Uploads are on disk only from 1.11.0

content/docs/host/cli.mdx

@@ -134,8 +134,10 @@ follows its output, which is where to look when voice stops working. `x` stops
 it for every server at once, and says so.
 
 A machine with servers made by an older version of `gryt` has a second row,
-**Object store**. That's the MinIO those servers keep their uploads in, and it
-stays for as long as one of them uses it.
+**Object store**. That's the MinIO those servers kept their uploads in. A
+server on 1.11.0 copies its uploads out of it on its first start, and the next
+`gryt start` or `gryt pull` switches that server over to disk. The row goes once
+the last one has moved. MinIO's volume stays until `gryt remove`.
 
 A media worker runs beside each server rather than being shared, because it
 reads that server's own database. It has no row: `l` on the server already

content/docs/host/configuration.mdx

@@ -188,17 +188,18 @@ One for a single reverse proxy. More only if you genuinely chained them.
 
 ## Uploads
 
-Uploads are files under `DATA_DIR`, in a folder named by `S3_BUCKET`. The
-compose files and the `gryt` CLI set both of these:
-
-```bash
-STORAGE_BACKEND=filesystem
-S3_BUCKET=gryt
-```
-
-The server still defaults to `STORAGE_BACKEND=s3` when it's unset, so set it if
-you run the server some other way. The folder's setting is called `S3_BUCKET`
-because it used to name a bucket.
+Uploads are files under `DATA_DIR`, in a folder named by `S3_BUCKET`, which is
+`gryt` if you leave it out. There's nothing else to choose: since 1.11.0 the
+server only stores files on disk. The setting is called `S3_BUCKET` because it
+used to name a bucket.
+
+A server that still has S3 settings (`S3_ENDPOINT`, an S3 key, or
+`STORAGE_BACKEND=s3`) copies its bucket into that folder on its first start on
+1.11.0, before it accepts connections. It checks each file's size, keeps each
+content type, and leaves the bucket alone. If the copy fails it doesn't start,
+and the log says why. Once it's done you can remove the S3 settings.
+
+`DISABLE_UPLOADS=true` turns uploads off.
 
 Keys look like `uploads/<uuid>.<ext>`, and each file has a `.meta` file beside
 it holding its content type. The media worker reads and writes the same folder,

content/docs/host/docker-compose.mdx

@@ -527,6 +527,11 @@ discovery matters for both, put them on separate machines or move them to the br
 Older copies of this compose file ran MinIO for uploads. The new one moves them
 out by itself the first time it starts, so upgrading is the usual pull and up.
 
+If you only pull the new images and keep the old compose file, the server
+does the same copy itself on its first start on 1.11.0, through the MinIO that's
+still running. Either way the result is the same folder and the same
+`.migrated-from-minio` marker, so the other step sees it's done.
+
 <Steps>
 
 <Step>

content/docs/host/quick-start.mdx

@@ -152,7 +152,7 @@ cd gryt
 ops/start_dev.sh
 ```
 
-That runs MinIO, the SFU, two servers and the Vite client in a tmux session, and
+That runs the SFU, two servers and the Vite client in a tmux session, and
 the client is on [localhost:3666](http://localhost:3666).
 
 ## Behind a tunnel

content/docs/host/server.mdx

@@ -78,9 +78,8 @@ for what a shorter window costs a device that's been offline.
 
 | Variable | Default | What it does |
 |----------|---------|--------------|
-| `STORAGE_BACKEND` | `s3` | Set it to `filesystem`. The compose files and the CLI do |
-| `S3_BUCKET` | — | The folder under `DATA_DIR` that uploads go in, `gryt` in the compose files |
-| `DISABLE_S3` | — | Turn uploads off entirely |
+| `S3_BUCKET` | `gryt` | The folder under `DATA_DIR` that uploads go in |
+| `DISABLE_UPLOADS` | — | Turn uploads off entirely. `DISABLE_S3` still works as the old name |
 | `IMAGE_WORKER_URL` | — | Where the media worker listens |
 | `MEDIA_SWEEP_INTERVAL_MS` | `600000` | How often orphaned media is swept |
 | `MEDIA_SWEEP_GRACE_MS` | `1800000` | How long an orphan is left alone first |