Description of problem on github:
The downstream nfs-ganesha.spec-in.cmake file must be changed to fix this issue.
nfs-ganesha should be able to start with systemd on RHEL 7.4. SELinux should be on, which it is by default.
Ganesha service is not coming up with build: "nfs-ganesha-rgw-2.5.2-11.el7cp.x86_64"
Steps followed to verify the bug:
1. Configured Ceph RGW
2. Installed ganesha and configured ganesha as in this document: https://access.qa.redhat.com/documentation/en-us/red_hat_ceph_storage/2/html-single/object_gateway_guide_for_red_hat_enterprise_linux/#exporting_the_namespace_to_nfs_ganesha
3. Started ganesha service using command "/usr/bin/ganesha.nfsd -f /etc/ganesha/ganesha.conf" without "sudo". Ganesha service not started.
[ubuntu@host118 ~]$ /usr/bin/ganesha.nfsd -f /etc/ganesha/ganesha.conf
[ubuntu@host118 ~]$ !ps
ps aux | grep nfs
ubuntu 15874 0.0 0.0 112660 972 pts/1 S+ 09:28 0:00 grep --color=auto nfs
As per the github issue 212 in bug description /var/log/ganesha should be owned by ganesha:root for the ganesha service to start.
But after nfs installation "/var/log/ganesha" folder is still owned by ganesha:ganesha
[ubuntu@host118 ~]$ ll /var/log/ | grep ganesha
drwxr-xr-x. 2 ganesha ganesha 4096 Oct 25 14:50 ganesha
Planning to move this bug to assigned state. Let me know your view on this. Thanks
There is no info about nfs-ganesha in systemctl or in systemd
[ubuntu@host118 ~]$ systemctl --all | grep ganesha
user-10103.slice loaded active active User Slice of ganesha
[ubuntu@host118 systemd]$ find . | grep ganesh
[ubuntu@host118 systemd]$ ls
bootchart.conf coredump.conf journald.conf logind.conf system system.conf user user.conf
Ceph Build: ceph version 12.2.1-25.el7cp (899fcaceb921cbeb5d875ee24dc4ceb0ef0280a1) luminous (stable)
[ubuntu@host118 systemd]$ sudo rpm -qa | grep nfs
Tried some more Tests:
1. Start/Stop/Restart of "sudo systemctl <> nfs-ganesha" worked without any issues.
2. After node reboot ganesha service not comes up --> Is this expected or separate bug needs to filed ?
3. Ceph-object-gateway document need to be updated for nfs service start command.
Moving this bug to verified state.
As mentioned in comment12 service is starting without any issues.
Filed doc bug for command update bz:1507208
Ran NFS Teuthology Automation scripts on the NFS mount which worked fine without any issues.
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.
For information on the advisory, and where to find the updated
files, follow the link below.
If the solution does not work for you, open a new bug report.