diff --git a/erpnext/manufacturing/report/work_order_stock_report/test_work_order_stock_report.py b/erpnext/manufacturing/report/work_order_stock_report/test_work_order_stock_report.py
new file mode 100644
index 00000000000..583a89e9664
--- /dev/null
+++ b/erpnext/manufacturing/report/work_order_stock_report/test_work_order_stock_report.py
@@ -0,0 +1,24 @@
+# Copyright (c) 2017, Velometro Mobility Inc and contributors
+# For license information, please see license.txt
+
+import frappe
+
+from erpnext.tests.utils import ERPNextTestSuite
+
+
+class TestWorkOrderStockReport(ERPNextTestSuite):
+	def test_report_executes_and_lists_work_order(self):
+		# get_item_list computes build_qty by multiplying bin/bom/bom_item columns that are not
+		# functionally dependent on the grouped item_code; they must be in the GROUP BY for the
+		# report to run on Postgres. This exercises that query on both engines.
+		from erpnext.manufacturing.doctype.work_order.test_work_order import make_wo_order_test_record
+		from erpnext.manufacturing.report.work_order_stock_report.work_order_stock_report import execute
+
+		wo = make_wo_order_test_record(
+			production_item="_Test FG Item", qty=1, source_warehouse="_Test Warehouse - _TC"
+		)
+
+		columns, data = execute(frappe._dict(warehouse="_Test Warehouse - _TC"))
+
+		self.assertTrue(columns)
+		self.assertIn(wo.name, {row["work_order"] for row in data})
diff --git a/erpnext/manufacturing/report/work_order_stock_report/work_order_stock_report.py b/erpnext/manufacturing/report/work_order_stock_report/work_order_stock_report.py
index 998b0e4bccd..0637b14ccc7 100644
--- a/erpnext/manufacturing/report/work_order_stock_report/work_order_stock_report.py
+++ b/erpnext/manufacturing/report/work_order_stock_report/work_order_stock_report.py
@@ -47,7 +47,10 @@ def get_item_list(wo_list, filters):
 						& (bom_item.item_code == wo_item_details.item_code)
 						& (bom.name == wo_details.bom_no)
 					)
-					.groupby(bom_item.item_code)
+					# build_qty multiplies columns from bin/bom/bom_item that aren't functionally
+					# dependent on the grouped item_code, so postgres requires them in the GROUP BY.
+					# The WHERE pins bom, item and warehouse to single rows, so this stays one row.
+					.groupby(bom_item.item_code, bom.quantity, bom_item.stock_qty, bin.actual_qty)
 				).run(as_dict=1)
 
 				stock_qty = 0