Forum Discussion
Cannot refresh or save dataflow - Cannot acquire lock for model
NAOS I'm experiencing something similar with one particular dataflow that has two entities connecting to an Azure SQL Database without using a gateway. It started failing on Thursday 02/01. The dataflow refresh status shows as "Failed" but the Entity refresh status for both entities shows as "Completed". There is no error message in the "Error" column. It fails on the scheduled refresh and a manual refresh.
When I try to modify the dataflow and save the changes I get the "cannot acquire lock for model" error message.
If I export the JSON and create a new replica dataflow in the same workspace by importing the JSON it works fine.
I will be interested to know if you find a solution!
NAOSRather than restart the whole capacity (which we would need to be done out of hours to minimise disruption) I went into Admin Portal >> Capacity Settings and under Workloads >> Dataflows increased the Max Memory by 1%, applied the change and then changed it back to its original setting and applied the change again. This only took a few seconds. The theory was that this would force the dataflow workload to restart which might clear the problem.
It appears to have worked as I am now able to refresh the dataflow ok, so this might be something worth suggesting to your IT Department as being less disruptive than restarting the whole capacity.
NAOSI didn't really consider creating a new dataflow to be a solution, more a troubleshooting step to rule out any connection issues etc. Whilst we do sometimes use parameters in reports to hold the workspace and dataflow id (so that we can easily switch between a dev and production version of a dataflow), if you have a large number of reports created by users of mixed skillsets like we have then changing the id on all of them isn't really practical.
Did your IT Department try restarting the capacity? We will try that tonight, but in the meantime I guess I will log a case as well.
Hi PaulKn ,
We haven't yet restarted our capacity (IT dep. still hasn't done it...), but I'll post here once we do to let you know if that fixed the issue or not.
Kind regards,
NAOS
NAOSRather than restart the whole capacity (which we would need to be done out of hours to minimise disruption) I went into Admin Portal >> Capacity Settings and under Workloads >> Dataflows increased the Max Memory by 1%, applied the change and then changed it back to its original setting and applied the change again. This only took a few seconds. The theory was that this would force the dataflow workload to restart which might clear the problem.
It appears to have worked as I am now able to refresh the dataflow ok, so this might be something worth suggesting to your IT Department as being less disruptive than restarting the whole capacity.
PaulKn , venal , v-eachen-msft
That worked! I'll try change the last comment to be the solution.
Thanks PaulKn! Great work.NAOS