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

Posted to hdfs-dev@hadoop.apache.org by "Akira AJISAKA (JIRA)" <ji...@apache.org> on 2014/06/20 00:31:25 UTC

[jira] [Created] (HDFS-6571) NameNode should delete intermediate fsimage.ckpt when checkpoint fails

Akira AJISAKA created HDFS-6571:
-----------------------------------

             Summary: NameNode should delete intermediate fsimage.ckpt when checkpoint fails
                 Key: HDFS-6571
                 URL: https://issues.apache.org/jira/browse/HDFS-6571
             Project: Hadoop HDFS
          Issue Type: Bug
    Affects Versions: 2.4.0
            Reporter: Akira AJISAKA


When checkpoint fails in getting a new fsimage from standby NameNode or SecondaryNameNode, intermediate fsimage (fsimage.ckpt_<txid>) is left and never to be cleaned up.
If fsimage is large and fails to checkpoint many times, the growing intermediate fsimage may cause out of disk space.



--
This message was sent by Atlassian JIRA
(v6.2#6252)