Bug 1004287

Summary:	quota: ping_pong issue
Product:	Red Hat Gluster Storage	Reporter:	Saurabh <saujain>
Component:	glusterd	Assignee:	Susant Kumar Palai <spalai>
Status:	CLOSED ERRATA	QA Contact:	Saurabh <saujain>
Severity:	medium	Docs Contact:
Priority:	unspecified
Version:	2.1	CC:	kparthas, mzywusko, rgowdapp, rhs-bugs, saujain, shaines, spalai, vagarwal, vbellur
Target Milestone:	---	Keywords:	ZStream
Target Release:	---
Hardware:	x86_64
OS:	Linux
Whiteboard:
Fixed In Version:		Doc Type:	Bug Fix
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2013-10-23 06:54:33 UTC	Type:	Bug
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:

Description Saurabh 2013-09-04 10:33:58 UTC

Description of problem:
vol type  :- 6x2 

present volume info

Volume Name: lockvol
Type: Distributed-Replicate
Volume ID: 5f0b92a8-99ec-4de0-8cae-198fc54c2b06
Status: Started
Number of Bricks: 6 x 2 = 12
Transport-type: tcp
Bricks:
Brick1: 10.70.37.213:/rhs/bricks/d1r1-2
Brick2: 10.70.37.145:/rhs/bricks/d1r2-2
Brick3: 10.70.37.68:/rhs/bricks/d2r1-2
Brick4: 10.70.37.76:/rhs/bricks/d2r2-2
Brick5: 10.70.37.213:/rhs/bricks/d3r1-2
Brick6: 10.70.37.145:/rhs/bricks/d3r2-2
Brick7: 10.70.37.68:/rhs/bricks/d4r1-2
Brick8: 10.70.37.76:/rhs/bricks/d4r2-2
Brick9: 10.70.37.213:/rhs/bricks/d5r1-2
Brick10: 10.70.37.145:/rhs/bricks/d5r2-2
Brick11: 10.70.37.68:/rhs/bricks/d6r1-2
Brick12: 10.70.37.76:/rhs/bricks/d6r2-2
Options Reconfigured:
features.quota: off
storage.batch-fsync-delay-usec: 0



Version-Release number of selected component (if applicable):
glusterfs-3.4.0.30rhs-2.el6rhs.x86_64
client,
RHEL 6.4

How reproducible:
seeing on this build

Steps to Reproduce:
1. volume 6x2 
2. mount on a client using nfs
3. ping_pong a 5, from two instances of same mount on same client and one more on a different client.
4. enable quota
5. ctrl+c to stop 3rd instance of ping_pong.
7. ctrl+c first two instances of ping_pong
8. start ping_pong test again this time atleat one instance
9. disable quota. and retry step 8

  
Actual results:
step 3, works fine.
step 4, works fine. ping_pong on all three instances works fine.
step 5, the ping_pong does not change values on first two instances
step 8, it does not respond with intended output like "locks/sec"
step 9, the result is same as step8.


Expected results:

ping pong should have worked properly in all the mentioned conditions.


Additional info:
nfs.log doesn't point out any errors all through the sequence of operations.

Comment 2 Raghavendra G 2013-09-12 07:23:18 UTC

Saurabh,

I am not able to reproduce the bug in my setup. ping_pong works as expected. Couple of observations:
* enabling/disabling quota had no effect on lock rate
* when only third instance of ping_pong is killed, lock rate doesn't shoot up significantly. However, when 2 out of 3 instances is killed, the remaining instance achieves locking rate equal to what a single instance of ping_pong achieves.
* I am using the same version of glusterfs as you.

Can you please confirm this is the behaviour in your setup too?

regards,
Raghavendra.

Comment 4 Susant Kumar Palai 2013-10-04 10:59:32 UTC

Tried to reproduce the issue in my setup (rhs-2.1). Ping-pong worked as expected.

It passed both lock coherency and I/O coherency test.

Comment 7 Susant Kumar Palai 2013-10-23 06:54:33 UTC

Closing the bug as not reproducible in current release-glusterfs3.4.0.35rhs.

Comment 8 Scott Haines 2013-10-23 12:24:13 UTC

Since the problem described in this bug report should be resolved in a recent
advisory, it has been closed with a resolution of ERRATA. 

For information on the advisory, and where to find the updated files, follow
the link below.

If the solution does not work for you, open a new bug report.

http://rhn.redhat.com/errata/RHBA-2013-1262.html