You are viewing a plain text version of this content. The canonical link for it is here.
Posted to notifications@mynewt.apache.org by "Christopher Collins (JIRA)" <ji...@apache.org> on 2016/02/26 22:17:18 UTC
[jira] [Closed] (MYNEWT-164) NFFS - Restore fails when an object
from a GCed area is referenced
[ https://issues.apache.org/jira/browse/MYNEWT-164?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Christopher Collins closed MYNEWT-164.
--------------------------------------
Resolution: Fixed
> NFFS - Restore fails when an object from a GCed area is referenced
> ------------------------------------------------------------------
>
> Key: MYNEWT-164
> URL: https://issues.apache.org/jira/browse/MYNEWT-164
> Project: Mynewt
> Issue Type: Bug
> Components: NFFS
> Reporter: Christopher Collins
> Assignee: Christopher Collins
> Fix For: v0_8_0_beta2
>
>
> (Credit for debugging this goes to Paul and Marko)
> At the end of file system detection, a final sweep through the hash is performed. During this sweep, any deleted inodes and their children are removed from the hash. If an inode's child (data block if file; child inode if directory) is not present in the hash, the procedure terminates prematurely, leaving the hash in a corrupt state.
> The sweep process should expect some child objects to be missing from the hash. Since the child objects belong to a deleted inode, they are subject to garbage collection. When an NFFS area is garbage collected, it may leave some objects on disk with invalid references (this is not a problem, because these objects are themselves deleted).
> The fix is to ignore this error during the sweep.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)