Setting Up Novell Storage Services

Updating NSS Volumes

After you upgrade to NetWare^® 6 and reboot the server, you need to update your NSS volumes. NSS volumes from previous releases will not function until you update them to NetWare 6 volumes-they do not upgrade automatically.

Checking the NDS Status

Before you update NSS volumes, you must ensure that NDS^® is running properly; otherwise, the upgrade process might remove your trustees. To check ensure NDS is running, you need to check the time stamp for the backlink process.

Do the following to ensure NDS is running:

1. Use iMonitor to determine that the upgrade process is complete.

   1. Open iMonitor in a network browser by entering the IP address of the server followed by /nds. For example: http://155.555.1/nds.

      1. Click Agent Process Status > External Reference Status.

         1. Make sure the time stamp (under the Time column) shows a time after the time you began the upgrade-specifically when NDS unloaded during the upgrade. The following graphic provides and example of the time stamp:

         
         

      If the time stamp shows a time earlier than when you began the upgrade, the backlink process is not complete. Repeat Step 1.a through Step 1.a until the backlink is complete.

Upgrading NSS Volumes

To upgrade NSS volumes, enter one of the following commands at the NetWare 6 server console:

• NSS /ZLSSVOLUMEUPGRADE=ALL
• NSS /ZLSSVOLUMEUPGRADE=volumename

Transaction Tracking System (TTS)

Transaction Tracking protects database applications by backing out transactions that are incomplete due to a system failure. TTS can be available on either all traditional or all NSS logical volumes. It cannot be available on all volumes if you use both types of volumes. If your system uses both traditional and logical volumes, by default it can only work on traditional volumes. If you want to use it on logical volumes on a system that has both, you must enter DISABLE TTS on the AUTOEXEC.NCF then enable the feature on each logical volume you want it on. Unlike traditional volumes, TTS does not apply to logical volumes by default. To enable it on logical volumes, enter the following command at the server console:

nss/transaction=<volname>.

Setting the Cache Buffers

A cache buffer is a block (4 Kb) of NetWare server memory (RAM) that temporarily stores data. When you install NSS, it consumes 60 percent of the the cache buffers by default.

The following are options for changing the cache buffers:

• You can set a maximum of 1048576 cache buffers.
• You can set the buffers in percentages rather than integers.

Use the following commands at the server console to set your cache buffers or tune your cache parameters. You can set or tune cache buffers at anytime.

Using ConsoleOne to Configure Novell Storage Services

ConsoleOne^TM is the primary utility you use to configure and maintain partitions, storage pools, and volumes in NSS. To open ConsoleOne and access the options you need, complete the following:

1. Open ConsoleOne.

2. Open the tree you want to work with.

3. Right-click on the server object you want.

4. Click Properties.

   This opens the ConsoleOne interface where you can access the snap-ins for managing NSS. The NSS configuration features are in the Media snap-in.

Setting Up ConsoleOne on a Workstation

To use ConsoleOne on a client workstation, you need to load ConsoleOne 1.3 or later. Then you must copy the NSS JAR files from your server to your client machine.

1. Copy these three NSS JAR files to your local LIB directory:

   public\mgmt\consoleone\1.3\lib\nssadmin.jar

   public\mgmt\consoleone\1.3\lib\nssc1lib.jar

   public\mgmt\consoleone\1.3\lib\nssjavalib.jar

2. Copy these two JAR files to your local RESOURCES directory:

   public\mgmt\consoleone\1.3\resources\nssadminres.jar

   public\mgmt\consoleone\1.3\resources\nssc1libres.jar

3. Copy the following JAR file to your local SNAPINS directory:

   public\mgmt\consoleone\1.3\snapins\nssadminreg.jar

Configuring Disk Space

To set up the NSS storage and file system, complete the following tasks in the given order:

1. Create a RAID device (optional).

   See Creating a RAID Device.

2. Create partitions.

   See Creating a Partition.

3. Mirror partitions.

   See Creating a Partition.

4. Create storage pools.

   See Creating a Storage Pool.

5. Create logical and traditional volumes.

   See Creating a Logical Volume and Creating a Traditional Volume.

Creating a RAID Device

NSS lets you create a RAID 0 device by striping data across multiple drives on your system. This RAID configuration occurs at the software level. You can use this RAID feature for both logical and traditional volumes.

Important information about NSS RAID

• Each element in the RAID configuration should come from a different device. NSS will let you obtain RAID elements from the same device, but this will severely impede the performance of your file system.
• Currently, the NSS RAID supports RAID 0. This means there is no disk redundancy. If one element in the RAID device malfunctions, the entire configuration is unusable.
• NSS RAID can accommodate a maximum of eight segments (amount of space used from each device) on a device per RAID 0 logical device.
• The stripe size (amount of data the file system places on one device before moving to the next device) ranges between 4 KB and 256 KB in increments of 2 KB.

1. From ConsoleOne, open the tree you want.

2. Right-click on the server object and select Properties.

3. Click Media > Raid Devices > New.

4. Enter the amount of space you want to obtain from each storage device in megabytes.

5. Select a segment size.

6. Select a stripe size.

7. Select a RAID type.

   The RAID type indicates the method of how data is striped across the different disks.

8. Click Next.

9. Select the devices you want to use for the RAID configuration.

   You can select one or more devices, but select only the ones you want to use for the RAID configuration.

10. Click Finish.

    The file system first creates a virtual device unit, then creates a RAID device.

Deleting a RAID Device

When you delete a RAID device, the file system deletes all the partitions and volumes from all physical storage devices used in the RAID configuration.

1. From ConsoleOne, open the tree you want.

2. Right-click on the server object and select Properties.

3. Click Media > Raid Devices.

4. Select the Raid device you want to delete.

5. Click Delete.

   The file system displays a warning message to ensure you want to delete the RAID device.

6. Click Yes or No.

Increasing the Size of the RAID Device

You can increase the size of a RAID device by adding more segments from another storage device. You should not place more than one RAID segment on a disk; this will severely impede the performance of your file system.

1. From ConsoleOne, click Media > Raid Devices.

2. Select the RAID device you want to increase.

3. Click Increase Size.

4. Select the device you want to add to the RAID configuration.

5. Click Finish.

   After you add another physical storage device to your RAID device, you must restripe the configuration. This stripes the data to the other device you added.

   NOTE:  The file system performance might slow down during the restriping process. Make sure you add all the physical storage devices you want to the RAID device before you click Restripe.

6. Click Restripe.

   The Restripe option recognizes new devices in your RAID configuration and distributes the data across all the devices.

Marking a Device as Sharable for Clustering

Some storage devices exist in a storage area network (SAN). These devices can be shared by multiple servers in a cluster.

IMPORTANT:  The system hardware does not specify disk drives to come up automatically as sharable for clustering or not sharable for clustering. You must manually set different devices according to the configuration of your storage system.

The following is important information about making drives Sharable for Clustering:

• All portions of an NSS pool must be on a storage device using the Sharable for Clustering feature. This means if you have a storage pool that spans multiple devices, all these devices must be either sharable for clustering or not.
• The Sharable for Clustering option is not available for devices that contain the storage pool SYS: or Volume SYS:.
• ConsoleOne does not provide the Sharable for Clustering option for devices that contain traditional partitions and volumes.
• When you add partitions to a mirror group, all partitions in that group must be either sharable for clustering or not.

1. From ConsoleOne, open the tree you want.

2. Right-click on the server object and select Properties.

3. Click Media > Devices.

4. Select a device.

5. Select Sharable for Clustering.

6. Click OK.

Creating a Partition

The first task for setting up NSS is to create partitions on your storage devices for the NSS storage pools and logical volumes.

1. From ConsoleOne, open the tree you want.

2. Right-click the server object and select Properties.

3. Click Media > Partitions > New.

4. Select a device to create a partition on.

5. Select the type of partition you want to create (NSS, Traditional, Remote Storage Device).

6. Enter the size of the partition in bytes (B), kilobytes (KB), megabytes (MB), or gigibytes (GB).

7. To reserve space for the Hot Fix error correction feature, select Hot Fix and enter the amount of space or percentage you want to reserve.

   Mirrored partitions must be compatible in data area size. This means the new partition must be at least the same size or slightly larger than the other partitions in the group. The physical size (combined data and Hot Fix size) of the partition must be at least 100 KB, but no more than 120 MB larger than the data size of the existing partitions in the mirror group.

8. To mirror the partition, select Mirror and select one of the following options:

   • Create New Mirror. This option means you are making the partition capable of being part of a mirror group. You do not actually create the group until you add another mirrored partition to the partition you are creating.
   • Existing Mirror Group. (If you select this option, also select the ID of the mirrored partition.) This shows a list of existing mirror groups that are compatible in data area size. This option lets you add this new partition to one of the mirror groups in the list.

9. Select a label for the partition (optional).

10. Click OK.

Deleting a Partition

When you delete a partition you remove all volumes and data on that partition. If the partition is mirrored, the other partitions in the mirror group will retain the data from the deleted partition. If you want to delete a mirrored partition, you must unmirror that partition before you delete it.

Unmirror a partition

1. From ConsoleOne, open the tree you want.

2. Right-click the server object and select Properties.

3. Click Media > Partitions > Mirror.

4. Select the partition you plan to delete, click Remove.

   This removes the partition from the mirror group. You can now delete the partition.

   
   

   Delete a Partition

1. From ConsoleOne, open the tree you want.

2. Right-click the server object and select Properties.

3. Click Media > Partitions.

4. Select the partition you want to delete, click Delete.

Creating a Storage Pool

After creating and setting up your partitions, you can create storage pools within the partitions. When you create a storage pool, NSS lets you determine how much space to use from each storage device on your server.

When you create storage pools, you use partitioned disk space. However, you can select unpartitioned space. This will create a partition and make the storage pool the same size of the partition because you can have only one partition per storage pool.

1. From ConsoleOne, open the tree you want.

2. Right-click on the server object and select Properties.

3. Click Media > NSS Pools > New.

4. Enter a name for the new storage pool and click Next.

5. Select the existing NSS partitions or unpartitioned free space that you want to use for the pool.

   If you select an unpartitioned device, you can double-click on the number in this column to edit the current amount of space you want to obtain from the storage devices. The storage pool size will be the total of all devices in the pool.

6. Enter the size you want from each NSS partitioned space or unpartitioned space and click Next.

   You can select any of these options:

   • Create a Logical Volume with this same name. Automatically creates a logical volume when you create the pool.
   • Sharable for Clustering. Available if Clustering is installed and you want to use the feature.
   • Activate. Activates your pool and any logical volumes when you create the pool. Otherwise, you can activate it later.

7. Click Finish.

Creating a Logical Volume

After creating storage pools, you can create any number of logical volumes for the storage pools according to the physical space available. When you create a logical volume, you can either assign it a space or allow it to expand to the pool size.

The size of a single volume cannot exceed the size of a storage pool. However, the combined sizes of multiple volumes can exceed the size of the storage pool. This is called overbooking. For example, you might have an 800 MB storage pool with eight volumes set at 100 MB each. That is the maximum amount of space you have in the pool. To overbook the pool, you can add volumes or increase the size of existing volumes knowing that not all the volumes in the storage pool will fill up. If the volumes fill up, you need to add more disk space.

NSS recognizes DOS, Macintosh*, UNIX*, and long name spaces.

1. From ConsoleOne, open the tree you want.

2. Right-click on the server object and select Properties.

3. Click Media > NSS Logical Volumes > New.

4. Enter a name for the volume and click Next.

5. Select the storage pool where you want to store the unpartitioned space or NSS partitioned space.

   If you select unpartitioned space, NSS creates an NSS partition, then the storage pool, then the volume.

6. Do one of the following:

   • Enter a quota size for the volume if you want to limit the size of the volume.
   • Mark "Allow volume quota to grow to the pool size" if you want the volume expand to the size of the pool.

7. Click Next.

8. Select the attribute settings you want for the new logical volume from the following list:

   • Backup. Indicates if the volume should be backed up. Set this flag if the volume contains data you want to back up. Clear this flag if the volume is empty or if backing up the data is unnecessary. This backup option is separate from the third-party backup vendor you use. This means your backup system might not recognize this option whether you select it or not.
   • Compression. Activates file compression in the logical volumes.
   • Data Shredding. Activates the Data Shredding security feature. This feature scrambles any data that you delete to prevent anyone from accessing the information with a disk reader. Enter the number of times you want the data shredder to scramble your deleted files. You can select between one and seven times.
   • Directory Quotas. Activates a feature that lets you assign a maximum quota of space a directory can have
   • Flush Files Immediately. Activates the feature that flushes all data in a file to disk immediately when you close the file.
   • Modified File List (MFL). This provides a list of all files modified since the previous backup. A third-party vendor must implement the Modified File List feature in order for you to use it.
   • Salvage Files. Activates the feature that lets you salvage deleted files. The file system keeps all deleted files in an allocated space until that space is needed for other data. The Salvage feature tracks the files and lets you retrieve the data for a time until the space is needed for other data. You must select this option to salvage the files in the volume.
   • Snapshot-File Level (File Snapshot). Activates the Snapshot feature at the file level. The Snapshot feature allows the backup utility to capture a snapshot of the last closed version of a file. For example, if your system backs up or crashes while you have a file open, this feature will save a copy of the file before you opened it. You might lose some new information, but you will retain all the previous information.
   • User Space Restrictions. Activates the user space restrictions feature on the volumes you create.
   • Activate. Activates logical volumes as soon as you create them.
   • Mount. Mounts logical volumes as soon as you create them.

9. Click Finish.

You can now store and manage files in the NSS system.

Renaming a Storage Pool

You can rename existing storage pools. For example, you might want to assign a storage pool name that relates to a department name change.

1. From ConsoleOne, open the tree you want.

2. Right-click on the server object and select Properties.

3. Click Media > NSS Pools.

4. Select a pool to rename.

5. Click Rename.

6. Enter a new name for the pool.

7. Click Finish.

Renaming a Logical Volume

You can rename logical volumes. For example, you might want to change the name of a volume to reflect the department or organization that uses it.

1. From ConsoleOne, open the tree you want.

2. Right-click on the server object and select Properties.

3. Click Media > Logical Volumes.

4. Click Rename.

5. Enter a new name for the logical volume.

6. Click Finish.

Increasing the Size of a Storage Pool

You can increase the size of your storage pools, but you cannot reduce their size.

1. From ConsoleOne, open the tree you want.

2. Right-click on the server object and select Properties.

3. Click Media > NSS Pools.

4. Select the storage pool you want to increase, then click Increase Size.

5. Select the storage objects you want to obtain the space from.

6. Enter the amount of space you want from each storage object in the Used column.

7. Click Finish.

Deleting NSS Pools or Logical Volumes

You might need to delete a storage pool or a logical volume to create more free space for other storage pools. When you delete a logical volume or a traditional volume from a partition, that partition still owns the space, but you can reassign the space to another storage pool or traditional file system volume. When you delete a partition, ownership of the space in that partition is removed.

After you delete a logical or traditional volume, free space becomes available. You can assign this space to other storage pools to increase their size. You can also use the free space (if you removed NSS ownership) for a traditional file system volume. If necessary, you can restore a volume, see Reviewing and Restoring Deleted Logical Volumes. If you delete a storage pool, you cannot restore any of the volumes in that pool.

1. From ConsoleOne, open the tree you want.

2. Right-click on the server object and select Properties.

3. Click Media > NSS Pools or Logical Volumes.

4. Select the pool or volume you want and click Delete.

Reviewing and Restoring Deleted Logical Volumes

If you delete a volume, the file system removes it from the storage pool. However, for a specified amount of time, called the Purge Delay time, you can review and even restore the contents of the volume you deleted. For example, you might want to restore information in a volume that was deleted in error. You must retrieve the volume before the delay time elapses; otherwise, the volume is removed from the system and you can no longer access it.

If you delete a storage pool, you delete all the volumes in that pool. You cannot restore volumes.

The default setting for the Purge Delay time is two days. After this time expires, NSS automatically purges the volume.You can change the Purge Delay time to extend or reduce the time for the automatic purging cycle. See Changing the Purge Delay Setting. You can also manually purge deleted volumes. You purge or restore volumes in the "Deleted Logical Volumes on Pool" screen.

1. From ConsoleOne, open the tree you want.

2. Right-click on the server object and select Properties.

3. Click Media > Logical Volumes.

4. Click Deleted Volumes.

5. Select one of the following options:

   • Purge. Click Purge to immediately purge all the deleted volumes.
   • Prevent Purge/Allow Purge. Click Prevent Purge to stop the volume purging process. For example, you might want to access the deleted volume again, or wait until there is less activity on the server. After you click Pause Purge, the button changes to Allow Purge. Click this button to purge the volume.
   • Salvage. Click Salvage to restore the deleted volume.
   • Refresh. Click Refresh to rescan the volumes that have been deleted and update the list in the panel.

6. Click Close when you are finished.

Changing the Purge Delay Setting

The default for the Purge Delay setting is two days. This is the amount of time that you can still access the volume before it is removed from the system. To change the Purge Delay time, enter the following at the server console:

NSS/logicalVolumePurgeDelay=ddd

In this command, you replace ddd with the actual number of seconds to delay the auto purge. The default is 172800 (2 days). For example, if you want to change the Purge Delay time for one day, enter 86400.

The Purge Delay change command is not permanent. You must enter the command each time you restart the server. If you want to make the new setting permanent, enter the command in the AUTOEXEC.NCF file.

Creating a Traditional Volume

NSS is the default file storage and management system in NetWare 6. However, you might want to continue maintaining traditional partitions and volumes in NSS. The following is a list of guidelines for creating traditional volumes in NetWare 6:

• NWCONFIG and NSS Menu are not compatible with NetWare 6. Use ConsoleOne to create, mount, modify, and rename traditional volumes.
• You can use VREPAIR to fix traditional volumes, but you need to use VERIFY and REBUILD for all logical volumes. For instructions on using VREPAIR, see VREPAIR in the Utilities Reference.
• You cannot put traditional volumes in an NSS storage pool.
• If you create a traditional volume in NetWare 6, you cannot access that volume from previous releases of NetWare.

You can use ConsoleOne to create, mount, and manage traditional volumes.

1. From ConsoleOne, open the tree you want.

2. Right-click on the server object and select Properties.

3. Click Media > Traditional Volumes.

4. Click New.

5. Enter a name for the volume and click Next.

6. Select either partitioned or unpartitioned.

7. Enter a size for the volume in the Used column and click Next.

8. Select a block size from the pull-down menu.

9. Select any of the following options:

   • Compression. Enables the file system to compress the files in the volume. You set up file compression when you create volumes. Once you enable file compression, you cannot turn it off without creating the volume again.
   • Migration. Enables the operating system to move infrequently accessed data to remote areas on your server. This creates space for new and more commonly accessed data. Selecting this option only enables the attribute. The data migration feature uses a third-party software package that does the migration function.
   • Suballocation. Enables the file system to divide partially used disk blocks into suballocation blocks of 512 bytes. These blocks can be used by other data files.
   • Mount Volume on Creation. Instructs the operating system to mount the volume when you create it. Otherwise, you can mount it later.

10. Click Finish.

Increasing the Size of a Traditional Volume

To increase the size of a traditional volume, you need to add another segment to that volume.

1. From ConsoleOne, open the tree you want.

2. Right-click on the server object and select Properties.

3. Click Media > Traditional Volumes.

4. Select a volume and click Increase size.

5. Select the storage device you want to obtain space from.

6. Double-click on the Used column next to the selected storage device.

7. Enter the amount of space you want.

8. Click Finish.

Renaming a Traditional Volume

You can rename a traditional volume.

1. From ConsoleOne, open the tree you want.

2. Right-click on the server object and select Properties.

3. Click Media > Traditional Volumes.

4. Select the volume you want to rename and click Rename.

5. Enter a name.

6. Click Finish.

Deleting a Traditional Volume

You can delete traditional volumes.

1. From ConsoleOne, open the tree you want.

2. Right-click on the server object and select Properties.

3. Click Media > Traditional Volumes.

4. Select the volume you want to delete and click Delete.

Dismounting a Traditional Volume

To repair a traditional volume, you need to dismount that volume.

1. From ConsoleOne, open the tree you want.

2. Right-click on the server object and select Properties.

3. Click Media > Traditional Volumes.

4. Select the volume you want to dismount and click Dismount.

   The button label changes to mount.

Repairing a Traditional Volume

Typically, you can't mount a volume if it has even minor damage. Occasionally, a damaged volume mounts and causes errors in the process.

Use VREPAIR to correct volume problems or to remove name space entries from File Allocation Tables (FATs) and Directory Entry Tables (DETs). For instructions on using VREPAIR, see VREPAIR in the Utilities Reference.

NOTE:  If you need to repair an NSS volume, see Rebuilding NSS Storage Pools and Volumes.

You can run VREPAIR on a damaged volume while other volumes are mounted. Following are typical instances when VREPAIR can help:

• A hardware failure either prevented a volume from mounting or caused a disk read error.

  NOTE:  Although VREPAIR can't fix hardware problems, VREPAIR can sometimes fix related volume damage.

• A power failure caused a corrupted volume.
• The server console displays a mirroring error when the server boots. This mirroring refers to the two copies of FATs and DETs that the operating system keeps (if disks are mirrored, NetWare keeps four copies).

If a volume fails to mount as the server is booting, VREPAIR loads automatically and attempts to repair the volume.

When VREPAIR autoloads, it uses the default options. If you want to use an alternate option, load VREPAIR manually and set the alternate option before running VREPAIR.

NOTE:  If you don't want VREPAIR to automatically repair a volume that won't mount, use the SET parameter named Automatically Repair Bad Volumes to change the default.

Prerequisites

• The volume you want to repair must be dismounted.
• If the volume to be repaired has name space support, the corresponding VREPAIR name space module (V_namespace.NLM) must be located in either the SYS:SYSTEM directory or in a search path directory.

  Example modules include V_MAC.NLM and V_LONG.NLM.

Procedure

1. At the server console prompt, enter

   VREPAIR [volume_name] [logfile_name]

   (Optional) Replace volume name with the name of the volume to repair. If there is only one volume that is dismounted, you don't need to specify this parameter, since VREPAIR will attempt to repair that volume.

   (Optional) If you want to save the error log, replace the logfile_name with the name of the file you want VREPAIR to create. VREPAIR creates a log of errors it finds. VREPAIR displays the errors on screen and will write them to a file if you specify a filename.

   When you load VREPAIR, an Options menu is displayed.

2. Accept the default options, or select alternate options, as appropriate.

   The first time you try to repair a volume, accept the default options. If the default options fail to repair the volume, select alternate options.

   1. To accept the default options, continue with Step 3.

   2. To set alternate options at the Options menu, choose Set VRepair Options by typing

      2 Enter

3. To begin the repair process, choose Repair A Volume from the Options menu.

   • If more than one volume is dismounted, select the volume to repair from those listed.
   • If only one volume is dismounted, VREPAIR assumes it is the volume that needs repairing and begins the repair.

   As the volume is being repaired, the server console screen displays a message indicating VREPAIR activity.

4. (Optional) Modify error log settings after the repair has started.

   If VREPAIR finds many errors during the repair process, you might want to change some of the run-time error settings. To modify these settings after the repair has started, press F1 to display the Current Error Settings menu.

   • Select Option 1 if you do not want VREPAIR to pause after each error.
   • Select Option 2 if you want VREPAIR to log errors in a text file.
   • Select Option 3 to stop the repair of the volume.
   • Select Option 4 to continue with a volume repair after you have stopped it.

5. When the repair is complete, answer Y when prompted to write repairs to the disk.

6. If VREPAIR has found errors, run VREPAIR again by repeating Step 2 through Step 6. Repeat until VREPAIR finds no errors.

   If you are unable to mount the volume after running VREPAIR several times, you must delete the volume, re-create the volume using ConsoleOne.