You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@flink.apache.org by "jiwei (JIRA)" <ji...@apache.org> on 2019/01/18 07:20:00 UTC

[jira] [Created] (FLINK-11387) Flink task can not restart from failed after running two days.

jiwei created FLINK-11387:
-----------------------------

             Summary: Flink task can not restart from failed after running two days.
                 Key: FLINK-11387
                 URL: https://issues.apache.org/jira/browse/FLINK-11387
             Project: Flink
          Issue Type: Bug
          Components: Table API &amp; SQL, YARN
    Affects Versions: 1.7.0
         Environment:  

 
            Reporter: jiwei
         Attachments: image-2019-01-18-15-15-36-281.png, image-2019-01-18-15-17-57-250.png

The sql statement is as below.

SELECT k AS key, SUM(p) AS pv, SUM(u) AS uv
FROM (
 SELECT concat_ws('|', udf_pcmplat(site_id), action, position_mark, udf_hour(_time)) AS k
 , COUNT(*) AS p
 , udf_countdistinct(cuid) AS u
 FROM xxxx
 WHERE action NOT IN (
 'unknown', 
 'nil', 
 'null', 
 'UNKNOWN', 
 'NIL', 
 'NULL'
 )
 AND udf_pcmplat(site_id) IS NOT NULL
 GROUP BY md5_2(cuid),udf_hour(_time), udf_pcmplat(site_id), action, position_mark
) t
GROUP BY k

 

This is the error log.

!image-2019-01-18-15-17-57-250.png!

Do you have any idea?



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