10.2.2 Node switching caused by an error in JP1/AJS3 - Agent
The following figure shows the processing if node switching occurs in JP1/AJS3 - Agent during operation.
|
|
![[Figure]](GRAPHICS/ZJ110035.GIF)
The flow of events in system processing is as follows:
-
An error occurs on the primary node system agent, and the JP1/AJS3 service for JP1/AJS3 - Agent stops.
The status of the job being executed when node switching occurs remains Now Running. This status is managed by JP1/AJS3 - Manager.
-
The contents of the shared disk are inherited by the secondary node system agent.
-
The JP1/AJS3 service of the secondary node system agent starts.
-
JP1/AJS3 - Manager is notified that the JP1/AJS3 service of the secondary node system agent has started.
-
JP1/AJS3 - Manager reports the status of the job that was being executed to JP1/AJS3 - Agent on the secondary node system agent.
The job that is reported at this point is the one being executed when the error occurred (the status of this job was not reported to JP1/AJS3 - Manager previously). Note that the status of jobs being executed when failure occurs will be changed to Killed or Unknown end status, and the jobnet containing the job will be placed in an abnormal status.
For details about the job status inherited when a failover occurs, see 6.2.1(4) Job statuses on the manager host when an agent host is restarted.
This concludes the processing by the system. After this procedure, re-execute the Killed jobs and the jobnet containing such jobs and continue operation.