Reinstalling or restoring the agent

Document ID : KB000004391
Last Modified Date : 14/02/2018
Show Technical Document Details
Issue:

A previously working agent was somehow damaged leaving it in a stated where some files were missing from the agent directory.

 

Environment:
SolarisWorkload Automation AE 11.3.x agent
Resolution:

If a backup of the agent directory is available one may restore it.Or if another machine is available which has the same agent and platform type one may tar up the entire WA_AGENT directory, as root, move it to the problem host and extract the contents there.

 

Additional notes to be aware of:

- remove all files (and sub directories) within the following directories: 

<agent>/database 

<agent>/log 

<agent>/spool 

 

- remove the files: 

<agent>/status 

<agent>/nohup* 

 

- also copy the /etc/init.d/waae_agent-WA_AGENT file from the working system to the non-working one as needed and identify the file name and locations for the agent's start up script in the /etc/rc* locations so you can mimic them on the problem system. 

Example: creating links like S11waae_agent-WA_AGENT in /etc/rc3.d to point back to /etc/init.d/waae_agent-WA_AGENT.  This way the agent can start at boot time. 

 

- The above also assumes the agent_names are identical, as are the paths.  If they are not adjustments can be made to the agentparm.txt file and cryptkey.txt files.

 

- The above also is performing what is called a cold start of the agent.  Meaning any jobs that were actively running on the host would be lost and need to be restarted.  To get the listing of the job run "autorep -M <machine> -d".  Toggle the status of the jobs to INACTIVE and restart or re-save them as needed to get them to run or schedule themselves for their next start time.