R help - Jul 2016 - glmmPQL crashes on inclusion of corSpatial object

Link to data: <https://www.dropbox.com/s/yi3vf0bmqvydr8h/data.Rd?dl=0>  
(1170 obs, 9 variables, .Rd file) [plain link in case sth goes wrong 
with the hyperlink: https://www.dropbox.com/s/yi3vf0bmqvydr8h/data.Rd?dl=0]

Simply read it in using `readRDS(file)`.

I?m trying to setup a GLMM using the `glmmPQL` function from the `MASS` 
package including a random effects part and accounting for spatial 
autocorrelation. However, R (Version: 3.3.1) crashes upon execution.

     library(nlme)

     # setup model formula
     fo <- hail ~ prec_nov_apr + t_min_nov_apr + srad_nov_apr + age

     # setup corSpatial object
     correl = corSpatial(value = c(10000, 0.1), form = ~ry + rx, nugget 
= TRUE,
                         fixed = FALSE, type = "exponential")
     correl = Initialize(correl, data = d)

     # fit model
     fit5 <- MASS::glmmPQL(fo, random = ~1 | date, data = d,
                     correl = correl, family = binomial)

What I tried so far:

- reduce number of observation
- play with `corSpatial` parameters (range and nugget)
- reduce number of fixed predictors
- execute code on Windows, Linux (Debian) and Mac R installations


While I get no error message on my local pc (RStudio just crashes), 
running the script on a server returns the following error message:

`R: malloc.c:3540: _int_malloc: Assertion (fwd->size & 0x4) == 0' 
failed. Aborted`

Debugging leads me to a "glibc" c++ library problem. I also run
valgrind
on it. If you need the output, just ask!

Help ist highly appreciated!
Cheers, Patrick

	[[alternative HTML version deleted]]

Patrick Johann Schratz <patrick.schratz <at> gmail.com> writes:
> 
> Link to data:
<https://www.dropbox.com/s/yi3vf0bmqvydr8h/data.Rd?dl=0>
> (1170 obs, 9 variables, .Rd file) [plain link in case sth goes wrong 
> with the hyperlink: https://www.dropbox.com/s/yi3vf0bmqvydr8h/data.Rd?dl=0]
> 
 By the way, this has been cross-posted both to StackOverflow
and to r-sig-mixed-models at r-project.org. Cross-posting between
R lists is explicitly deprecated; cross-posting between R lists
and StackOverflow is not *explicitly* deprecated anywhere, but is
probably a bad idea (tends to waste the time of answerers who don't
know the question has been commented on and/or answered already
elsewhere).  Try to pick the single best venue and stick to it.

  Ben Bolker