ecrypts file corruption

Bug #655615 reported by Arne Caspari
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
eCryptfs
Incomplete
High
Tyler Hicks
ecryptfs-utils (Ubuntu)
Incomplete
High
Unassigned

Bug Description

Binary package hint: ecryptfs-utils

I am trying to build an Angstrom/Openembedded image. The process involves a lot of CPU and disk activity. The build home is on my home partition which is using ecrypt fs.

During the build I get occasional messages in the kernel log like:
[440102.824861] Valid eCryptfs headers not found in file header region or xattr region
[440102.824867] Either the lower file is not in a valid eCryptfs format, or the key could not be retrieved. Plaintext passthrough mode is not enabled; returning -EIO

Consequently the build stops with an IO error. Continuing the build usually fixes this issue ( ie. the files can be correctly read the second time ).

But unfortunately from time to time, the files get corrupted which means that they only contain binary data garbage!

ProblemType: Bug
DistroRelease: Ubuntu 10.04
Package: ecryptfs-utils 83-0ubuntu3
ProcVersionSignature: Ubuntu 2.6.32-24.43-generic 2.6.32.15+drm33.5
Uname: Linux 2.6.32-24-generic i686
NonfreeKernelModules: wl
Architecture: i386
Date: Wed Oct 6 11:51:17 2010
EcryptfsInUse: Yes
InstallationMedia: Ubuntu 9.10 "Karmic Koala" - Release i386 (20091028.5)
ProcEnviron:
 SHELL=/bin/bash
 PATH=(custom, user)
 LANG=de_DE.utf8
SourcePackage: ecryptfs-utils

Revision history for this message
Arne Caspari (arne-datafloater) wrote :
Revision history for this message
Arne Caspari (arne-datafloater) wrote :

I moved the build environment out of the encrypted home folder to an unencrypted part of the same harddisk and made a complete rebuild of the Angstrom image. This time the build went through without any issues. There were no IO errors and more importantly, there was no file corruption.

I think this pretty much confirms that ecryptfs is the cause for the file corruption

Revision history for this message
Dustin Kirkland  (kirkland) wrote :

I suspect that this might be a duplicate of Bug #509180.

Subscribing upstream.

Tyler, can you take a look at this and see if you think this is a dupe of our other corruption issue?

Thanks!

Changed in ecryptfs-utils (Ubuntu):
importance: Undecided → High
status: New → Incomplete
Changed in ecryptfs:
status: New → Incomplete
importance: Undecided → High
assignee: nobody → Tyler Hicks (tyhicks)
Revision history for this message
Tyler Hicks (tyhicks) wrote :

I'm fairly certain that this is a dupe of Bug #509180, which has a fix released.

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.