Root Cause Analysis
If a Not Predicted to Finish (NPTF) alert occurs, users can now see the actual event(s) or root cause that generated the alert to be sent by clicking on one of the provided links. The root cause links can be found in the Monitoring Tab Alerts panel and the real-time Gantt chart. The Root Cause Analysis dialog lists all the jobs that are causing the jobstream to be predicted not to finish. The information provided includes the job name, the reason, and any other available details.
This table describes the possible reasons a jobstream would be predicted to not finish:
Code | Description | Explanation |
---|---|---|
1 |
Job start condition is not met |
One or more of the job's start conditions have not been met. |
2 |
Exit code condition is not met |
The job's exit code did not match the condition specified in the job definition. |
3 |
Run window is not met |
The job's start conditions were not met during the run window defined on the job. |
4 |
N/A (this code is not currently used) |
-- |
5 |
Calendar is expired |
The calendar specified in the job's start conditions has expired. |
6 |
Job is on hold |
The job cannot start because it has been put into On Hold status. |
7 |
Job is on ice |
The job cannot start because it has been put into On Ice status. |
8 |
Box already completed |
The job cannot start because its parent Box is not running. |
9 |
Box success/failure conditions not met |
The job is a Box and the success/failure conditions defined for it have not been met. |
10 |
Predecessor could not be analyzed |
The job's start conditions include a reference to a job that does not exist in the scheduler. |
11 |
Target Job didn’t complete with success status |
The Target Job ran but did not complete successfully. |
12 |
Machine is offline (supported in AutoSys r11) |
The job cannot run because the machine on which the job has been defined to run is not active. |
13 |
Exceeded Lookback seconds, stale status |
The job has a predecessor that has not run to satisfy a start condition within the lookback secs setting in the job definition. |
14 |
Job depends on itself |
The job's start conditions contain a condition on the job itself, so it cannot start. |
15 |
Root cause cannot be determined |
A cause could not be found for the failure to predict the Target Job. |
16 |
Job timed out |
The Target Job has timed out. |
17 |
Job has no start conditions |
The scheduler will not run the job because its definition contains no start conditions. |
18 |
Predecessor job has been trimmed from the jobstream |
The job has a predecessor that has been trimmed from the jobstream. |
19 |
Dependency on job in an instance not defined to AAI |
The job depends on a job in an instance that is not defined to AAI. |
20 |
Target Job not predicted to complete with Success status |
The Target of the jobstream is predicted to finish unsuccessfully. |
21 |
Global variable condition not met |
The job's global variable condition is not satisfied. |
When a jobstream is not predicted to finish, the Early/Late criteria in the monitoring grid is shown as N/A. The Predicted End Time column will show that the jobstream is Not Predicted to Finish, and that link can be clicked on to reveal the Root Cause dialog showing the job(s) that are causing the jobstream to be interrupted. Or, you may click on the alert in the Alerts Panel and it will reveal the Root Cause Job(s) and you can click on those links to get to the Real-time Gantt chart view. When the Gantt chart view appears, it will go directly to the job(s) that are interrupting the jobstream.