Bug 37161 - [ext2fs] ext2 linux file system, error handling large files
Summary: [ext2fs] ext2 linux file system, error handling large files
Status: Closed FIXED
Alias: None
Product: Base System
Classification: Unclassified
Component: kern (show other bugs)
Version: Unspecified
Hardware: Any Any
: Normal Affects Only Me
Assignee: freebsd-emulation (Nobody)
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2002-04-16 23:20 UTC by Ariel Florio
Modified: 2005-07-31 13:17 UTC (History)
0 users

See Also:


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Ariel Florio 2002-04-16 23:20:01 UTC
I've got a ext2 file system with a 30Gb tar file, I need untar this file but the system cut the operation in the first 2,9Gb aprox.
The tar file is being untared on FBSD File System on JBOD Mylex DAC960.
I tried to copy the tar file ( cp ) and the same problem is obtained, the operation cuts in the 10% aprox.

How-To-Repeat: Trying untar the file, copy, list the file contents etc.
Comment 1 Bruce Evans 2002-04-18 09:20:30 UTC
On Tue, 16 Apr 2002, Ariel Florio wrote:

> >Description:
> I've got a ext2 file system with a 30Gb tar file, I need untar this file but the system cut the operation in the first 2,9Gb aprox.
> The tar file is being untared on FBSD File System on JBOD Mylex DAC960.
> I tried to copy the tar file ( cp ) and the same problem is obtained, the operation cuts in the 10% aprox.

I have planned to fix this for years but haven't done anything.  read()s
on ext2fs filesystems don't work for offsets larger than 2G, but the usual
maxfilesize check is ifdefed out at the beginning of ext2_read(), so
such reads blunder on for longer than they should instead of doing the usual
(wrong) error handling very early (read() should return EOF, not EFBIG).

Similarly in ext2 in ext2_write(), except blundering on may damage the
filesystm and the usual error handling is correct.

Bruce
Comment 2 Tim Robbins freebsd_committer freebsd_triage 2004-02-19 11:55:16 UTC
State Changed
From-To: open->patched

This is believed to be fixed in -current & is awaiting MFC.
Comment 3 Bruce M Simpson freebsd_committer freebsd_triage 2004-06-18 13:10:23 UTC
Responsible Changed
From-To: freebsd-bugs->emulation

One for the emulation team
Comment 4 Matteo Riondato 2005-04-12 08:02:33 UTC
This was fixed in RELENG_5 too, long time ago (Feb 2004).
No fix in RELENG_4 yet and I don't know if it is possible to merge to
that branch. If it's not, then this PR can be closed.
Thank you.
Best Regards
-- 
Rionda aka Matteo Riondato
Disinformato per default
G.U.F.I. Staff Member (http://www.gufi.org)
FreeSBIE Developer (http://www.freesbie.org)
Comment 5 Matteo Riondato freebsd_committer freebsd_triage 2005-07-31 13:17:40 UTC
State Changed
From-To: patched->closed

Fixed but won't be MFCed to RELENG_4