[jira] [Created] (FLINK-21739) TopN with TUMBLE window get Incorrect result

2021-03-11 Thread wxmimperio (Jira) 
wxmimperio created FLINK-21739:
--

 Summary: TopN with TUMBLE window get Incorrect result
 Key: FLINK-21739
 URL: https://issues.apache.org/jira/browse/FLINK-21739
 Project: Flink
  Issue Type: Bug
  Components: Table SQL / Runtime
Affects Versions: 1.11.1
Reporter: wxmimperio


{code:java}
SELECT * FROM (
  SELECT * ,
  ROW_NUMBER() OVER (
PARTITION BY window_start
ORDER BY gap_transactions desc
  ) AS rownum
  FROM (
SELECT TUMBLE_START(event_time, INTERVAL '5' MINUTE) AS window_start, 
itemId, sum(reply_gap) AS gap_transactions
FROM ItemTransactions
GROUP BY itemId, TUMBLE(event_time, INTERVAL '5' MINUTE)
  )
)
WHERE rownum <=5;
{code}
I want to get the maximum sum of every 5 minutes window.

The expected result is that there are only 5 records every 5 minutes, But the 
reality is that there are more than 5 pieces of data every 5 minutes.

 



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

[jira] [Created] (FLINK-21528) Rest Api Support Wildcard

2021-02-28 Thread wxmimperio (Jira) 
wxmimperio created FLINK-21528:
--

 Summary: Rest Api Support Wildcard
 Key: FLINK-21528
 URL: https://issues.apache.org/jira/browse/FLINK-21528
 Project: Flink
  Issue Type: Improvement
  Components: Runtime / REST
Affects Versions: 1.11.1
Reporter: wxmimperio
 Attachments: image-2021-03-01-12-00-36-269.png, 
image-2021-03-01-12-05-05-881.png, image-2021-03-01-12-06-15-340.png

!image-2021-03-01-12-00-36-269.png!

To obtain detailed metrics information, you need to specify the complete 
metrics name.

But these names are automatically generated, only by obtaining a list of all 
metrics names, and then filtering by keywords.

For example, the content of the red box is automatically generated without 
knowing in advance.

!image-2021-03-01-12-06-15-340.png!
I can only get all the metrics names, and then filter TableSourceScan and 
join-time-max.

If I can http://xxx/metrics/get=*join-time-max, I can easily filter what i want.

 



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

[jira] [Created] (FLINK-20466) Table 'EXPR$1' not found with UNION ALL

2020-12-03 Thread wxmimperio (Jira) 
wxmimperio created FLINK-20466:
--

 Summary: Table 'EXPR$1' not found with UNION ALL
 Key: FLINK-20466
 URL: https://issues.apache.org/jira/browse/FLINK-20466
 Project: Flink
  Issue Type: Bug
  Components: Table SQL / Planner
Affects Versions: 1.11.1
Reporter: wxmimperio


 
{code:java}
CREATE TABLE table_01 (  aaa varchar,  bbb varchar) WITH(...);

CREATE TABLE table_02 (  aaa varchar,  bbb varchar) WITH(...);

create view my_view as
select aaa,bbb from (
   select aaa,bbb from table_01
   union all
   select aaa,bbb from table_02
);
create table bsql_log (  aaa varchar,  bbb varchar) with (  'connector' = 
'log');
insert into bsql_log SELECT aaa,bbb FROM my_view
{code}
 
Run the above code will report an error：
{code:java}
org.apache.calcite.runtime.CalciteContextException: From line 1, column 8 to 
line 1, column 15: Table 'EXPR$1' not found
{code}
But if I assign an alias to the result of union all, it can be normal.
{code:java}
create view my_view as
select aaa,bbb from ( 
 select aaa,bbb from table_01 
 union all 
 select aaa,bbb from table_02
) as union_result;
{code}
 
 



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

[jira] [Created] (FLINK-20378) Watermark generation check TIMESTAMP_WITHOUT_TIME_ZONE

2020-11-26 Thread wxmimperio (Jira) 
wxmimperio created FLINK-20378:
--

 Summary: Watermark generation check TIMESTAMP_WITHOUT_TIME_ZONE
 Key: FLINK-20378
 URL: https://issues.apache.org/jira/browse/FLINK-20378
 Project: Flink
  Issue Type: Improvement
  Components: Table SQL / Planner
Affects Versions: 1.11.1
Reporter: wxmimperio


 

 
{code:java}
 def generateWatermarkGenerator(
  config: TableConfig,
  inputType: RowType,
  watermarkExpr: RexNode): GeneratedWatermarkGenerator = {
// validation
val watermarkOutputType = 
FlinkTypeFactory.toLogicalType(watermarkExpr.getType)
if (watermarkOutputType.getTypeRoot != 
LogicalTypeRoot.TIMESTAMP_WITHOUT_TIME_ZONE) {
  throw new CodeGenException(
"WatermarkGenerator only accepts output data type of TIMESTAMP," +
  " but is " + watermarkOutputType)
}
{code}
 

Why does watermark generation need to be detected as 
TIMESTAMP_WITHOUT_TIME_ZONE?

If I remove this check, what effect will it have on the watermark?

 

 

 

 



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

[jira] [Created] (FLINK-11228) Flink web UI can not read stdout logs

2018-12-28 Thread wxmimperio (JIRA) 
wxmimperio created FLINK-11228:
--

 Summary: Flink web UI can not read stdout logs
 Key: FLINK-11228
 URL: https://issues.apache.org/jira/browse/FLINK-11228
 Project: Flink
  Issue Type: Bug
  Components: Web Client
Affects Versions: 1.7.1
Reporter: wxmimperio
 Attachments: image-2018-12-28-18-58-41-111.png, 
image-2018-12-28-19-00-17-652.png

I start flink as local cluster.

$ tar xzf flink-*.tgz
$ cd flink-1.6.1
$ ./bin/start-cluster.sh  # Start Flink

I createRemoteEnvironment to submit a remote job and job run successful.

However when I read stdout log from web ui, I find nothing.
I can find logs on linux server.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)