Agents can be run as single runs or as job runs. A job run consists of multiple agent runs that are tracked as a single entity. A job run will not complete before all agent runs associated with the job have completed, and a job run is only completed successfully if all agent runs have completed successfully.
Jobs usually start agent runs using job schedules. A job can consist of one or more job schedules and the job will not complete before all schedules have completed. Each job schedule can start multiple agent runs, and a job schedule is not completed before all agent runs have completed.
Jobs can also be started directly from the Agent Repository page. If a job is started with performance sessions, the job will not complete before all sessions have completed, and exported data from all sessions will be combined into a single data file. If you start an agent as a single run with performance sessions, each session will be considered an individual run and each session will export its own data.
Only one job can be configured for each agent, and one agent can only have one active job run.
Managing Jobs in ACC Web Portal
Once you deployed your agent to a cluster, you can create Jobs and Job Schedules, configure Job Settings, view Job History, etc.
The 'Jobs' which you can find at dashboard actually refer to the running, failed, scheduled and succeeded jobs. From here you will see if a job has run successfully or has failed and what is the 'Last Run/Next Run/Started/Completed' time.
You can click the icon that comes at the end of each job to check out the 'Agent Run History', 'Job History' and 'Job Schedules'. The button of 'View in Repository' provides you with the key info of the selected job, such as the Version of the Agent, the Server/Cluster that the job is associated to, the Proxy Pool bundled with the job, whether Success Criteria is enabled, Rate limit applied or not, when was it Last updated, status of Job and Run, what are the Tickets created for the job.
To configure an agent job with the proper setting, select a specific agent from the directory and click on the action menu on the right side of the window. From the drop-down list, click on Job Settings and it will display a window from where you can set up a job.
Run ID - This is a unique identifier in the form of integer value which will show the number of job execution. It will increase as per the number of times the run will get executed. You can set it from job settings if you need to run the agent for a specific Run ID.
Continue on failure- You can select Yes, No or Default from drop down options required for a particular job.
Success Criteria – For setting success criteria, check to enable success criteria and then enter the value in Min data count and Max errors. If you check on the Percentage of minimum data count or maximum errors, then it will take the value as a percentage and compare the current job with the last executed successful job. If the success criteria meets the expectation, then the job status will be "Successful", else "Unsuccessful".
One Job per schedule - If you check on one job per schedule, then it will execute one job for each schedule. Basically, it comes into consideration when multiple schedules are set for an agent, then it will execute each schedule as a separate job. If it is unchecked, then it will consider overall/multiple schedules as one job.
A job can include one or more job schedules that start agent runs on a schedule. All schedules in a job must complete before the job is completed, and the job is only completed successfully if all its job schedules are completed successfully.
To view and add job schedules for an agent, select Job Schedules from the agent context menu in the Agent Repository.
The Job Schedules page will show status information for all schedules.
Click Add Schedule to add a new job schedule.
A basic schedule can run an agent every specified number of days, hours, minutes or seconds. Start time and time zone specifies the first time an agent will run.
A CRON expression can be used to add a more complex schedule. Please see CRON expressions for more information. CRON expressions cannot contain time zone information, so if a CRON expression specifies a specific run time, you can use the Time zone field to specify the time zone for that run time. A CRON schedule will start on or after the time specified by the Start time field.
Log level is the log level used when the agent is run. The log level can be High, Medium, Low or None.
Session ID can be used to start multiple instances of the same agent. The Session ID field is disabled if the agent does not support sessions. See Performance Sessions for more information.
Input parameters can be used to provide input data for the agent. See Input Parameters for more information.
View All Schedules on a Cluster
A user with the Admin role can view all schedules on a cluster by selecting Schedules from the Agents menu, and then selecting a cluster from the Select Server or Cluster drop down.
The information on this page can help you make sure you have not scheduled too many agents at the same time on the same cluster.
Job history page displays the execution history of the jobs that were scheduled to run on a cluster. You can navigate to the job history page via below option:
- Agents→Schedules→Context Menu→Job History
The values displayed under each column of the job history page is explained as below:
RunId: This column shows an increasing number for each row w.r.t the jobs executed.
Start Time: This column displays the date and time at which the job started executing on the cluster.
End Time: This column displays the date and time at which the job finished its execution on the cluster.
Schedules: This column displays the count of job schedules that got executed. This column will display 1/1 for each row if the Job Settings is set with "one job per schedule" and will display the count of all the schedules that got executed again the job schedules for a particular agent.
Runs: This column displays the count of runs that got executed for a particular job schedule.
Actions: The successfully executed action commands.
Pages: Actual page loads. This is all pages that are loaded into the main frame of the browser or pages loaded by the HTML/JSON/XML parser.
Requests: All requests made to the target web server, including HTML, images, styles, script, etc. For HTML/JSON/XML parsers there's one request per page load.
Errors: All action errors. Note that a page load error that's retired and then successful does not result in an error, but the error counts as a page load if a page was actually loaded.
Data: The number of records that have been extracted by the agent. This count can be different if the "Increase Data Count" property is manually configured else the count will be based on the increased count on each extracted content element.
Delivery: Actual count of the data rows delivered to the destination endpoint.
Error Message: This column displays the error message of the job failure.
Status: This column displays the execution status of the job which can be In Progress, Succeeded, Failed.
Additionally, you can navigate to the run history page by going to View Run History link available under the action menu for each row under the job history page.
Also, you can manually set the job status to failed or success which allows the users to stop the job execution or continue the job execution basis the last job execution status.
A user can Download Data by selecting this option from the action menu of each row under job history page. Partial data can also be downloaded with option 'Download Partial Data' in case of Failed status.
Download Logs - A user can also download logs from Job history and Run history page to check execution log that provides time-based information of the actions performed by the agent.
NOTE - The Download/Get log feature depends on the size of the log file. A log file can only be viewed on web browser if the file size is less than 5 MB. The file can't be downloaded by the user in this case. If the logs file is more than say 5 MB size then it can only be downloaded and cannot be viewed on web browser.
Delete option can be used to delete the job from Job history page.
Managing Jobs in Sequentum Enterprise (Desktop)
Agent Jobs can be managed/configured in Sequentum Enterprise Editor via the Job context menu. Simply right-click on an agent in the Agent Repository window to show the context menu.
To access Job Settings configuration in Sequentum Enterprise Editor, use Job → Job Settings from the context menu. Refer to the section above on Job Settings for more details.
To access Job Schedules in Sequentum Enterprise Editor, use Job → Job Schedules from the context menu. Refer to the section above on Job Schedules for more details.
To access Job History in Sequentum Enterprise Editor, use Job → Job History from the context menu. Refer to the section above on Job History for more details.
Set Failed Status
To set Job status as Failed in Sequentum Enterprise Editor, use Job → Set Failed Status from the context menu. A user can manually set the job status to Failed which allows the users to stop the job execution on the basis of last job execution status.
Set Success Status
To set Job status as Success in Sequentum Enterprise Editor, use Job → Set Success Status from the context menu. A user can manually set the job status to Success which allows the users to continue the job execution on the basis of last job execution status.
Delete Job Settings
To delete Job settings in Sequentum Enterprise Editor, use Job → Delete Job Settings from the context menu. A user can use this option to delete all job settings including all associated schedules and job history of the agent.