# NFS Share - chown problem

**URL:** <https://community.unix.com/t/nfs-share-chown-problem/246178>\
**Category:** Solaris\
**Created:** [August 3, 2009, 11:03pm UTC](https://community.unix.com/t/nfs-share-chown-problem/246178 "2009-08-03T23:03:13Z")\
**Posts on this page:** 9\
**Page:** 1

<div class="post-metadata">

**Author:** ![SmartAntz](https://community.unix.com/letter_avatar/smartantz/32/5_5575768a8748004e209b776fc1b2916d.png) [@SmartAntz](https://community.unix.com/u/SmartAntz)\
**Post date:** [August 3, 2009, 11:03pm UTC](https://community.unix.com/t/nfs-share-chown-problem/246178/1 "2009-08-03T23:03:13Z")

</div>

Hi all,

I had share the server Gemini /u10 to the server Centaurus.  
/etc/dfs/dfstab

```nohighlight
share -F nfs -o root=centaurus /u10

```

My problem is:  
in the server Gemini the the owner for /U10 is oracle:dba  
but when in the server Centaurus, I am not able to change the owner become oracle:dba, it show me oracle:nobody in Server Centaurus.

```nohighlight
Server Gemini
drwxr-xr-x 4 oracle dba 512 Aug 3 08:26 u10

/etc/group
dba:*:101:root,oracle,ora10g

```

```nohighlight
Server Centaurus
drwxr-xr-x 4 oracle nobody 512 Aug 3 08:26 u10

etc/group
dba::101:oracle,ora10g,mysql,lss

```

How i need to do,to change the owner become oracle:dba for /u10 in server Centaurus?

---

<div class="post-metadata">

**Author:** ![rram\_k](https://community.unix.com/letter_avatar/rram_k/32/5_5575768a8748004e209b776fc1b2916d.png) [@rram\_k](https://community.unix.com/u/rram_k)\
**Post date:** [August 3, 2009, 11:47pm UTC](https://community.unix.com/t/nfs-share-chown-problem/246178/2 "2009-08-03T23:47:57Z")

</div>

Please checkthe GID of the dba group in both the servers.

My guess is that the "nobody" user in centauras is having the same GID of the "dba" user in Gemini.

Usually group permissions are set based on the name maching the GID of the NFS share directory on the client side.

If this is the case, simply you can change the GID in the NFS client in /etc/group

Regards  
Ramkumar

---

<div class="post-metadata">

**Author:** ![SmartAntz](https://community.unix.com/letter_avatar/smartantz/32/5_5575768a8748004e209b776fc1b2916d.png) [@SmartAntz](https://community.unix.com/u/SmartAntz)\
**Post date:** [August 4, 2009, 8:25pm UTC](https://community.unix.com/t/nfs-share-chown-problem/246178/3 "2009-08-04T20:25:04Z")

</div>

Thanks for ur reply.  
below is the both /etc/group

Gemini

```nohighlight
root::0:
other::1:root
bin::2:root,daemon
sys::3:root,bin,adm
adm::4:root,daemon,pongla
uucp::5:root
mail::6:root
tty::7:root,adm
lp::8:root,adm
nuucp::9:root
staff::10:
daemon::12:root
sysadmin::14:pongla
smmsp::25:
gdm::50:
webservd::80:
postgres::90:
nobody::60001:
noaccess::60002:
nogroup::65534:
sasl::100:
dovecot::104:
nagios::102:
nagioscm::103:nagios
dba:*:101:root,oracle,ora10g

```

Centaurus

```nohighlight
root::0:
other::1:root
bin::2:root,daemon
sys::3:root,bin,adm
adm::4:root,daemon
uucp::5:root
mail::6:root
tty::7:root,adm
lp::8:root,adm
nuucp::9:root
staff::10:
daemon::12:root
sysadmin::14:
nobody::60001:
noaccess::60002:
nogroup::65534:
smmsp::25:
gdm::50:
webservd::80:
postgres::90:
users::100:
onsite::105:
softdev::102:lss,dss,netsys,css,scss,amts,packout,dcm,arch,hsa,eco,ecoman,can
guest::102:devp,guest
oper::103:coherent,brtv,oracle
dba::101:oracle,ora10g,mysql,lss
ftp::30000:ftp,coherent,brtv
sib::60090:tibco,patrol
systems::60095:harvest

```

The nobody have the different GID.. any idea?

---------- Post updated 08-05-09 at 08:25 AM ---------- Previous update was 08-04-09 at 11:55 AM ----------

Hi all , the problem solve  
solution: unshare the filesystem, and redo again the prosess to share the filesystem.

---

<div class="post-metadata">

**Author:** ![ilikecows](https://community.unix.com/user_avatar/community.unix.com/ilikecows/32/1562_2.png) [@ilikecows](https://community.unix.com/u/ilikecows)\
**Post date:** [August 5, 2009, 12:02am UTC](https://community.unix.com/t/nfs-share-chown-problem/246178/4 "2009-08-05T00:02:52Z")

</div>

Are you using any naming services? Please post your /etc/nsswitch.conf from both machines.

---

<div class="post-metadata">

**Author:** ![jlliagre](https://community.unix.com/letter_avatar/jlliagre/32/5_5575768a8748004e209b776fc1b2916d.png) [@jlliagre](https://community.unix.com/u/jlliagre)\
**Post date:** [August 5, 2009, 1:31am UTC](https://community.unix.com/t/nfs-share-chown-problem/246178/5 "2009-08-05T01:31:42Z")

</div>

Centaurus and geminy are likely on different NFSv4 domains, probably because they have a different DNS domain defined in /etc/resolv.conf. In such case, groups ids are ignored as they might not match.

You can force the domain on both servers by setting NFSMAPID\_DOMAIN in /etc/default/nfs.

Search NFSMAPID\_DOMAIN in "man nfs" for details.

---

<div class="post-metadata">

**Author:** ![SmartAntz](https://community.unix.com/letter_avatar/smartantz/32/5_5575768a8748004e209b776fc1b2916d.png) [@SmartAntz](https://community.unix.com/u/SmartAntz)\
**Post date:** [August 5, 2009, 9:03pm UTC](https://community.unix.com/t/nfs-share-chown-problem/246178/6 "2009-08-05T21:03:10Z")

</div>

> [@jlliagre](#):
>
> Centaurus and geminy are likely on different NFSv4 domains, probably because they have a different DNS domain defined in /etc/resolv.conf. In such case, groups ids are ignored as they might not match.
> 
> You can force the domain on both servers by setting NFSMAPID\_DOMAIN in /etc/default/nfs.
> 
> Search NFSMAPID\_DOMAIN in "man nfs" for details.

Both the resolv.conf have the same domain.🙂

---

<div class="post-metadata">

**Author:** ![jlliagre](https://community.unix.com/letter_avatar/jlliagre/32/5_5575768a8748004e209b776fc1b2916d.png) [@jlliagre](https://community.unix.com/u/jlliagre)\
**Post date:** [August 6, 2009, 4:14am UTC](https://community.unix.com/t/nfs-share-chown-problem/246178/7 "2009-08-06T04:14:12Z")

</div>

These boxes are very likely on different NFSv4 domains. That would explain the group being ignored. Don't focus on resolv.conf and fix your configuration if you want this to be corrected.

---

<div class="post-metadata">

**Author:** ![incredible](https://community.unix.com/letter_avatar/incredible/32/5_5575768a8748004e209b776fc1b2916d.png) [@incredible](https://community.unix.com/u/incredible)\
**Post date:** [August 7, 2009, 12:10am UTC](https://community.unix.com/t/nfs-share-chown-problem/246178/8 "2009-08-07T00:10:09Z")

</div>

check the nfs version on both nodes.  
nfsstat -a or nfsstat -m \<mount point name\>  
If they differ between the 2 systems, mount them using the same NFS verison and try  
mount -F nfs -o vers=3 hostname:directory mountpoint

---

<div class="post-metadata">

**Author:** ![jlliagre](https://community.unix.com/letter_avatar/jlliagre/32/5_5575768a8748004e209b776fc1b2916d.png) [@jlliagre](https://community.unix.com/u/jlliagre)\
**Post date:** [August 7, 2009, 3:34am UTC](https://community.unix.com/t/nfs-share-chown-problem/246178/9 "2009-08-07T03:34:51Z")

</div>

This issue is obviously related to NFSv4 being used. I don't get what you mean by different versions at both sides. A single protocol can be used for this mount, it can't be mixed.

Downgrading to NFSv3 will fix the problem but it would be better in my opinion to properly configure NVSv4.
