In addition to globally configuring PureLytics streaming to send real time UEM data to an external data source, you can also configure AppMon to stream data to an external target for a specific timeframe. This gives you the convenience of performing more simplified and precise, time range-based analyses of steamed data. For example, you can set up streams during a weekly peak traffic period and use the streamed data to spot specific traffic trends that occur during that period.
AppMon can stream Purelytics user action and visit data in an on-demand time range regardless of whether a global PureLytics stream has been configured for the system profile. On-demand streamed data includes user visits and user actions, but does not include client errors.
Configure On-demand PureLytics Streaming
Click Settings > Configure Purelytics Stream to open the Purelytics Stream dialog box.
Under Timeframe, select the date and time to start data streaming in the From fields, and a date and time to end data streaming on the to fields.
Choose the target you want to stream the JSON documents to. You can either stream them to an Elasticsearch Cluster (Version 1 to 5), Elasticsearch 6 Cluster or stream them to an generic HTTP endpoint using the same Elasticsearch bulk format.
Enter the URL to your Elasticsearch cluster or generic HTTP endpoint (REST port) in the URL field. The PureLytics Stream appends
/_bulk to the entered URL upon successful connection testing.
The on-demand Purelytics stream by default streams user action and visit data. You can clear the User Actions or Visits check boxes to refine the stream to only user action data or visit data. This is useful when streaming large amounts of specific data over longer time periods.
AppMon uses the following predefined index patterns:
- For an Elasticsearch Cluster (Version 1 to 5) or a generic HTTP Post, you can use the index pattern to organize data by date (
- For an Elasticsearch 6 cluster, you can organize by visit and date (
dt_visit_YYYY-MM), or by user action and date (
Use the Index field to customize these index patterns as needed.
See the PureLytics Stream Technical details and limitations for more information about using index patterns.
Optionally select Allow untrusted SSL if you want to use HTTPS to export to the server specified in the URL field without certificate checking.
Optionally select Basic HTTP Authentication to authenticate with a user name and password, then enter the user name and password required to authenticate. You can also verify the connection to the Elasticsearch cluster or generic endpoint by clicking Test connection.
Use Queue Size to set the amount of data sent with every bulk request. A request is sent when the internal document queue hits the size specified in Queue Size. Request size is approximately the number of documents times 2 kB, depending on factors such as number of user actions per visit.
Click Stream to start the data stream at the configured date and time.