redundant reverse DNS on NFS mount or umount

redundant reverse DNS on NFS mount or umount

[Pádraig Brady]

I've a problem where reverse DNS is being done by the NFS server.
This is very problematic when the DNS server goes away.
As far as I can see there should be no lookups being performed
because I've nothing in /etc/hosts.{allow,deny} and just
numeric hosts in /etc/exports.

I can simulate this easily by putting a bad nameserver in /etc/resolv.c=
onf
and restarting the nfs server. Any mounts will take then take ages.
Others seem to have had this issue as well, but possibly with
non numeric names in /etc/exports:
http://marc.info/?l=3Dlinux-nfs&m=3D110977757630297&w=3D2
http://marc.info/?l=3Dlinux-nfs&m=3D111015343008456&w=3D2

Here are my config details:

# rpm -q nfs-utils
nfs-utils-1.1.0-6.fc8

# cat /etc/resolv.conf
search company.com
nameserver 192.168.2.110

# cat /etc/exports
/home   192.168.2.25(async,rw,all_squash,anonuid=3D500,anongid=3D500)

Any ideas appreciated.

cheers,
P=C3=A1draig.

Re: redundant reverse DNS on NFS mount or umount

[Chuck Lever]

On Feb 17, 2009, at Feb 17, 2009, 10:22 AM, P=E1draig Brady wrote:
> I've a problem where reverse DNS is being done by the NFS server.
> This is very problematic when the DNS server goes away.
> As far as I can see there should be no lookups being performed
> because I've nothing in /etc/hosts.{allow,deny} and just
> numeric hosts in /etc/exports.

rpc.statd performs reverse DNS lookups to verify the hostname of the =20
remote that is about to be monitored.... perhaps that is one place to =20
start looking.

--
Chuck Lever
chuck[dot]lever[at]oracle[dot]com

Re: redundant reverse DNS on NFS mount or umount

[J. Bruce Fields]

On Tue, Feb 17, 2009 at 03:22:35PM +0000, P=C3=A1draig Brady wrote:
> I've a problem where reverse DNS is being done by the NFS server.
> This is very problematic when the DNS server goes away.
> As far as I can see there should be no lookups being performed
> because I've nothing in /etc/hosts.{allow,deny} and just
> numeric hosts in /etc/exports.
>=20
> I can simulate this easily by putting a bad nameserver in /etc/resolv=
=2Econf
> and restarting the nfs server. Any mounts will take then take ages.
> Others seem to have had this issue as well, but possibly with
> non numeric names in /etc/exports:
> http://marc.info/?l=3Dlinux-nfs&m=3D110977757630297&w=3D2
> http://marc.info/?l=3Dlinux-nfs&m=3D111015343008456&w=3D2
>=20
> Here are my config details:
>=20
> # rpm -q nfs-utils
> nfs-utils-1.1.0-6.fc8
>=20
> # cat /etc/resolv.conf
> search company.com
> nameserver 192.168.2.110
>=20
> # cat /etc/exports
> /home   192.168.2.25(async,rw,all_squash,anonuid=3D500,anongid=3D500)
>=20
> Any ideas appreciated.

I agree that DNS lookups shouldn't be required in this case.

It's rpc.mountd that's responsible for this, so if you can figure out
where in the mountd code this is happening, it shouldn't be hard to fix=
=2E

--b.

Re: redundant reverse DNS on NFS mount or umount

[Chuck Lever]

On Mar 2, 2009, at Mar 2, 2009, 12:41 PM, J. Bruce Fields wrote:
> On Tue, Feb 17, 2009 at 03:22:35PM +0000, P=E1draig Brady wrote:
>> I've a problem where reverse DNS is being done by the NFS server.
>> This is very problematic when the DNS server goes away.
>> As far as I can see there should be no lookups being performed
>> because I've nothing in /etc/hosts.{allow,deny} and just
>> numeric hosts in /etc/exports.
>>
>> I can simulate this easily by putting a bad nameserver in /etc/=20
>> resolv.conf
>> and restarting the nfs server. Any mounts will take then take ages.
>> Others seem to have had this issue as well, but possibly with
>> non numeric names in /etc/exports:
>> http://marc.info/?l=3Dlinux-nfs&m=3D110977757630297&w=3D2
>> http://marc.info/?l=3Dlinux-nfs&m=3D111015343008456&w=3D2
>>
>> Here are my config details:
>>
>> # rpm -q nfs-utils
>> nfs-utils-1.1.0-6.fc8
>>
>> # cat /etc/resolv.conf
>> search company.com
>> nameserver 192.168.2.110
>>
>> # cat /etc/exports
>> /home   192.168.2.25(async,rw,all_squash,anonuid=3D500,anongid=3D500=
)
>>
>> Any ideas appreciated.
>
> I agree that DNS lookups shouldn't be required in this case.
>
> It's rpc.mountd that's responsible for this, so if you can figure out
> where in the mountd code this is happening, it shouldn't be hard to =20
> fix.

That would most likely be matchhostname().

--
Chuck Lever
chuck[dot]lever[at]oracle[dot]com