frigate/frigate/storage.py
Claude 6248dbf12f
Delete preview files when emergency storage cleanup removes recordings
When reduce_storage_consumption deletes old recording segments to free
disk space, it now also deletes preview files that overlap the same
time range. Without this, preview mp4 files on the same disk continued
to consume space, causing the storage maintainer to delete progressively
newer recordings while old previews accumulated — resulting in archives
where older periods had previews but no video.

This is particularly impactful for multi-path setups where each camera's
preview directory shares a disk with its recordings.

https://claude.ai/code/session_016bxjbVpx8DqpjysnGYmXdx
2026-03-13 08:51:42 +00:00

443 lines
18 KiB
Python

"""Handle storage retention and usage."""
import logging
import re
import shutil
import threading
from pathlib import Path
from peewee import SQL, fn
from frigate.config import FrigateConfig
from frigate.const import RECORD_DIR, REPLAY_CAMERA_PREFIX
from frigate.models import Event, Previews, Recordings
from frigate.util.builtin import clear_and_unlink
logger = logging.getLogger(__name__)
bandwidth_equation = Recordings.segment_size / (
Recordings.end_time - Recordings.start_time
)
MAX_CALCULATED_BANDWIDTH = 10000 # 10Gb/hr
class StorageMaintainer(threading.Thread):
"""Maintain frigates recording storage."""
def __init__(self, config: FrigateConfig, stop_event) -> None:
super().__init__(name="storage_maintainer")
self.config = config
self.stop_event = stop_event
self.camera_storage_stats: dict[str, dict] = {}
def calculate_camera_bandwidth(self) -> None:
"""Calculate an average MB/hr for each camera."""
for camera in self.config.cameras.keys():
# Skip replay cameras
if camera.startswith(REPLAY_CAMERA_PREFIX):
continue
# cameras with < 50 segments should be refreshed to keep size accurate
# when few segments are available
if self.camera_storage_stats.get(camera, {}).get("needs_refresh", True):
self.camera_storage_stats[camera] = {
"needs_refresh": (
Recordings.select(fn.COUNT("*"))
.where(Recordings.camera == camera, Recordings.segment_size > 0)
.scalar()
< 50
)
}
# calculate MB/hr from last 100 segments
try:
# Subquery to get last 100 segments, then average their bandwidth
last_100 = (
Recordings.select(bandwidth_equation.alias("bw"))
.where(Recordings.camera == camera, Recordings.segment_size > 0)
.order_by(Recordings.start_time.desc())
.limit(100)
.alias("recent")
)
bandwidth = round(
Recordings.select(fn.AVG(SQL("bw"))).from_(last_100).scalar()
* 3600,
2,
)
if bandwidth > MAX_CALCULATED_BANDWIDTH:
logger.warning(
f"{camera} has a bandwidth of {bandwidth} MB/hr which exceeds the expected maximum. This typically indicates an issue with the cameras recordings."
)
bandwidth = MAX_CALCULATED_BANDWIDTH
except TypeError:
bandwidth = 0
self.camera_storage_stats[camera]["bandwidth"] = bandwidth
logger.debug(f"{camera} has a bandwidth of {bandwidth} MiB/hr.")
def calculate_camera_usages(self) -> dict[str, dict]:
"""Calculate the storage usage of each camera."""
usages: dict[str, dict] = {}
for camera in self.config.cameras.keys():
# Skip replay cameras
if camera.startswith(REPLAY_CAMERA_PREFIX):
continue
camera_storage = (
Recordings.select(fn.SUM(Recordings.segment_size))
.where(Recordings.camera == camera, Recordings.segment_size != 0)
.scalar()
)
camera_key = (
getattr(self.config.cameras[camera], "friendly_name", None) or camera
)
usages[camera_key] = {
"usage": camera_storage,
"bandwidth": self.camera_storage_stats.get(camera, {}).get(
"bandwidth", 0
),
}
return usages
def calculate_camera_usages_by_root(self) -> dict[str, dict]:
"""Calculate camera storage usage grouped by recordings root."""
root_usages: dict[str, dict] = {}
root_camera_stats = (
Recordings.select(
Recordings.camera,
Recordings.path,
fn.SUM(Recordings.segment_size).alias("usage"),
)
.where(Recordings.segment_size != 0)
.group_by(Recordings.camera, Recordings.path)
.namedtuples()
.iterator()
)
for stat in root_camera_stats:
camera = stat.camera
# Skip replay cameras
if camera.startswith(REPLAY_CAMERA_PREFIX):
continue
root_path = self._get_recordings_root_from_path(stat.path, camera)
if not root_path:
continue
camera_usage = stat.usage or 0
camera_config = self.config.cameras.get(camera)
camera_friendly_name = (
getattr(camera_config, "friendly_name", None) if camera_config else None
)
if root_path not in root_usages:
root_usages[root_path] = {
"path": root_path,
"is_default": root_path == RECORD_DIR,
"recordings_size": 0,
"cameras": [],
"configured_cameras": [],
"camera_usages": {},
}
root_usages[root_path]["recordings_size"] += camera_usage
if camera not in root_usages[root_path]["camera_usages"]:
root_usages[root_path]["cameras"].append(camera)
root_usages[root_path]["camera_usages"][camera] = {
"usage": 0,
"bandwidth": self.camera_storage_stats.get(camera, {}).get(
"bandwidth", 0
),
"friendly_name": camera_friendly_name,
}
root_usages[root_path]["camera_usages"][camera]["usage"] += camera_usage
for root_path, root in root_usages.items():
root["cameras"] = sorted(root["cameras"])
root["configured_cameras"] = sorted(
[
camera
for camera in self.config.cameras.keys()
if self.config.get_camera_recordings_path(camera) == root_path
]
)
return root_usages
def _get_recordings_root_from_path(self, recording_path: str, camera: str) -> str:
# Prefer configured recording roots when available.
for configured_root in sorted(
self.config.get_recordings_paths(), key=len, reverse=True
):
if recording_path == configured_root or recording_path.startswith(
f"{configured_root}/"
):
return configured_root.rstrip("/") or "/"
# Support layouts like /root/YYYY-MM-DD/HH/... and normalize to /root.
date_hour_match = re.match(
r"^(?P<root>.+?)/\d{4}-\d{2}-\d{2}/\d{2}(?:/|$)", recording_path
)
if date_hour_match:
return date_hour_match.group("root").rstrip("/") or "/"
camera_segment = f"/{camera}/"
if camera_segment in recording_path:
return recording_path.split(camera_segment, 1)[0].rstrip("/") or "/"
# Fallback for unexpected path layouts; expected format is root/camera/date/file
path = Path(recording_path)
if len(path.parents) >= 3:
return str(path.parents[2]).rstrip("/") or "/"
return str(path.parent).rstrip("/") or "/"
def _get_path_bandwidths(self) -> dict[str, float]:
bandwidth_per_path: dict[str, float] = {}
for camera, stats in self.camera_storage_stats.items():
if camera not in self.config.cameras:
continue
path = self.config.get_camera_recordings_path(camera)
bandwidth_per_path[path] = bandwidth_per_path.get(path, 0) + stats.get(
"bandwidth", 0
)
return bandwidth_per_path
def check_storage_needs_cleanup(self, recordings_root: str) -> bool:
"""Return True if the given recordings root path needs cleanup."""
# currently runs cleanup if less than 1 hour of space is left
# disk_usage should not spin up disks
hourly_bandwidth = self._get_path_bandwidths().get(recordings_root, 0)
if not hourly_bandwidth:
return False
try:
remaining_storage = round(
shutil.disk_usage(recordings_root).free / pow(2, 20), 1
)
except (FileNotFoundError, OSError):
return False
logger.debug(
f"Storage cleanup check: {hourly_bandwidth} hourly with remaining storage: {remaining_storage} for path {recordings_root}."
)
return remaining_storage < hourly_bandwidth
def reduce_storage_consumption(self, recordings_root: str) -> None:
"""Remove oldest hour of recordings."""
logger.debug("Starting storage cleanup.")
deleted_segments_size = 0
hourly_bandwidth = self._get_path_bandwidths().get(recordings_root, 0)
recordings: Recordings = (
Recordings.select(
Recordings.id,
Recordings.camera,
Recordings.start_time,
Recordings.end_time,
Recordings.segment_size,
Recordings.path,
)
.where(Recordings.path.startswith(f"{recordings_root}/"))
.order_by(Recordings.start_time.asc())
.namedtuples()
.iterator()
)
retained_events: Event = (
Event.select(
Event.start_time,
Event.end_time,
)
.where(
Event.retain_indefinitely == True,
Event.has_clip,
)
.order_by(Event.start_time.asc())
.namedtuples()
)
event_start = 0
deleted_recordings = []
for recording in recordings:
# check if 1 hour of storage has been reclaimed
if deleted_segments_size > hourly_bandwidth:
break
keep = False
# Now look for a reason to keep this recording segment
for idx in range(event_start, len(retained_events)):
event = retained_events[idx]
# if the event starts in the future, stop checking events
# and let this recording segment expire
if event.start_time > recording.end_time:
keep = False
break
# if the event is in progress or ends after the recording starts, keep it
# and stop looking at events
if event.end_time is None or event.end_time >= recording.start_time:
keep = True
break
# if the event ends before this recording segment starts, skip
# this event and check the next event for an overlap.
# since the events and recordings are sorted, we can skip events
# that end before the previous recording segment started on future segments
if event.end_time < recording.start_time:
event_start = idx
# Delete recordings not retained indefinitely
if not keep:
try:
clear_and_unlink(Path(recording.path), missing_ok=False)
deleted_recordings.append(recording)
deleted_segments_size += recording.segment_size
except FileNotFoundError:
# this file was not found so we must assume no space was cleaned up
pass
# check if need to delete retained segments
if deleted_segments_size < hourly_bandwidth:
logger.error(
f"Could not clear {hourly_bandwidth} MB, currently {deleted_segments_size} MB have been cleared. Retained recordings must be deleted."
)
recordings = (
Recordings.select(
Recordings.id,
Recordings.camera,
Recordings.start_time,
Recordings.end_time,
Recordings.path,
Recordings.segment_size,
)
.where(Recordings.path.startswith(f"{recordings_root}/"))
.order_by(Recordings.start_time.asc())
.namedtuples()
.iterator()
)
for recording in recordings:
if deleted_segments_size > hourly_bandwidth:
break
try:
clear_and_unlink(Path(recording.path), missing_ok=False)
deleted_segments_size += recording.segment_size
deleted_recordings.append(recording)
except FileNotFoundError:
# this file was not found so we must assume no space was cleaned up
pass
else:
logger.info(f"Cleaned up {deleted_segments_size} MB of recordings")
logger.debug(f"Expiring {len(deleted_recordings)} recordings")
# delete up to 100,000 at a time
max_deletes = 100000
# Update has_clip for events that overlap with deleted recordings
if deleted_recordings:
# Group deleted recordings by camera
camera_recordings = {}
for recording in deleted_recordings:
if recording.camera not in camera_recordings:
camera_recordings[recording.camera] = {
"min_start": recording.start_time,
"max_end": recording.end_time,
}
else:
camera_recordings[recording.camera]["min_start"] = min(
camera_recordings[recording.camera]["min_start"],
recording.start_time,
)
camera_recordings[recording.camera]["max_end"] = max(
camera_recordings[recording.camera]["max_end"],
recording.end_time,
)
# Find all events that overlap with deleted recordings time range per camera
events_to_update = []
for camera, time_range in camera_recordings.items():
overlapping_events = Event.select(Event.id).where(
Event.camera == camera,
Event.has_clip == True,
Event.start_time < time_range["max_end"],
Event.end_time > time_range["min_start"],
)
for event in overlapping_events:
events_to_update.append(event.id)
# Update has_clip to False for overlapping events
if events_to_update:
for i in range(0, len(events_to_update), max_deletes):
batch = events_to_update[i : i + max_deletes]
Event.update(has_clip=False).where(Event.id << batch).execute()
logger.debug(
f"Updated has_clip to False for {len(events_to_update)} events"
)
# Also delete preview files that overlap with deleted recordings so they
# don't continue to consume space on the same disk after the recordings
# are gone (especially important for multi-path setups where preview and
# recordings share the same disk).
if deleted_recordings:
deleted_previews = []
for camera, time_range in camera_recordings.items():
overlapping_previews = (
Previews.select(Previews.id, Previews.path)
.where(
Previews.camera == camera,
Previews.start_time < time_range["max_end"],
Previews.end_time > time_range["min_start"],
)
.namedtuples()
)
for preview in overlapping_previews:
clear_and_unlink(Path(preview.path), missing_ok=True)
deleted_previews.append(preview.id)
logger.debug(f"Expiring {len(deleted_previews)} previews")
for i in range(0, len(deleted_previews), max_deletes):
Previews.delete().where(
Previews.id << deleted_previews[i : i + max_deletes]
).execute()
deleted_recordings_list = [r.id for r in deleted_recordings]
for i in range(0, len(deleted_recordings_list), max_deletes):
Recordings.delete().where(
Recordings.id << deleted_recordings_list[i : i + max_deletes]
).execute()
def run(self):
"""Check every 5 minutes if storage needs to be cleaned up."""
self.calculate_camera_bandwidth()
while not self.stop_event.wait(300):
if not self.camera_storage_stats or True in [
r["needs_refresh"] for r in self.camera_storage_stats.values()
]:
self.calculate_camera_bandwidth()
logger.debug(f"Default camera bandwidths: {self.camera_storage_stats}.")
for recordings_root in self.config.get_recordings_paths():
if self.check_storage_needs_cleanup(recordings_root):
logger.info(
f"Less than 1 hour of recording space left for {recordings_root}, running storage maintenance..."
)
self.reduce_storage_consumption(recordings_root)
logger.info("Exiting storage maintainer...")