Add unified operations realtime dashboard

- add /portal/ws/operations WebSocket snapshot endpoint - add shared operations-live.js with reconnect and operations:snapshot events - update Operations, Jobs, and Workers pages to use realtime snapshots - keep job detail live logs on dedicated /ws/jobs/{job_id}/logs endpoint - add Jobs and Workers pagination/filtering - remove old page auto-refresh behavior from jobs/deployments - normalize main navigation labels - move New App CTA from menu to Apps page - add backup download action for admins - add Workers pages and operations dashboard data helpers
2026-05-29 11:17:17 +02:00
parent 0dcf9b7b05
commit d145161e4e
9 changed files with 519 additions and 67 deletions
@@ -248,6 +248,7 @@ def get_next_queued_job():

 def get_jobs(
    limit: int = 100,
+    offset: int = 0,
    status: str | None = None,
    job_type: str | None = None,
    target: str | None = None,
@@ -282,32 +283,76 @@ def get_jobs(
            END,
            id DESC
        LIMIT ?
+        OFFSET ?
        """,
-        (*params, limit),
+        (*params, limit, offset),
    ).fetchall()

    con.close()
    return [dict(row) for row in rows]


+def count_jobs(
+    status: str | None = None,
+    job_type: str | None = None,
+    target: str | None = None,
+):
+    run_migrations()
+    con = get_connection()
+
+    filters = []
+    params: list[Any] = []
+    if status:
+        filters.append("status = ?")
+        params.append(status)
+    if job_type:
+        filters.append("type = ?")
+        params.append(job_type)
+    if target:
+        filters.append("(target_type LIKE ? OR target_id LIKE ?)")
+        target_pattern = f"%{target}%"
+        params.extend([target_pattern, target_pattern])
+
+    where_sql = f"WHERE {' AND '.join(filters)}" if filters else ""
+    row = con.execute(
+        f"""
+        SELECT COUNT(*) AS count
+        FROM jobs
+        {where_sql}
+        """,
+        params,
+    ).fetchone()
+
+    con.close()
+    return row["count"] if row else 0
+
+
 def get_job_stats():
    run_migrations()
    con = get_connection()

    rows = con.execute(
        """
-        SELECT status, COUNT(*) AS count
+        SELECT
+            SUM(CASE WHEN status = 'queued' THEN 1 ELSE 0 END) AS queued,
+            SUM(CASE WHEN status = 'running' THEN 1 ELSE 0 END) AS running,
+            SUM(CASE WHEN status = 'failed' THEN 1 ELSE 0 END) AS failed,
+            SUM(CASE WHEN status = 'success' THEN 1 ELSE 0 END) AS success,
+            SUM(CASE WHEN status = 'failed' AND created_at >= datetime('now', '-24 hours') THEN 1 ELSE 0 END) AS failed_24h,
+            SUM(CASE WHEN status = 'success' AND created_at >= datetime('now', '-24 hours') THEN 1 ELSE 0 END) AS success_24h
        FROM jobs
-        WHERE status IN ('queued', 'running', 'failed', 'success')
-        GROUP BY status
        """
-    ).fetchall()
+    ).fetchone()

    con.close()
-    stats = {"queued": 0, "running": 0, "failed": 0, "success": 0}
-    for row in rows:
-        stats[row["status"]] = row["count"]
-    return stats
+    return dict(rows) if rows else {
+        "queued": 0,
+        "running": 0,
+        "failed": 0,
+        "success": 0,
+        "failed_24h": 0,
+        "success_24h": 0,
+    }


 def get_job(job_id: int):
@@ -2,6 +2,10 @@ from app.db.database import get_connection
 from app.db.migrations import run_migrations
 from app.db.workers import is_worker_online

+DEPLOYMENT_SUCCESS_STATUSES = ("ok", "success", "succeeded", "done", "deployed", "completed")
+DEPLOYMENT_FAILED_STATUSES = ("failed", "failure", "error", "cancelled", "canceled")
+DEPLOYMENT_RUNNING_STATUSES = ("running", "pending", "queued", "in_progress", "starting")
+

 def _table_columns(con, table_name: str) -> set[str]:
    rows = con.execute(f"PRAGMA table_info({table_name})").fetchall()
@@ -142,3 +146,100 @@ def get_recent_audit_events(limit: int = 20):

    con.close()
    return [dict(row) for row in rows]
+
+
+def get_operations_snapshot():
+    run_migrations()
+    con = get_connection()
+
+    workers = con.execute(
+        """
+        SELECT id, status, started_at, last_seen_at, current_job_id, metadata_json
+        FROM workers
+        ORDER BY last_seen_at DESC, id
+        LIMIT 20
+        """
+    ).fetchall()
+    recent_workers = []
+    workers_online = 0
+    for row in workers:
+        worker = dict(row)
+        worker["online"] = is_worker_online(worker.get("last_seen_at"))
+        workers_online += 1 if worker["online"] else 0
+        recent_workers.append(worker)
+
+    total_workers = con.execute("SELECT COUNT(*) AS count FROM workers").fetchone()["count"]
+    online_all = con.execute("SELECT last_seen_at FROM workers").fetchall()
+    workers_online = sum(1 for worker in online_all if is_worker_online(worker["last_seen_at"]))
+
+    jobs_summary = con.execute(
+        """
+        SELECT
+            SUM(CASE WHEN status = 'queued' THEN 1 ELSE 0 END) AS queued,
+            SUM(CASE WHEN status = 'running' THEN 1 ELSE 0 END) AS running,
+            SUM(CASE WHEN status = 'failed' AND created_at >= datetime('now', '-24 hours') THEN 1 ELSE 0 END) AS failed_24h,
+            SUM(CASE WHEN status = 'success' AND created_at >= datetime('now', '-24 hours') THEN 1 ELSE 0 END) AS success_24h
+        FROM jobs
+        """
+    ).fetchone()
+    recent_jobs = con.execute(
+        """
+        SELECT id, type, target_type, target_id, status, source, created_at, started_at, finished_at
+        FROM jobs
+        ORDER BY id DESC
+        LIMIT 20
+        """
+    ).fetchall()
+
+    deployments_summary = con.execute(
+        f"""
+        SELECT
+            SUM(CASE WHEN LOWER(COALESCE(status, '')) IN ({','.join('?' for _ in DEPLOYMENT_RUNNING_STATUSES)}) THEN 1 ELSE 0 END) AS running,
+            SUM(CASE WHEN LOWER(COALESCE(status, '')) IN ({','.join('?' for _ in DEPLOYMENT_FAILED_STATUSES)}) AND started_at >= datetime('now', '-24 hours') THEN 1 ELSE 0 END) AS failed_24h,
+            SUM(CASE WHEN LOWER(COALESCE(status, '')) IN ({','.join('?' for _ in DEPLOYMENT_SUCCESS_STATUSES)}) AND started_at >= datetime('now', '-24 hours') THEN 1 ELSE 0 END) AS success_24h
+        FROM deployments
+        """,
+        (*DEPLOYMENT_RUNNING_STATUSES, *DEPLOYMENT_FAILED_STATUSES, *DEPLOYMENT_SUCCESS_STATUSES),
+    ).fetchone()
+    recent_deployments = con.execute(
+        """
+        SELECT id, app_id, kind, status, started_at, finished_at, trigger_source
+        FROM deployments
+        ORDER BY id DESC
+        LIMIT 20
+        """
+    ).fetchall()
+    recent_audit = con.execute(
+        """
+        SELECT id, username, action, target_type, target_id, source, created_at
+        FROM audit_events
+        ORDER BY id DESC
+        LIMIT 20
+        """
+    ).fetchall()
+
+    con.close()
+    return {
+        "type": "operations.snapshot",
+        "jobs": {
+            "queued": jobs_summary["queued"] or 0,
+            "running": jobs_summary["running"] or 0,
+            "failed_24h": jobs_summary["failed_24h"] or 0,
+            "success_24h": jobs_summary["success_24h"] or 0,
+            "recent": [dict(row) for row in recent_jobs],
+        },
+        "workers": {
+            "online": workers_online,
+            "offline": max(0, total_workers - workers_online),
+            "recent": recent_workers,
+        },
+        "deployments": {
+            "running": deployments_summary["running"] or 0,
+            "failed_24h": deployments_summary["failed_24h"] or 0,
+            "success_24h": deployments_summary["success_24h"] or 0,
+            "recent": [dict(row) for row in recent_deployments],
+        },
+        "audit": {
+            "recent": [dict(row) for row in recent_audit],
+        },
+    }