We tested these models with the EEMS10.3.2 release and found that they are working normally, we have attached two images.
In Figure 1 the left panel is 2DH of WSE and dye of the original model displayed at day 4 after the model run. The right panel model uses a restart file on day 4. You can see that the value of dye and WSE are the same. We are not sure which component you are referring to has the zero value.
In Figure 2 you can see the setting up of the hydraulic structure. We tested the model with two boundaries: upstream flow boundary and downstream open boundary. Dye is assigned at the upstream boundary, and after few days of calculation has flowed to the downstream, so EEMS10.3.2 works normally.
Your model setup needs to ensure that the water level upstream of the hydraulic structure is greater than 235 m. If the water level upstream is less than 235 m, it is clear that water cannot pass through the structure.