@ehsanT what you should do is make sure RServe does run under R / Studio alone and will be started when library(“Rserve”) is called. Then you could tray the integration into KNIME.
In order to do a fresh installation you might try these steps if the ones from the R integration guide would not work. The latest Version 1.8-10 should be available as a Windows binary so there would be no need to compile:
https://CRAN.R-project.org/package=Rserve
For Windows:
install.packages("Rserve", repos = "https://cran.r-project.org", type="win.binary")
If this is not successful you could try and do a installation via the installr package like this:
# install.packages("installr")
library(installr)
install.packages.zip("https://cran.r-project.org/bin/windows/contrib/4.1/Rserve_1.8-10.zip")
or:
install.packages("https://cran.r-project.org/bin/windows/contrib/4.1/Rserve_1.8-10.zip", repos = NULL, type ="win.binary")
You might want to check the latest binary version on the Rserve website:
The same thing for MacOS would look like this:
install.packages("Rserve", repos = "https://cran.r-project.org", type="mac.binary")
Or you could do with the specific package version:
install.packages("https://cran.r-project.org/bin/macosx/contrib/4.1/Rserve_1.8-10.tgz", repos = NULL, type ="mac.binary")
Besides the official guide to R and KNIME integration I try to keep track of several ways and quirks around KNIME and R/RStudio here:
You could check the version and packages of your R installation in KNIME here: