It may happen that an agent shuts down with the following message:
Fatal: DB Lower disk threshold of (null) breached. Shutting down
How can this be prevented? Do we have any control over the threshold?
The agent uses a dedicated directory, the database directory, to keep track of the jobs it handles and their statuses. The database is located under the agent directory.
By default, agents do monitor the database directory, and if a 10M threshold is reached, the agent shuts down.
This behavior can be adjusted by configuring the agent to monitor (or stop monitoring) the amount of available disk space for the database directory.
All the parameters needed to go about the changes are located in the agentparm.txt file. There are 2 actions that can be undertaken:
Disabling the monitoring: agent.resourcemon.enable allows enabling or disabling monitoring. If set to "false", then the agent will stop monitoring and will no longer shutdown. The drawback is that there will be no alerts until the space is exhausted. This may not be recommended reason why the default is true.
Tailoring the monitoring thresholds. First monitoring should be enabled (default):
Next 3 thresholds can be set:
agent.resourcemon.threshold.disk.warning.notice=n where n is the limit where the agent sends a message, but keeps running. By default if less that 21M is left.
agent.resourcemon.threshold.disk.warning.severe=n where n is the limit where the agent issues a critical message, keeps running but stops receiving messages; once the disk space gets lower than this limit, the agent resumes normal processing. Default 20M
agent.resourcemon.threshold.disk.critical=n If this limit is reached, then the agent shuts down. Default 10M.
To make the changes effective, you need to:
Shut the agent.
Change agentparm.txt files to adjust the above parameters.
Restart the agent
Release: ATSYHA99000-11.3.5-Workload Automation AE-High Availability Option