Bug 773187

Summary:	[glusterfs-3.3.0qa19]: directory entry not present in first subvolume
Product:	[Community] GlusterFS	Reporter:	Rahul C S <rahulcs>
Component:	distribute	Assignee:	shishir gowda <sgowda>
Status:	CLOSED CURRENTRELEASE	QA Contact:	amainkar
Severity:	medium	Docs Contact:
Priority:	high
Version:	3.3-beta	CC:	amarts, gluster-bugs, nsathyan, rfortier, vbellur
Target Milestone:	---
Target Release:	---
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:
Fixed In Version:	glusterfs-3.4.0	Doc Type:	Bug Fix
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2013-07-24 17:30:20 UTC	Type:	---
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:	3.3.0beta	Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:
Bug Depends On:
Bug Blocks:	817967

Description Rahul C S 2012-01-11 07:28:02 UTC

Description of problem:
I created a distributed-replicate volume & created 2 mounts. Removed all performance translators. 
On 1 client started, untarring ltp-testsuite in a loop & on another client started rm -rf of the ltp directory in a loop.

Got directory not empty error during rm which was fine. Stopped all operations & then reset the volume.

Next i tried removing the ltp directory. I got these errors:
rm: cannot remove `ltp-full-20091031/testcases/ballista/': Directory not empty
rm: cannot remove `ltp-full-20091031/testcases/kernel/syscalls': Directory not empty
rm: cannot remove `ltp-full-20091031/testcases/open_posix_testsuite/conformance': Directory not empty
rm: cannot remove `ltp-full-20091031/testcases/network/stress': Directory not empty


Checked the backend:
The first dht subvolume did not have any entries under ltp-full-20091031/testcases/ballista/ but the second subvolume had directories under it.

echo 3 > /proc/sys/vm/drop_caches on the mountpoint & `stat`d from the mountpoint. Found nothing. 

Therefore it keeps telling. directory not empty.

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


How reproducible:


Steps to Reproduce:
1.
2.
3.
  
Actual results:


Expected results:


Additional info:

Comment 2 Anand Avati 2012-05-09 01:32:59 UTC

CHANGE: http://review.gluster.com/3202 (cluster/dht: rmdir should succeed last on the hashed subvol) merged in master by Anand Avati (avati)