bug in the csv files

Summary:

The CSV files generated by the sampling function contain duplicated lines.

Description:

I run sampling with the sample_file argument. This generates CSV files containing the simulations. These files contain duplicated lines, just after the warm-up.

Reproducible Steps:

library(rstan)
rstan_options(auto_write = TRUE)
options(mc.cores = parallel::detectCores())

# stan model
stancode <- 'data {
int<lower=0> J; // number of schools 
real y[J]; // estimated treatment effects
real<lower=0> sigma[J]; // s.e. of effect estimates 
}
parameters {
real mu; 
real<lower=0> tau;
real eta[J];
}
transformed parameters {
real theta[J];
for (j in 1:J)
theta[j] = mu + tau * eta[j];
}
model {
eta ~ normal(0, 1);
y ~ normal(theta, sigma);
}'

# compile model  
stanmodel <- stan_model(model_code = stancode, model_name="schools")

# stan data 
standata <- list(J = 4, 
                 y = c(28,  8, -3,  7), 
                 sigma = c(15, 10, 16, 11))

# sampling
samples <- sampling(stanmodel, data = standata, 
                    iter = 1500, warmup = 500, chains = 4,
                    sample_file="stansamples")

Current Output:

RStan Version:

_rstan version 2.12.1_

R Version:

R version 3.3.1 (2016-06-21)

Operating System:

Ubuntu 14.04

Reproduce the issue with the provided RStan example, focusing on sampling(..., sample_file="stansamples") and the generated CSV files immediately after warm-up. Compare the output across the four chains and confirm that the completed files contain no duplicated simulation lines.