Error reading error output of external R process.

[abadoi]

Hi,

I’m a complete knime noob but slowly making progress with the R integration. My issue is:

I am running the R example of R snippet workflow from the site here to test R integration. All nodes work and provide output including plots and predictions of iris dataset.

BUT i also get these errors. The errors don;t seem to stop the process (i.e. - all nodes go to green status) but I want to resolve them as to avoid issues further down the line.

ERROR RBinUtil Error reading error output of external R process.
ERROR RBinUtil Error reading error output of external R process.
ERROR RBinUtil Error reading output of external R process.
ERROR RBinUtil Error reading output of external R process.

I am running:

Rserve 1.8-6
R version 3.5.2 (2018-12-20)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows 10 x64 (build 17763)

tying in in R console →
> R.home()
[1] “C:/PROGRA~1/R/R-35~1.2”
> file.path(R.home(“bin”), “R”)
[1] “C:/PROGRA~1/R/R-35~1.2/bin/x64/R”

I also ran KNIME in debug mode and the log is attached.

Many thanks,
Alex

errors.txt (24.4 KB)

[JanDuo]

Hi @abadoi

Welcome to the KNIME Community!

I guess you’re using this example workflow in your test.

I ran the same without any errors, which makes it a bit difficult to grasp what might be going wrong on your machine. Maybe the issue is caused by the version of R?

In my R setup in Knime I followed the description I found in the topic Install R alongside KNIME on Windows and MacOS.
The actual path to R Home in the preferences (File > Preferences > Knime > R) I’m using is the one which refers to an R-environment I created using Anaconda.

This gives me the possibility to maintain installation of packages via RStudio (under Anaconda) which is easier than from within the R snippets in Knime.
I use this environment too when “just using R” (so not from within Knime). This way I have the same set of packages available in both RStudio and Knime.

The R version in this environment is R3.6.1. As said, this version doesn’t give the errors you noticed.

Hope this helps you a bit further.

Kind regards,
Jan

[abadoi]

Hi Jan,

thank you for taking the time to write your solution. To asnwer some of your questions:
I already did the set-up steps and manually installed RServe 1.8-6
I am running the same R workflow you linked in.

As for the versioning I am running this in work and we are restricted within this version and set-up. Anaconda distributed won’t be possible hence I have a fairly outdated version of R.

As stated before the workflow appears to run:

I also have my R path set-up:

[JanDuo]

Hi @abadoi

This is what I get running this workflow.

The two warnings are the only things shown by this workflow in the console.

Which version of Knime are you using? I’m using 4.1.3.

[abadoi]

Yes, I don’t get any warnings but other than that all are in green status the same as over here…

KNIME version is 3.6.1

Regards,
Alex

[JanDuo]

Hi Alex,
I’m just another Knime user like yourself, but I fear the errors in the console relate to the older versions, or the combination of the older versions, of Knime and R.

Maybe one of the Knime Teammembers could help you out?

[ipazin]

Hi @abadoi,

welcome to KNIME Community!

3.6.1? Why? Can you update KNIME to newest version and then try?

Br,
Ivan