When using fitglme, what does this warning mean: "Final linear predictor from PL iterations is not feasible"?

I’m using function fitglme to estimate a mixed effects panel probit model with fixed slope, fixed effects for the intercept plus random effects for the intercept for each level of the grouping variable according to a formula:

cr ~ 1 + indep.var1+ indep.var2+ indep.var3+ indep.var4+ indep.var5+ (1|id_country),

where cr is a binary variable and id_country is a grouping variable. As a FitMethod I choose Laplace.

When estimating the model I get a warning: "Final linear predictor from PL iterations is not feasible". What predictor does this warning refer to? Could the model estimates still be reliable?