[
https://issues.apache.org/jira/browse/HADOOP-3030?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Chris Douglas updated HADOOP-3030:
----------------------------------
Resolution: Fixed
Status: Resolved (was: Patch Available)
I just committed this. Thanks, Devaraj!
> InMemoryFileSystem.reserveSpaceWithChecksum does not look at failures while
> reserving space for the file in question
> --------------------------------------------------------------------------------------------------------------------
>
> Key: HADOOP-3030
> URL: https://issues.apache.org/jira/browse/HADOOP-3030
> Project: Hadoop Core
> Issue Type: Bug
> Components: fs
> Reporter: Devaraj Das
> Assignee: Devaraj Das
> Fix For: 0.17.0
>
> Attachments: 3030.patch, hadoop-3030.patch
>
>
> The return statement code in InMemoryFileSystem.reserveSpaceWithCheckSum
> looks like
> {noformat}
> return (mfs.reserveSpace(f, size) && mfs.reserveSpace(getChecksumFile(f),
> checksumSize));
> {noformat}
> This should be broken up to check for successful reserveSpace for each of the
> components. In some cases, we might reserve space for the first component and
> fail while doing the same for the second (checksum file). This will lead to
> wastage of space since we don't un-reserve the space we got for the first
> component. This usually won't happen due to the policy associated with
> creating a file in the InMemoryFileSystem (since the checksum component is
> usually very small) but still it should be fixed.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
