ESMValTool memory issue in parallel computation #160
Description
On the eWaterCycle teaching machines when generating forcing using LumpedMakkinkForcing the kernel freezes when the number of years > 4 (the juypter cell keeps running without producing a result) . It works fine up to 4 years of forcing.
After changing the ESMValTool configuration to use a sequential runner, the result is produced fast without issues.
stacktraces
We should figure out if this is an ESMValTool problem, or if it's related to our (dask) configuration.