Add config options for media retention (#12732)

Author: anoadragon453

changelog.d/12732.feature

@@ -0,0 +1 @@
+Add new `media_retention` options to the homeserver config for routinely cleaning up non-recently accessed media.

docs/usage/configuration/config_documentation.md

@@ -1459,7 +1459,7 @@ federation_rr_transactions_per_room_per_second: 40
 ```
 ---
 ## Media Store ##
-Config options relating to Synapse media store.
+Config options related to Synapse's media store.
 
 ---
 Config option: `enable_media_repo` 
@@ -1563,6 +1563,33 @@ thumbnail_sizes:
     height: 600
     method: scale
 ```
+---
+Config option: `media_retention`
+
+Controls whether local media and entries in the remote media cache
+(media that is downloaded from other homeservers) should be removed
+under certain conditions, typically for the purpose of saving space.
+
+Purging media files will be the carried out by the media worker
+(that is, the worker that has the `enable_media_repo` homeserver config
+option set to 'true'). This may be the main process.
+
+The `media_retention.local_media_lifetime` and
+`media_retention.remote_media_lifetime` config options control whether
+media will be purged if it has not been accessed in a given amount of
+time. Note that media is 'accessed' when loaded in a room in a client, or
+otherwise downloaded by a local or remote user. If the media has never
+been accessed, the media's creation time is used instead. Both thumbnails
+and the original media will be removed. If either of these options are unset,
+then media of that type will not be purged.
+
+Example configuration:
+```yaml
+media_retention:
+    local_media_lifetime: 90d
+    remote_media_lifetime: 14d
+```
+---
 Config option: `url_preview_enabled`
 
 This setting determines whether the preview URL API is enabled.

synapse/config/repository.py

@@ -223,6 +223,22 @@ def read_config(self, config: JsonDict, **kwargs: Any) -> None:
                 "url_preview_accept_language"
             ) or ["en"]
 
+        media_retention = config.get("media_retention") or {}
+
+        self.media_retention_local_media_lifetime_ms = None
+        local_media_lifetime = media_retention.get("local_media_lifetime")
+        if local_media_lifetime is not None:
+            self.media_retention_local_media_lifetime_ms = self.parse_duration(
+                local_media_lifetime
+            )
+
+        self.media_retention_remote_media_lifetime_ms = None
+        remote_media_lifetime = media_retention.get("remote_media_lifetime")
+        if remote_media_lifetime is not None:
+            self.media_retention_remote_media_lifetime_ms = self.parse_duration(
+                remote_media_lifetime
+            )
+
     def generate_config_section(self, data_dir_path: str, **kwargs: Any) -> str:
         assert data_dir_path is not None
         media_store = os.path.join(data_dir_path, "media_store")

synapse/rest/media/v1/media_repository.py

@@ -65,7 +65,12 @@
 logger = logging.getLogger(__name__)
 
 
-UPDATE_RECENTLY_ACCESSED_TS = 60 * 1000
+# How often to run the background job to update the "recently accessed"
+# attribute of local and remote media.
+UPDATE_RECENTLY_ACCESSED_TS = 60 * 1000  # 1 minute
+# How often to run the background job to check for local and remote media
+# that should be purged according to the configured media retention settings.
+MEDIA_RETENTION_CHECK_PERIOD_MS = 60 * 60 * 1000  # 1 hour
 
 
 class MediaRepository:
@@ -122,11 +127,36 @@ def __init__(self, hs: "HomeServer"):
             self._start_update_recently_accessed, UPDATE_RECENTLY_ACCESSED_TS
         )
 
+        # Media retention configuration options
+        self._media_retention_local_media_lifetime_ms = (
+            hs.config.media.media_retention_local_media_lifetime_ms
+        )
+        self._media_retention_remote_media_lifetime_ms = (
+            hs.config.media.media_retention_remote_media_lifetime_ms
+        )
+
+        # Check whether local or remote media retention is configured
+        if (
+            hs.config.media.media_retention_local_media_lifetime_ms is not None
+            or hs.config.media.media_retention_remote_media_lifetime_ms is not None
+        ):
+            # Run the background job to apply media retention rules routinely,
+            # with the duration between runs dictated by the homeserver config.
+            self.clock.looping_call(
+                self._start_apply_media_retention_rules,
+                MEDIA_RETENTION_CHECK_PERIOD_MS,
+            )
+
     def _start_update_recently_accessed(self) -> Deferred:
         return run_as_background_process(
             "update_recently_accessed_media", self._update_recently_accessed
         )
 
+    def _start_apply_media_retention_rules(self) -> Deferred:
+        return run_as_background_process(
+            "apply_media_retention_rules", self._apply_media_retention_rules
+        )
+
     async def _update_recently_accessed(self) -> None:
         remote_media = self.recently_accessed_remotes
         self.recently_accessed_remotes = set()
@@ -835,6 +865,45 @@ async def _generate_thumbnails(
 
         return {"width": m_width, "height": m_height}
 
+    async def _apply_media_retention_rules(self) -> None:
+        """
+        Purge old local and remote media according to the media retention rules
+        defined in the homeserver config.
+        """
+        # Purge remote media
+        if self._media_retention_remote_media_lifetime_ms is not None:
+            # Calculate a threshold timestamp derived from the configured lifetime. Any
+            # media that has not been accessed since this timestamp will be removed.
+            remote_media_threshold_timestamp_ms = (
+                self.clock.time_msec() - self._media_retention_remote_media_lifetime_ms
+            )
+
+            logger.info(
+                "Purging remote media last accessed before"
+                f" {remote_media_threshold_timestamp_ms}"
+            )
+
+            await self.delete_old_remote_media(
+                before_ts=remote_media_threshold_timestamp_ms
+            )
+
+        # And now do the same for local media
+        if self._media_retention_local_media_lifetime_ms is not None:
+            # This works the same as the remote media threshold
+            local_media_threshold_timestamp_ms = (
+                self.clock.time_msec() - self._media_retention_local_media_lifetime_ms
+            )
+
+            logger.info(
+                "Purging local media last accessed before"
+                f" {local_media_threshold_timestamp_ms}"
+            )
+
+            await self.delete_old_local_media(
+                before_ts=local_media_threshold_timestamp_ms,
+                keep_profiles=True,
+            )
+
     async def delete_old_remote_media(self, before_ts: int) -> Dict[str, int]:
         old_media = await self.store.get_remote_media_before(before_ts)