Bug 486768

Summary: connectathon test5 failing when run against rhel4 server
Product: Red Hat Enterprise Linux 4 Reporter: Jeff Layton <jlayton>
Component: kernelAssignee: Jeff Layton <jlayton>
Status: CLOSED DUPLICATE QA Contact: Red Hat Kernel QE team <kernel-qe>
Severity: medium Docs Contact:
Priority: low    
Version: 4.8CC: rwheeler, steved
Target Milestone: rc   
Target Release: ---   
Hardware: All   
OS: Linux   
Whiteboard: cthon09
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2010-03-29 12:49:20 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Attachments:
Description Flags
binary capture showing the problem packet in frame 1716 none

Description Jeff Layton 2009-02-22 01:31:40 UTC
Created attachment 332835 [details]
binary capture showing the problem packet in frame 1716

To reproduce:

# mount -t nfs4 rhel4-server:/export /mnt/cthon
# cd /mnt/cthon
# env NFSTESTDIR=`pwd` /root/cthon04/basic/test5 -s
/root/cthon04/basic/test5: read and write
	/root/cthon04/basic/test5: (/mnt/cthon/redhat-11.test) can't create 'bigfile' : Input/output error

...after this, the directory FH goes stale:

# ls
ls: cannot open directory .: Stale NFS file handle

...the problem seems to be that after an open, we try to do a setattr on the file and the client is sending a bogus stateid (zeroed out).

Comment 1 Jeff Layton 2009-02-22 01:39:28 UTC
I only see this problem when the test is run from recent clients. A RHEL5 client didn't trigger it. So I think the issue is that the client is now sending a zero stateid for this SETATTR call and older clients didn't. The server is throwing back an error, but according to the CITI guys it probably shouldn't be.

I'll have to verify this in the RFC...

Comment 2 Jeff Layton 2010-03-29 12:49:20 UTC

*** This bug has been marked as a duplicate of bug 517162 ***