as being ALIEN, please contact IBM support to find out how to handle these arrays. 15. In the event a critical array fails over to the peer controller (or becomes critical while owned by the non-preferred controller), the array will not complete a rebuild until the array is manually moved back to its preferred controller. Critical arrays that are in the process of rebuilding will failover and continue to rebuild if the preferred controller goes down. 16. If a rebuild of an array halts after a controller or path failover (all LUNs end up on a single controller), the rebuild may restart on its own when the array is failed back to the original or 'replaced' controller. 17. If a drive is inserted into the enclosure from another DS300 or DS400 enclosure, that drive may contain metadata from the previous configuration. This can result in foreign or alien arrays being presented in the array list. These arrays must be destroyed before the drive can be used. 18. If a drive is removed, causing an array to go into a degraded or critical state, only replace the removed drive if a hot spare has not been assigned to that array or a global hot spare has not rebuilt for that drive. 19. Logical drives mapped to multiple initiators that are installed in the same host system need to be mapped with the same logical drive / LUN combination on all initiators. 20. The amber LED on the disk drive may not light for a defunct drive nor flash for a rebuild on a hot-spare disk drive. The ServeRAID Manager GUI will display the rebuild status. 21. If a hard drive fails while the array is still synchronizing and the controller goes offline causing the array to failover to the alternate controller, a rebuild will not begin until the array has been failed back to the original controller and the synchronization has completed. 22. You must set the Cluster service to be dependent on the iSCSI service when using the MS iSCSI software initiator. For directions on how to do this refer to MS Technet article 883397. 23. When connecting Cluster shared disks using a MS iSCSI software initiator the drives must be mapped as persistent targets. To do this select the logical drive that you want to connect and then click logon. Then check the box that say "Automatically restore this connection when the system reboots". 24. LUN snapshots, when using the Flashcopy option can grow to 2x the size of the parent LUN, plus some additional 10-30MB. Please be aware that this space must be left as free space on the array to allow the LUN growth. 25. When a controller is in a failed state (all arrays moved to the remaining controller), do not do any array/LUN operations on the arrays/LUNS on the non-preferred controller. This includes such things as creating LUNs, creating snapshots, creating hotspares, etc. Wait until the bad controller or network is replaced and the arrays failed back to the 'preferred' controller before doing these array/LUN operations. 26. If a LUN or array or disk drive 'disappear' from the GUI or CLI or appear multiple times in the GUI or CLI, reboot the controllers. 27. Snapshot logical drives can not be mapped to the same host on a Windows 2000 host. Since the Snapshot logical drive contains the same disk signature as the base logical drive the Windows Disk Management tool will not know how to resolve this conflict. On a Windows 2003 host the disk signature will be changed if the snapshot logical drive is a read\write logical drive. It is always recommended to map the snapshot logical drives to a second host before accessing them. 28. To format an existing hard disk or partition on a shared cluster hard disk or partition, please refer to Microsoft support article 257937, at URL: http://support.microsoft.com/default.aspx?scid=kb;en-us;257937 29. For a dual controller DS400, if an array is made up of logical drives and snapshots and mapped to a Linux host. During a failover/failback condition one or more of the LUNs may become inaccessible because the logical drive or snapshot have gone offline. Rebooting the storage subsystem will restore the item to an online status. 30. After completing an Online Capacity Expansion (OCE) and additional logical drives are created, it is possible that when mapped to an initiator all newly created LUNs will not be seen by the initiator. If this occurs, delete the logical drives that the initiator does not see and recreate them. 31. When the size of the global hotspare is larger than the capacity of the disk drives in the array it rebuilds into, the global hotspare after completion of the array rebuild will be represented in the SRM and CLI as a dedicated hotspare. This is a reporting issue, the dedicated hotspare is an online disk drive and a valid member of the array. The rebuilt disk drive is treated as an online member of the array. Failovers, failbacks and additional rebuilds with a hotspare all will work correctly. 32. When the OS of the SRM management station is a Japanese DBCS OS, one cannot create arrays with the SRM. The CLI must be used to create the array. 33. Logical drive names must start with an Alpha character. 34. When updating the date/time in the ServeRAID Manager, manageability may be temporarily lost. Manageability will return on its own within 1 minute. If the date/time update is still missing after the ServeRAID Manager becomes manageable again, hit PF5 to refresh the GUI. 35. RedHat v4, Update1 does not support any iSCSI software driver solution. This is expected to be resolved in Update2. 36. ServeRAID Manager will expect a minimum of 4 disk drives for an online capacity expansion (OCE) of a RAID-10 array. 37. DS300 and DS400 only support U320 hard drives. ======================================================================= 1.2 Enhancements ----------------- DS300 ----------- Combined single-controller and dual-controller firmware into a single image. Added online capacity expansion (OCE) function Added RAID level migration (RLM) function Added support for 1GB system DIMM upgrade Added RAID-50 function to dual controller systems Added global hot-spares to dual controller systems Added support for 4 snapshots in base firmware without FlashCopy option Added SnapShot >4 function to dual controller systems Added support for Flashdrive HDD firmware upgrade Added support for Red Hat Linux v3 & v4 Added support for Suse Linux v8 & v9 Added support for EXP400 expansion to single controller systems Added MSCS support for single controller systems DS400 ----------- Combined single-controller and dual-controller firmware into a single image. Added online capacity expansion (OCE) function Added RAID level migration (RLM) function Added support for 1GB system DIMM upgrade Added RAID-50 function to dual controller systems Added global hot-spares to dual controller systems Added support for 4 snapshots in base firmware without FlashCopy option Added SnapShot >4 function to dual controller systems Added support for Flashdrive HDD firmware upgrade Added support for Red Hat Linux v3 & v4 Added support for Suse Linux v8 & v9 Added support for EXP400 expansion to single controller systems Added MSCS support for single controller systems ======================================================================= 1.3 Prerequisites ------------------ The system can be a DS400, Machine/Type: 1700/1RS,2RD,1ES,2ES or DS300, Machine/Type: 1701/1RL,1RS,2RD,1ES,2ES running either the 4.15,5.78 ,6.02, 6.03 or 6.04 controller firmware. The management system must be running ServeRAID Manager v8.20. ======================================================================= 1.4 Dependencies ----------------- ServeRaid Manager provides the monitoring and Management Functions for the DS300 & DS400. This utility must be installed to provide notification of alerts to user level SNMP Trap Managers. ======================================================================= 2.0 Change History ----------------------------- ====================================================================== 2.1 Version History ----------------------------- DS300 ----------- 7.00 10/2005 New Function & Maintenance Release Fixed multiple CMVC defects associated with the following issues: Array/logical drive offline Controller offline Luns in transition MSCS cluster issues ServeRAID Manager reporting Linux iSCSI support Linux failover support 6.04 07/2005 Maintenance Release Fixed issue with data scrubbing function 6.03 05/2005 Maintenance Release Fixed multiple CMVC defects associated with the following issues: Array/logical drive offline Background data scrubbing off Blink LED on controller Controller hangs Luns in transition MSCS cluster issues MS device logo issues ServeRAID Manager event reporting ServeRAID Manager info/status SnapShot creation Snapshot LUN issues 6.02 03/2005 Initial Dual Controller Model Release 4.15 - 10/29/04 Single Controller Maintenance Release - Fixed several error handling conditions for blink LED state. - Added additional performance tuning. - Fixed several issues related to how HDD firmware download is handled by the controller. - Added HDD buffer test and tracert function to the CLI diag menu. - Provided fixes for Microsoft HCT testing. - Changed that way controller deletes arrays when initiators are assigned. - Updated Event Monitor messages. 4.04 - 09/28/04 Initial Single Controller Model Release DS400 ----------- 7.00 10/2005 New Function & Maintenance Release Fixed multiple CMVC defects associated with the following issues: Array/logical drive offline Controller offline Luns in transition MSCS cluster issues ServeRAID Manager reporting Linux iSCSI support Linux failover support 6.04 07/2005 Maintenance Release Fixed issue with data scrubbing function 6.03 05/2005 Maintenance Release Fixed multiple CMVC defects associated with the following issues: Array/logical drive offline Background data scrubbing off Blink LED on controller Controller hangs Luns in transition MSCS cluster issues MS device logo issues ServeRAID Manager event reporting ServeRAID Manager info/status SnapShot creation Snapshot LUN issues 6.02 03/2005 Maintenance Release Fixed multiple CMVC defects associated with the following issues: Array/Logical Drive offline Array/Logical Drive creation Array/Logical Drive controller ownership change Blink LED on controller Controller info/status Controller boot issue Controller hangs Diagnostic fix Duplicate Array/Logical Drives Error handling of HDD events Expansion fix and license key ISCSI initiator connectivity LUNs assessable to host LUNs in transition ServeRAID Manager event reporting ServeRAID Manager info/status reporting 5.87 12/2004 Initial Dual Controller Model Release 4.15 - 10/29/04 Single Controller Maintenance Release - Fixed several error handling conditions for blink LED state. - Added additional performance tuning. - Fixed several issues related to how HDD firmware download is handled by the controller. - Added HDD buffer test and tracert function to the CLI diag menu. - Provided fixes for Microsoft HCT testing. - Changed that way controller deletes arrays when initiators are assigned. - Updated Event Monitor messages. 4.04 - 09/28/04 Initial Single Controller Model Release ====================================================================== 2.2 ======================================================================= 2.3 Code levels at time of release are as follows ------------------------------------------------- DS300 : 7.00 DS400 : 7.00 Refer to the http://www.ibm.com/servers/storage/support/disk IBM TotalStorage™ Disk Storage Systems Technical Support web site for the latest released code levels. ======================================================================= 3.0 Installation and Setup Instructions ----------------------------------------- Notice: Backup SnapShots prior to upgrade from v4.xx firmware If you are updating the controller in a single-controller DS300 or DS400 storage subsystem and the firmware on the controller is version 4 (any release), any existing snapshots (from the FlashCopy® option) will be inaccessible with the new replacement controller. After the update the controller firmware, you must delete the snapshot LUNs and re-create them at the new firmware level. If the firmware on the controller is version 5.0 or later, the snapshots are preserved with the new firmware level. If the snapshots are unavailable, you are notified in ServeRAID Manager that those LUNs are offline, invalid, or unavailable. Notice: Need to updated license key after upgrade from v4.xx firmware If the firmware on the controller is version 4 (any release), existing license keys might not be valid and must be regenerated. Contract IBM Support for an updated option license key. See http://www.ibm.com/planetwide/ for support telephone numbers. Notice: Import of foreign arrays after upgrade from v4.xx firmware After upgrading a controller from v4.xx firmware all arrays will be shown as Foreign. Foreign arrays can be imported via the GUI by right clicking on the array and selecting Import foreign array. Notice: Loading v7.00 image to both primary and secondary image bank After upgrading a controller from v4.xx firmware, reboot the system. When the system reboots, download the v7.00 firmware image a second time. This ensures that the controller can not be backleveled to v4.xx firmware. After upgrading to v7.00 firmware, moving back to firmware v4.xx will render all arrays/logical drives inaccessible. 3.1 Installation from ServeRAID Manager GUI To download firmware, do the following: a. In the Enterprise view, right-click the Networked storage icon and select Update controller images. The IBM ROM Update wizard opens. b. Click Next. The ROM Update Image selection window opens. c. Click Add and select the firmware update image file. Then click Next. d. Select the check box for the enclosure controllers that you want to update, and then click Next. Note: To force an update, hold down the Ctrl key while selecting the enclosure to be updated. e. Review the update summary, and then click Apply. ServeRAID Manager applies the firmware update to the selected enclosure controllers. The progress meter shows the update status. When the update is completed, the progress meter automatically closes. f. Restart the controller by right-clicking on the storage subsystem in the Enterprise view and selecting Restart enclosure. 3.2 Installation from Command Line Interface (CLI) To download firmware, do the following: a. Open a Telnet session to the storage controller. Note: The default IP settings for the management port are, IP Address: 192.168.70.123, Subnet Mask: 255.255.255.0 Password: passw0rd (where the letter o is replaced with a zero 0 ) b. Place the DS400 firmware image on an FTP server that is also accessible to the storage subsystem. c. type ADMIN , then Enter d. type IMAGE UPLOAD ftp:///xx.xx.xx.upgrade, then Enter e. When asked if you want to replace IMAGE xxx type Y, then Enter f. Once the download is complete type System Reboot, then Enter g. When asked if you want to reboot, type Y, then Enter ======================================================================= 3.3 Helpful Hints ------------------ 1. If you lose connectivity to a controller, right click on the management station and select remove agent. Once the controller has been removed select add agent and add back the controller agent. 2. You will need to have the IQN of your iSCSI initiators before assigning the logical drives. Please use the proper software utility for finding this information before attempting to assign LUNs. If you need additional assistance, please refer to the "DS300 / DS400 Problem Determination Guide". 3. In some instances when a number of LUNs are assigned to a host, the Windows operating system will attempt to access the LUNs via the alternate path (if available) which can result in a long delay before the LUNs are available to the operating system. Assigning all the Arrays/Logical Drives to the A controller can minimize this from occurring. 4. IPfailover policies: There are four options to set for IPfailover; remote, local, local-remote and none. The default option is remote. This allows for: Eth2[A] to failover to Eth2[B] and Eth3[A] to failover to Eth3[B] and vice versa. The local option allows for failover between ethernet ports Eth2 and Eth3 within the same controller. There is no failover between controllers if this option is used. If this option is used you must also have ip forwarding or some other type of routing enabled within your network. The local-remote option allows for initial failover first between ethernet ports Eth2 and Eth3 on the same controller. Subsequent failovers will failover from ethernet port Eth2 to ethernet port Eth2 on the alternate controller or ethernet port Eth3 to ethernet port Eth3 on the alternate controllers. This option also requires support for ipforwarding or routing. The none option disables failover so that if a port fails those sessions will be dropped. ======================================================================= 4.0 Configuration Information ----------------------------- 4.1 Configuration Settings -------------------------- 1. By default the ServeRAID Manager software does not automatically map logical drives. This means that the logical drives are not automatically presented to host systems. All logical drives created in ServeRaid Manager must be assigned an access control list. The logical drive(s) may be assigned as unrestricted in iSCSI but a third party management software tool must then be used to control access to the assigned logical drive(s). 2. Access control is maintained per logical drive. You may assign the same logical drive to multiple hosts but you must then have some type of third party management software to control access to the logical drive. 3. When you are configuring IBM DS400 storage subsystems as a boot device, contact IBM support for supported configurations and instructions for configuring LUNs in the IBM storage subsystems as boot devices. ======================================================================= 4.2 Best practices ------------------------------ Use the following Best Practices to properly manage and maintain the IBM TotalStorage™ DS300 and DS400 1. Hot spare hard disk drives (HDD) should be globally assigned. All hot spare HDDs should be at least as large as the largest drive in any array. 2. Data scrubbing should be enabled and set to the highest setting that a user's performance will allow. Data scrubbing provides a means for bad HDD blocks to be discovered and corrected and thus reducing the risk of potential I/O errors. The default value is high. If the data scrubbing rate is set to Off, after a reboot data scrubbing will be enabled and set to Low. 3. Verify storage configurations before and after repair actions. Ensure arrays and logical drives are online and optimal prior to performing a failback operation. 4. Backup data regularly. ======================================================================= 4.3 Unsupported configurations ------------------------------ The following lists configurations that are currently not being supported with IBM TotalStorage™ DS400 1. Fibre Channel loop environments with the IBM Fibre Channel Hub/switch, machine type 3523 and 3534, in conjunction with the IBM Fibre Channel Switch, machine types 2109-S16, 2109-F16 or 2109-S8. In this configuration, the hub is connected between the switch and the IBM Fibre Channel RAID Controllers. 2. The IBM Fibre Channel Hub, machine type 3523, connected to IBM machine type 1700. 3. The McData 4300 fabric switch, connected to IBM machine type 1700. ======================================================================= 5.0 Unattended Mode --------------------- N/A ======================================================================= 6.0 WEB Sites and Support Phone Number ---------------------------------------- 6.1 IBM TotalStorage™ Disk Storage Systems Technical Support web site: http://www.ibm.com/servers/storage/support/disk 6.2 IBM TotalStorage™ Marketing Web Site: http://www.ibm.com/servers/storage/disk/ds/index.html 6.3 If you have any questions about this update, or problem applying the update go to the following HelpCenter World Telephone Numbers URL: http://www.ibm.com/planetwide ======================================================================= 7.0 Trademarks and Notices ---------------------------- 7.1 The following terms are trademarks of the IBM Corporation in the United States or other countries or both: IBM ToTalStorage™ the e-business logo xSeries pSeries HelpCenter Microsoft, Windows, and Windows NT are trademarks of Microsoft Corporation in the United States, other countries, or both. Java and all Java-based trademarks and logos are trademarks or registered trademarks of Sun Microsystems, Inc. in the United states, other countries, or both. UNIX is a registered trademark of The Open Group in the United States and other countries. Linux is a registered trademark of Linus Torvalds. Other company, product, and service names may be trademarks or service marks of others. ======================================================================= 8.0 Disclaimer ---------------- 8.1 THIS DOCUMENT IS PROVIDED "AS IS" WITHOUT WARRANTY OF ANY KIND. IBM DISCLAIMS ALL WARRANTIES, WHETHER EXPRESS OR IMPLIED, INCLUDING WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF FITNESS FOR A PARTICULAR PURPOSE AND MERCHANTABILITY WITH RESPECT TO THE INFORMATION IN THIS DOCUMENT. BY FURNISHING THIS DOCUMENT, IBM GRANTS NO LICENSES TO ANY PATENTS OR COPYRIGHTS. 8.2 Note to U.S. Government Users -- Documentation related to restricted rights -- Use, duplication or disclosure is subject to restrictions set forth in GSA ADP Schedule Contract with IBM Corporation.