Practical self-hosting · beginner
06 · Recreate the container, keep the story
Distinguish replaceable processes from durable application state.
What you can take away
- Map each data item to its storage and recovery method.
- Make a data map for a fictional practice site with uploads, a database, configuration, cache, and generated thumbnails
Your trail through this idea
Checking saved progress…
A container is replaced and the family’s uploads disappear. The service restarted, but the information did not survive. This is the difference between process recovery and data recovery.
Prerequisite: complete the preceding lesson or bring its worksheet.
Understand the system
Docker volumes persist independently of a container’s writable layer. Bind mounts connect a selected host path to a container path. Both can hold important data, but they differ in who manages the location and how portable the configuration is. A mount can also obscure files already present at the target path.
Persistence is not backup. A volume on the same failed disk does not help recover that disk. Nor does a copy of a running database directory necessarily represent an application-consistent backup. Identify the application’s supported backup mechanism before treating a folder copy as sufficient.
Work through the case
The current learning-loader source provides a concrete design example: modules are read from a configured directory outside the application image. That separates editable lesson content from application builds. It does not verify which host path is mounted on the live site. The exact live mount remains an operator observation.
Make something useful
Try this
Original exercise
Make a data map for a fictional practice site with uploads, a database, configuration, cache, and generated thumbnails. For every row, decide whether it is authoritative, reproducible, or disposable; name its storage location abstractly; and specify a restore check. Use labels such as 'uploads volume' rather than private filesystem paths. Challenge one another’s classification: can thumbnails really be regenerated without the original photos?
Check your reasoning
A quick check · for your own learning
Why is a persistent volume not automatically a backup?
A self-study prompt. You can retry; this is not an exam.
Keep the result and follow the sources
Success is a map that survives a container replacement and explains what survives a disk failure. Never use volume deletion as a casual cleanup step while learning. The backup module will turn this map into a recovery experiment.
Sources and scope
- Primary documentation or archival source, reviewed 14 November 2026. Additional direct links appear beside the claims they support.
- Editorial basis: Ian's historical self-hosting questions and current repository inspection informed the case selection. Historical requests do not establish current deployment. Private chat records, credentials and host details are excluded.
- Examples and exercises are original educational synthesis. External sources support the stated technical or historical context, not every hypothetical scenario.