Bug 1690254

Summary:	Volume create fails with "Commit failed" message if volumes is created using 3 nodes with glusterd restarts on 4th node.
Product:	[Community] GlusterFS	Reporter:	Kshithij Iyer <kiyer>
Component:	glusterd	Assignee:	bugs <bugs>
Status:	CLOSED NOTABUG	QA Contact:
Severity:	medium	Docs Contact:
Priority:	unspecified
Version:	mainline	CC:	amukherj, bugs
Target Milestone:	---
Target Release:	---
Hardware:	x86_64
OS:	Linux
Whiteboard:
Fixed In Version:		Doc Type:	If docs needed, set a value
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2019-04-01 14:30:01 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:
Embargoed:

Description Kshithij Iyer 2019-03-19 07:18:29 UTC

Description of problem
On a 4 node cluster(N1, N2, N3 and N4), create a volume using the first 3 nodes(N1, N2 and N3) and restart glusterd on N4. Volume create fails with commit failed message.  

Version-Release number of selected component (if applicable):
Not sure whatever is there in upstream 

How reproducible:
4/4

Steps to Reproduce:
1. Create a cluster with 4 nodes.
2. Create volume using the first three nodes say N1, N2 and N3.
3. While the create is happening restart the fourth node N4.

Actual results:
Volume create fails with the error "volume create: testvol_distributed: failed: Commit failed on 172.19.2.166. Please check log file for details."

Expected results:
Volume should be successfully created. 

Additional info:

Comment 1 Atin Mukherjee 2019-04-01 14:30:01 UTC

The current behavior is as per design. Please remember in GD1, every nodes have to participate in the transaction and the commit phase should succeed irrespective of if the bricks are hosted on m out of n nodes in the trusted storage pool.