Trouble accessing R Studio

If your R Studio won’t load, particilarly first thing in the morning, it is likely that the virtual machine has not been started for the day. You will need to log in to AWS at https://sso-int-sce-network.awsapps.com/start/#/ and then complete the following steps:

  1. Access the EC2 app. This may be in your “Recently visited” apps pane, but if not use the search bar at the top and search “EC2”

  2. In the Resources pane, select “Instances”. NOTE: don’t select Instances (running) as the instance of the VM is not currently running

  3. In the search bar which says “Find instance by atribute or tag”, type the name of the virtual machine (most likely ranch-793) and press enter.

  4. Tick the checkbox on the left of the instance name, click “Instance state” on the top right side of the screen and select “Start instance”.

  5. Wait about 30 seconds and refresh your R Studio, it should now connect. If it doesn’t try again after waiting a few seconds.

Troubleshooting Problems Loading/Writing data to S3

If you encounter the error:

Error in parse_aws_s3_response(r, Sig, verbose = verbose)
Moved Permanently (HTTP 301).

While using functions from the “aws.s3” package such as “aws.s3::s3read_using()“

  1. Ensure that the following package is installed:

    install.packages("aws.ec2metadata")
library(aws.ec2metadata)

    You can check this in the bottom right under the packages tab:

    to ensure you install to a share library you need to specify the location of the shared library like so:

    install.packages("aws.ec2metadata", lib = "/home/R_package_lib")

  2. Also check that the packages aws.signature and aws.s3 are also installed.

    Run the following code:

    Sys.setenv("AWS_DEFAULT_REGION" = 'eu-west-1')

  3. If there are still issues with the package accessing the data, check the Virtual machine can access the Ranch bucket you need to pull data from.

    You can do this from the Linux Console with a command like this (Replace the XXX with the relevant bucket’s ranch number):

    aws s3 ls s3://s3-ranch-XXX/

    If the output shows you a list of folder names, the VM has access, eg:

  4. If step 3 shows you there is no access, request that your VM be given read/Write access to the relevant buckets on the SCE Slack server. You will need the owner of the VM and the Buckets to approve the request.