Use > Software management > Remediate and installing software > Timeout handling for installation or uninstallation and remediation jobs

Timeout handling for remediation and installation jobs

A timeout can occur during a software remediation or installation job. When this happens the job execution stops to prevent further changes to the server.

The error handling option set for the job (see Step 2 (Optional) - Specify reboot, error handling, and script options for remediation) does not apply to timeouts. A timeout, unlike a general error, describes a situation where the job had been running successfully until a specified time interval passed. This is not considered an error, and the job does not attempt to run the remaining phases irrespective of the above flag.

Execution exceeds maximum defined duration

The main reason remediation and installation jobs time out is, when processing a job phase or job item exceeds the maximum defined duration for the specific phase or item (Example: package, script or patch).

These timeout values can be adjusted in the Command Engine section of System Configuration (for more information contact your HPE Server Automation Administrator).

When encountering this type of timeout, the job attempts to stop as soon as possible. This behavior depends on phase active when the timeout occurred:

  • Analyze: the analyze phase gets completed, but the next phases will not be started
  • Download: the download stops after the current package is downloaded. Pending packages will not be downloaded and the next phases will not be started.
  • Action: the installation that is currently running will complete but no remaining items will be processed.

After the timeout occurs and until the job execution stops, the status of the server will be changed to Stopping. After the job execution stops the server will be marked as Timed Out.

Communication problems between the agent and the core

An SA core considers a job as timed out, if communication problems prevent it from reaching the agent during job execution.

Timeouts and server reboots

When a timeout is detected during the Action phase the job prevents the server reboots requested by the packages imported into SA.

If the timeout occurs during a reboot, then after restarting, the agent will not continue with the job. This addresses the situation when servers stay offline for a long time before managing to start again.