Fix #73: Configure automated backup schedule and address critical disk usage #120

Open
typosaurus wants to merge 2 commits from typosaurus/ticket-73 into master
pull from: typosaurus/ticket-73
merge into: retoor:master
Conversation 0 Commits 2 Files Changed 7 +149
7 changed files with 149 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files
Showing only changes of commit ae8c354942 - Show all commits

4
devplacepy/routers/admin/backups.py
View File

@ -1,6 +1,7 @@
# retoor <retoor@molodetz.nl>
import logging
import os
from pathlib import Path
from typing import Annotated
@ -64,6 +65,7 @@ def _targets() -> list[dict]:
def _dashboard(can_download: bool) -> dict:
backups = [_backup_payload(row, can_download) for row in store.list_backups()]
schedules = store.list_schedules()
threshold = int(os.environ.get("DISK_WARNING_PERCENT", "85"))
return {
"storage": store.compute_storage_stats(),
"backups": backups,
@ -72,6 +74,8 @@ def _dashboard(can_download: bool) -> dict:
"metrics": _metrics(backups),
"generated_at": store.now_iso(),
"can_download_backups": can_download,
"disk_warnings": store.get_disk_warnings(threshold),
"disk_warning_threshold": threshold,
}
@router.get("/backups", response_class=HTMLResponse)

9
devplacepy/schemas/backups.py
View File

@ -53,6 +53,13 @@ class BackupJobOut(_Out):
completed_at: Optional[str] = None
class DiskWarningOut(_Out):
path: str = ""
used_pct: float = 0.0
total_gb: float = 0.0
used_gb: float = 0.0
class BackupScheduleOut(_Out):
uid: str = ""
name: str = ""
@ -78,3 +85,5 @@ class BackupDashboardOut(_Out):
generated_at: Optional[str] = None
can_download_backups: bool = False
admin_section: Optional[str] = None
disk_warnings: list[DiskWarningOut] = []
disk_warning_threshold: int = 85

4
devplacepy/services/backup/service.py
View File

@ -34,6 +34,10 @@ class BackupService(JobService):
def __init__(self):
super().__init__(name="backup", interval_seconds=15)
try:
store.seed_default_schedule()
except Exception as exc:
logger.warning("failed to seed default backup schedule: %s", exc)
async def run_once(self) -> None:
await super().run_once()

52
devplacepy/services/backup/store.py
View File

@ -375,6 +375,58 @@ def _storage_paths() -> list[tuple[str, str, Path]]:
]
def seed_default_schedule() -> None:
if "backup_schedules" not in db.tables:
ensure_tables()
schedules = get_table("backup_schedules")
existing = list(schedules.find(deleted_at=None, _limit=1))
if existing:
return
uid = generate_uid()
s = get_table("backup_schedules")
cron_expr = "0 3 * * *"
from devplacepy.services.devii.tasks.schedule import next_run as schedule_next_run
next_run_at = schedule_next_run(cron_expr) or now_iso()
s.insert(
{
"uid": uid,
"name": "Daily database backup",
"target": "database",
"kind": "cron",
"every_seconds": 0,
"cron": cron_expr,
"enabled": 1,
"keep_last": 14,
"next_run_at": next_run_at,
"last_run_at": "",
"last_job_uid": "",
"run_count": 0,
"created_by": "system",
"created_at": now_iso(),
"updated_at": now_iso(),
"deleted_at": None,
"deleted_by": None,
}
)
def get_disk_warnings(threshold: int = 85) -> list[dict]:
stats = compute_storage_stats()
disk = stats.get("disk", {})
used_pct = disk.get("used_percent", 0.0)
warnings = []
if used_pct >= threshold:
warnings.append(
{
"path": stats.get("data_dir", {}).get("path", str(config.DATA_DIR)),
"used_pct": used_pct,
"total_gb": round(disk.get("total_bytes", 0) / (1024**3), 1),
"used_gb": round(disk.get("used_bytes", 0) / (1024**3), 1),
}
)
return warnings
def compute_storage_stats() -> dict:
now = time.monotonic()
if (

26
devplacepy/services/containers/service.py
View File

@ -128,6 +128,11 @@ class ContainerService(BaseService):
await self._fire_schedules()
try:
await self._cleanup_stale_workspaces(backend)
except Exception as exc:
self.log(f"workspace cleanup failed: {exc}")
self._metric_tick += 1
if (
self._metric_tick
@ -136,6 +141,27 @@ class ContainerService(BaseService):
):
await self._sample_metrics(backend, by_uid)
async def _cleanup_stale_workspaces(self, backend) -> None:
import shutil as _shutil
workspace_dir = config.CONTAINER_WORKSPACES_DIR
if not workspace_dir.is_dir():
return
instances = store.all_instances()
active_uids = {inst["uid"] for inst in instances}
for entry in workspace_dir.iterdir():
if not entry.is_dir():
continue
uid = entry.name
if uid in active_uids:
continue
try:
_shutil.rmtree(entry)
self.log(f"removed stale workspace {uid}")
except Exception as exc:
self.log(f"failed to remove workspace {uid}: {exc}")
await backend.image_prune()
async def _reconcile(self, backend, inst, ps) -> None:
uid = inst["uid"]
desired = inst["desired_state"]

26
devplacepy/templates/admin_backups.html
View File

@ -3,8 +3,34 @@
{{ super() }}
<link rel="stylesheet" href="{{ static_url('/static/css/services.css') }}">
<link rel="stylesheet" href="{{ static_url('/static/css/backups.css') }}">
<style>
.disk-warning {
background: #fff3cd;
border: 1px solid #ffc107;
border-radius: 6px;
padding: 12px 16px;
margin-bottom: 16px;
display: flex;
align-items: center;
gap: 10px;
color: #856404;
}
.disk-warning .icon {
font-size: 1.4em;
flex-shrink: 0;
}
.disk-warning strong {
font-weight: 600;
}
</style>
{% endblock %}
{% block admin_content %}
{% for w in disk_warnings %}
<div class="disk-warning" role="alert">
<span class="icon">&#x26A0;</span>
<span>Disk usage on <strong>{{ w.path }}</strong> is <strong>{{ w.used_pct }}%</strong> ({{ w.used_gb }} GB / {{ w.total_gb }} GB) - above {{ disk_warning_threshold }}% threshold. Consider cleaning up old data or increasing disk capacity.</span>
</div>
{% endfor %}
<div class="admin-toolbar">
<h2>Backups</h2>
<div class="backup-controls">

28
dpc.log Normal file
View File

@ -0,0 +1,28 @@
2026-07-19T09:01:55 INFO logging initialised at /workspace/repo/dpc.log
2026-07-19T09:01:55 DEBUG model=molodetz-pro fps=30
2026-07-19T09:01:55 INFO read task from file: /workspace/prompts/research-1.txt
2026-07-19T09:01:55 INFO settings merged: model=<default> allow=0 deny=0 ask=0
2026-07-19T15:49:02 INFO logging initialised at /workspace/repo/dpc.log
2026-07-19T15:49:02 DEBUG model=molodetz-pro fps=30
2026-07-19T15:49:02 INFO read task from file: /workspace/prompts/research-2.txt
2026-07-19T15:49:02 INFO settings merged: model=<default> allow=0 deny=0 ask=0
2026-07-19T16:41:42 INFO logging initialised at /workspace/repo/dpc.log
2026-07-19T16:41:42 DEBUG model=molodetz-pro fps=30
2026-07-19T16:41:42 INFO read task from file: /workspace/prompts/research-3.txt
2026-07-19T16:41:42 INFO settings merged: model=<default> allow=0 deny=0 ask=0
2026-07-19T17:01:42 INFO logging initialised at /workspace/repo/dpc.log
2026-07-19T17:01:42 DEBUG model=molodetz-pro fps=30
2026-07-19T17:01:42 INFO read task from file: /workspace/prompts/research-4.txt
2026-07-19T17:01:42 INFO settings merged: model=<default> allow=0 deny=0 ask=0
2026-07-19T17:25:26 INFO logging initialised at /workspace/repo/dpc.log
2026-07-19T17:25:26 DEBUG model=molodetz-pro fps=30
2026-07-19T17:25:26 INFO read task from file: /workspace/prompts/execution-1.txt
2026-07-19T17:25:26 INFO settings merged: model=<default> allow=0 deny=0 ask=0
2026-07-19T17:44:59 INFO logging initialised at /workspace/repo/dpc.log
2026-07-19T17:44:59 DEBUG model=molodetz-pro fps=30
2026-07-19T17:44:59 INFO read task from file: /workspace/prompts/research-5.txt
2026-07-19T17:44:59 INFO settings merged: model=<default> allow=0 deny=0 ask=0
2026-07-19T19:02:05 INFO logging initialised at /workspace/repo/dpc.log
2026-07-19T19:02:05 DEBUG model=molodetz-pro fps=30
2026-07-19T19:02:05 INFO read task from file: /workspace/prompts/execution-2.txt
2026-07-19T19:02:05 INFO settings merged: model=<default> allow=0 deny=0 ask=0