[ http://issues.apache.org/jira/browse/HADOOP-518?page=comments#action_12434785 ] Doug Cutting commented on HADOOP-518: -------------------------------------
Could this be related to HADOOP-320? In other words, is the bug perhaps that 'dfs -cp' didn't copy the checksum file? > hadoop dfs -cp foo/bar/bad-file mumble/new-file copies a file with a bad > checksum > --------------------------------------------------------------------------------- > > Key: HADOOP-518 > URL: http://issues.apache.org/jira/browse/HADOOP-518 > Project: Hadoop > Issue Type: Bug > Components: dfs > Environment: red hat > Reporter: Dick King > > I have a file that reliably generates a checksum error when it's read, > whether by a map/reduce job as input or by a "dfs -get" command. > However... > if I do a "dfs -cp" from the file with the bad checksum the copy can be read > in its entirety without a checksum error. > I would consider it reasonable for the command to fail, or for the new file > to be created but to also have a checksum error in the same place, but this > behavior is unsettling. > -dk -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://issues.apache.org/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira
