Send updated Symphony file to an FTA with a corrupt Symphony file
How can you send a current Symphony file from a UNIX Tivoli Workload Scheduler (TWS) master to a Fault Tolerant Agent (FTA) that has a corrupted Symphony file?
The script sym_update.sh and procedures exist that facilitate sending an updated Symphony file to an FTA from a UNIX master.
The sym_update.sh utility script is designed to run from the Tivoli Workload Scheduler Version 8.2.0 or greater workstation with the following UNIX operating systems:
The sym_update.sh is used generate an updated Sinfonia file that may be sent to a TWS agent that has a compromised Symphony file. The sym_update.sh utility script may be run from the UNIX prompt on either the TWS MDM or DM workstation.
The script may exist in any directory. The sym_update.sh be executed on DM or MDM by the MDM or DM TWS user and must have execute permissions. This requirement is necessary since TWS operations performed by the script require TWS user access.
The sym_update.sh is not designed to be run as a defined job. If it is launched via a defined job
Attention:The sym_update.sh is not designed to be run as a defined job. The script may complete successfully but the job may show an incorrect job state.
The sym_update.sh must run after the agent with compromised Symphony file has been stopped, and the Sinfonia and Symphony have been renamed and a TWSHome/StartUp has been executed.
The /tmp directory must exist on the workstation where sym_update.sh will execute since the /tmp directory is where temporary files will be generated. The temporary files will have a prefix of sym_ and will be removed when script completes.
The “thiscpu” variable in the TWSHome/localopts and the entries in the /etc/TWS/TWSRegistry.dat file for the TWSuser must be correct and valid.
The sym_update.sh requires the MDM/DM TWS user name and TWS agent (workstation that will receive the updated Symphony file). The agent is required since the script will set the cpu limit to “0” before performing the Symphony update procedures, this is necessary to prevent jobs or job streams from executing until the user has verified that TWS operations may resume on the TWS agent.
NOTE: The job limit must be raised for the TWS agent before normal operations can resume on the TWS agent. This may be performed via JSC or conman.
More support for:
IBM Workload Scheduler
Software version: 8.5, 8.5.1
Operating system(s): AIX, HP-UX, Linux, Platform Independent, Solaris, Windows
Reference #: 1306334
Modified date: 27 May 2016