Our organization has a Tableau server which is very extract heavy. Lots of extracts running daily and hourly.
We have set our Backgrounder query limit at 12000. Any task taking more than 12000 seconds would time out making way for other refresh extracts in the queue.
We have a particular extract which runs weekly once or twice and has a very long runtime. Approx 25000 seconds and fluctuates. When we had a long Backgrounder query limit to accommodate long extracts, we would see that when an extracts fails randomly(Data issue, connections issue etc) it would occasionally be running until that timeout is reached. Causing the backgrounder to be taken up and a huge queue of extracts waiting for free backgrounders.
Since having huge time out isn't a good practice we have set it to 12000 seconds now. But I would like to provide exceptions to a few Refresh tasks on the server.
Is it possible to set an exception like this.
If not, what would be the best way to run such long extracts occasionally without tableau terminating them.
P.S: We have taken the Hyper leap and are on 10.5.1