Bug 788837 - NFS mount hangs when the server closes connection.
Summary: NFS mount hangs when the server closes connection.
Keywords:
Status: CLOSED INSUFFICIENT_DATA
Alias: None
Product: Red Hat Enterprise Linux 6
Classification: Red Hat
Component: nfs-utils
Version: 6.2
Hardware: Unspecified
OS: Unspecified
unspecified
unspecified
Target Milestone: rc
: ---
Assignee: Steve Dickson
QA Contact: Filesystem QE
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2012-02-09 05:37 UTC by Raghavendran Ramanathan
Modified: 2021-01-11 02:28 UTC (History)
5 users (show)

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of:
Environment:
Last Closed: 2013-10-27 11:39:35 UTC
Target Upstream Version:
r.raghavendran: needinfo-
r.raghavendran: needinfo-
r.raghavendran: needinfo-


Attachments (Terms of Use)
packets captured between nfs server and client during mount. (8.09 KB, application/octet-stream)
2012-02-09 16:15 UTC, Raghavendran Ramanathan
no flags Details

Description Raghavendran Ramanathan 2012-02-09 05:37:19 UTC
Description of problem:
I have a NFSv3 server, which closes nfs connection to the client when there is a mis-configuration in the server. But the RHEL6.2 client doesnot recognize this, but keeps sending a FSINFO request on that connection there by hanging the mount command.

Version-Release number of selected component (if applicable):
proto-howler(US):~:rpm -q nfs-utils
nfs-utils-1.2.3-7.el6.i686
proto-howler(US):~:

How reproducible:
Always reproducible.

Steps to Reproduce:
1.Execute a mount command from the client:
mount -t nfs -o vers=3 nfs-server:/ /mnt/
2.Force close the network connection to the client due to a mis configuration.
3.The mount hangs for ever. The network packet trace will show continuos FSINFO calls being sent to the server, even after the server has closed the connection.

Actual results:
Mount hangs.

Expected results:
Mount should error out saying the server closed connection.

Additional info:
When nfsv4 is used, it returns an error code under these conditions, which is recognized by the client and the mount command throws up a Remote-I/O error.

Comment 2 Steve Dickson 2012-02-09 13:23:48 UTC
Could you please attached (in a binary format) the packet trace you refer to in the Description?

Comment 3 Raghavendran Ramanathan 2012-02-09 16:15:20 UTC
Created attachment 560629 [details]
packets captured between nfs server and client during mount.

Comment 4 Steve Dickson 2012-03-01 18:44:35 UTC
(In reply to comment #0)
> Description of problem:
> I have a NFSv3 server, which closes nfs connection to the client when there is
> a mis-configuration in the server. But the RHEL6.2 client doesnot recognize
> this, but keeps sending a FSINFO request on that connection there by hanging
> the mount command.
> 
> Version-Release number of selected component (if applicable):
> proto-howler(US):~:rpm -q nfs-utils
> nfs-utils-1.2.3-7.el6.i686
> proto-howler(US):~:
> 
> How reproducible:
> Always reproducible.
> 
> Steps to Reproduce:
> 1.Execute a mount command from the client:
> mount -t nfs -o vers=3 nfs-server:/ /mnt/
> 2.Force close the network connection to the client due to a mis configuration.
> 3.The mount hangs for ever. The network packet trace will show continuos FSINFO
This is by designed.... Its not really a hung, its retrying in the hope the server will come back...  

> calls being sent to the server, even after the server has closed the
> connection.
> 
> Actual results:
> Mount hangs.
How long does it hang? Does the mount ever fail after a number of mins?
See the timeo value in the nfs(5) man page.

> 
> Expected results:
> Mount should error out saying the server closed connection.
> 
> Additional info:
> When nfsv4 is used, it returns an error code under these conditions, which is
> recognized by the client and the mount command throws up a Remote-I/O error.
This might actually be the a bug... since I am thinking the mount should retry like the v3 mount does...

Comment 7 yanfu,wang 2012-08-08 10:10:21 UTC
hi Raghavendran,
Could you tell me what's the mis-configuration in your server since I want to reproduce the problem? thanks.

Comment 12 Raghavendran Ramanathan 2013-05-07 03:36:18 UTC
Essentially, our in built nfs server simply sent a tcp reset to the client under this condition in NFSv3. To be exact, it is a Netapp Filer. The NFSv4 returned a NFS error.

The word mis-configuration might be misleading, I apologize for that. You can try a nfs server that you can run the nfs server in the foreground and once the client starts a mount command, interrupt the nfs server and kill it.

Comment 13 Steve Dickson 2013-08-14 13:58:11 UTC
Is this still a problem? Because it really looks like a server issued... 

Have you run this by Netapp support?

Comment 14 RHEL Program Management 2013-10-14 00:48:11 UTC
This request was evaluated by Red Hat Product Management for
inclusion in the current release of Red Hat Enterprise Linux.
Because the affected component is not scheduled to be updated
in the current release, Red Hat is unable to address this
request at this time.

Red Hat invites you to ask your support representative to
propose this request, if appropriate, in the next release of
Red Hat Enterprise Linux.

Comment 15 Ric Wheeler 2013-10-27 11:39:35 UTC
Closing this - it does not seem to be an issue with RHEL.

Please reopen if you see still see this.

Best regards,

Ric


Note You need to log in before you can comment on or make changes to this bug.