No, that's the newest ntirpc. Do you have steps to reproduce this? (I guess you do, since you re-tested with 1.7.1). Daniel On 10/16/2018 02:31 PM, Naresh Babu wrote:

The same crash happens with Ganesha 2.7.1 and libntirpc 1.7.1. Is there any particular version of libntirpc that fixes this issue?

Thanks, Naresh

On Tue, Oct 16, 2018 at 3:26 AM William Allen Simpson <william.allen.simpson@gmail.com <mailto:william.allen.simpson@gmail.com>> wrote:

On 10/16/18 1:55 AM, Soumya Koduri wrote: > This list has been deprecated. Please subscribe to the new devel list at lists.nfs-ganesha.org <http://lists.nfs-ganesha.org>. > There is similar crash (in 'svc_release_it') hit when rpc callback channel is  used. Jiffin reported the same in github [1]. Somewhere in cbk paths probably, xprt->ref is not taken. > > Thanks, > Soumya > > > [1] https://github.com/nfs-ganesha/ntirpc/issues/153 > Please test Ganesha 2.7.1 (rpc 1.7.1).  Not everything has been backported, even though I flagged stuff for backport to rpc 1.6 more than half a year ago.  As I've already mentioned....

I've given up on Ganesha 2.6.  I'm working on rebasing for 2.8.

The log appears to be fairly standard nfsv3 writes, so it's not hugely helpful. Can you get a packet dump of the last minute or 2 before the crash? That would help a lot. If that doesn't work, there appear to be some free levels of veeam, so depending on what you're doing with it, I may be able to try to reproduce. Or I may be able to get them to give me a temporary license to reproduce. Thanks for helping with this. Daniel On 10/16/2018 03:47 PM, Naresh Babu wrote:

nfs.log <https://drive.google.com/file/d/1oEChKhNdzhYWuUOcAAtYNN771AQaZCmj/view?usp=drive_web> It happens with veeam backup copy job to NFS mount. That may be difficult to reproduce in your environment. Please find the attached log that includes all protocol requests before the crash. Unfortunately, none of the standard benchmark tests reproduce this crash. If you need packet traces or any other information, please let me know.

Thanks, Naresh

On Tue, Oct 16, 2018 at 11:35 AM Daniel Gryniewicz <dang@redhat.com <mailto:dang@redhat.com>> wrote:

No, that's the newest ntirpc.

Do you have steps to reproduce this?  (I guess you do, since you re-tested with 1.7.1).

Daniel

On 10/16/2018 02:31 PM, Naresh Babu wrote: > The same crash happens with Ganesha 2.7.1 and libntirpc 1.7.1. Is there > any particular version of libntirpc that fixes this issue? > > Thanks, > Naresh > > On Tue, Oct 16, 2018 at 3:26 AM William Allen Simpson > <william.allen.simpson@gmail.com <mailto:william.allen.simpson@gmail.com> > <mailto:william.allen.simpson@gmail.com <mailto:william.allen.simpson@gmail.com>>> wrote: > >     On 10/16/18 1:55 AM, Soumya Koduri wrote: >      > This list has been deprecated. Please subscribe to the new devel >     list at lists.nfs-ganesha.org <http://lists.nfs-ganesha.org> <http://lists.nfs-ganesha.org>. >      > There is similar crash (in 'svc_release_it') hit when rpc >     callback channel is  used. Jiffin reported the same in github [1]. >     Somewhere in cbk paths probably, xprt->ref is not taken. >      > >      > Thanks, >      > Soumya >      > >      > >      > [1] https://github.com/nfs-ganesha/ntirpc/issues/153 >      > >     Please test Ganesha 2.7.1 (rpc 1.7.1).  Not everything has been >     backported, even though I flagged stuff for backport to rpc 1.6 >     more than half a year ago.  As I've already mentioned.... > >     I've given up on Ganesha 2.6.  I'm working on rebasing for 2.8. >