Home / server / Questions / 367107
Accepted
Steve
Asked: 2012-03-07 22:51:55 +0800 CST

Can't mount NFS share over TCP

  • 772

I'm beating my head against a wall with this one. Environment is 2 x CentOS 6 64bit installs. Both NFS client and server are fully up to date as of 1 hour ago.

I've set up an NFS export on the server:

/opt/nfs     10.1.1.0/24(rw,sync,no_root_squash,no_all_squash)

AFAICT, all relevant NFS services on the server are running:

(2) (0 Jobs) [root@lb01-cbr01-au ~]$ service rpcbind status
rpcbind (pid  20079) is running...
(2) (0 Jobs) [root@lb01-cbr01-au ~]$ service nfslock status
rpc.statd (pid  19986) is running...
(2) (0 Jobs) [root@lb01-cbr01-au ~]$ service nfs status
rpc.svcgssd is stopped
rpc.mountd (pid 20034) is running...
nfsd (pid 20031 20030 20029 20028 20027 20026 20025 20024) is running...
(2) (0 Jobs) [root@lb01-cbr01-au ~]$

On the client, both rpcbind and nfslock report as running.

On the server, the output of rpcinfo for localhost looks good:

[root@lb01-cbr01-au ~]# rpcinfo -p localhost
   program vers proto   port  service
    100000    4   tcp    111  portmapper
    100000    3   tcp    111  portmapper
    100000    2   tcp    111  portmapper
    100000    4   udp    111  portmapper
    100000    3   udp    111  portmapper
    100000    2   udp    111  portmapper
    100024    1   udp  39893  status
    100024    1   tcp  59014  status
    100003    2   tcp   2049  nfs
    100003    3   tcp   2049  nfs
    100003    4   tcp   2049  nfs
    100227    2   tcp   2049  nfs_acl
    100227    3   tcp   2049  nfs_acl
    100003    2   udp   2049  nfs
    100003    3   udp   2049  nfs
    100003    4   udp   2049  nfs
    100227    2   udp   2049  nfs_acl
    100227    3   udp   2049  nfs_acl
    100021    1   udp  44725  nlockmgr
    100021    3   udp  44725  nlockmgr
    100021    4   udp  44725  nlockmgr
    100021    1   tcp  40736  nlockmgr
    100021    3   tcp  40736  nlockmgr
    100021    4   tcp  40736  nlockmgr
    100005    1   udp  55385  mountd
    100005    1   tcp  55481  mountd
    100005    2   udp  46027  mountd
    100005    2   tcp  59968  mountd
    100005    3   udp  45069  mountd
    100005    3   tcp  33231  mountd
[root@lb01-cbr01-au ~]#

Similarly, rpcinfo -p localhost on the client indicates good state:

   program vers proto   port  service
    100000    4   tcp    111  portmapper
    100000    3   tcp    111  portmapper
    100000    2   tcp    111  portmapper
    100000    4   udp    111  portmapper
    100000    3   udp    111  portmapper
    100000    2   udp    111  portmapper
    100024    1   udp  59519  status
    100024    1   tcp  39715  status

The firewall is open between the client and server - an allow rule exists for the interface on both the input and output chain on each host.

From the client, when I issue showmount -e <server_ip>, it hangs for 20 seconds until eventually producing the export list. Issuing rpcinfo -p <server_ip> it also hangs for 20 seconds until eventually returning "rpcinfo: can't contact portmapper: RPC: Remote system error - Connection timed out".

When I attempt to actually mount the export from the client, using:

mount -t nfs 10.1.1.33:/opt/nfs /opt/test/nfs

It hangs for 3m 30 seconds, returning "mount.nfs: Connection timed out".

However, if I try and mount over UDP:

mount -o udp -t nfs 10.1.1.33:/opt/nfs /opt/test/nfs

It instantly succeeds and the mount is accessible.

I haven't done anything to hosts.allow or hosts.deny (both are empty, which from my reading of man 5 hosts_access indicates access will be allowed).

What am I missing here?

Edit: SELinux is permissive on both hosts.

centos nfs tcp

2 Answers

  1. Best Answer

    Turns out there was a "security" feature enabled on our PowerConnect switch that took offense to NFS SYN packets with source ports < 1024 (dos-control tcpflag). Suffice it to say, disabling the feature solved the issue.

    • 2

  2. Although SELinux is permissive, try : setsebool -P nfs_export_all_rw 1 Restart rpcbind, nfs and nfslock And then exportfs -a

    • 0