ShinyProxy - error: "there is no package called ‘tidyverse’"

Hi everyone,

I’m trying to host a shiny-app on digitalocean using shiny proxy. (I should add as a disclaimer that my programming experience until now is limited to data cleaning in R/python and I’m a bit overwhelmed by this whole process.)

I’ve downloaded the template described in and tried to apply it to my own app by creating a local directory and replacing the “euler” app with my own app and any references to it in the Dockerfile with the name of my own app directory. My app uses quite a few additional packages, and I have added those to the Dockerfile in two separate locations as follows:

# basic shiny functionality
RUN R -e "install.packages(c('shiny', 'rmarkdown', 'tidyverse', 'ggplot2', 'plotly', 'leaflet', 'dplyr', 'sf', 'htmlwidgets', 'shinythemes', 'shinyWidgets', 'shinydashboard', 'maps', 'RColorBrewer', 'jcolors', 'rsconect', 'data.table', 'shinyjs'), repos='')"

# install dependencies of the app
RUN R -e "install.packages(c('Rmpfr', 'tidyverse', 'ggplot2', 'plotly', 'leaflet', 'dplyr', 'sf', 'htmlwidgets', 'shinythemes', 'shinyWidgets', 'shinydashboard', 'maps', 'RColorBrewer', 'jcolors', 'rsconect', 'data.table', 'shinyjs'), repos='')"

I’ve been able to create a Docker image using this file. But every time I try to open it in my browser the following error appears:

An error has occurred!

there is no package called ‘tidyverse’

I’m not sure why this is happening or how to resolve it since I did ask the Docker file to install tidyverse and all the other packages (tidyverse is the first package my App tries to load which I presume is the reason it’s the one referenced in the error).

Hi @pk94,

I would check the Docker build logs carefully. This usually means the tidyverse package could not be installed because one of its dependencies could not be installed. Usually this happens when a system dependency is not available in the Docker image (e.g. libxml2-dev etc.). The confusion probably comes from the fact that the Docker build will just continue even though there were errors for some R package installs (but you should be able to see that in the Docker build logs).

Hope this helps!