Maintenance Pack NB_SMU_50_5_M.hp_11.tar provides fixes to the VERITAS Storage Migrator (tm) for UNIX.

Article:TECH38296  |  Created: 2005-01-12  |  Updated: 2006-01-14  |  Article URL http://www.symantec.com/docs/TECH38296
Article Type
Technical Solution


Environment

Issue



Maintenance Pack NB_SMU_50_5_M.hp_11.tar provides fixes to the VERITAS Storage Migrator (tm) for UNIX.

Solution



 SMU 5.0GA Pack NB_SMU_50_5_M README                               April 12, 2005
================================================================================
This Maintenance Pack provides fixes to the VERITAS Storage Migrator (VSM) for UNIX platforms.


=================
PACK DEPENDENCIES
=================

    -- Installation of this Maintenance Pack requires version 1.17.4.25 of  
       the Vrts_pack.install.

    -- NetBackup Maintenance Pack NB_50_5_M must be installed prior to  
       installing this Maintenance Pack.


I.   DOWNLOAD INSTRUCTIONS
II.  INSTALLATION INSTRUCTIONS
III. UNINSTALL INSTRUCTIONS
IV.  DESCRIPTIONS OF PROBLEMS FIXED
       Current Pack  
       NB_SMU_50_4_M
       NB_SMU_50_3_M
       NB_SMU_50_1_M


=========================
I. DOWNLOAD INSTRUCTIONS
=========================
1. Download NB_SMU_50_5_M_<6 digit number>.<platform>.tar into the /tmp  
  directory.

    where <6 digit number> is an internal tracking identifier

    where <platform> is hp_11, sgi_dm, or solaris_dm

2. untar NB_SMU_50_5_M_<6 digit number>.<platform>.tar

       /bin/tar xvf NB_SMU_50_5_M_<6 digit number>.<platform>.tar

   This will create the files:
       Vrts_pack.install
       VrtsNB_SMU_50_5_M.README
       VrtsNB_SMU_50_5_M.<platform>.tar.Z


==============================
II. INSTALLATION INSTRUCTIONS
==============================
As root on the Storage Migrator server:

1. Stop the VSM daemons and all VSM activity by running:  

  /usr/openv/hsm/bin/migVSMshutdown

2. Install Maintenance Pack binaries.

  cd /tmp
  /bin/sh Vrts_pack.install

3. Restart daemons

  /usr/openv/hsm/bin/migVSMstartup


============================
III. UNINSTALL INSTRUCTIONS
============================
1. Stop the VSM daemons and all VSM activity by running:  

  /usr/openv/hsm/bin/migVSMshutdown

2. Change directory to the patch save directory.  
  Substitute the pack name for $PACK in the following command:

  cd /usr/openv/pack/$PACK/save

3. Run the uninstall script:

  ./Vrts_pack.uninstall

4. Verify that the pack uninstalled successfully by checking  

  /usr/openv/pack/pack.history

5. Restart daemons

  /usr/openv/hsm/bin/migVSMstartup


==================================
IV. DESCRIPTION OF PROBLEMS FIXED
==================================
The following are descriptions of the problems fixed.
Please read the entire document before installing.

README Conventions:

Description
    Describes particular problem fix or feature contained in this  
    Maintenance Pack.

** Description **  
    Describes a problem that can lead to potential data loss. Please  
    read these problem descriptions carefully.

Workaround
    Any available workarounds to a problem are also listed. Workarounds  
    can be used INSTEAD of applying the patch, however, VERITAS strongly  
    recommends the "best practice" of being at the latest patch level.

Additional Notes  
    Any additional information regarding this problem or feature is included.


=============  
Current pack  
=============  

================================================================================
Description:  
   After a system crash, the VOLDB may not reflect recently copied files that  
   the FHDB records as having been successfully copied.  The VOLDB record  
   changes may not have been synchronized to the disk yet.  

   (All VSM Servers)
--------------------------------------------------------------------------------
Description:  
   Using the Java GUI to delete a hierarchy sometimes caused migrd
   to dump core.  

   (All VSM Servers)
--------------------------------------------------------------------------------
Description:  
   migdbcheck sometimes terminated with a SEGV and core dump.  

   (All VSM Servers)
--------------------------------------------------------------------------------
Description:  
   migrd could hang in a CPU loop while deleting hierarchies that had
   scheduler items.  

Workaround:  
   Prior to deleting hierarchy in the VSM GUI, use the GUI's scheduler
   dialog to delete scheduler items.  

   (All VSM Servers)
--------------------------------------------------------------------------------
Description:  
   migrd could hang waiting for the completion of an fsclustadm command.  

   (All VSM Servers)
================================================================================

=============
NB_SMU_50_4_M
=============
Description:  
   migstage would unnecessarily cache from secondary media if there were no  
   "dk" entry in the FHDB for a migrated file.  

   (All VSM Servers)
--------------------------------------------------------------------------------
Description:  
   VSM-Java did not recognize the correct selected file name after column
   sorting had been used to change the order of file names in the File Browser.  

   (All VSM Servers)
--------------------------------------------------------------------------------
Description:  
   migmerge fails to back up a large FHDB, instead logging perror 72,
   "value too large to be stored in datatype".  

   (All VSM Servers)
--------------------------------------------------------------------------------
Description:  
   VSM did unnecessary repeated unmounts and mounts (one for each granule of  
   data) of secondary media when the primary media copy was unavailable.  

Workaround:  
   If there are continuing problems with a given VSM copy level (such as  
   copy 1), use the LEVEL_PREFERRED file to force VSM to begin with a different
   copy.  
   
   Example:  Assume the database directory for the HSM named "music" is  
   /usr/openv/hsm/database/music/database.  Assume that mounts from level 1  
   are failing and that we want to always try level 2 first.
   
   Then the command will make copy 2 the preferred cache level for this HSM.
   
   echo 2 > /usr/openv/hsm/database/music/database/music.LEVEL_PREFERRED  

   (All VSM Servers)
--------------------------------------------------------------------------------
Description:  
   VCS Agents that are built with the agent framework can now run with
   SFCFS 4.0MP1 systems.  

   (All NetBackup Servers: Solaris DMAPI)
--------------------------------------------------------------------------------
Description:  
   In attempting to filter out file system names that are not appropriate for
   VSM management (for example, /var), VSM-Java filtered out some file system
   names that it should not have filtered (such as: /variety).  

   (All VSM Servers)



=============  
NB_SMU_50_3_M
=============  
Description:  
   migVSMshutdown did not consistently terminate all VSM and VSM Java
   processes correctly.  

   (All VSM Servers)
--------------------------------------------------------------------------------
Description:  
   Under certain circumstances the scheduler can erroneously reverse the order
   of the next two scheduled jobs, causing some scheduled jobs to be run later
   than intended.  

   (All VSM Servers)
--------------------------------------------------------------------------------
Description:  
   The migcopy executable delivered for HP-UX was RISC2.0 instead of RISC1.1,
   which caused a problem on some older HP hardware.  

   (All VSM Servers)
--------------------------------------------------------------------------------
Description:  
   When migcopy is called to recall (cache) a file, it can select a granule  
   on a volume that is marked corrupt.  That volume will be mounted, and then  
   the corrupt flag will be detected.  The corrupt volume will be unmounted  
   and an alternative volume (if one exists) will be selected for the reload.  

   (All VSM Servers)
--------------------------------------------------------------------------------
Description:  
   When the VSM File Browser has directory names in the left window which
   embed whitespace into the directory names, clicking these names would
   not bring up directory contents to the right window.  

   (All VSM Servers)
--------------------------------------------------------------------------------
Description:  
   miglogscan could core dump if a caching operation was in progress at the
   time that the log files were restarted.  

   (All VSM Servers)
--------------------------------------------------------------------------------
Description:  
   When a Ctrl-C function is used to interrupt a  migstage -w  command, the
   files being staged could be left in inconsistent states.  In particular, the
   data blocks were often on-line, using up file system disk space even though
   the file VSM state was still "purged." The file cannot be purged because
   VSM thinks it is already purged.
   
   In addition, this inconsistent state is not detected by migdbcheck.  

Workaround:  
   The VERITAS Support group has a script and a TechNote that identifies files
   in the inconsistent state.  These files have to be manually cached and
   then they can be purged.  

   (All VSM Servers)
================================================================================


=============
NB_SMU_50_1_M
=============
Description:  
   migrd could hang if VxFS is not version 4.0.  

   (All VSM Servers)
--------------------------------------------------------------------------------
Description:  
   migdbcheck would sometimes dump core due to a BUS error.  

   (All VSM Servers)
--------------------------------------------------------------------------------
Description:  
   There was no way to turn off migstop quota checking for non-root users.
   Now a value of -1 (bytes) will be recognized as meaning "no quota."

Workaround:  
   Use a global migstop file.  

   (All VSM Servers)
--------------------------------------------------------------------------------
Description:  
   migdbrpt sometimes reported an incorrect usage percentage for optical
   media.  

   (All VSM Servers)
--------------------------------------------------------------------------------
Description:  
   mignospace processing did not always make space available to the purge
   mark even though it was possible to do so.  This fix will cause migsweep
   to run if migpsweep released some space but not enough to reach the purge
   mark.  

   (All VSM Servers)
--------------------------------------------------------------------------------
Description:  
   The migrd command core dumps when trying to reload its configuration from
   disk because too many files are open.  

   (All VSM Servers)
--------------------------------------------------------------------------------
Description:  
   When the command migdbcheck was used with the '-p' option to update
   FNDB pathnames, the command also repaired other FHDB and filesystem  
   inconsistencies without asking whether the repairs should be performed.  

   (All VSM Servers)
--------------------------------------------------------------------------------
Description:  
   The migmdclean command (clean VSM migration media) failed because the list  
   of granules associated with the volume was larger than the amount of memory
   available.  

   (All VSM Servers)
================================================================================




Attachments

NB_SMU_50_5_M_276305.hp_11.tar (8.9 MBytes)


Legacy ID



276305


Article URL http://www.symantec.com/docs/TECH38296


Terms of use for this information are found in Legal Notices