[Rd] makeCluster Stall on 32-bit Windows
Henric Winell
nil@@on@henric @ending from gm@il@com
Thu Jun 14 14:49:44 CEST 2018
Hi Dario,
On 2018-06-14 14:00, Dario Strbenac wrote:
> Good day,
>
> I'm trying the example cl <- makeCluster(2, type = "SOCK")
Why use the 'snow' package and not 'parallel'?
> from the makeCluster documentation and the R command prompt never
> returns. I am using a 32-bit Windows 10 computer. The problem doesn't
> happen on another 64-bit Windows 10 computer but does happen on
> another 32-bit Windows 7 computer. Can anyone reproduce it? Are there
> any other options that can be passed to makeCluster which could get
> it going?
Sorry, I don't have access to a 32-bit W10 machine for testing.
Since the worker processes communicate using sockets I guess that the
Windows firewall on your machine is blocking R from using the necessary
ports (11000-11999). Upon first use on a freshly installed 64-bit W10 a
window pops-up -- "Windows Firewall has blocked this app", or something
like that -- where you can allow access. As long as this window is
open, the command prompt is not returned.
Henric Winell
P.S. This topic is probably not appropriate for R-devel and follow-ups
should likely be directed to R-help instead. D.S.
> R version 3.5.0 (2018-04-23)
> Platform: i386-w64-mingw32/i386 (32-bit)
> Running under: Windows >= 8 (build 9200)
>
> Matrix products: default
>
> locale:
> [1] LC_COLLATE=English_Australia.1252 LC_CTYPE=English_Australia.1252
> [3] LC_MONETARY=English_Australia.1252 LC_NUMERIC=C
> [5] LC_TIME=English_Australia.1252
>
> attached base packages:
> [1] parallel stats graphics grDevices utils datasets methods
> [8] base
>
> other attached packages:
> [1] snow_0.4-2
>
> loaded via a namespace (and not attached):
> [1] compiler_3.5.0 tools_3.5.0 yaml_2.1.19
>
> --------------------------------------
> Dario Strbenac
> University of Sydney
> Camperdown NSW 2050
> Australia
>
> ______________________________________________
> R-devel using r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-devel
>
More information about the R-devel
mailing list