Bug 1469292

Description of problem:
NFS File System resource reports all metrics as NaN if NFS server does not expose NFSv2 mounts.

This is a big problem for modern operating systems such as RHEL 7 which no longer supports NFSv2 and disables it by default. 

The result is that `Free Files` ('fileSystemUsage.freeFiles`), `Free Space` ('fileSystemUsage.free`), `Used` ('fileSystemUsage.used`), `Used Percentage` ('fileSystemUsage.usePercent`) and `Capacity` ('fileSystemUsage.total`) always report a metric value of `NaN`.

Version-Release number of selected component (if applicable):
3.3.8

How reproducible:
Always

Steps to Reproduce:
1.  Configure NFS server to export NFS mounts:

----
yum install -y nfs-utils libnfsidmap
systemctl enable rpcbind
systemctl enable nfs-server
systemctl start rpcbind
systemctl start nfs-server
systemctl start rpc-statd
systemctl start nfs-idmapd
firewall-cmd --permanent --zone public --add-service mountd
firewall-cmd --permanent --zone public --add-service rpc-bind
firewall-cmd --permanent --zone public --add-service nfs
firewall-cmd --reload

# where our exports are mounted and shared from
mkdir -p /export/data
# a simple example directory to make available via NFS
mkdir -p /data
# manual mount
# mount --bind /data /export/data
# mount on boot or mount -a
echo '/data    /export/data   none    bind  0  0' >>/etc/fstab
# create the export entries for our subnet/clients
echo '/export      10.10.0.0/21(rw,fsid=0,insecure,no_subtree_check,async)' >>/etc/exports
echo '/export/data 10.10.0.0/21(rw,nohide,insecure,no_subtree_check,async)' >>/etc/exports
# refresh/reload the NFS exports
exportfs -r

# if not using RHEL 7, explicitly disable NFSv2:
echo "RPCNFSDARGS=\"--no-nfs-version 2\"" >>/etc/sysconfig/nfs
----

2.  Install, configure and start JBoss ON 3.3 system.
3.  Configure agent host to mount NFS mounts exported from NFS server:

----
yum install -y nfs-utils libnfsidmap
systemctl enable rpcbind
systemctl start rpcbind
mkdir -p /mnt/data
# mount $HOSTNAME:/export/data /mnt/data
echo "${HOSTNAME}"':/export/data /mnt/data nfs rw,sync,hard,intr 0 0' >>/etc/fstab
mount -a
----

4.  From JBoss ON system inventory, perform service/detail discovery scan for agent so new NFS mounts can be discovered and imported.
5.  From imported NFS file system resource, enable all metrics with a collection interval of 30 seconds.
6.  View the NFS file system resource's Metrics Monitoring page and Traits Monitoring page.

Actual results:
All Live Values report `--no data available--` and the most recent minimum, maximum, and average report `NaN`.

Expected results:
Measurements Free Files, Free Space, Used, Used Percentage and trait Capacity should reflect a numeric value of >= 0.

Additional info:
This issue is a direct result of the RPC ping that is performed by the Sigar code. A hard-coded version is used for the ping[1].

If version 2 of the RPC program is not available, the ping will return (9) RPC: Program/version mismatch. 

Please note that if the ping used version 3 or version 4 or if NFSv2 was enabled on the remote NFS server (if it is an option), all begins to work fine. Basically, the version used by ping is an exact match. Meaning that whatever version is used by ping must be what is running on the remote server. In legacy RHEL versions NFSv2, v3, and v4 were started by default, As of RHEL 7, only v3 and v4 are started as v2 is no longer supported. It is possible that other supported platforms have various other version restrictions.

[1]: https://github.com/hyperic/sigar/blob/sigar-1.6/bindings/java/src/org/hyperic/sigar/NfsFileSystem.java#L50