DMP 5.0GA Pack NB_DMP_50_7_M README                             December 5, 2006
Requirement: NB_CLT_50_7_M
================================================================================
** All 5.0 MP7 Maintenance Packs for NetBackup server, client, all Add-on  
products, Database Agents, and Java GUI must be applied so that the entire  
system is at the same Maintenance Pack level. **

This Maintenance Pack should only be applied to Veritas NetBackup (tm) UNIX  
servers running NetBackup for Network Data Management Protocol (NDMP).

================================================================================


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

      -- Installation of this Maintenance Pack requires version 1.17.4.40 of  
         the Vrts_pack.install.

      -- NB_CLT_50_7_M Maintenance Pack must be installed prior to  
         installing this Maintenance Pack.  


I.   DOWNLOAD INSTRUCTIONS
II.  KNOWN ISSUES
III. INSTALLATION INSTRUCTIONS
IV.  UNINSTALL INSTRUCTIONS
V.   DESCRIPTION OF PROBLEMS FIXED
       Current Pack
          NB_DMP_50_7_M

       Pack History
          NB_DMP_50_6_M
          NB_DMP_50_5_M
          NB_DMP_50_4_M
          NB_DMP_50_1_M

=========================
I. DOWNLOAD INSTRUCTIONS
=========================
1) Download NB_DMP_50_7_M_<6 digit number>.<platform>.tar into the  
  /tmp directory.

   where <6 digit number> is an internal tracking identifier

   where <platform> is:  hp_ux, linux, rs6000, solaris

2) Extract NB_DMP_50_7_M_<6 digit number>.<platform>.tar
       /bin/tar xvf NB_DMP_50_7_M_<6 digit number>.<platform>.tar

       This will create the files
       Vrts_pack.install
       VrtsNB_DMP_50_7_M.README
       VrtsNB_DMP_50_7_M.<platform>.tar.Z



=================
II. KNOWN ISSUES
=================
The following known issue exists in this pack.

- Duplicating a backup image using NDMP tape drives may fail if the new image
 spans tapes.  The bptm log will contain the following message:

    ndmp_mover_continue failed, status = NDMP_PRECONDITION_ERR
 
 To avoid this issue, use a non-NDMP tape drive on the write side of the
 duplication, that is, specify a non-NDMP target storage unit for the duplication.
 Or, call Symantec technical support to request an emergency engineering binary
 to fix this problem.




==============================
III. INSTALLATION INSTRUCTIONS
==============================
UNIX:
As root on the NetBackup Server with NetBackup for NDMP:
1) Install Maintenance Pack binaries.

   cd /tmp
  /bin/sh Vrts_pack.install


============================
IV. UNINSTALL INSTRUCTIONS
============================
Note:  This will ONLY uninstall the Maintenance Pack from your local machine.

   1) Close the NetBackup user interfaces.

       Make sure the NetBackup server has no active jobs running (for
       example, backups, restores, or duplications).

       If a database agent is being used, such as Oracle,  
       ensure that the database services are stopped.  

   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 un-install script:

         ./Vrts_pack.uninstall

   4) Verify that the pack uninstalled successfully by checking  
       /usr/openv/pack/pack.history.

   5) If update_clients was run after the pack was originally INSTALLED,  
       run it again after that pack is successfully UNINSTALLED.

   6) If necessary, restart the NetBackup and Media Manager daemons:
       /usr/openv/netbackup/bin/goodies/netbackup start



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

README Conventions:

Description
    Describes a particular problem contained in this 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, Symantec strongly  
    recommends that you are at the latest patch level.

Additional Notes  
    Any additional information regarding a problem is included.



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

================================================================================
Etrack Incident = ET533118

Description:  
   NDMP restore of different files from different backup directories may fail.
   The restore would fail with a status 5:
   
    "The restore failed to recover the requested files (5)"
   
   However, the files are restored properly. The failure was caused by an
   internal miscalculation of backup image lengths.  

Workaround:  
   After verifying that the files were restored properly, you can ignore
   the error.    
--------------------------------------------------------------------------------
Etrack Incident = ET577815

Description:  
   The selected host did not have the BPRD service operating. (Error 25)
   
   This problem occurred on Windows platforms. The handle count for the
   tldcd.exe process was very high (several thousand) and resulted in  
   a lack of resources that caused other processes, such as bprd, to fail.  

Workaround:  
   To avoid this issue, stop and start the Device Manager Services to begin
   a new tldcd process.  The old tldcd process will shutdown and release all
   of its unclosed handles.  
--------------------------------------------------------------------------------
Etrack Incident = ET579377

Description:  
   The verification of a multiplexed, non-NDMP backup image on NDMP drives
   failed.  The progress log contained the following:
   
   INF - from host kazaam.min.veritas.com, ERR - Unexpected EOF reading image
   The image information is not complete.
   
   This failure was intermittent. Some images could be verified successfully
   and others failed.  

Workaround:  
   To avoid this issue, add the following line to the file
   <installdir>/netbackup/db/config/ndmp.cfg
   
   NDMP_MOVER_CLIENT_DISABLE
   
   Note: The above work-around resulted in slower performance for all  
   duplicate, verify, and import operations from NDMP drives.  
   
   A preferred workaround is to read all multiplexed images from non-NDMP  
   Drives, if possible.  
--------------------------------------------------------------------------------
Etrack Incident = ET575347 ET609338 ET609487 ET609540 ET608376 ET609623 ET614668
ET612687 ET616976 ET617837 ET616940 ET622225 ET624432 ET633297 ET641183 ET699951
ET701953 ET781198 ET597210 ET598796 ET575338 ET596635 ET612761 ET787052 ET788824
ET786896 ET814652 ET575351 ET575352 ET575097 ET575330 ET575331 ET575345 ET575337
ET575335 ET577778 ET575333 ET596395 ET575334 ET575313 ET583770 ET575353 ET581028
ET575346 ET575329 ET575343 ET575308 ET575732 ET575309 ET575341 ET577306 ET575350
ET596637 ET596130 ET597780 ET601849 ET596188 ET596645 ET596654 ET596697 ET597281
ET597219 ET597298 ET596342 ET597379 ET598052 ET598865 ET599823 ET598813 ET600384
ET601396 ET601454 ET601344 ET600735 ET602205 ET600844 ET602349 ET603659 ET605285
ET602850 ET575311 ET607666 ET608687    

Description:  
   Buffer overflow areas have been identified and corrected in daemons running
   on Veritas NetBackup master, media, and client servers.  Successful access
   to a vulnerable Veritas NetBackup server and the ability to successfully
   execute arbitrary code could potentially result in unauthorized access with
   elevated privilege on a system.  
================================================================================



============
Pack History
============

=============
NB_DMP_50_6_M
=============
Etrack Incident = ET416208 ET414212

Description:  
   A condition existed that resulted in excessive hosts_equal calls in the
   bptm log.  
--------------------------------------------------------------------------------
Etrack Incident = ET309015

Description:  
   An issue existed that caused backups to fail when writing tir.  The  
   bptm log contained the following information.
   
   15:46:57.554 [29981] <2> setup_tir_read: successfully connected to bpdbm to
   read TIR file, size is 577960 bytes + 0 GB
   15:46:57.554 [29981] <2> io_write_back_header: drive index 12,
   grzfs03_nafsp02_vol0_1108477090, file num = 50, mpx_headers = 0, copy 1
   15:46:57.658 [29981] <2> io_write_block: ndmp_tape_write_func returned 1024
   15:46:57.758 [29981] <2> NdmpMoverClient: ndmp_mover_set_record_size(64512)
   15:46:57.858 [29981] <2> NdmpMoverClient: ndmp_mover_set_window(0,
   18446744073709551615)
   15:46:58.058 [29981] <2> NdmpMoverClient: ndmp_mover_listen returned addr
   10.200.12.10 port 12091
   ------------
   15:50:42.667 [29981] <16> NdmpMoverClient: ERROR ConnectSocket - connect
   failed: 145 Connection timed out
   15:50:42.667 [29981] <16> NdmpMoverClient: ERROR ConnectSocket failed
   15:50:42.668 [29981] <16> NdmpMoverClient: ERROR Start failed
   15:50:42.947 [29981] <2> NdmpMoverClient: mover halted reason
   NDMP_MOVER_HALT_ABORTED  

Additional Notes:  
   To disable NdmpMoverClient in netbackup/db/config/ndmp.cfg, add the
   string, NDMP_MOVER_CLIENT_DISABLE.  
================================================================================


=============
NB_DMP_50_5_M
=============
================================================================================
Etrack Incident = ET271314

Description:  
   Using the encryption algorithm, blowfish, on Solaris 9 platforms caused
   set_ndmp_attr of ndmpmoveragent to fail.  The following data is an example
   of this failure.
   
   [root@galvatron]:/:>set_ndmp_attr -verify
   Verify Host name: galvatron
   Connecting to host "galvatron" as user "root"...
   Waiting for connect notification message...
   Opening session--attempting with NDMP protocol version 2...
   Opening session--successful with NDMP protocol version 2
     host supports TEXT authentication
   Logging in using TEXT method...
   17:44:53.152 [03044] ndmp_net_read: connection 0x3e0d0: (5) end of file
   detected
   17:44:53.153 [03044] ndmp_message_receive: xdr_ndmp_header failed, unable  
   to
   XDR decode  
   ndmp_connect_client_auth failed
   set_ndmp_attr: host "galvatron" failed
   NDMP failed to verify host
     

   (All NetBackup Servers: HP Intel-NT Linux RS6000 Solaris)
--------------------------------------------------------------------------------
Etrack Incident = ET266316

Description:  
   The "set_ndmp_attr -verify server" for remote NDMP would fail, if the
   server's root/administrator password was more than eight characters long.  

   (All NetBackup Servers: HP Intel-NT Linux RS6000 Solaris)
--------------------------------------------------------------------------------
Etrack Incident = ET266303

Description:  
   For NDMP restores on NetApp filers, more than 1024 files can be restored  
   using DAR.
     
   (All NetBackup Servers: HP Intel-NT Linux RS6000 Solaris)
--------------------------------------------------------------------------------
Etrack Incident = ET303121

Description:  
   The NetBackup bptm process could core dump if NDMP processes file history  
   with inodes with the 32nd bit set as code paths used signed integets for  
   this value, making the inode appear negative.  

Additional Notes:  
   This problem was encountered while running a third party file system, the  
   SGI
   open source XFS file system. This file system actually uses 64-bit inodes
   however in this instance the NDMP file history was altered which caused the
   32nd bit to be set.  

   (All NetBackup Servers: HP Intel-NT Linux RS6000 Solaris)
================================================================================



=============
NB_DMP_50_4_M
=============
Description:  
   An NDMP backup in the ASCII catalog format would fail and produce the  
   following error in the bptm log:
       
   08:42:17.202 [5704.5624] <2> write_data_ndmp: calling  
   ndmp_bpfsmap_create
   08:42:18.304 [5704.5624] <2> write_data_ndmp: bpfsmap longjmp: setjmp  
   returned 6; aborting backup
   ...
   08:42:18.764 [5704.5624] <2> write_data: write_data_ndmp() returned -8,  
   EXIT_Status 6, Kbytes = 1359099
   08:42:18.774 [5704.5624] <2> write_backup: write_data() returned,  
   exit_status = 6, CINDEX = 0, TWIN_INDEX = 0, backup_status = -8
     
   (All NetBackup Servers: HP Intel-NT Linux RS6000 Solaris)
--------------------------------------------------------------------------------
Description:  
   When attempting to restore from a Hitachi Lightning NAS Blade 9900V image  
   that was backed up with the file list "set files=dir1 dir2 dir3" directive  
   specifying multiple files, the restore fails with status 5.
     
   (All NetBackup Servers: HP Intel-NT Linux RS6000 Solaris)
--------------------------------------------------------------------------------
Description:  
   The following log entry in the bptm log was confusing (in this patch, the  
   word DEBUG has been removed):  
       
   09:37:46.035 [16536] <4> ndmp_internal_open_and_auth: NDMP server DEBUG
   suggests protocol version 3
     
   (All NetBackup Servers: HP Intel-NT Linux RS6000 Solaris)
--------------------------------------------------------------------------------



=============  
NB_DMP_50_1_M
=============  
Description:  
   NDMP backup fails writing TIR data.  
   
   bptm log shows:
   19:56:36.854 [2969] <16> write_data_tir: ERR - Cannot write TIR data to
   media, NDMP return code 9.
   
   ndmp logs shows:
   
   19:56:36.801 [02969] 162988 ->  (8) 267 [0] 19:56:36 NDMP_TAPE_WRITE
   (0x304) NDMP_NO_ERR (0x0)
   19:56:36.801 [02969] 162988 .   data_out(0)=
   19:56:36.801 [02969] 162988 **  End NDMP_TAPE_WRITE
   19:56:36.850 [02969] 162988 <-  (8) 599 [267] 19:57:51 NDMP_TAPE_WRITE
   (0x304) NDMP_NO_ERR (0x0)
   19:56:36.850 [02969] 162988 .   error=NDMP_ILLEGAL_ARGS_ERR (0x9)
   19:56:36.850 [02969] 162988 .   count=0 (0x0)
   19:56:36.850 [02969] 162988 **  End NDMP_TAPE_WRITE
   
   (All NetBackup Servers: HP Intel-NT Linux RS6000 Solaris)
================================================================================