Daniel, I don't have this available from my previous attempt but I'm going through this now: https://wiki.centos.org/TipsAndTricks/ABRT to get this particular server set up for crash handling and reporting, and then I'll test again. If there's anything else that I should do to improve what I can provide to you folks, please let me know, or point me at what I should read, and I'll do what I can. Thanks, Todd On Fri, 17 Apr 2020, Daniel Gryniewicz wrote:

An upgrade should never trigger a segfault. Do you have backtraces saved somewhere?

Daniel

On 4/15/20 12:34 PM, Todd Pfaff wrote:

...

Thanks for the feedback, Daniel.  I quickly tried upgrading to from 2.8 to 3.2 on our CentOS 7 nfs-ganesha server.  The 'yum upgrade' process appeared to go just fine - it was simply a matter of:

  systemctl stop nfs-ganesha   yum install centos-release-nfs-ganesha30   yum update   systemctl start nfs-ganesha

to go from nfs-ganesha-2.8.3-4.el7.x86_64 to nfs-ganesha-3.2-6.el7.x86_64. Unfortunately that was the extent of my joy.  Upon startup, ganesha.nfsd crashed with these log entries:

  Apr 13 22:48:17 ganesha systemd: Started NFS-Ganesha file server.   Apr 13 22:48:17 ganesha rpc.statd[30746]: Received SM_UNMON_ALL request   from ganesha while not monitoring any hosts   Apr 13 22:48:52 ganesha kernel: ganesha.nfsd[30820]: segfault at   7f561b35d26f ip 00007f561b35d26f sp 00007f561b35bc18 error 15   Apr 13 22:48:52 ganesha systemd: nfs-ganesha.service: main process   exited, code=killed, status=11/SEGV

I quickly downgraded since I wasn't in a position to try to debug this at that time.

I admit that I did not dig for documentation that describes potential issues when upgrading from 2.8 to 3.2.  Are there any that I should be aware of before trying this again?  For example, are there any significant ganesha.conf changes required for this upgrade?

Thanks, Todd

On Thu, 9 Apr 2020, Daniel Gryniewicz wrote:

...

3.2 is newer, and therefore has more bug fixes, than 2.8.3.  It also has much more development, so it has more potential bugs.  I would consider the newest stable version (3.2 in this case) to be the most stable version.

That said, we are about to release new versions of both 2.8.x and 3.x in the next week or two, so you may want to wait for that.

Daniel

On 4/8/20 10:40 PM, Todd Pfaff wrote:

...

I'm having stability problems with nfs-ganesha 2.8 and PROXY FSAL on CentOS 7.

I haven't yet tried nfs-ganesha 3.0 but I could probably quickly and easily move to that if there is an advantage in doing so.

Before I spend time switching from 2.8 to 3.0, can anyone tell me whether either of these versions of nfs-ganesha is considered to be more stable than the other?

   centos-release-nfs-ganesha28.noarch : NFS-Ganesha 2.8 packages from the    CentOS                                        : Storage SIG repository    centos-release-nfs-ganesha30.noarch : NFS-Ganesha 3.0 packages from the    CentOS                                        : Storage SIG repository

What about PROXY FSAL in either of these versions - is one or the other known to be more stable?

Is there a more robust version of nfs-ganesha and PROXY FSAL that I should consider trying instead?

Thanks, Todd _______________________________________________ Support mailing list -- support@lists.nfs-ganesha.org To unsubscribe send an email to support-leave@lists.nfs-ganesha.org