Error specifying formula and model fit about metida.jl HOT 4 CLOSED

julia> fm4a = fit(MixedModel, @formula(strength ~ 1 + (1|batch/cask)), pastes, REML = true)
Linear mixed model fit by REML
 strength ~ 1 + (1 | batch) + (1 | batch & cask)
 REML criterion at convergence: 246.99074585344238

Variance components:
                Column   Variance Std.Dev.
batch & cask (Intercept)  8.433665 2.904077
batch        (Intercept)  1.657311 1.287366
Residual                  0.678000 0.823408
 Number of obs: 60; levels of grouping factors: 30, 10

  Fixed-effects parameters:
─────────────────────────────────────────────────
               Coef.  Std. Error      z  Pr(>|z|)
─────────────────────────────────────────────────
(Intercept)  60.0533     0.67687  88.72    <1e-99
─────────────────────────────────────────────────

Linear Mixed Model: strength ~ 1
Random 1: 
    Model: 1|batch
    Type: SI (1), Subjects: 10
Random 2: 
    Model: 1|batch & cask
    Type: SI (1), Subjects: 30
Repeated:
    Model: 1|1
    Type: SI (1)
    Blocks: 10, Maximum block size: 6
Status: converged (No Errors)
    -2 logREML: 246.991    BIC: 259.223

    Fixed-effects parameters:
Name          Estimate   SE        z         Pr(>|z|)   
(Intercept)   60.0533    0.67687   88.7221   0.0

    Variance components:
    θ vector: [1.28736, 2.90408, 0.823408]
Random 1   σ²    var   1.6573
Random 2   σ²    var   8.43368
Residual   σ²    var   0.678001