Bug 1154017

Summary:	Fix race between rdma_disconnect and other dependant operations
Product:	[Community] GlusterFS	Reporter:	Raghavendra Talur <rtalur>
Component:	rdma	Assignee:	bugs <bugs>
Status:	CLOSED EOL	QA Contact:
Severity:	unspecified	Docs Contact:
Priority:	unspecified
Version:	pre-release	CC:	bugs
Target Milestone:	---	Keywords:	Triaged
Target Release:	---
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:
Fixed In Version:		Doc Type:	Bug Fix
Doc Text:		Story Points:	---
Clone Of:
Clones:	1154018 (view as bug list)		Environment:
Last Closed:	2015-10-22 15:40:20 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:
Bug Depends On:
Bug Blocks:	1154018

Description Raghavendra Talur 2014-10-17 09:27:32 UTC

Description of problem:
rdma_disconnect function takes longer than socket_shutdown,
this has uncovered cases where we run into race between this and other
dependent functions.

Examples below are using gfapi but the implications are equally applicable to fuse and nfs clients. 


Example:
glfs_fini should wait for child_down event otherwise could lead to a crash
with rdma transport.


How reproducible:
Always

Steps to Reproduce:
1. Use gfapi with rdma transport and connect to a volume
2. Perform some IO
3. Call glfs_fini.

Actual results:
Program using gfapi crashes.

Expected results:
Connection to close successfully without crash.

Comment 1 Kaleb KEITHLEY 2015-10-22 15:40:20 UTC

pre-release version is ambiguous and about to be removed as a choice.

If you believe this is still a bug, please change the status back to NEW and choose the appropriate, applicable version for it.