Bug 662782

Summary: Bug fixes to the 2.6.37 NFS Client
Product: Red Hat Enterprise Linux 6 Reporter: Steve Dickson <steved>
Component: kernelAssignee: Steve Dickson <steved>
Status: CLOSED ERRATA QA Contact: Filesystem QE <fs-qe>
Severity: urgent Docs Contact:
Priority: urgent    
Version: 6.1CC: arozansk, dhoward, dtian, eguan, fhrbata, jjarvis, kzhang, qcai, rwheeler, tsmetana, yanwang
Target Milestone: rcKeywords: ZStream
Target Release: 6.1   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: kernel-2.6.32-94.el6 Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2011-05-19 12:07:44 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:
Bug Depends On:    
Bug Blocks: 678026    
Attachments:
Description Flags
ltp-nfstest log none

Description Steve Dickson 2010-12-13 20:40:08 UTC 
Description of problem:
These are the bug fixes that were posted in upstream 
for the 2.6.37 NFS client. We need to continue to stay 
up with upstream for any chance of supporting pNFS 
in 6.1
Comment 2 RHEL Program Management 2010-12-17 04:20:12 UTC 
This request was evaluated by Red Hat Product Management for inclusion
in a Red Hat Enterprise Linux maintenance release. Product Management has 
requested further review of this request by Red Hat Engineering, for potential
inclusion in a Red Hat Enterprise Linux Update release for currently deployed 
products. This request is not yet committed for inclusion in an Update release.
Comment 6 Aristeu Rozanski 2011-01-07 12:41:09 UTC 
Patch(es) available on kernel-2.6.32-94.el6
Comment 10 IBM Bug Proxy 2011-02-09 22:41:34 UTC 
------- Comment From tpnoonan.com 2011-02-09 17:33 EDT-------
Hi Red Hat, Please consider for RHEL6.0.z. as this is critical for perfrmance of workloads with Cloud. Thanks
Comment 15 IBM Bug Proxy 2011-02-25 18:00:45 UTC 
------- Comment From pbadari.com 2011-02-25 12:52 EDT-------
I would like to suggest that instead of considering all NFS changes that are targetted for RHEL6.1,
lets focus on specific fixes we need for RHEL6.0-z stream.

I will provide list of fixes we need for z-stream:

1) fix similar to the one descrined in bug 67632 (for RHEL5 series)

@@ -709,7 +709,7 @@ static ssize_t nfs_direct_write(struct k
return -ENOMEM;
nfs_alloc_commit_data(dreq);

-       if (dreq->commit_data == NULL || count<  wsize)
+       if (dreq->commit_data == NULL || count<= wsize)
sync = FLUSH_STABLE;

dreq->inode = inode;

2) we may have another one to improve O_SYNC performance. let me look at that also.
Comment 16 IBM Bug Proxy 2011-02-25 20:30:55 UTC 
------- Comment From tpnoonan.com 2011-02-25 15:28 EDT-------
(In reply to comment #5)
> I would like to suggest that instead of considering all NFS changes that are
> targetted for RHEL6.1,
> lets focus on specific fixes we need for RHEL6.0-z stream.
> I will provide list of fixes we need for z-stream:
> 1) fix similar to the one descrined in bug 67632 (for RHEL5 series)
> @@ -709,7 +709,7 @@ static ssize_t nfs_direct_write(struct k
>                 return -ENOMEM;
>         nfs_alloc_commit_data(dreq);
> -       if (dreq->commit_data == NULL || count<  wsize)
> +       if (dreq->commit_data == NULL || count<= wsize)
>                 sync = FLUSH_STABLE;
>         dreq->inode = inode;
> 2) we may have another one to improve O_SYNC performance. let me look at that
> also.

ltcbz 67632 is rhbz 643441/677172 ltc bz 67632  -  RIT00404862- pure nfs client performance using odirect
Comment 17 yanfu,wang 2011-03-02 11:11:21 UTC 
verified these patch applied in kernel-2.6.32-118.el6.
run connectathon test suite against RHEL6.1 Beta kernel 2.6.32-118.el6 and results passed: 
https://beaker.engineering.redhat.com/jobs/56442 https://beaker.engineering.redhat.com/jobs/56443 https://beaker.engineering.redhat.com/jobs/56449 https://beaker.engineering.redhat.com/jobs/56450

run ltp-nfstest test suite against kernel 2.6.32-118.el6 and results passed, will attached log later
Comment 18 yanfu,wang 2011-03-02 11:12:52 UTC 
Created attachment 481840 [details]
ltp-nfstest log

run it on nfsv{2,3,4}+{tcp,udp}
Comment 19 yanfu,wang 2011-03-02 11:13:42 UTC 
if need to run other testsuites pls let me know.
Comment 20 yanfu,wang 2011-03-17 08:43:27 UTC 
Testing NFS client handling writes with O_DIRECT against https://bugzilla.redhat.com/show_bug.cgi?id=643441#c17

reproduced on RHEL6.0 GA:
[root@hp-p6100z-02 ~]# cat /proc/mounts 
...
hp-dl360g5-02.rhts.eng.bos.redhat.com:/export/ /mnt nfs4 rw,relatime,vers=4,rsize=131072,wsize=131072,namlen=255,hard,proto=tcp,port=0,timeo=600,retrans=2,sec=sys,clientaddr=10.16.44.27,minorversion=0,addr=10.16.64.101 0 0

[root@hp-p6100z-02 ~]# ./write /mnt/file 1 131072

[root@hp-p6100z-02 ~]# tshark -R nfs
Running as user "root" and group "root". This could be dangerous.
Capturing on eth0
  8.060457  10.16.44.27 -> 10.16.64.101 NFS V4 COMP Call <EMPTY> PUTFH;ACCESS ACCESS;GETATTR GETATTR
  8.060844 10.16.64.101 -> 10.16.44.27  NFS V4 COMP Reply (Call In 568) <EMPTY> PUTFH;ACCESS ACCESS;GETATTR GETATTR
  8.063700  10.16.44.27 -> 10.16.64.101 NFS V4 COMP Call <EMPTY> SETCLIENTID SETCLIENTID
  8.063993 10.16.64.101 -> 10.16.44.27  NFS V4 COMP Reply (Call In 571) <EMPTY> SETCLIENTID SETCLIENTID
  8.064081  10.16.44.27 -> 10.16.64.101 NFS V4 COMP Call <EMPTY> SETCLIENTID_CONFIRM SETCLIENTID_CONFIRM;PUTROOTFH PUTROOTFH;GETATTR GETATTR
  8.064645 10.16.64.101 -> 10.16.44.27  NFS V1 CB_NULL Call
  8.064738  10.16.44.27 -> 10.16.64.101 NFS V1 CB_NULL Reply (Call In 577)
  8.065215 10.16.64.101 -> 10.16.44.27  NFS V4 COMP Reply (Call In 573) <EMPTY> SETCLIENTID_CONFIRM SETCLIENTID_CONFIRM;PUTROOTFH PUTROOTFH;GETATTR GETATTR
  8.065320  10.16.44.27 -> 10.16.64.101 NFS V4 COMP Call <EMPTY> PUTFH;SAVEFH SAVEFH;OPEN OPEN;GETFH GETFH;GETATTR GETATTR;RESTOREFH RESTOREFH;GETATTR GETATTR
  8.065669 10.16.64.101 -> 10.16.44.27  NFS V4 COMP Reply (Call In 582) <EMPTY> PUTFH;SAVEFH SAVEFH;OPEN OPEN;GETFH GETFH;GETATTR GETATTR;RESTOREFH RESTOREFH;GETATTR GETATTR
  8.065753  10.16.44.27 -> 10.16.64.101 NFS V4 COMP Call <EMPTY> PUTFH;OPEN_CONFIRM OPEN_CONFIRM
  8.096465 10.16.64.101 -> 10.16.44.27  NFS V4 COMP Reply (Call In 584) <EMPTY> PUTFH;OPEN_CONFIRM OPEN_CONFIRM
  8.105602  10.16.44.27 -> 10.16.64.101 NFS V4 COMP Call <EMPTY> PUTFH;WRITE WRITE;GETATTR GETATTR
  8.109234 10.16.64.101 -> 10.16.44.27  NFS V4 COMP Reply (Call In 708) <EMPTY> PUTFH;WRITE WRITE;GETATTR GETATTR
  8.109342  10.16.44.27 -> 10.16.64.101 NFS V4 COMP Call <EMPTY> PUTFH;COMMIT COMMIT;GETATTR GETATTR
  8.138476 10.16.64.101 -> 10.16.44.27  NFS V4 COMP Reply (Call In 723) <EMPTY> PUTFH;COMMIT COMMIT;GETATTR GETATTR
  8.138589  10.16.44.27 -> 10.16.64.101 NFS V4 COMP Call <EMPTY> PUTFH;CLOSE CLOSE;GETATTR GETATTR
  8.138853 10.16.64.101 -> 10.16.44.27  NFS V4 COMP Reply (Call In 725) <EMPTY> PUTFH;CLOSE CLOSE;GETATTR GETATTR

verified on RHEL6.1 kernel 2.6.32-118.el6:
[root@sun-x8450-01 ~]# tshark -R nfs -i eth6
Running as user "root" and group "root". This could be dangerous.
Capturing on eth6
  4.262935  10.16.64.67 -> 10.16.64.101 NFS V4 COMP Call <EMPTY> PUTFH;SAVEFH SAVEFH;OPEN OPEN;GETFH GETFH;GETATTR GETATTR;RESTOREFH RESTOREFH;GETATTR GETATTR
  4.263320 10.16.64.101 -> 10.16.64.67  NFS V4 COMP Reply (Call In 71) <EMPTY> PUTFH;SAVEFH SAVEFH;OPEN OPEN;GETFH GETFH;GETATTR GETATTR;RESTOREFH RESTOREFH;GETATTR GETATTR
  4.263417  10.16.64.67 -> 10.16.64.101 NFS V4 COMP Call <EMPTY> PUTFH;OPEN_CONFIRM OPEN_CONFIRM
  4.263680 10.16.64.101 -> 10.16.64.67  NFS V4 COMP Reply (Call In 74) <EMPTY> PUTFH;OPEN_CONFIRM OPEN_CONFIRM
  4.268239  10.16.64.67 -> 10.16.64.101 NFS V4 COMP Call <EMPTY> PUTFH;WRITE WRITE;GETATTR GETATTR
  4.279462 10.16.64.101 -> 10.16.64.67  NFS V4 COMP Reply (Call In 101) <EMPTY> PUTFH;WRITE WRITE;GETATTR GETATTR
  4.279549  10.16.64.67 -> 10.16.64.101 NFS V4 COMP Call <EMPTY> PUTFH;CLOSE CLOSE;GETATTR GETATTR
  4.279962 10.16.64.101 -> 10.16.64.67  NFS V4 COMP Reply (Call In 104) <EMPTY> PUTFH;CLOSE CLOSE;GETATTR GETATTR
Comment 21 yanfu,wang 2011-03-17 08:46:30 UTC 
verified on RHEL6.1 kernel 2.6.32-122.el6 and result is also ok:
hp-dl360g5-02.rhts.eng.bos.redhat.com:/export/ /mnt nfs4 rw,relatime,vers=4,rsize=131072,wsize=131072,namlen=255,hard,proto=tcp,port=0,timeo=600,retrans=2,sec=sys,clientaddr=10.16.64.57,minorversion=0,local_lock=none,addr=10.16.64.101 0 0

[root@dell-pe800-01 ~]# ./write /mnt/file1 1 131072

  1.656516  10.16.64.57 -> 10.16.64.101 NFS V4 COMP Call <EMPTY> SETCLIENTID SETCLIENTID
  1.656785 10.16.64.101 -> 10.16.64.57  NFS V4 COMP Reply (Call In 28) <EMPTY> SETCLIENTID SETCLIENTID
  1.656851  10.16.64.57 -> 10.16.64.101 NFS V4 COMP Call <EMPTY> SETCLIENTID_CONFIRM SETCLIENTID_CONFIRM;PUTROOTFH PUTROOTFH;GETATTR GETATTR
  1.657110 10.16.64.101 -> 10.16.64.57  NFS V4 COMP Reply (Call In 31) <EMPTY> SETCLIENTID_CONFIRM SETCLIENTID_CONFIRM;PUTROOTFH PUTROOTFH;GETATTR GETATTR
  1.657169  10.16.64.57 -> 10.16.64.101 NFS V4 COMP Call <EMPTY> PUTFH;SAVEFH SAVEFH;OPEN OPEN;GETFH GETFH;GETATTR GETATTR;RESTOREFH RESTOREFH;GETATTR GETATTR
  1.657501 10.16.64.101 -> 10.16.64.57  NFS V4 COMP Reply (Call In 33) <EMPTY> PUTFH;SAVEFH SAVEFH;OPEN OPEN;GETFH GETFH;GETATTR GETATTR;RESTOREFH RESTOREFH;GETATTR GETATTR
  1.657551  10.16.64.57 -> 10.16.64.101 NFS V4 COMP Call <EMPTY> PUTFH;OPEN_CONFIRM OPEN_CONFIRM
  1.660841 10.16.64.101 -> 10.16.64.57  NFS V1 CB_NULL Call
  1.660870  10.16.64.57 -> 10.16.64.101 NFS V1 CB_NULL Reply (Call In 39)
  1.683694 10.16.64.101 -> 10.16.64.57  NFS V4 COMP Reply (Call In 35) <EMPTY> PUTFH;OPEN_CONFIRM OPEN_CONFIRM
  1.687534  10.16.64.57 -> 10.16.64.101 NFS V4 COMP Call <EMPTY> PUTFH;WRITE WRITE;GETATTR GETATTR
  1.700134 10.16.64.101 -> 10.16.64.57  NFS V4 COMP Reply (Call In 73) <EMPTY> PUTFH;WRITE WRITE;GETATTR GETATTR
  1.700204  10.16.64.57 -> 10.16.64.101 NFS V4 COMP Call <EMPTY> PUTFH;CLOSE CLOSE;GETATTR GETATTR
  1.700448 10.16.64.101 -> 10.16.64.57  NFS V4 COMP Reply (Call In 77) <EMPTY> PUTFH;CLOSE CLOSE;GETATTR GETATTR

[root@dell-pe800-01 ~]# ./write /mnt/file1 1 131073
  2.372354  10.16.64.57 -> 10.16.64.101 NFS V4 COMP Call <EMPTY> PUTFH;ACCESS ACCESS;GETATTR GETATTR
  2.372659 10.16.64.101 -> 10.16.64.57  NFS V4 COMP Reply (Call In 67) <EMPTY> PUTFH;ACCESS ACCESS;GETATTR GETATTR
  2.372749  10.16.64.57 -> 10.16.64.101 NFS V4 COMP Call <EMPTY> PUTFH;SAVEFH SAVEFH;OPEN OPEN;GETFH GETFH;GETATTR GETATTR;RESTOREFH RESTOREFH;GETATTR GETATTR
  2.373099 10.16.64.101 -> 10.16.64.57  NFS V4 COMP Reply (Call In 71) <EMPTY> PUTFH;SAVEFH SAVEFH;OPEN OPEN;GETFH GETFH;GETATTR GETATTR;RESTOREFH RESTOREFH;GETATTR GETATTR
  2.373143  10.16.64.57 -> 10.16.64.101 NFS V4 COMP Call <EMPTY> PUTFH;OPEN_CONFIRM OPEN_CONFIRM
  2.373414 10.16.64.101 -> 10.16.64.57  NFS V4 COMP Reply (Call In 73) <EMPTY> PUTFH;OPEN_CONFIRM OPEN_CONFIRM
  2.377200  10.16.64.57 -> 10.16.64.101 NFS V4 COMP Call <EMPTY> PUTFH;WRITE WRITE;GETATTR GETATTR
  2.377907 10.16.64.101 -> 10.16.64.57  NFS V4 COMP Reply (Call In 97) <EMPTY> PUTFH;WRITE WRITE;GETATTR GETATTR
  2.378013 10.16.64.101 -> 10.16.64.57  NFS V4 COMP Reply (Call In 97) <EMPTY> PUTFH;WRITE WRITE;GETATTR GETATTR
  2.378054  10.16.64.57 -> 10.16.64.101 NFS V4 COMP Call <EMPTY> PUTFH;COMMIT COMMIT;GETATTR GETATTR
  2.406684 10.16.64.101 -> 10.16.64.57  NFS V4 COMP Reply (Call In 102) <EMPTY> PUTFH;COMMIT COMMIT;GETATTR GETATTR
  2.406737  10.16.64.57 -> 10.16.64.101 NFS V4 COMP Call <EMPTY> PUTFH;CLOSE CLOSE;GETATTR GETATTR
  2.406982 10.16.64.101 -> 10.16.64.57  NFS V4 COMP Reply (Call In 104) <EMPTY> PUTFH;CLOSE CLOSE;GETATTR GETATTR
Comment 22 errata-xmlrpc 2011-05-19 12:07:44 UTC 
An advisory has been issued which should help the problem
described in this bug report. This report is therefore being
closed with a resolution of ERRATA. For more information
on therefore solution and/or where to find the updated files,
please follow the link below. You may reopen this bug report
if the solution does not work for you.

http://rhn.redhat.com/errata/RHSA-2011-0542.html