Description of problem: Each ganesha export with FSAL_GLUSTER consumes 3-10% CPU. I have just found these two notions of this issue on the internet with no solution yet. http://lists.gluster.org/pipermail/gluster-users.old/2016-November/028939.html https://github.com/nfs-ganesha/nfs-ganesha/issues/124 Version-Release number of selected component (if applicable): nfs-ganesha-2.3.3-1.el7.x86_64 nfs-ganesha-gluster-2.3.3-1.el7.x86_64 glusterfs-ganesha-3.8.13-1.el7.x86_64 Steps to Reproduce: 1. Create a gluster volume, disable gnfs 2. Create ganesha configuration 3. Start nfs-ganesha service 4. Observe a thread which is consuming CPU Additional info: Sample export config: EXPORT{ Export_Id= 19 ; Path = "/clientmounts/replicated/test"; FSAL { name = GLUSTER; hostname="localhost"; volume="storage-replicated"; volpath="/clientmounts/test"; } Access_type = RW; Disable_ACL = true; Squash="No_root_squash"; Pseudo="/clientmounts/replicated/test"; Protocols = "3", "4" ; Transports = "UDP","TCP"; SecType = "sys"; }
We have the same problem with nfs-ganesha 2.5.1
(In reply to renaud.fortier from comment #3) > We have the same problem with nfs-ganesha 2.5.1 Can u try with latest ganesha. IMO we resolved this issue on current nfs-ganesha
Indeed! the issue is solved with the latest version. Thanks.
Thanks closing this bug as fixed in nfs-ganesha 2.6