JAGS not saving all output for long model runs

Question

I have a JAGS model that I am running through R, using the jagsUI package (which I believe is a wrapper for rjags).

My model runs fine for a reasonably large number of iterations (400,000), but when I try a long overnight run (4,000,000) it no longer returns all the output. Additionally, I get the message

Warning message:
In order.params(samples, parameters.to.save, DIC, verbose = verbose) :
  JAGS did not monitor deviance.

There are a number of [n.iter x 173 x 2] arrays it should be returning; most are completely missing, and a few are collapsed to [n.iter x 2]. (Perhaps it's only storing the first row of these?) I notice that deviance is missing as well.

I'm running 6 chains in parallel, discarding the first half as burn-in, and thinning such that I save 1000 samples per chain.

Is it possible that I'm running out of available memory before thinning & burn-in?