Variables, setting the VHDX path for the image, for SQL Server full backups
For a sample, see the comments in the dockerfile below.
dockerfile
FROM mssql-2016
# Puts the VHDs and clones for this image in the specified directory
# ENV IMAGE_VHDX_DIRECTORY=path\to\vhds\for\this\image
# This line defines the image as a full backup only image and cannot be updated with transaction log backups
# Full backup images allow you to run SQL scripts at image build time # Remove this line if you want to update the image periodically with transaction log backups
# For "transaction log backup images", SQL scripts must be run at container create time (RUN must be below ENV USE_DOCKERFILE_TO_CREATE_CONTAINER=1)
ENV FULL_BACKUPS_ONLY=1
USE_DOCKERFILE_TO_CREATE_CONTAINER=1 # The $ variables listed below are populated by Windocks automatically
# You can use them anywhere in the docker file below the USE_DOCKERFILE_TO_CREATE_CONTAINER=1
# $ContainerDir $ContainerPort $ContainerId, $ImageName, $ContainerName
# You may also define your own variables and use them in the dockerfile.
# Users creating the containers can provide the values for those variables in the web application UI or in the command line -e or in the REST API(enterprise edition only)
# The convention is to put the variable name in between the delimiters STARTENV_ and _ENDENV
RUN script.ps1 STARTENV_VARIABLE1_ENDENV, STARTENV_VARIABLE2_ENDENV