Bug 765388 (GLUSTER-3656) - Bricks running out of space while rebalance is happening shows sticky bits on the mountpoint
Summary: Bricks running out of space while rebalance is happening shows sticky bits on...
Keywords:
Status: CLOSED CURRENTRELEASE
Alias: GLUSTER-3656
Product: GlusterFS
Classification: Community
Component: core
Version: pre-release
Hardware: x86_64
OS: Linux
medium
medium
Target Milestone: ---
Assignee: Amar Tumballi
QA Contact:
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2011-09-28 10:02 UTC by shylesh
Modified: 2013-12-19 00:06 UTC (History)
2 users (show)

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of:
Environment:
Last Closed:
Regression: ---
Mount Type: ---
Documentation: ---
CRM:
Verified Versions:


Attachments (Terms of Use)

Description shylesh 2011-09-28 10:02:56 UTC
while running rebalance if one of the brick runs out of space log says "no space left" where as rebalance status says "Rebalance completed". 
But mount point contains many files with sticky bit on.

Comment 1 Anand Avati 2011-09-30 04:55:23 UTC
CHANGE: http://review.gluster.com/540 (* cluster/distribute: while rebalance, differentiate between valid) merged in master by Vijay Bellur (vijay)

Comment 2 Anand Avati 2011-09-30 04:57:08 UTC
CHANGE: http://review.gluster.com/541 (* don't set 'sticky' and 'sgid' bits to 0 without checking if the) merged in master by Vijay Bellur (vijay)

Comment 3 Amar Tumballi 2011-09-30 05:47:25 UTC
With the patches merged all the below issues should be fixed:

* if a data-migration fails inbetween copying data, the destination file is reset back to 0 size, with only stickybit (so it will be a empty linkfile)

* if data-migration fails inbetween, glusterd status will be set to 'failed', earlier it could have been 'completed'.

* if source file had stickybit (or sgid bit) set on it, after migration, it was removed at destination. Now, same permission is preserved.


Note You need to log in before you can comment on or make changes to this bug.