I’m having the problem with my /tmp directory being set with noexec (so scripts inside it can’t be executed). This is causing pkgbuild::compile_dll() to fail when building an R package with Stan models. I’ve tried to fix this a number of ways.
I created a new folder called ~/rtmp and set TMPDIR=~/rtmp. This worked in the sense that temporary files were created inside the new folder, but compile_dll() still fails with the same error.
I did sudo chmod 777 ~/rtmp and sudo chmod -R 777 ~/rtmp and those also failed—the change didn’t apply to the new file that was created by compile_dll().
Here’s what it all looks like after creating a fresh rtmp directory and giving it full permissions with chmod as above:
setwd(“/home/connor/dev/spstan”)
pkgbuild::compile_dll()
Re-compiling spstan
─ installing source package ‘spstan’ …
** using staged installation
ERROR: ‘configure’ exists but is not executable – see the ‘R Installation and Administration Manual’
─ removing ‘/home/connor/rtmp/RtmpPVXJ8C/devtools_install_5d8d388b0a4b/spstan’
Error in (function (command = NULL, args = character(), error_on_status = TRUE, :
System command error
Then looking at the contents and permissions of ~/rtmp:
connor@xps:~$ ls -l rtmp/RtmpPVXJ8C
total 40
drwxrwxrwx 2 connor connor 4096 Jun 24 17:42 devtools_install_5d8d388b0a4b
-rw-rw-rw- 1 connor connor 327 Jun 24 17:42 file5d8d1562eae6
-rw-rw-rw- 1 connor connor 327 Jun 24 17:42 file5d8d1a3a36b
-rw-rw-rw- 1 connor connor 200 Jun 24 17:42 file5d8d22b9e7a9
-rw-rw-rw- 1 connor connor 2305 Jun 24 17:42 file5d8d382ad0bc
-rw-rw-rw- 1 connor connor 780 Jun 24 17:42 file5d8d4b931938
-rw-rw-rw- 1 connor connor 200 Jun 24 17:42 file5d8d5a60119b
-rw-rw-rw- 1 connor connor 2304 Jun 24 17:42 file5d8d5eb353c1
-rw-rw-rw- 1 connor connor 781 Jun 24 17:42 file5d8d776cca6b
-rw-rw-rw- 1 connor connor 2608 Jun 24 17:42 foo.o
Any ideas for how to address this?
The need to manually set TMPDIR was previously discussed here RStan and PyStan may fail to compile model if TMPDIR not manually set
and I also posted this with some other details at stackoverflow but haven’t gotten any response
I also tried this with ~/rtmp mounted as a proper temporary directory (by adding it to my /etc/fstab file and rebooting), but that didn’t help.
sessionInfo()
R version 3.6.0 (2019-04-26)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 18.04.2 LTS
Matrix products: default
BLAS: /usr/lib/x86_64-linux-gnu/blas/libblas.so.3.7.1
LAPACK: /usr/lib/x86_64-linux-gnu/lapack/liblapack.so.3.7.1
locale:
[1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C
[3] LC_TIME=en_US.UTF-8 LC_COLLATE=en_US.UTF-8
[5] LC_MONETARY=en_US.UTF-8 LC_MESSAGES=en_US.UTF-8
[7] LC_PAPER=en_US.UTF-8 LC_NAME=C
[9] LC_ADDRESS=C LC_TELEPHONE=C
[11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C
attached base packages:
[1] stats graphics grDevices utils datasets methods base
loaded via a namespace (and not attached):
[1] Rcpp_1.0.1 compiler_3.6.0 pillar_1.4.1
[4] prettyunits_1.0.2 tools_3.6.0 pkgbuild_1.0.3
[7] tibble_2.1.2 gtable_0.3.0 lattice_0.20-38
[10] pkgconfig_2.0.2 rlang_0.3.4 Matrix_1.2-17
[13] cli_1.1.0 parallel_3.6.0 loo_2.1.0
[16] gridExtra_2.3 dplyr_0.8.1 desc_1.2.0
[19] stats4_3.6.0 rprojroot_1.3-2 grid_3.6.0
[22] tidyselect_0.2.5 inline_0.3.15 glue_1.3.1
[25] R6_2.4.0 processx_3.3.1 BH_1.69.0-1
[28] rstan_2.18.2 ggplot2_3.2.0 callr_3.2.0
[31] purrr_0.3.2 magrittr_1.5 matrixStats_0.54.0
[34] backports_1.1.4 scales_1.0.0 ps_1.3.0
[37] StanHeaders_2.18.1-10 assertthat_0.2.1 colorspace_1.4-1
[40] lazyeval_0.2.2 munsell_0.5.0 crayon_1.3.4
[43] RcppEigen_0.3.3.5.0