Novell NetWare - Volume and Pool Maintenance Suggestions

  • 3579327
  • 25-Jun-2007
  • 27-Apr-2012


Novell NetWare 6.5
Novell NetWare 6.0
Novell NetWare 5.1


NetWare File Systems have evolved through the last several iterations of the Operating System. There exist some powerful tools to repair corruption on these systems. Unfortunately, there is no tool to perform repairs while a file system is mounted and accessible. Here are some light instructions and suggestions for reactionary repairs and preventative schedules.



The traditional file system has been discontinued and is no longer being enhanced.

There are specific procedures to follow regarding reactionary(WC) VOLUME maintenance or repair such as with a VREPAIR. However, administrators can schedule specific procedures such as a PURGE or a VREPAIR on a continuing basis as preemptive measures to prevent VOLUME or Directory Entry corruption.


To perform a VREPAIR, please do the following:

1. Make sure the volume is dismounted.

2. Type VREPAIR at the console prompt.


4. Note the Current VREPAIR Options.
Choosing any of the numbers and pressing enter will change the current options to reflect the choice.
Most administrators will choose to have the following parameters:

* Quit if a required VRepair name space support NLM is not loaded.
* Write all directory and FAT entries out to disk.
* Write changes immediately to disk.
* Purge all deleted files.

Return to main menu and continue with volume repair

5. The next screen is another that will need some negotiating.
Choose the following options unless a log file is absolutely necessary.

* Do not pause after errors
* Do not log errors to a file

6. Begin the VREPAIR. Note that there are several items that the repair will check and repair if needed:

* FAT Blocks>
* Mirror Mismatches>
* Directories>
* Files>
* Trustees>
* Deleted Files>
* Free Blocks>

In each instance, ... (dots) will increment across the screen as the repair progresses. If the repair starts to take a large amount of time or seems to be hung and isn't progressing, stop the VREPAIR by typing F1 to change the error settings. Changing these will give the administrator the option to cancel the repair and start over if they so choose. The restarted VREPAIR in almost all instances continues at a much quicker pace than the previous one that may have been hung at some point.
However, stopping a vrepair is not a recommended practice. Only stop a vrepair as a last resort.


To perform a purge on a volume without dismounting it as in a VREPAIR, try the following:


1. Load TOOLBOX -NL to load it without authenticating.

2. Type PURGE SYS: -A to purge all the files at the root and in all subdirectories.


1. Right click on the Red 'N' in the SYSTRAY.

2. Move to NetWare Utilities

3. Choose PURGE

4. Browse to the location and run the purge.
Choose to purge all and/or purge the subdirectories once the files are found.


1. Go to Remote Manager

2. Click on Volumes

3. The the information logo to the side of the desired volume

4. Find the PURGE DELETED FILES button. This will open a new window and several paths will scroll down the screen

5. When the page stops scrolling the purge is complete.


The NSS file system is a much more robust and expandable file system. It mounts more quickly and has more capability for maintenance and repair under various circumstances.

Different from the Traditional File System, NSS utilizes a complex metadirectory tree that makes it possible to expand the size of the associated POOL's and VOLUME's to very large capacities.
Note that NSS has VOLUME's, but utilizes POOL's as its primary storage entity. For more information on the differences between NSS and the Traditional File System, please see the documentation at:

NSS employs several different modes for everything from creation to repair.

MAINTENANCE MODE: the default mode for running a VERIFY or a REBUILD
ACTIVE: active does not mean mounted, but the POOL is ready to mount the volumes
DEACTIVE: not active and not able to mount volumes until in the ACTIVE state
NOTE: deactivated POOL's will NOT show any volumes within ...

Administrators should follow roughly the same type of maintenance schedule for NSS POOL's as they follow for the Traditional VOLUME's, though the monthly checks and the maintenance processes are, of course, going to differ.

If a POOL's metadata is corrupt or if an administrator suspects corruption on the POOL or on a VOLUME, the following strategy has been shown to be effective under differing circumstances:


If the POOL will not activate, REBUILD the POOL by using the following commands:

1. Deactivate the POOL and put it in MAINTENANCE mode by typing:

2. Next, REBUILD the POOL by typing:

3. It may be necessary to delete and recreate the metadata Salvage Tree ... to do this type:

The REBUILD will progress over time. REBUILD's of very corrupt file systems can take many hours. Expect to have a POOLREBUILD on a 100 GB POOL run for at least 1.5 hours if the POOL was nearly full.

Once the REBUILD is finished, the POOL may or may not revert to the ACTIVE state. If it does not, attempt to activate it by typing NSS /POOLACTIVATE=. If it will not activate, contact Novell Technical Support. If it does activate, attempt to mount it.

NOTE: The REBUILD will write a log to the DOS partition. This log will contain references to fixed problems. It will have the extension .RLF. If files have been deleted, they may show up in an FP directory at the root of the POOL. These files may or may not be recoverable.


If the POOL is active and the administrator is simply running a normal maintenance schedule, a full REBUILD is not recommended. Instead, run a VERIFY. A POOLVERIFY scans the POOL for errors and does not fix them. This is much less invasive, though the POOL has to be deactivated for this operation to take place.

If the POOL is in a normal state do the following to VERIFY the contents of the POOL:

1. Deactivate the POOL and put it in MAINTENANCE mode by typing:

2. Next, VERIFY the POOL by typing:

The VERIFY will take a shorter time than a REBUILD and is noninvasive. Note that the VERIFY will create a log just as the REBUILD did, but with a different extension. a .VLF. If the log shows warnings, this is a sign of impending corruption in the metadata, which should be followed by close watch on the POOL. If the log shows errors, the POOL should be rebuilt at the earliest convenience of the administrator.