I've setup export policy like in printout on the bottom.
When I mount on a client all root created files are owned by nobody user and group.
Can this be changed to be listed as "root" owned? If I choose nfsv3 then this works as I want.
ClusterT::vserver security file-directory> vserver export-policy rule show -vserver SVM_xxxx -policyname Vol_xxxx -instance
Policy Name: Vol_xxxx
Rule Index: 4
Access Protocol: nfs4
List of Client Match Hostnames, IP Addresses, Netgroups, or Domains: 10.xx.yy.zz
RO Access Rule: sys
RW Access Rule: sys
User ID To Which Anonymous Users Are Mapped: 65534
Superuser Security Types: any
Honor SetUID Bits in SETATTR: true
Allow Creation of Devices: true
The reason for this is because NFSv4 clients send symbolic user/group names rather than numeric userid/groupid as it was in NFSv2 and NFSv3 and the filer needs some way to map this symbolic names to numeric IDs. If the information in /etc/passwd and /etc/group information between the filer and the Linux host does not match, the filer will use nobody:nobody for the user/group file ownership.
Cause: The export policy rule applied to the volume has the superuser set to 'none', which squashes the root user to anonymous user.The anonymous user by default is set to uid 65534, therefore, the files created are owned by uid 65534. UID 65534 is interpreted by some Linux clients such as RedHat as 'nfsnobody'.