flink sql升级到1.12 基于eventtime的window聚合没有watermark导致没有数据输出

[bigjar3344]

String ddl =
        "CREATE TABLE orders (\n"
                + "  user_id INT,\n"
                + "  product STRING,\n"
                + "  amount INT,\n"
                + "  `time` bigint,\n"
                + "  `ts` AS TO_TIMESTAMP(FROM_UNIXTIME(`time`)),\n"
                + "  WATERMARK FOR ts AS ts - INTERVAL '3' SECOND\n"
                + ") WITH (\n"
                + "    'connector' = 'kafka',\n"
                + "    'topic' = 'test',\n"
                + "    'properties.group.id' = 'flink_ods1',\n"
                + "    'properties.bootstrap.servers' = ‘172.21.11.11:9092‘\n"
                + "    'format' = 'csv',\n"
                + "    'csv.ignore-parse-errors' = 'true',\n"
                + "    'csv.field-delimiter' = ',',\n"
                + "    'scan.startup.mode' = 'group-offsets'\n"
                + ")";


String ddl2 = "CREATE TABLE print_check2 (\n"
        + "    `window_start` STRING,\n"
        + "    `order_num` BIGINT,\n"
        + "    `total_amount` BIGINT,\n"
        + "    `unique_products` BIGINT\n"
        + ") WITH (\n"
        + "    'connector' = 'print'\n"
        + ")";


String ddl3 =
        "insert into print_check2 SELECT\n"
                + "  CAST(TUMBLE_START(ts, INTERVAL '5' SECOND) AS
STRING) window_start,\n"
                + "  COUNT(*) order_num,\n"
                + "  SUM(amount) total_amount,\n"
                + "  COUNT(DISTINCT product) unique_products\n"
                + "FROM orders\n"
                + "GROUP BY TUMBLE(ts, INTERVAL '5' SECOND)";

执行上述语句后，在1.12中print_check2表没有结果输出，而在1.11.3中则有结果输出。上述的对比在相同的kafka输入数据环境下。
<https://maas.mail.163.com/dashi-web-extend/html/proSignature.html?ftlId=1&name=bigjar3344&uid=bigjar3344%40gmail.com&iconUrl=https%3A%2F%2Fmail-online.nosdn.127.net%2Fqiyelogo%2FdefaultAvatar.png&items=%5B%22bigjar3344%40gmail.com%22%5D>