You are viewing a plain text version of this content. The canonical link for it is here.
Posted to common-dev@hadoop.apache.org by "Ivan A. Veselovsky (JIRA)" <ji...@apache.org> on 2013/02/27 17:11:16 UTC

[jira] [Created] (HADOOP-9337) org.apache.hadoop.fs.DF.getMount() does not work on Mac OS

Ivan A. Veselovsky created HADOOP-9337:
------------------------------------------

             Summary: org.apache.hadoop.fs.DF.getMount() does not work on Mac OS
                 Key: HADOOP-9337
                 URL: https://issues.apache.org/jira/browse/HADOOP-9337
             Project: Hadoop Common
          Issue Type: Bug
         Environment: Mac OS 10.8
            Reporter: Ivan A. Veselovsky
            Assignee: Ivan A. Veselovsky


test org.apache.hadoop.fs.TestLocalFileSystem.testReportChecksumFailure() (added in HADOOP-9067) appears to fail on MacOS because 
method org.apache.hadoop.fs.DF.getMount() does not work correctly.
The problem is that "df -k <path>" command returns on MacOS output like the following:
-------
Filesystem   1024-blocks      Used Available Capacity  iused    ifree %iused  Mounted on
/dev/disk0s4   194879828 100327120  94552708    52% 25081778 23638177   51%   /Volumes/Data
-------
while the following is expected:
-------
Filesystem         1024-blocks      Used Available Capacity Mounted on
/dev/mapper/vg_iveselovskyws-lv_home 420545160  15978372 383204308       5% /home
-------
So, we see that Mac's output has 3 additional tokens.

I can suggest 2 ways to fix the problem.
(a) use "-P" (POSIX) option when invoking df command. This will probably ensure unifirm output on all Unix systems;
(b) move Mac branch to specific "case" branch and treat it specifically (like we currently have for AIX, DF.java, line 214)

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira