Bug 762714 (GLUSTER-982)

Summary: Three replica config under nfsx blocks IO when one replica goes down
Product: [Community] GlusterFS Reporter: Shehjar Tikoo <shehjart>
Component: coreAssignee: Shehjar Tikoo <shehjart>
Status: CLOSED NOTABUG QA Contact:
Severity: high Docs Contact:
Priority: low    
Version: nfs-alphaCC: gluster-bugs
Target Milestone: ---   
Target Release: ---   
Hardware: All   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: Type: ---
Regression: RTP Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Attachments:
Description Flags
nfsx log file
none
nfsx state dump none

Description Shehjar Tikoo 2010-06-03 07:38:34 UTC
Created attachment 230 [details]
Test program to show the bug.

Comment 1 Shehjar Tikoo 2010-06-03 08:03:48 UTC
My bad. Didnt apply  patch on a clean git clone.

Comment 2 Shehjar Tikoo 2010-06-03 10:37:39 UTC
The server side volfiles are simple posixes+ac+locks+iot+server.

On the gl client side we have clients+afr+nfsx and three replicas under afr.

The test starts with a dd of a 1g file over the nfs mount point. When one of the replicas' glusterfsd is killed, the nfsx stops responding to IO requests. The log file and nfsx glusterfsd state dump is attached. The log shows no CHILD_DOWN and the protocol/client continuing reconnection attempts when it should not.