In-depth Troubleshooting Guide for Exit Status Code 50

Problem

In-depth Troubleshooting Guide for Exit Status Code 50 in NetBackup.

Solution

See the flowchart in the Word document link, below.

1. Resource/System level issues

It is often necessary to examine a Status Code 50 event from a Non-NetBackup perspective.  Often, an OS or environmental event is a factor in the Status Code 50.  Examine the following non-NetBackup related information on all hosts involved with the failure:

1. Available Memory on the system.
2. Available Hard Disk space on the system.
3. OS System Logs

     
Solaris: /var/adm/messages

HP-UX: /var/adm/syslog/syslog.log

AIX: The syslog log is not turned on by default.  Check /etc/syslog.conf file for location of *.debug and *.error log file.  If it is commented out (if there is a # in front of *.debug and *.error), then syslog is not recording errors at the OS level. 

The AIX errpt command can also be used to look for hardware errors: 
# errpt -a 

Linux: /var/log/syslog
Windows:   View the Event Viewer Application and System logs for more details on the failure.
 

2. Terminated Imports (Windows)

An import of an image has the potential to take a long time. This is based on hardware performance, multiplexing and whether the image spans multiple media.  In a Windows environment, launching either a phase 1 or phase 2 import causes the bpimport.exe process to spawn, using the credentials of the currently logged on user.  This means that if the currently logged on user logs out prior to the completion of the import job, bpimport.exe will terminate.  This will generate a Status Code 50 for the import job in the NetBackup Activity Monitor.  This is also the case if a Terminal Services or Remote Desktop client was used to log in to the Master Server and initiate the import.  If the Terminal Server session is ended, the active bpimport.exe process will terminate also.  Therefore, the user that initiates the import must stay logged in to ensure successful import completion.

3. Database extension backups and Status Code 50:

3.1  Lotus Notes (Linux)
A Status 50 error can be seen with Domino Server 6.0.2 and later on a Linux Platform.  Lotus Support has acknowledged a new variable in the Lotus C API called BACKUP_TIMEOUT. When the backup of a (usually large) database exceeds the default BACKUP_TIMEOUT setting of 15 minutes, the API will return a Status 50 and log the following in the bpbkar log:

nbex_DebugLog: ERR - NBLN_ChgInfoBackupOpen() NSFBackupGetChangeInfoSize() Failed:0x3810:Backup terminated due to timeout

tar_base::V_vTarMsgW: ERR - failure reading file: Lotus Notes:\C:\[path_to_Domino_Data_Dir]\[database_name].nsf (WIN32 1: Incorrect function.)

The timeout setting may be adjusted to a value that is suitable for systems with very large databases by adding the following variable to the Domino server's notes.ini file:

BACKUP_TIMEOUT=[number of minutes allowed for backup]

Alternatively, the timeout setting can be disabled altogether by adding the following variable to the Domino server's notes.ini file:

BACKUP_NO_TIMEOUT=1

The Domino server must reread the notes.ini file (usually done by restarting the associated Domino partition) to activate the new settings.

3.2  All other Veritas Netbackup Database Agents
The Status 50 error is not common for other database agents.  If this issue occurs with other agents, please contact VERITAS support for further details in attempting to diagnose the issue.