Normal distribution using the zeros trick...
model {
for (i in 1:8) {
z[i] <- 0
z[i] ~ dpois(phi[i])
phi[i] <- log(sigma) + 0.5*pow((y[i] - mu)/sigma, 2)
}
y[8] ~ dflat()
sigma ~ dunif(0, 100)
mu ~ dunif(-100, 100)
}

Data:
list(y = c(-1, -0.3, 0.1, 0.2, 0.7, 1.2, 1.7, NA))

Inits:
list(y = c(NA, NA, NA, NA, NA, NA, NA, 0))

   node   mean   sd   MC error   2.5%   median   97.5%   start   sample
   mu   0.365   0.4758   0.006864   -0.5948   0.3693   1.316   4001   10000
   sigma   1.18   0.481   0.01139   0.6216   1.067   2.415   4001   10000
   y[8]   0.3499   1.355   0.03415   -2.345   0.3564   3.095   4001   10000

Normal distribution specified directly...
model {
for (i in 1:8) {
y[i] ~ dnorm(mu, tau)
}
tau <- 1 / pow(sigma, 2)
sigma ~ dunif(0, 100)
mu ~ dunif(-100, 100)
}

Data:
list(y = c(-1, -0.3, 0.1, 0.2, 0.7, 1.2, 1.7, NA))

Inits:
list(y = c(NA, NA, NA, NA, NA, NA, NA, 0))

   node   mean   sd   MC error   2.5%   median   97.5%   start   sample
   mu   0.3734   0.4716   0.004705   -0.5638   0.3784   1.316   4001   10000
   sigma   1.169   0.4544   0.007378   0.6268   1.066   2.371   4001   10000
   y[8]   0.3566   1.355   0.01284   -2.346   0.359   3.038   4001   10000