Increased CPQ and memory usage since 1.71.0

[purnhar]

Since the update to version 1.71.0 the borg-ui container uses more memory and CPU (see attached screenshor). Yesterday it went totally erratic and allocated all the memory it could get on the server (see attached screenshot). Unfortunately, I don't have any logs for this timeframe anymore.

[karanhudia]

@purnhar Nothing has changed in the last release which could cause memory issues/leaks. Could you please try restarting your docker as well?

[purnhar]

I restarted the complete host yesterday after the container consumed it's whole memory. I just updated to 1.72.2 and will give you an update tomorrow.

Are there any debug settings I could activate?

[purnhar]

The behavior is still the same with the current 1.72.2 version. It does not seem to be a serious problem so it's up to you to just close the issue.

[karanhudia]

@purnhar Do you have logs available of this time? Or did you perform some actions in borg-ui during this time? For example- "Archive browsing", "Mount", etc. The operations which could potentially consume RAM.

[purnhar]

Here are the log entries from one of the CPU spikes. It happens once every hour. Seems not to be a real problem.

borg  | INFO:app.services.stats_refresh_scheduler:{"count": 4, "time": "datetime.datetime(2026, 3, 1, 20, 39, 43, 137932)", "event": "Starting repository stats refresh", "logger": "app.services.stats_refresh_scheduler", "level": "info", "timestamp": "2026-03-01T20:39:43.138744Z"}
borg  | INFO:app.core.borg:{"command": "borg list ssh://XXXXXX@XXXXXX.your-storagebox.de:23/home/docker --json", "cwd": null, "event": "Executing command", "logger": "app.core.borg", "level": "info", "timestamp": "2026-03-01T20:39:43.140737Z"}
borg  | INFO:app.core.borg:{"command": "borg list ssh://XXXXXX@XXXXXX.your-storagebox.de:23/home/docker --json", "event": "Command executed successfully", "logger": "app.core.borg", "level": "info", "timestamp": "2026-03-01T20:39:46.148508Z"}
borg  | INFO:app.api.repositories:{"repository": "docker", "archive_count_old": 15, "archive_count_new": 15, "size_old": "1.74 GB", "size_new": "1.74 GB", "last_backup_old": "datetime.datetime(2026, 3, 1, 4, 0, 47)", "last_backup_new": "datetime.datetime(2026, 3, 1, 4, 0, 47)", "event": "Updated repository stats", "logger": "app.api.repositories", "level": "info", "timestamp": "2026-03-01T20:39:49.640681Z"}
borg  | INFO:app.core.borg:{"command": "borg list ssh://XXXXXX@XXXXXX.your-storagebox.de:23/home/fotos --json", "cwd": null, "event": "Executing command", "logger": "app.core.borg", "level": "info", "timestamp": "2026-03-01T20:39:49.647594Z"}
borg  | INFO:app.core.borg:{"command": "borg list ssh://XXXXXX@XXXXXX.your-storagebox.de:23/home/fotos --json", "event": "Command executed successfully", "logger": "app.core.borg", "level": "info", "timestamp": "2026-03-01T20:39:53.769873Z"}
borg  | INFO:app.api.repositories:{"repository": "fotos", "archive_count_old": 16, "archive_count_new": 16, "size_old": "245.82 GB", "size_new": "245.82 GB", "last_backup_old": "datetime.datetime(2026, 3, 1, 4, 15, 47)", "last_backup_new": "datetime.datetime(2026, 3, 1, 4, 15, 47)", "event": "Updated repository stats", "logger": "app.api.repositories", "level": "info", "timestamp": "2026-03-01T20:39:57.839817Z"}
borg  | INFO:app.core.borg:{"command": "borg list ssh://XXXXXX@XXXXXX.your-storagebox.de:23/home/nextcloud --json", "cwd": null, "event": "Executing command", "logger": "app.core.borg", "level": "info", "timestamp": "2026-03-01T20:39:57.847067Z"}
borg  | INFO:app.core.borg:{"command": "borg list ssh://XXXXXX@XXXXXX.your-storagebox.de:23/home/nextcloud --json", "event": "Command executed successfully", "logger": "app.core.borg", "level": "info", "timestamp": "2026-03-01T20:40:00.638206Z"}
borg  | INFO:app.main:{"method": "GET", "path": "/", "client_ip": "127.0.0.1", "event": "request_received", "logger": "app.main", "level": "info", "timestamp": "2026-03-01T20:40:01.864911Z"}
borg  | INFO:app.main:{"method": "GET", "path": "/", "status_code": 200, "event": "request_completed", "logger": "app.main", "level": "info", "timestamp": "2026-03-01T20:40:01.939747Z"}
borg  | INFO:app.api.repositories:{"repository": "nextcloud", "archive_count_old": 16, "archive_count_new": 16, "size_old": "7.50 GB", "size_new": "7.50 GB", "last_backup_old": "datetime.datetime(2026, 3, 1, 1, 31, 9)", "last_backup_new": "datetime.datetime(2026, 3, 1, 1, 31, 9)", "event": "Updated repository stats", "logger": "app.api.repositories", "level": "info", "timestamp": "2026-03-01T20:40:03.877932Z"}
borg  | INFO:app.core.borg:{"command": "borg list ssh://XXXXXX@XXXXXX.your-storagebox.de:23/home/archiv --json", "cwd": null, "event": "Executing command", "logger": "app.core.borg", "level": "info", "timestamp": "2026-03-01T20:40:03.881701Z"}
borg  | INFO:app.core.borg:{"command": "borg list ssh://XXXXXX@XXXXXX.your-storagebox.de:23/home/archiv --json", "event": "Command executed successfully", "logger": "app.core.borg", "level": "info", "timestamp": "2026-03-01T20:40:06.987952Z"}
borg  | INFO:app.api.repositories:{"repository": "archiv", "archive_count_old": 1, "archive_count_new": 1, "size_old": "289.16 GB", "size_new": "289.16 GB", "last_backup_old": "datetime.datetime(2025, 12, 1, 6, 9, 35)", "last_backup_new": "datetime.datetime(2025, 12, 1, 6, 9, 35)", "event": "Updated repository stats", "logger": "app.api.repositories", "level": "info", "timestamp": "2026-03-01T20:40:09.948692Z"}
borg  | INFO:app.services.stats_refresh_scheduler:{"event": "Synced MQTT state after stats refresh", "logger": "app.services.stats_refresh_scheduler", "level": "info", "timestamp": "2026-03-01T20:40:09.964651Z"}
borg  | INFO:app.services.stats_refresh_scheduler:{"total": 4, "success": 4, "errors": 0, "duration_repos": 4, "event": "Completed repository stats refresh", "logger": "app.services.stats_refresh_scheduler", "level": "info", "timestamp": "2026-03-01T20:40:09.965241Z"}

[karanhudia]

The hourly CPU spikes are the stats refresh scheduler running borg list + borg info for each of your 4 repos. This is expected behavior that can be tuned or disabled under Settings → System → Stats Refresh Interval.

The gradual memory increase is Python's normal allocator behavior (freed memory isn't returned to the OS immediately), not a leak.