Can't add NFS share even though dataset exists?

[pyrodex]

UPDATE: I had to do a "systemctl restart nfs-ganesha" for this to show up. Why doesn't a reload catch this issue?

So I created a simple generic dataset and then went to create a corresponding NFS share and getting the following errors from NFS:

Code:

29/09/2021 08:02:41 : epoch 614f44a6 : tardis.lan : ganesha.nfsd-3577127[sigmgr] vfs_create_export :FSAL :CRIT :resolve_posix_filesystem(/mnt/storage/zoneminder) returned No such file or directory (2)
29/09/2021 08:02:41 : epoch 614f44a6 : tardis.lan : ganesha.nfsd-3577127[sigmgr] mdcache_fsal_create_export :FSAL :MAJ :Failed to call create_export on underlying FSAL VFS
29/09/2021 08:02:41 : epoch 614f44a6 : tardis.lan : ganesha.nfsd-3577127[sigmgr] fsal_cfg_commit :CONFIG :CRIT :Could not create export for (/zoneminder) to (/mnt/storage/zoneminder)
29/09/2021 08:02:41 : epoch 614f44a6 : tardis.lan : ganesha.nfsd-3577127[sigmgr] config_errs_to_log :CONFIG :CRIT :Config File (/etc/ganesha/ganesha.conf:183): 1 validation errors in block FSAL
29/09/2021 08:02:41 : epoch 614f44a6 : tardis.lan : ganesha.nfsd-3577127[sigmgr] config_errs_to_log :CONFIG :CRIT :Config File (/etc/ganesha/ganesha.conf:183): Errors processing block (FSAL)
29/09/2021 08:02:41 : epoch 614f44a6 : tardis.lan : ganesha.nfsd-3577127[sigmgr] config_errs_to_log :CONFIG :CRIT :Config File (/etc/ganesha/ganesha.conf:168): 1 validation errors in block EXPORT
29/09/2021 08:02:41 : epoch 614f44a6 : tardis.lan : ganesha.nfsd-3577127[sigmgr] config_errs_to_log :CONFIG :CRIT :Config File (/etc/ganesha/ganesha.conf:168): Errors processing block (EXPORT)

Here is the config from NFS:

Code:

EXPORT {
    Export_Id = 8;
    Path = /mnt/storage/zoneminder;
    Protocols = 3, 4;
    Pseudo = /zoneminder;
    Transports = TCP, UDP;
    Access_Type = None;
    CLIENT {
        Clients = 192.168.24.70;
        Access_Type = RW;
    }
    Squash = AllSquash;

    Anonymous_Uid = 0;
    Anonymous_Gid = 0;
    FSAL {
        Name = VFS;
    }
}

And the file system does exist...

Code:

tardis# cd /mnt/storage/zoneminder
tardis# ls
tardis# pwd
/mnt/storage/zoneminder
tardis# df -h .
Filesystem          Size  Used Avail Use% Mounted on
storage/zoneminder   45T  256K   45T   1% /mnt/storage/zoneminder
tardis# touch test
tardis# ls -la test
-rw-r--r-- 1 root root 0 Sep 29 08:05 test
tardis#

Any thoughts?

 

[sretalla]

Any thoughts?
UPDATE: I had to do a "systemctl restart nfs-ganesha" for this to show up. Why doesn't a reload catch this issue?

Some kind of caching being done for datasets by the daemon that isn't updated with a reload...? Just a guess.