hadoop-hdfs-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Lei (Eddy) Xu (JIRA)" <j...@apache.org>
Subject [jira] [Created] (HDFS-7332) Fix TestMultipleNNDataBlockScanner failures in jenkin slaves.
Date Tue, 04 Nov 2014 00:22:34 GMT
Lei (Eddy) Xu created HDFS-7332:

             Summary: Fix TestMultipleNNDataBlockScanner failures in jenkin slaves.
                 Key: HDFS-7332
                 URL: https://issues.apache.org/jira/browse/HDFS-7332
             Project: Hadoop HDFS
          Issue Type: Bug
          Components: datanode
    Affects Versions: 2.5.1
            Reporter: Lei (Eddy) Xu
            Assignee: Lei (Eddy) Xu

{{BlockPoolSliceScanner}} sorts {{BlockScanInfo}} by its {{lastScanTime}}. Each time, the
scanner picks a {{BlockScanInfo}} with the smallest {{lastScanTime}} to scan and then updates
its {{lastScanTime = Time.mononicNow()}}. 

Since the Jenkins test slave VM is usually rebooted for each job, the {{Time.mononicNow()}}
in the VM returns a small number, which is smaller than the initial values of {{BlockScanInfo}}.
Thus {{BlockPoolSliceScanner}} stops at the first block that has been scanned and could not
finish scanning all blocks. The result of it is that {{TestMultipleNNDataBlockScanner#testDataBlockScanner}}
times out due to unfinished scanning job.

This message was sent by Atlassian JIRA

View raw message