SAP Data Intelligence - R Client Operator - issue in adding R package to the R code

[Dokon]

We are facing issue with R Client Operator where we need to add R packages to our R code.

We are trying to follow:

https://blogs.sap.com/2018/12/10/sap-data-hub-and-r-time-series-forecasting/

Details:

We have created the following pipeline:

We have created a dockerfile file in the repository:

We have added a tags:rforecast in the configuration. We are able to build the same as well.

To import r packages to the R code we have created group on the R Client Operator and added the above created tags. When we are trying to import rforecast and run pipeline we are getting the following error:

Error:

failed to prepare graph description: failed to select image: No matching image found for group group1 with runtime tags: map[opensuse: rforecast: rjsonlite: rmsgpack: rserve:] Alternatives are: {com.sap.opensuse.dq map[deprecated: opensuse: vflow_dh_dq:2.7.19]} {com.sap.python36 map[debian: opensuse: pandas: python36: scikit-learn: tornado:5.0.2]} {com.sap.sles.ml.functional-services map[python27: python36: requests:2.18.4 sles: tornado:5.0.2]} {com.sap.opensuse.streaming map[deprecated: opensuse: python27: python36: sapjvm: streaming_lite: tornado:5.0.2]} {com.sap.sles.golang map[python27: python36: sapgolang:1.12.1-bin sles: tornado:5.0.2]} {com.sap.opensuse.base map[deprecated: opensuse: python27: python36: tornado:5.0.2]} {com.sap.opensuse.flowagent-codegen map[deprecated: flowagent-codegen:2.7.84 hadoop:2.9.0 opensuse: python36: spark:2.4.0 tornado:5.0.2]} {com.sap.opensuse.ml.python map[automated-analytics:3.2.0.9 deprecated: grpcio:1.8.6 minio:3.0.3 ml-python: numpy:1.13.1 oauthlib:2.0.4 opencv:3.4.2 opensuse: pillow:5.0.0 pykalman: python27: python36: requests:2.18.4 requests-oauthlib:0.8.0 scipy:1.0.1 tensorflow:1.7.0 tensorflow-serving-api:1.4.0 textblob:0.12.0 tornado:5.0.2 tweepy:3.7.0]} {com.sap.opensuse.ml.rbase map[opensuse: python36: r:3.5.0 rjsonlite: rmsgpack: rserve: sles: tornado:5.0.2 tree:]} {com.sap.sles.dq map[sles: vflow_dh_dq:2.7.19]} {teched.dat649.zzforecast map[zzforecast:]} {com.sap.sles.base map[default: python27: python36: sles: tornado:5.0.2]} {com.sap.opensuse.golang.zypper map[deprecated: opensuse: python27: python36: sapgolang:1.12.1-bin tornado:5.0.2 zypper:]} {com.sap.opensuse.python36 map[deprecated: opensuse: python27: python36: tornado:5.0.2]} {com.sap.ml.functional-services map[deprecated: opensuse: python27: python36: requests:2.18.4 sles: tornado:5.0.2]} {com.sap.opensuse.flowagent-operator map[deprecated: flowagent:2.7.84 opensuse:]} {com.sap.opensuse.textanalysis map[deprecated: opensuse: python27: python36: tornado:5.0.2 vflow_textanalysis:]} {com.sap.sles.flowagent-operator map[flowagent:2.7.84 sles:]} {com.sap.sles.sapjvm map[python27: python36: sapjvm: sles: tornado:5.0.2]} {demo.Docker_R_Forecast map[rforecast: rjsonlite: rserve:]} {teched.dat649.marathon map[opensuse: python36: r:3.5.0 rjsonlite: rmsgpack: rserve: sles: tornado:5.0.2 tree: zforecast:]} {com.sap.opensuse.node map[deprecated: node: opensuse: python27: python36: tornado:5.0.2]} {com.sap.sles.flowagent-codegen map[flowagent-codegen:2.7.84 hadoop:2.9.0 python36: sles: spark:2.4.0 tornado:5.0.2]} {com.sap.sles.ml.python map[automated-analytics:3.2.0.9 grpcio:1.8.6 minio:3.0.3 ml-python: numpy:1.13.1 oauthlib:2.0.4 opencv:3.4.2 pillow:5.0.0 pykalman: python27: python36: requests:2.18.4 requests-oauthlib:0.8.0 scipy:1.0.1 sles: tensorflow:1.7.0 tensorflow-serving-api:1.4.0 textblob:0.12.0 tornado:5.0.2 tweepy:3.7.0]} {com.sap.sles.hana_replication map[hanareplication:0.0.101 python27: python36: sapjvm: sles: tornado:5.0.2]} {com.sap.sles.streaming map[python27: python36: sapjvm: sles: streaming_lite: tornado:5.0.2]} {com.sap.sles.textanalysis map[python27: python36: sles: tornado:5.0.2 vflow_textanalysis:]} {com.sap.dh.workflow map[dh-app-base:2.7.160 node:vflow-sub-node]} {com.sap.ml.dsp map[aiohttp:3.5.4 backoff:1.8.0 deprecated: hdfs:2.5.0 opensuse: python27: python36: requests:2.18.4 tornado:5.0.2 uvloop:0.12.2]} {com.sap.opensuse.golang map[deprecated: opensuse: python27: python36: sapgolang:1.12.1-bin tornado:5.0.2]} {com.sap.opensuse.hana_replication map[deprecated: hanareplication:0.0.101 opensuse: python27: sapjvm: tornado:5.0.2]} {com.sap.opensuse.sapjvm map[deprecated: opensuse: python27: python36: sapjvm: tornado:5.0.2]} {com.sap.scenariotemplates.customdataprocessing.pandas map[pandas: python27: python36: sles: tornado:5.0.2]}

Please let us know how we can provide the proper dockerfile or tags so that we are able to rectify the issue.

[AndreasForster]

Hello Do-kon, The current version of Data Hub / Data Intelligence also requires the rmsgpack library.

This is the Docker file I am currently using for this scenario

FROM debian:9.4


RUN apt-get update \
	&& apt-get install -y --no-install-recommends \
		locales \
	&& rm -rf /var/lib/apt/lists/*


## Install R
## Set a default CRAN repo
RUN apt-get update \
	&& apt-get install -y --no-install-recommends \
		r-base \
		r-base-dev \
		r-recommended \
		libnlopt-dev \
		curl \
        && echo 'options(repos = c(CRAN = "https://cran.rstudio.com/"), download.file.method = "libcurl")' >> /etc/R/Rprofile.site \
	&& rm -rf /tmp/downloaded_packages/ /tmp/*.rds \
	&& rm -rf /var/lib/apt/lists/*


## Install additional R packages
RUN apt-get update
RUN apt-get install libcurl4-openssl-dev


RUN echo 'options(repos="https://cran.rstudio.com/", download.file.method = "libcurl")' >> /etc/R/Rprofile.site \
    && Rscript  -e "install.packages(c('Rserve'), dependencies=TRUE)" \
    && Rscript  -e "install.packages(c('devtools'), dependencies=TRUE)" \
    && Rscript  -e "install.packages(c('jsonlite'), dependencies=TRUE)" \
    && Rscript  -e "install.packages(c('MLmetrics'), dependencies=TRUE)" \
    && Rscript  -e "install.packages(c('colorspace'), dependencies=TRUE)" \	
	&& Rscript  -e "install.packages(c('timeDate'), dependencies=TRUE)" \	
	&& Rscript  -e "install.packages(c('RcppArmadillo'), dependencies=TRUE)" \	
    && Rscript  -e "install.packages('https://cran.r-project.org/src/contrib/Archive/quadprog/quadprog_1.5-5.tar.gz', dependencies=TRUE, repos=NULL, method='libcurl')" \
	&& Rscript  -e "install.packages(c('tseries'), dependencies=TRUE)" \	
	&& Rscript  -e "install.packages(c('fracdiff'), dependencies=TRUE)" \		
    && Rscript  -e "install.packages(c('forecast'), dependencies=TRUE)" \
#    && Rscript  -e "install.packages('https://cran.r-project.org/src/contrib/Archive/forecast/forecast_7.3.tar.gz', dependencies=TRUE, repos=NULL, method='libcurl')" \
    && Rscript  -e "install.packages(c('lubridate'), dependencies=TRUE)"  \
    && Rscript  -e "install.packages('https://cran.r-project.org/src/contrib/Archive/msgpack/msgpack_1.0.tar.gz', repos=NULL, method='libcurl')"

And you also need to add the opensuse and rmsgpack tags. If I remember right the r / 3.3.3 tag in the screenshot is optional.

Many Greetings, Andreas

[former_member727550]

Hello Andreas,

I am also facing the same kind of issue while using the docker on R client Operator

Thank you! this would help to resolve.