How AutoSupport creates and sends event-triggered messages

AutoSupport creates event-triggered AutoSupport messages when the EMS processes a trigger event. An event-triggered AutoSupport message alerts recipients to problems that require corrective action and contains only information that is relevant to the problem. You can customize what content to include and who receives the messages.

AutoSupport uses the following process to create and send event-triggered AutoSupport messages:

  1. When the EMS processes a trigger event, EMS sends AutoSupport a request.

    A trigger event is an EMS event with an AutoSupport destination and a name that begins with a callhome. prefix.

  2. AutoSupport creates an event-triggered AutoSupport message.

    AutoSupport collects basic and troubleshooting information from subsystems that are associated with the trigger to create a message that includes only information that is relevant to the trigger event.

    A default set of subsystems is associated with each trigger. However, you can choose to associate additional subsystems with a trigger by using the system node autosupport trigger modify command.

  3. AutoSupport sends the event-triggered AutoSupport message to the recipients defined by the system node autosupport modify command with the -to, -noteto, -partner-address, and -support parameters.

    You can enable and disable delivery of AutoSupport messages for specific triggers by using the system node autosupport trigger modify command with the -to and -noteto parameters.

Example of data sent for a specific event

The storage shelf PSU failed EMS event triggers a message that contains basic data from the Mandatory, Log Files, Storage, RAID, HA, Platform, and Networking subsystems and troubleshooting data from the Mandatory, Log Files, and Storage subsystems.

You decide that you want to include data about NFS in any AutoSupport messages sent in response to a future storage shelf PSU failed event. You enter the following command to enable troubleshooting-level data for NFS for the callhome.shlf.ps.fault event:

cluster1::> system node autosupport trigger modify -node node1 -autosupport-message shlf.ps.fault -troubleshooting-additional nfs

Note that the callhome. prefix is dropped from the callhome.shlf.ps.fault event when you use the system node autosupport trigger commands, or when referenced by AutoSupport and EMS events in the CLI.