I use files in /etc/exports.d/ to control my exports. That has advantages when dealing with exports from remote machine using some orchestration FW (ansible in my case) - no need to parse/edit /etc/exports file directly. Consider I have 'backend' host specified in /etc/hosts file, I have something like: # cat /etc/exports # cat /etc/exports.d/test.exports /data/nfs-points/results backend(rw,no_root_squash) /data/nfs-points/lock-builder backend(rw,no_root_squash) # exportfs -va exportfs: No file systems exported! exporting backend:/data/nfs-points/lock-builder exporting backend:/data/nfs-points/results # echo $? 1 That ^^ exit value breaks 'systemctl start nfs-server'. During parsing of '/etc/exports', the 'exports_errno' is set (and warning is thrown to stderr) and then it is propagated as program exit value. # rm /etc/exports # exportfs -va exportfs: can't open /etc/exports for reading exportfs: No file systems exported! exporting backend:/data/nfs-points/lock-builder exporting backend:/data/nfs-points/results So no "easy" solution for this? Pavel
Created attachment 959651 [details] proposed solution
Created attachment 959656 [details] Proposed solution (fixed && usage)
*** This bug has been marked as a duplicate of bug 1164477 ***