Bug 1323895

Summary:	Can I configure data transfer network and ping check network separately?
Product:	[Community] GlusterFS	Reporter:	ariowner
Component:	rpc	Assignee:	bugs <bugs>
Status:	CLOSED EOL	QA Contact:
Severity:	low	Docs Contact:
Priority:	unspecified
Version:	3.5.6	CC:	bugs, ndevos, rgowdapp, sarumuga
Target Milestone:	---	Keywords:	Triaged
Target Release:	---	Flags:	ykaul: needinfo+
Hardware:	x86_64
OS:	Linux
Whiteboard:
Fixed In Version:		Doc Type:	Bug Fix
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2016-06-17 15:56:26 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 ariowner 2016-04-05 01:59:26 UTC

Description of problem:
Sometimes fuse client would be disconnected with server by ping timed out.
It would be helpful that seperate network between data transfer and ping check.


Version-Release number of selected component (if applicable):
Gluster 3.5.6
RHEL 6.6

How reproducible:
It occured randomly.

Steps to Reproduce:
1. make gluster volume
2. fuse client access gluster volume with heavy data load

Actual results:
Ping timeout occured unexpectedly. disconnect certain client and reconnect very soon. 

Expected results:
If I configure seperate network between data transfer and ping, It will helpful to stable connection.

Additional info:

Comment 1 Niels de Vos 2016-04-05 12:19:58 UTC

This would not really help with the intention of checking if a server is reachable. If the connection for PING and data are split, one could function while the other is really broken (router, gateway, firewall, ...).

Raghavendra, do you remember if PING over a transport has a higher priority than other FOPs? Or, did we discuss updating the last PING-reply when other FOPs over a connection arrived? Were there improvements in newer versions (> 3.5.6)?

Comment 2 Niels de Vos 2016-06-17 15:56:26 UTC

This bug is getting closed because the 3.5 is marked End-Of-Life. There will be no further updates to this version. Please open a new bug against a version that still receives bugfixes if you are still facing this issue in a more current release.