Whole-cluster backups + on-demand backup/restore endpoints

app/backup.py

@@ -60,9 +60,9 @@ def backup():
     name = f"backup-{_ts()}.sql.gz"
     path = os.path.join(BACKUP_DIR, name)
     try:
+        # whole-cluster dump: every database + roles (not just one DB)
         dump = subprocess.run(
-            ["pg_dump", "-h", PGHOST, "-p", PG_PORT, "-U", PG_USER,
-             "-d", PG_DB, "--clean", "--if-exists"],
+            ["pg_dumpall", "-h", PGHOST, "-p", PG_PORT, "-U", PG_USER],
             stdout=subprocess.PIPE, check=True,
         ).stdout
         with gzip.open(path, "wb") as f:
@@ -114,10 +114,11 @@ def restore():
         return False
 
     print(f"[backup] restoring from {src}", flush=True)
+    # connect to 'postgres'; the dumpall script \connects into each database itself
     with gzip.open(src, "rb") as f:
         subprocess.run(
             ["psql", "-h", PGHOST, "-p", PG_PORT, "-U", PG_USER,
-             "-d", PG_DB, "-v", "ON_ERROR_STOP=0", "-q"],
+             "-d", "postgres", "-v", "ON_ERROR_STOP=0", "-q"],
             input=f.read(), check=True,
         )
     print("[backup] restore complete", flush=True)

app/main.py

@@ -155,6 +155,26 @@ async def connection(_: None = Depends(require_auth)):
     }
 
 
+@api.post("/admin/backup")
+async def admin_backup(_: None = Depends(require_auth)):
+    try:
+        path = backup.backup()
+        return {"ok": True, "file": os.path.basename(path)}
+    except Exception as e:
+        return JSONResponse({"ok": False, "error": str(e)}, status_code=500)
+
+
+@api.post("/admin/restore")
+async def admin_restore(_: None = Depends(require_auth)):
+    """Reload the latest backup (whole cluster). Cleanest on a fresh/empty DB;
+    on a populated DB it best-effort merges and skips rows that already exist."""
+    try:
+        done = backup.restore()
+        return {"ok": bool(done)}
+    except Exception as e:
+        return JSONResponse({"ok": False, "error": str(e)}, status_code=500)
+
+
 # ------------------------------------------------- Gradio uptime watcher ----
 def snapshot():
     tunnel = get_tunnel()
@@ -211,6 +231,7 @@ with gr.Blocks(title="Remote Postgres — Uptime Watcher", theme=gr.themes.Soft(
     with gr.Row():
         refresh_btn = gr.Button("Refresh now", variant="primary")
         backup_btn = gr.Button("Back up now")
+        restore_btn = gr.Button("Restore latest")
 
     def backup_now():
         try:
@@ -219,12 +240,20 @@ with gr.Blocks(title="Remote Postgres — Uptime Watcher", theme=gr.themes.Soft(
             print(f"[ui] manual backup failed: {e}", flush=True)
         return snapshot()
 
+    def restore_now():
+        try:
+            backup.restore()
+        except Exception as e:
+            print(f"[ui] manual restore failed: {e}", flush=True)
+        return snapshot()
+
     outputs = [status_box, url_box, host_box, port_box,
                user_box, pw_box, db_box, history_box]
     timer = gr.Timer(10)
     timer.tick(snapshot, outputs=outputs)
     refresh_btn.click(snapshot, outputs=outputs)
     backup_btn.click(backup_now, outputs=outputs)
+    restore_btn.click(restore_now, outputs=outputs)
     demo.load(snapshot, outputs=outputs)