Through the Portal, you can configure tasks to be run by cxConnect for Data Analysis according to a schedule.
In the Configured Tasks window, you can review the set of currently configured tasks. For each task, you can review its start time and frequency (Task Type column), as well as whether it is enabled.
The ID column contains the internal Experience Analytics identifier for the task. Identifiers may be used by cxConnect for Data Analysis tasks, hidden internal tasks, and cxVerify tasks, if cxVerify is installed.
- Some Experience Analytics internal tasks are not displayed, which might result in gaps in the ID sequence.
- Because there are unique internal identifiers, you can have multiple tasks with the same name, but this practice is not recommended.
Notes on the General tab
Using the General tab, you can configure several elements of a task, including the name, description, scheduling information, daily extraction option, and post command.
- Invalid characters are removed from the Name when the task is saved.
- To account for the processes of indexing sessions and canister session timeout settings, it is recommended that any daily tasks be scheduled after 02:00:00 each night.
- For the Extraction period, you can apply additional filters to the data through the Data Filters tab.
- Some session data can be excluded as part of the data extract through the Data Set tab.
- Post commands run as separate tasks after the current task. These tasks can be tracked as scheduled tasks.
- The sessions to include in an extraction are defined by the session time for each session. The timestamp of a session is the timestamp of the last hit in a session. For example, if you configure an extract for sessions between 1:00 and 2:00, a session that begins at 1:50 and contains a final hit that is recorded at 2:10 is not included.
Depending on the type of extraction, cxConnect for Data Analysis waits the following time periods:
- For repeated tasks such as Daily or Hourly task, cxConnect for Data Analysis verifies that the session indexes were updated at least 1 hour after the end of the configured extract time. If an hourly job is unable to complete in the allotted number of tries, it is skipped and is not reattempted. The next hour, the job is run to collect that hour's data.
- For ad hoc tasks, cxConnect for Data Analysis does not attempt an extraction until 30 minutes past the end of the extract time.
- Run Now tasks do not check the indexes.
Notes on the Data Set tab
In the Data Set tab, you can specify search criteria to apply to the available session data for the selected time period. Additionally, you can specify custom commands to extract the data and to run after the task was completed.
For the Enable Custom Extract String option:
In the Portal or the CX RealiTea Viewer, you can search for specific sessions. For example, you can search for specific values in fields in the session data.
- In the RTV search page, specify the search.
- Click Copy to Advanced Page.
- On the Advanced tab, copy the specified search criteria and use it for your cxConnect for Data Analysis job.
Notes on the Destination tab
In the Destination tab, you can select and specify the connection information of the destination for the extract.
- Use the data files option for session export, which simplifies the process and increases the throughput performance.
- For any task, only one connection type can be active at any time.
At the scheduled time, cxConnect for Data Analysis runs the specified task. At that time, cxConnect for Data Analysis requires that basic CX services, such as Search Server, be available. Additionally, the Extractor Service must be operational.
- Errors are reported in the Extractor log and the cxConnect for Data Analysis Tasks log.
- The number of repeat attempts can be configured through the Extractor Service configuration in TMS.
If the number of failures reached the limit, then the task is resumed at the next scheduled time.