Skip to content

Latest commit

 

History

History
120 lines (94 loc) · 4.32 KB

UsingDocker.md

File metadata and controls

120 lines (94 loc) · 4.32 KB

Using rsqlserver with Docker

This file contains instructions on building a Docker container containing 'rsqlserver' side-by-side with an MS SQL Server database

If you have any problems or questions please raise an issue

Command Line R

Pull the latest Docker images

Due to the size of the Docker images, this may take some time depending on your internet speed however this "pull" step is a once-off and further updates to the images will make use of your existing local caches.

docker pull ruaridhw/rsqlserver:latest && docker pull microsoft/mssql-server-linux:latest

Start up the server container

Your Docker instance will need to be allocated at least 3-4GB of memory in order for the database to successfully start up.

docker run -e 'ACCEPT_EULA=Y' -e 'MSSQL_SA_PASSWORD=Password12!' -h mydockermsdb -p 1433:1433 --name mssqldb -d microsoft/mssql-server-linux

Edit the database password, hostname -h and container name --name parameters if required. It is possible (and advisable) to change the database password later once running to avoid showing it as plaintext.

Run a query against the server

docker exec -t mssqldb /opt/mssql-tools/bin/sqlcmd \
   -S localhost -U SA -P 'Password12!' \
   -Q "CREATE DATABASE rsqlserverdb;
       GO
       USE rsqlserverdb;
       CREATE TABLE Inventory (id INT, name NVARCHAR(50), quantity INT);
       INSERT INTO Inventory VALUES (1, 'banana', 150), (2, 'orange', 154);
       GO
       SELECT * FROM Inventory WHERE quantity > 152;"
#> Changed database context to 'rsqlserverdb'.
#>
#> (2 rows affected)
#> id          name                                               quantity
#> ----------- -------------------------------------------------- -----------
#>          2 orange                                                     154
#>
#> (1 rows affected)

Run a command in the rsqlserver R session container

docker run --name testrsqlserver --link=mssqldb --rm ruaridhw/rsqlserver Rscript \
   -e "library(rsqlserver)" \
   -e "con <- dbConnect('SqlServer', host = 'mydockermsdb', dbname = 'rsqlserverdb', user = 'SA', password = 'Password12!')" \
   -e "dbReadTable(con, 'Inventory')"
#> Loading required package: methods
#> Loading required package: rClr
#> Assembly '/usr/local/lib/R/site-library/rClr/libs/ClrFacade.dll' doesn't have an entry point.
#> Loading the dynamic library for Mono runtime...
#> Loaded Common Language Runtime version 4.0.30319.17020
#>   id   name quantity
#> 1  1 banana      150
#> 2  2 orange      154

The "Assembly entry point" warning message is a bug with rClr and can be ignored

Re-enter the R session interactively

docker run --name rsqlserver --link=mssqldb -i ruaridhw/rsqlserver
#> R version 3.4.2 (2017-09-28) -- "Short Summer"
#> Copyright (C) 2017 The R Foundation for Statistical Computing
#> Platform: x86_64-pc-linux-gnu (64-bit)
#> ...
#> >

RStudio

In order to use RStudio instead for easier interactivity over command line R, you can replace the Dockerfile in this repository with the two files located here

The "Pull" command now requires a build from the local Dockerfile:

docker build -t rsqlserver-rstudio . && docker pull microsoft/mssql-server-linux:latest

For the run command, it is possible to persist a local directory on your host machine through to the container and have any updates to either directory immediately reflected in the other instance:

docker run -d -p 8787:8787 --name rsqlstudio --link=mssqldb \
   --mount type=bind,source=/path/to/local/rsqlserver,destination=/home/rstudio/rsqlserver \
   rsqlserver-rstudio

In this example, /path/to/local/rsqlserver is a copy of the repository on the host machine which is replicated at /home/rstudio/rsqlserver on the container

The RStudio server will run as a service on the container so simply open a local browser window pointing to http://localhost:8787 and login using the username and password "rstudio"

When you are done, call docker stop rsqlstudio to stop the server and docker start rsqlstudio whenever you need it again. The run command is only for the container instantiation.

Tested in the following environments:

R version 3.4.1 (2017-06-30)
Platform: x86_64-apple-darwin15.6.0 (64-bit)
Operating System: macOS Sierra 10.12.6