Installation and replication of this Open TSDB is quite complex. Could anyone please provide any suggestion?
We are running Qlik Sense September 2017 and are having an issue loading data from the Open TSDB API due to the JSON format being interpreted in a way that means the time series data cannot be loaded for a filtered time period. Here is a link to the API documentation http://opentsdb.net/docs/build/html/api_http/index.html
The JSON returns a set of value pairs for time and metric but this is interpreted by Qlik as Field Name and Value. So when you create the REST SQL query you get: -
...(thousands of other fields)...
FROM "dps" FK "__FK_dps")
FROM JSON (wrap on) "root" PK "__KEY_root";
I am looking to create an on demand App that is driven from a summary view of the time series data that highlights anomalies and allows the user to highlight the time around the anomaly and launch an App with the detail metrics, value per second, around this event. As the REST SQL does not allow a select * statement the query filter on the API call will always return different time elements that are not predictable due to missing elements etc so there is no way that this will work as the SQL statement needs to be generated for each call.
This allows full integration with some back-end server like Open TSDB from both Qlik Scripts and directly from Qlik Sense Applications in the UI with something called Server Side Extensions which allow Qlik Expressions to include remote functions implemented on back-ends to be performed.
You have a choice of languages to implement the interface with examples in Python, R and more.