We’re slowly migrating our legacy Qlik Sense ETL to EasyMorph, and we are facing a scenario where an EasyMorph task should be scheduled between two QMC tasks: the first QMC task must complete before the EasyMorph task runs, and the third QMC task must wait for the EasyMorph task to finish. I found a discussion about this on GitHub : GitHub - easymorph/EasyMorph-Action: Qlik Sense extension for running tasks on EasyMorph server , but is there a way to run an EasyMorph task from a Qlik Sense load script?
Yes, the REST connector in Qlik is the right way. Alternatively, you could trigger a Server task from the command line in Qlik (the EXECUTE) statement, however it’s less secure and not available in Qlik Cloud.
Generally speaking, I’d do the whole orchestration in EasyMorph.
How is your experience with that so far? Any other challenges you faced in the process?
Indeed, we didn’t want to grant execute permissions on our on-prem Qlik server. Our goal is to gradually migrate all ETL processes to EasyMorph, rather than doing a big bang transition.
As I’ve always told you, EasyMorph is, in my 25 years of development experience, the best tool I’ve ever used — thanks to its no-code approach and instant results philosophy. You truly see what you get, like never before.
The main challenges we’re facing involve writing large QVD files, which can be time-consuming and memory-intensive, since it’s not possible to update these files incrementally as with a database. It’s also not easy to orchestrate the jobs overnight: some can run in parallel, while others have dependencies and must be executed sequentially. Our server often restarts during the night due to memory limitations, and when it does, we don’t always have complete information about failed jobs in the logs. We plan to upgrade to the Enterprise Server edition when we roll out our new ERP.
Nevertheless, we keep finding solutions, and we’re all very happy with how easy EasyMorph is to use — both on Desktop and on Server. We also have a new developer who had never written a single line of code before, former student, yet is now perfectly able to perform complex data transformation tasks.