Add config option to prevent media downloads from listed domains. (#15197)

This stops media (and thumbnails) from being accessed from the
listed domains. It does not delete any already locally cached media,
but will prevent accessing it.

Note that admin APIs are unaffected by this change.

3 files changed, 22 insertions, 0 deletions

diff --git a/synapse/config/repository.py b/synapse/config/repository.py
index ecb3edbe3a..655f06505b 100644
--- a/synapse/config/repository.py
+++ b/synapse/config/repository.py
@@ -137,6 +137,10 @@ class ContentRepositoryConfig(Config):
         self.max_image_pixels = self.parse_size(config.get("max_image_pixels", "32M"))
         self.max_spider_size = self.parse_size(config.get("max_spider_size", "10M"))
 
+        self.prevent_media_downloads_from = config.get(
+            "prevent_media_downloads_from", []
+        )
+
         self.media_store_path = self.ensure_directory(
             config.get("media_store_path", "media_store")
         )
diff --git a/synapse/media/media_repository.py b/synapse/media/media_repository.py
index b81e3c2b0c..e81c987b10 100644
--- a/synapse/media/media_repository.py
+++ b/synapse/media/media_repository.py
@@ -93,6 +93,7 @@ class MediaRepository:
         self.federation_domain_whitelist = (
             hs.config.federation.federation_domain_whitelist
         )
+        self.prevent_media_downloads_from = hs.config.media.prevent_media_downloads_from
 
         # List of StorageProviders where we should search for media and
         # potentially upload to.
@@ -276,6 +277,14 @@ class MediaRepository:
         ):
             raise FederationDeniedError(server_name)
 
+        # Don't let users download media from domains listed in the config, even
+        # if we might have the media to serve. This is Trust & Safety tooling to
+        # block some servers' media from being accessible to local users.
+        # See `prevent_media_downloads_from` config docs for more info.
+        if server_name in self.prevent_media_downloads_from:
+            respond_404(request)
+            return
+
         self.mark_recently_accessed(server_name, media_id)
 
         # We linearize here to ensure that we don't try and download remote
diff --git a/synapse/rest/media/thumbnail_resource.py b/synapse/rest/media/thumbnail_resource.py
index a6396fb05a..661e604b85 100644
--- a/synapse/rest/media/thumbnail_resource.py
+++ b/synapse/rest/media/thumbnail_resource.py
@@ -60,6 +60,7 @@ class ThumbnailResource(DirectServeJsonResource):
         self.media_storage = media_storage
         self.dynamic_thumbnails = hs.config.media.dynamic_thumbnails
         self._is_mine_server_name = hs.is_mine_server_name
+        self.prevent_media_downloads_from = hs.config.media.prevent_media_downloads_from
 
     async def _async_render_GET(self, request: SynapseRequest) -> None:
         set_cors_headers(request)
@@ -82,6 +83,14 @@ class ThumbnailResource(DirectServeJsonResource):
                 )
             self.media_repo.mark_recently_accessed(None, media_id)
         else:
+            # Don't let users download media from configured domains, even if it
+            # is already downloaded. This is Trust & Safety tooling to make some
+            # media inaccessible to local users.
+            # See `prevent_media_downloads_from` config docs for more info.
+            if server_name in self.prevent_media_downloads_from:
+                respond_404(request)
+                return
+
             if self.dynamic_thumbnails:
                 await self._select_or_generate_remote_thumbnail(
                     request, server_name, media_id, width, height, method, m_type