Note: This is a public test instance of Red Hat Bugzilla. The data contained within is a snapshot of the live data so any changes you make will not be reflected in the production Bugzilla. Email is disabled so feel free to test any aspect of the site that you want. File any problems you find or give feedback at bugzilla.redhat.com.
Bug 490628 - the quota is ineffective while mount ext4 fs with delalloc option
Summary: the quota is ineffective while mount ext4 fs with delalloc option
Keywords:
Status: CLOSED RAWHIDE
Alias: None
Product: Fedora
Classification: Fedora
Component: kernel
Version: rawhide
Hardware: x86_64
OS: Linux
low
medium
Target Milestone: ---
Assignee: Eric Sandeen
QA Contact: Fedora Extras Quality Assurance
URL: http://marc.info/?l=linux-ext4&m=1233...
Whiteboard:
Depends On:
Blocks: F11Target
TreeView+ depends on / blocked
 
Reported: 2009-03-17 11:39 UTC by Gui Xiaohua
Modified: 2013-01-10 05:06 UTC (History)
4 users (show)

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Quota on ext4 filesystems is not working correctly when the default delayed allocation mode is enabled. Mounting with -o nodelalloc should resolve this issue in the near term, and the upstream 2.6.30 will have this fixed properly for a future F11 kernel update.
Clone Of:
Environment:
Last Closed: 2009-05-11 21:57:08 UTC
Type: ---
Embargoed:

Attachments (Terms of Use)

Description Gui Xiaohua 2009-03-17 11:39:37 UTC 
Description of problem:
 the quota is ineffective while mount ext4 fs with delalloc option，
 but if I mount ext4 fs with nodelalloc,the quota is effective.

 Version-Release number of selected component (if applicable):
 Fedora-11-Alpha

 How reproducible:

 Steps to Reproduce:
 1.mkfs.ext4 /dev/sda7
 2.mount -t ext4 -o usrquota /dev/sda7 /mnt
 3.quotacheck -avug
 4.quotaon -avug
 5.setquota -u quser1 5 10 0 0 /mnt
 6.chmod 777 /mnt
 7.su quser1
 8.dd if=/dev/zero of=/mnt/file count=30
 9.sync
   
 Actual results:
 # quota -u quser1
 Disk quota for user quser1(uid 501)
 Filesystem  blocks  quota  limit  grace      files  quota   limit  grace
 /dev/sda7     16       5      10    7days      1       0       0       0    


 Expected results:
 # quota -u quser1
 Disk quota for user quser1(uid 501)
 Filesystem  blocks  quota  limit    grace      files  quota   limit  grace
 /dev/sda7     10       5      10    7days      1       0       0       0 

 Additional info:
Comment 1 Phil Knirsch 2009-03-17 12:06:38 UTC 
Reassigning to kernel component.

Thanks & regards, Phil
Comment 2 Chuck Ebbert 2009-04-24 02:21:02 UTC 
commit 60e58e0f30e723464c2a7d34b71b8675566c572d apparently fixes this
Comment 3 Gui Xiaohua 2009-04-24 05:15:12 UTC 
(In reply to comment #2)
> commit 60e58e0f30e723464c2a7d34b71b8675566c572d apparently fixes this  

I don't consider the patch fixes the problem already, i tested the patch
Comment 4 Eric Sandeen 2009-04-27 15:38:06 UTC 
That commit has a lot of dependencies....

You might test the latest Rawhide kernel to see if it's working better, it should have all of the upstream quota fixes to date.

-Eric
Comment 5 Eric Sandeen 2009-05-11 21:57:08 UTC 
From some brief testing, the 2.6.30 kernel should have this fixed (or at least working better?  For very small quota values I see some issues, but it generally seems to be working)

I won't be able to pull those chagnes into the initial F11 release, but when the .30 kernel hits F11 as an update it should be fixed.

Sorry for not having it working in GA, we'll add a release note about this issue.

Closing RAWHIDE for now, if you can test a rawhide kernel and still encounter quota bugs, please let me know.

Thanks,
-Eric
Comment 6 Eric Sandeen 2009-05-11 21:57:08 UTC 
Release note added. If any revisions are required, please set the 
"requires_release_notes" flag to "?" and edit the "Release Notes" field accordingly.
All revisions will be proofread by the Engineering Content Services team.

New Contents:
Quota on ext4 filesystems is not working correctly when the default delayed allocation mode is enabled.  Mounting with -o nodelalloc should resolve this issue in the near term, and the upstream 2.6.30 will have this fixed properly for a future F11 kernel update.
Note You need to log in before you can comment on or make changes to this bug.