Julio Ng created KAFKA-8362:
-------------------------------

             Summary: LogCleaner gets stuck after partition move between log 
directories
                 Key: KAFKA-8362
                 URL: https://issues.apache.org/jira/browse/KAFKA-8362
             Project: Kafka
          Issue Type: Bug
          Components: log cleaner
            Reporter: Julio Ng


When a partition is moved from one directory to another, their checkpoint entry 
in cleaner-offset-checkpoint file is not removed from the source directory.

As a consequence when we read the last firstDirtyOffset, we might get a stale 
value from the old checkpoint file.

Basically, we need clean up the entry from the check point file in the source 
directory when the move is completed

The current issue is that the code in LogCleanerManager:

{{def allCleanerCheckpoints: Map[TopicPartition, Long] = {}}
{{  inLock(lock) {}}
{{    checkpoints.values.flatMap(checkpoint => {}}
{{      try {}}
{{        checkpoint.read()}}
{{      } catch {}}
{{        case e: KafkaStorageException =>}}
{{          error(s"Failed to access checkpoint file ${checkpoint.file.getName} 
in dir ${checkpoint.file.getParentFile.getAbsolutePath}", e)}}
{{          Map.empty[TopicPartition, Long]}}
{{      }}}
{{    }).toMap}}
{{  }}}
{{}}}

collapses the offsets when multiple entries exist for the topicPartition



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

Reply via email to