This section shows which persistent data are exposed and where. It also recommends some ways to persist, share, and back up the data. For backup strategies and recommendations, see the Business Continuity & Disaster Recovery section.
Persistent data is different depending if you use the Linux version or the Windows version.
The Linux version has a more managed way to persist the database's data. Instead of only persisting the
.ldffile of the WorkflowGen database, you instead persist all of the SQL Server state data, including the master database files. For more information on how to persist the data in the Linux, see Configure SQL Server container images on Docker in the official Microsoft documentation for SQL Server.
Since the Windows version's base image hasn't been updated by Microsoft for quite some time, WorkflowGen's database files have to be persisted manually. Therefore, the database container has a defined volume that will contain only the WorkflowGen database files. As well, it's recommended to use SQL Server's contained database feature to keep the credentials at the database level instead of at the SQL Server system level. For more information about the contained database feature, see contained database authentication Server Configuration Option in the official Microsoft documentation.
Volumes are handled differently with Windows Containers compared to Linux Containers. The permission model changes and is different depending on whether you use process isolation or Hyper-V isolation. Before continuing with the procedures in this section, you should read the Microsoft documentation on Windows Server Container Storage.