Maintenance Pack NB_50_6_M.winnt.IA64.exe provides fixes to the VERITAS NetBackup (tm) Enterprise Server / Server 5.0 on Windows XP/2003 64 bit clients.

Article:TECH45413  |  Created: 2005-01-06  |  Updated: 2005-01-07  |  Article URL http://www.symantec.com/docs/TECH45413
Article Type
Technical Solution

Product(s)

Environment

Issue



Maintenance Pack NB_50_6_M.winnt.IA64.exe provides fixes to the VERITAS NetBackup (tm) Enterprise Server / Server 5.0 on Windows XP/2003 64 bit clients.

Solution



NB 5.0GA Pack NB_50_6_M README November 29, 2005
================================================================================
This Maintenance Pack provides fixes to the VERITAS NetBackup (tm)
Windows 64-bit clients.

Symantec recommends that the backing up of active file systems be avoided,
or the use of snapshot technologies be implemented. The directory structure
reported back from the file system (to NetBackup) may not contain all of the
files available during the time of backup. NetBackup will not report errors
in many cases where the file's existence is not known to NetBackup as reported
by the file system.



I. DOWNLOAD INSTRUCTIONS
II. KNOWN ISSUES
III. INSTALLATION INSTRUCTIONS
IV. UNINSTALL INSTRUCTIONS
V. DESCRIPTION OF PROBLEMS FIXED
Current Pack
NB_50_5_M
NB_50_4_M
NB_50_3_M
NB_50_2_M
NB_50_1_M


=========================
I. DOWNLOAD INSTRUCTIONS
=========================
This is an update to NetBackup/Media Manager products
with the latest fixes for Windows NT NetBackup servers.

Download instructions:
1) Download the NB_50_6_M.winnt.IA64_<6 digit number>.exe file
into a temporary directory.

where <6 digit number> is an internal tracking identifier

2) Extract the NB_50_6_M.winnt.IA64_<6 digit number>.exe
by double-clicking on it.

This will create a number of files that include:
README.NT
and
Setup.exe


==================
II. KNOWN ISSUES
==================
The following are known issues with NB_50_6_M. Please note that
these known issues will NOT be resolved after installing this pack.

Description:
For VSP users, a reboot will be required before the latest VSP fixes
will take effect.

--------------------------------------------------------------------------------
Description:
Anytime a backup job fails before the checkpoint is created, the resumed
backup will ignore the first file list entry, but successfully backup any
others and end with status 0 (assuming no other errors). This can result
in a data loss situation because the first file list entry does not make
it to tape. This issue is only present with UNIX Clients, using the file
directive of ALL_LOCAL_DRIVES.

--------------------------------------------------------------------------------
Description:
A backup that stops and resumes can successfully complete, but there may be
multiple duplicate fragments. This type of problem causes the import to not
be able to determine which duplicate fragment is correct. This is
technically not a data-loss issue (since the data is on tape), however the
data will be difficult to recover if the import is needed.

--------------------------------------------------------------------------------
Description:
A problem can occur with misconfigured drives or when a rogue rewind is
detected after a checkpoint has been made. In situations such as these, the
data written before the checkpoint is bad and on rare occasions, could cause
a loss of data.

--------------------------------------------------------------------------------
Description:
SQL Server advanced client images created with 5.0_MP1 cannot be imported.


===============================
III. INSTALLATION INSTRUCTIONS
===============================
NOTE: For Maintenance Pack installation on Microsoft Cluster Environment:

1) Install this Maintenance Pack on the inactive node(s) of the cluster
(follow steps 1-4 below)

2) Use the MSCS move group command to 'failover' the server to another node.
This should be done when the NetBackup system is quiet, i.e. no backups
or restores running.

3) Install this Maintenance Pack on the newly inactive node of the cluster

4) (Optional) Use the MSCS move group command to 'failover'
the server back to the original node.

--------------------------------------------------------------------------------

Installation steps:

1) Close all NetBackup Windows.
Make sure the NetBackup system has no active backups, restores,
duplications, etc., running. If there are NetBackup activities in process,
the Maintenance Pack will require a reboot if the file to update is busy.

If a database client is being used, such as Oracle, ensure that the
database services are stopped. Database services can be stopped via the
ControlPanel -> Services tool. Repeat this until all databases have been
stopped. These services must be restarted once the patch is successfully
installed.

2) Run Setup.exe (Double-click on the icon from File Manager or Explorer or run
it from the command prompt)

Setup.exe begins by stopping the appropriate NetBackup Services required for
the Maintenance Pack installation. Next, it will install the necessary files
into their correct locations. Lastly, Setup.exe will restart the appropriate
NetBackup services.

3) Examine the <TEMP>.Log file to ensure that no errors took
place during the installation.

4) Remove the temporary directory created in the download instructions.

Note: If you are installing the Maintenance Pack using the silent install script,
please review the <TEMP>.Log after the installation is finished,
a reboot maybe required to complete the installation of the Maintenance Pack. A
search of the log for the key words "in use" will indicate which files were busy
during the installation.


===========================
IV. UNINSTALL INSTRUCTIONS
===========================
**Important notice regarding un-installs on Windows**
Only the last Maintenance Pack installed on a Windows system
can be un-installed.

1) Close all NetBackup Windows.
Make sure the NetBackup server has no active backups, restores,
duplications, etc., running. If there are NetBackup activities in process,
the Maintenance Pack will not install if the file to update is busy.

If a database client is being used, such as Oracle, ensure that the
database services are stopped. Database services can be stopped via the
ControlPanel -> Services tool. Repeat this until all databases have been
stopped. These services must be restarted once the patch is successfully
installed.

2) Go to Add/Remove programs dialog box and select correct Pack to be
uninstalled.


=================================
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 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, Symantec 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
=============

================================================================================
Etrack Incident = ET412698

Description:
Special files (Such as ntuser.dat, ntuser.dat.log, and so on) could not
Be backed up when the backup job was started and the user logged off the
system. Now those files are backed up even if the user logs off after the
backup job is started.
================================================================================

=========
NB_50_5_M
=========
================================================================================
Etrack Incident = ET300775

Description:
Using the Java Console to restore UNC formatted images on the network to
any Windows master/media server, the restore operation would fail. The
directory/file names could not be found.

(NetBackup Clients: Intel-NT Intel64 )
================================================================================


=========
NB_50_4_M
=========
Description:
Restore job showed incomplete status (exit status 5) after a successful
client-initiated restore.

(NetBackup Clients: Intel-NT Intel64 )
--------------------------------------------------------------------------------
Description:
When a database agent gets queued multiple times in a very short period
of time, there were cases where the server would time out waiting for the
client to connect. This fix prevents the client from lagging behind the
server in these instances.

(NetBackup Clients: ALPHA HP Intel-NT Intel64 Linux2.4 RS6000_433 RS6000_51 Solaris )
--------------------------------------------------------------------------------
Description:
When performing an alternate client restore of RAW devices with the SAP
Agent, the restore would fail. The following errors were reported in
brrestore:

BR386E File /dev/vgsapiBWP/rBWPclui_1 reported as not found by backup
utility

The following information was reported in the backint logs if logging
is ON and VERBOSE=5:

11:13:28.129 [20169] <4> backint_find_master_time: WARNING: not found in
bplist list: /dev/rdsk/c1t4d0s1
11:13:28.129 [20169] <2> backint_find_master_time: Display inquire_info
structure
11:13:28.129 [20169] <2> backint_find_master_time: file_path_name =
>/dev/vgsapiBWP/rBWPclui_1<
11:13:28.129 [20169] <2> backint_find_master_time: actual_file_path =
>/dev/vgsapiBWP/rBWPclui_1<
11:13:28.130 [20169] <2> backint_find_master_time: bid =
>NS1096040689< dest_dir = >/dev/vgsapiBWC/rBWCclui_1<
11:13:28.130 [20169] <2> backint_find_master_time: start_unix_time =
1096040689 ; end_unix_time = 1096040689 ; image_time = 0
11:13:28.130 [20169] <2> backint_find_master_time: found = 0 ; raw_par
= 0 ; time_pos = 0 file_size = 0
11:13:28.130 [20169] <2> backint_find_master_time: exit status = 2
11:13:28.131 [20169] <4> backint_do_restore:
#NOTFOUND /dev/vgsapiBWP/rBWPclui_1

(NetBackup Clients: ALPHA HP Intel64 RS6000_433 Solaris )
--------------------------------------------------------------------------------
Description:
A problem was discovered when the following conditions were met:
1. Running SAP archiving in continuous mode, brarchive -f.
2. The policy_log and sched_log parameters are enabled in NetBackup for SAP
agent configuration file. (init<SID>.utl)

The policy_log and sched_log parameters enable NetBackup for SAP to perform
an additional backup of the archive logs. Having two backups of the archive
logs (usually to different media) is an additional measure to secure the
integrity of archive logs.

In situations where the brarchive command is run in continuous mode,
NetBackup for SAP performs two backups only for the initial set of archive
logs. All following sets of archive logs are backed up only once.

(NetBackup Clients: ALPHA HP Intel-NT Intel64 RS6000_433 RS6000_51 Solaris )




=========
NB_50_3_M
=========
Description:
When running on Windows Itanium 64-bit platforms, bpbkar32.exe may crash
with an alignment fault and generate a Dr Watson log when using true image
restore with move detection. The alignment fault is fixed.

(NetBackup Clients: Intel64 )
--------------------------------------------------------------------------------
Description:
The new keep-alive code is more efficient and prevents the server from
receiving a network timeout when the backup job is being resumed.

(NetBackup Clients: Intel-NT Intel64 )
--------------------------------------------------------------------------------
Description:
bpbkar32.exe no longer crashes during the restart of an incremental backup.

(NetBackup Clients: Intel-NT Intel64 )
--------------------------------------------------------------------------------
Description:
The command "backint -p <utl file> -f inquire" can fail.

When called without an input file list, the SAP Agent will fail.
If a user runs "backint -p <utl file> -f inquire", the reported error is:
"ERROR: couldn't open file: <file name>"

To avoid this issue, create an input file with file list and pass it with
a -i option. The following example can be used:
"backint -p <utl file> -f inquire -i <input file list path>

(NetBackup Clients: ALPHA HP Intel-NT Intel64 RS6000_433 RS6000_51 Solaris )
--------------------------------------------------------------------------------
Description:
SAP agent incorrectly sorts backup files by size on VxFS.

This problem appears only when backing up VxFS Quick I/O files on UNIX.

With the SAP database extension, VxFS Quick I/O file sizes are
incorrectly evaluated to a size of zero bytes. This creates a problem
when sorting backup streams by size. Therefore, the streams (drives)
are not equally loaded, which causes the backup to run longer than
expected.

(NetBackup Clients: ALPHA HP Intel-NT Intel64 RS6000_433 RS6000_51 Solaris )
--------------------------------------------------------------------------------
Description:
The SAP database (DB) environment agent reports a successful backup
despite an actual failure occurring.

This problem only occurs when the agent is called in SAP DB environment
with an adapter program. (This problem is not relevant for SAP on Oracle.)
The problem occurs if there were previously successful backups and current
backup just failed. The log of the failed backup would show input
files as saved, however, these files were saved from a previous successful
backup. This could mislead the user into thinking a failed backup was
successful when it was not.

(NetBackup Clients: ALPHA HP Intel-NT Intel64 RS6000_433 RS6000_51 Solaris )
--------------------------------------------------------------------------------
Description:
Archive bits were cleared for an entire incremental backup job when the job
was resumed and completed successfully.

(NetBackup Clients: Intel-NT Intel64 )
--------------------------------------------------------------------------------
Description:
The SAP Agent is unable to read output file of restore_filter script. This
failure occurs intermittently and produces a log similar to the following:

"14:30:00.381 [7082] <16> backint_restore_filter: cannot open restore
working file:
/oracle/BD1/sapbackup/backint_dir/.backint.filter.1085772595.7082
14:30:00.381 [7082] <16> backint_do_restore: ERROR: running restore filter"

In addition, the restore_filter script was unable to handle a use case
where the source client (a backed-up client) did not have soft links to the
actual file paths of the SAP data files. During a restore of a target
client (a new destination client), the SAP agent did not adjust the file
paths to match the originally backed up paths. The following is a sample
output of this type of problem:

************************************************************************
Program: ./backint 5.0GA
Input File: ./in_re
Profile: ./initSAP.utl
Function: RESTORE
*************************************************************************
#NOTFOUND <input file name>
*************************************************************************

Also, in the backint log, the following is recorded:

"10:56:07.093 [2838] <2> backint_find_master_time: exit status = 2
10:56:07.094 [2838] <4> backint_do_restore: #NOTFOUND /tmp/bd1/file1"


Workaround:
In case where restore_filter fails intermitently, try repeating restore
operation.

(NetBackup Clients: ALPHA HP Intel-NT Intel64 RS6000_433 Solaris )
================================================================================


=========
NB_50_2_M
=========
Description:
When running on Windows Itanium, 64-bit platforms, bpbkar32.exe may crash
with an alignment fault and generate a Dr Watson log when using true image
restore with move detection. This alignment fault issue has been fixed.

(NetBackup Clients: intel64 )
--------------------------------------------------------------------------------
Description:
An issue existed that caused bpbkar32.exe to crash during the restart of
incremental backups.

(NetBackup Clients: Intel-NT intel64 )
================================================================================


=========
NB_50_1_M
=========
Description:
When resuming a suspended backup job with Windows Open File Backup enabled,
the resumed backup will not re-use the original backup job's volume
snapshots. Instead, if the WOFB Error Control configuration setting was set
to Abort on Error, the resumed backup job will abort and a status 156-
snapshot error will be encountered. If the WOFB Error Control configuration
setting was set to Disable Snapshots & Continue, the resumed backup job
will be resumed without any volume snapshots. This behavior is not correct
and the resumed backup job should re-use the original, suspended backup
job's snapshot volumes if they were successfully created.

(NetBackup Clients: Intel-NT intel64 )
--------------------------------------------------------------------------------
Description:
Added support to delete VSP snapshots (cach files) when WOFB is enabled,
this prevents deleting VSP snapshots for XBSA backups.

(NetBackup Clients: Intel-NT intel64 )
--------------------------------------------------------------------------------
Description:
If the checkpoint name is in the DOS style, now it gets converted to
the UNIX style.

(NetBackup Clients: Intel-NT intel64 )
================================================================================

Attachments

NB_50_6_M_280772.winnt.IA64.exe (15.6 MBytes)


Legacy ID



280772


Article URL http://www.symantec.com/docs/TECH45413


Terms of use for this information are found in Legal Notices