Files
nas-tools/dashboard/backend/routers/media_mgmt.py
T
Gan, Jimmy 9f193ec133
Deploy Dashboard (Main) / Backend Tests (push) Failing after 18s
Deploy Dashboard (Main) / Frontend Tests (push) Successful in 10s
Deploy Dashboard (Main) / Build Main Image (push) Failing after 7m26s
Deploy Dashboard (Main) / Deploy to Production (push) Has been skipped
sprint: Media Ops — shuttle status endpoint, graceful offline mode, 18 backend tests, shuttle detail in pipeline card, error states with retry, 13 frontend vitest tests
2026-07-12 15:36:13 +08:00

210 lines
7.3 KiB
Python

import json
import logging
import os
import re
from datetime import datetime
import asyncssh
from fastapi import APIRouter, Depends, HTTPException, BackgroundTasks, Query
import config
router = APIRouter()
logger = logging.getLogger(__name__)
# Absolute paths to media-mgmt scripts on NAS host
HEALTH_CHECK_SCRIPT = "/volume1/homes/zjgump/media-mgmt/immich/scripts/health-check.sh"
STORAGE_REPORT_SCRIPT = "/volume1/homes/zjgump/media-mgmt/diagnostics/storage-report.sh"
SCAN_LIBRARY_SCRIPT = "/volume1/homes/zjgump/media-mgmt/immich/scripts/scan-library.sh"
THUMBNAIL_CLINIC_SCRIPT = "/volume1/homes/zjgump/media-mgmt/diagnostics/thumbnail-clinic.sh"
SHUTTLE_SCRIPT = "/volume1/homes/zjgump/media-mgmt/transfer/ssd-shuttle.sh"
PHONE_INGEST_SCRIPT = "/volume1/homes/zjgump/media-mgmt/transfer/phone-to-nas.sh"
LOG_DIR = "/volume1/homes/zjgump/media-mgmt/logs"
UNAVAILABLE_RESPONSE = {"status": "unavailable", "error": "NAS unreachable"}
async def run_host_command(cmd: str) -> tuple[int, str, str]:
"""Execute a command on the NAS host via SSH.
Returns (exit_code, stdout, stderr). On SSH connection failure,
exit_code=999 and stderr contains the error message.
"""
try:
full_cmd = f'PATH="/volume1/homes/zjgump/bin:$PATH" NAS_HOST="naslan" {cmd}'
async with asyncssh.connect(
config.SSH_HOST,
username=config.SSH_USER,
client_keys=[config.SSH_KEY_PATH],
known_hosts=None,
) as conn:
result = await conn.run(full_cmd)
return result.exit_status, result.stdout, result.stderr
except Exception as e:
logger.error(f"SSH command failed ({cmd}): {e}")
return 999, "", str(e)
@router.get("/health")
async def get_health():
"""Runs health-check.sh and returns structured status."""
code, stdout, stderr = await run_host_command(f"bash {HEALTH_CHECK_SCRIPT}")
if code == 999:
return UNAVAILABLE_RESPONSE
return {
"status": "ok" if code == 0 else "warning" if code == 1 else "error",
"exit_code": code,
"stdout": stdout,
"stderr": stderr,
}
@router.get("/storage")
async def get_storage():
"""Runs storage-report.sh -j and returns the parsed JSON storage report."""
code, stdout, stderr = await run_host_command(f"bash {STORAGE_REPORT_SCRIPT} -j")
if code == 999:
return UNAVAILABLE_RESPONSE
if code != 0:
return {"error": "Storage report failed", "stderr": stderr}
try:
return json.loads(stdout)
except json.JSONDecodeError:
return {"error": "Failed to parse JSON from storage report", "raw_output": stdout}
@router.get("/shuttle/status")
async def get_shuttle_status():
"""Shuttle pipeline status: queue depth, last transfer time, recent errors.
Reads the shuttle log file to extract pipeline state.
"""
log_path = f"{LOG_DIR}/shuttle.log"
code, stdout, stderr = await run_host_command(f"cat {log_path} 2>/dev/null || echo 'NO_LOG'")
if code == 999:
return UNAVAILABLE_RESPONSE
result = {
"running": False,
"queue_depth": 0,
"last_transfer": None,
"last_transfer_size": None,
"recent_errors": [],
"current_stage": None,
}
content = stdout or ""
if content.strip() == "NO_LOG":
return result
# Check if shuttle is currently running via pgrep
running_code, running_out, _ = await run_host_command(f"pgrep -f {SHUTTLE_SCRIPT}")
result["running"] = running_code == 0
if running_code == 0:
result["current_stage"] = "active"
result["pid"] = running_out.strip()
# Parse log for last transfer, errors, current stage
lines = content.strip().split("\n")
error_lines = []
transfer_lines = []
for line in lines:
if "❌" in line or "error" in line.lower() or "failed" in line.lower():
error_lines.append(line.strip())
if "copied" in line.lower() or "transferred" in line.lower() or "rsync" in line.lower():
transfer_lines.append(line.strip())
if "Stage" in line and ":" in line:
result["current_stage"] = line.strip()
if error_lines:
result["recent_errors"] = error_lines[-10:]
if transfer_lines:
last = transfer_lines[-1]
result["last_transfer"] = last
# Try to extract size from rsync-like output
size_match = re.search(r"(\d+\.?\d*)\s*(K|M|G)B", last)
if size_match:
result["last_transfer_size"] = size_match.group(0)
# Count queued source directories
queue_matches = [l for l in lines if "--source" in l or "--stage" in l]
result["queue_depth"] = len(queue_matches)
return result
@router.post("/maintenance/scan")
async def run_scan(background_tasks: BackgroundTasks, library: str = Query(None)):
"""Triggers scan-library.sh in the background."""
cmd = f"bash {SCAN_LIBRARY_SCRIPT}"
if library:
cmd += f" {library}"
log_file = f"{LOG_DIR}/scan.log"
run_cmd = f"mkdir -p {LOG_DIR} && echo '=== Scan Started at $(date) ===' > {log_file} && nohup {cmd} >> {log_file} 2>&1 &"
code, stdout, stderr = await run_host_command(run_cmd)
if code == 999:
return UNAVAILABLE_RESPONSE
if code != 0:
return {"error": "Failed to start scan", "stderr": stderr}
return {"message": "Scan started in background", "log_file": log_file}
@router.post("/maintenance/requeue")
async def run_requeue(background_tasks: BackgroundTasks):
"""Triggers thumbnail-clinic.sh -r in the background."""
cmd = f"bash {THUMBNAIL_CLINIC_SCRIPT} -r"
log_file = f"{LOG_DIR}/requeue.log"
run_cmd = f"mkdir -p {LOG_DIR} && echo '=== Requeue Started at $(date) ===' > {log_file} && nohup {cmd} >> {log_file} 2>&1 &"
code, stdout, stderr = await run_host_command(run_cmd)
if code == 999:
return UNAVAILABLE_RESPONSE
if code != 0:
return {"error": "Failed to start thumbnail requeue", "stderr": stderr}
return {"message": "Thumbnail requeue started in background", "log_file": log_file}
@router.get("/logs/{task_name}")
async def get_logs(task_name: str):
"""Reads logs for a task from the media-mgmt logs directory."""
if task_name not in ["scan", "requeue", "shuttle", "phone-ingest"]:
raise HTTPException(status_code=400, detail="Invalid task name")
log_path = f"{LOG_DIR}/{task_name}.log"
code, stdout, stderr = await run_host_command(f"cat {log_path} 2>/dev/null || echo 'NO_LOG'")
if code == 999:
return UNAVAILABLE_RESPONSE
if stdout.strip() == "NO_LOG":
return {"content": "No log file found."}
lines = stdout.strip().split("\n")
return {"content": "\n".join(lines[-200:])}
@router.get("/status")
async def get_active_status():
"""Checks if any ingestion/maintenance processes are currently running on the host."""
processes = {
"scan": SCAN_LIBRARY_SCRIPT,
"requeue": THUMBNAIL_CLINIC_SCRIPT,
"shuttle": SHUTTLE_SCRIPT,
"phone-ingest": PHONE_INGEST_SCRIPT,
}
status = {}
for name, script_path in processes.items():
code, stdout, stderr = await run_host_command(f"pgrep -f {script_path}")
if code == 999:
return UNAVAILABLE_RESPONSE
status[name] = {
"running": code == 0,
"pid": stdout.strip() if code == 0 else None,
}
return status