An update on the above:
We have been dealing directly with one of the Cloud Success Rep's at Qlik and are happy to say that we have resolved the above security and automation issue.
The automated system that we have set up is therefore as follows:
- QlikView (Personal Edition) has been set up to extract & transform data - generating cloud ready qvd's on the client's local server.
- The QlikView model(s) have been scheduled (via Task Scheduler) to load all new data every night.
- The client has set up FTP access to the above local server directory containing the cloud ready qvd's.
- Qlik were able to provide a list of IP addresses that the client could whitelist for access through their firewall.
- Qlik Sense Cloud Apps were set up to load the optimised qvd data directly from the client's local server directory, using FTP Web File connections (passing login credentials).
- Qlik Sense Cloud Apps were then scheduled to refresh published data every morning.
So far this system seems to work pretty well.
The client is able to manage & store the raw & optimised data on their local server, while Qlik Sense Cloud provides the analysis platform.
The IT guys are happy as the connections are as secure as they can be for now, until SFTP becomes an option.
The above system uses direct FTP access, no 3rd party or Dropbox required.
Users can log in from anywhere and the Qlik Sense App data is refreshed automatically every day.
The only limitation at this point is that users are not able to export data/visualisations from published apps, which Qlik are looking at for future functionality - possibly first half of 2018: Qliksense Cloud export on shared app
Users will therefore be accessing:
- Published Apps for analysis - screen clipping charts, dashboards & small table visualisations, etc.
- Workspace Apps to export large tables into Excel.
One important consideration with this approach is that only data in Published Apps can be scheduled to refresh automatically, so any users accessing Apps in Workspace will need to perform manual reloads.
For the cost and convenience, I think the client is pretty happy with the solution provided so far.
Any feedback, comments and questions are welcome.