Summary of changes in release 2.1

This section describes the enhancements made to SAN File System in release 2.1.

The following list describes the technical changes and enhancements made to SAN File System for release 2.1.
  • New Web address for the user interface — The Web address that is used to access the SAN File System console has changed to:

    https://IP_address:7979/sfs

  • Software-only solution — SAN File System is now a software-only solution that leverages the strengths of the standard SuSE Linux™ Enterprise Server platform.
  • Heterogeneous SAN environment — SAN File System now supports a wide variety of SAN configurations, which eases scaling to large numbers of SAN configurations and heterogeneous storage systems.
  • Storage pool access — The metadata servers require access to the metadata storage (system storage pool). They should not have access to the user data storage (user storage pools). Conversely, the clients require access to the user storage pool. They should not have access to the system storage pools.
  • Non-ASCII Unicode characters — SAN File System supports both uppercase and lowercase non-ASCII Unicode characters in file names. SAN File System policies also support non-ASCII Unicode characters.
  • Additional client platforms — SAN File System supports these additional client platforms:
    • AIX® 5.2 (32-bit and 64-bit)
    • Red Hat Enterprise Linux 3.0 Advanced Server
    • Sun Solaris 9 (64-bit)

    SAN File System also supports AIX and Solaris clients running in a clustered environment.

  • FIFO file-system objects — SAN File System supports FIFO file-system objects in the global namespace for UNIX-based clients.
  • Opportunistic locks (oplocks) — Windows-based clients can create and use opportunistic locks (oplocks). SAN File System supports level 1, 2, batch, and filter locks.
  • Application binary support — Application binaries for UNIX-based clients use the POSIX definition of three sets of three file modes bits: one set of reach user, group, and other.
  • Additional storage subsystem — SAN File System supports heterogeneous, simultaneously connected storage subsystems on clients with HBA sharing (subject to limitations of the client platform, drivers, and storage vendors). SAN File System conforms to FCP standards and is designed to work with any FCP-compliant storage subsystems for user data storage, including:
    • EMC Symmetrix
    • Hitachi 9900 Series and 9900V Series
    • IBM® FAStT 600T, 700, and 900 running firmware version 8.4 and software version 8.41
    Note: Only IBM storage subsystem are supported for the system storage pool.

    SAN File System supports an unlimited number of LUNs for user data storage. However, the amount of user data storage that you can have in your environment is determined by the amount of storage that is supported by the storage subsystems and the client operating systems.

  • High availability, non-disruptive maintenance, and serviceability — SAN File System has been improved to provide greater availability of the cluster.
    • Automatic workload failover (filesets and master role) — SAN File System supports the non-disruptive, automatic failover of the workload. SAN File System automatically redistributes the filesets of a failed or manually stopped metadata server and, if necessary, reassigns the master role to another metadata server in the cluster. SAN File System uses automatic workload failover to provide non-disruptive maintenance for the metadata servers
    • Automated failback of workload — If you statically assigned any filesets to a specific metadata server, SAN File System will automatically assign those filesets back to their statically assigned metadata server after the engine hosting the failed metadata server comes back online.
    • Non-disruptive LUN additions — SAN File System supports adding new LUNs to an existing system without restarting the system.
    • Non-disruptive fileset movement — You can manually reassign a fileset to another metadata server without disrupting service to the clients.
  • Master console improvements — In addition to updated software levels, the master console allows you to configure Windows® and IBM Director and to modify the machine name of the storage engines.
  • Serviceability improvements — Improvements have been made to various serviceability components, including:
    • Metadata checker support of a lost-and-found directory for unreferenced objects.
    • One button data collection utility that has a consistent interface across all platforms.
    • Performance improvements that reduce the impact of logging and tracing.
    • LUN and volume reporting to allow for easy comparison between the SAN configurations of each client.
    • Target Machine Validation Tool (TMVT) hardware and software verification tool. The TMVT runs before the SAN File System setup script and provides a report of installed levels of hardware and software, as well as the required levels of hardware and software.
    • IBM is providing you an optional script to customize your SuSE Linux Enterprise operating system. This script enables you to remove unnecessary packages from your system and to keep on your system only those packages required to run SAN File System. The script, named ktl.sh, can be found on the top level directory of the SAN File System Software CD.
  • Usability improvements — SAN File System has many usability improvements, including:
    • A Refresh button is on many SAN File System console panels to enable you to update the panel with latest information.
    • The dashboard has been simplified to provide you with a quick summary of the entire system.
    • The message IDs in the SAN File System console logs are linked to the Information Center to help you quickly resolve problems.
    • The helps for messages have been improved.
    • You can filter the SAN File System console logs by message date and severity.
    • You can display statistics for fileset transactions, policy sets, and file metadata.
    • You can add and remove individual privileged clients from the administrative command-line interface and SAN File System console.
    • SAN File System supports additional configuration options for the metadata servers and cluster.
    • You can start the cluster and metadata servers in the offline state.
  • Terminology changes — There are two major terminology changes in SAN File System:
    • The term container has been change to the more intuitive term, fileset.
    • The administrative command-line interface tool changed from tanktool to sfscli.
    • The string tank in most references has been replaced with sfs or sanfs.

Parent topic: Summary of changes

Library | Support | Terms of use | Feedback
(C) Copyright IBM Corporation 2003, 2004. All Rights Reserved.
IBM TotalStorage SAN File System v2.2