a wierd connecting Issue, root reachable but oracle not

A server's netbackup failed with below error:
RMAN-03009: failure of backup command on CH1 channel at 03/12/2012 23:16:59
ORA-19506: failed to create sequential file, name="VIGP_TAPECOPY_03122012_1_11668", parms=""
ORA-27028: skgfqcre: sbtbackup returned error
ORA-19511: Error received from media manager layer, error text:
VxBSAValidateFeatureId: Failed with error:
Server Status: cannot connect on socket
ORA-19600: input file is backup piece (/prod/backup/oracle/vigp/oracle/vigp1/arch/VIGP_ARCH_03022012_1_11668)

After checking the reason is netbackup server not reachable:
[oracle@node1 ~]$ ping netback_b
ping: unknown host netback_b

The settng for the /etc/hosts is correct:
[oracle@node1 ~]$ cat /etc/hosts|grep netback_b
2.1.0.10 netback_b netback_b.corporate.ge.com

And the wierd thing coming:
[oracle@node1 ~]$ su - root
[root@node1 ~]# ping netback_b
PING netback_b (2.1.0.10) 56(84) bytes of data.
64 bytes from netback_b (2.1.0.10): icmp_seq=1 ttl=252 time=1.11 ms
64 bytes from netback_b (2.1.0.10): icmp_seq=2 ttl=252 time=0.401 ms

Oracle it is not ping-able, but via root it is ping-able.
Why?

Let's trace the session:
[oracle@node1 ~]$strace -f -o /tmp/strace.log ping netback_b

Below is in the trace file:
..........
4347 open("/etc/nsswitch.conf", O_RDONLY) = -1 EACCES (Permission denied)
4347 open("/opt/CA/SharedComponents/lib/libnss_dns.so.2", O_RDONLY) = -1 ENOENT (No such file or directory)
4347 open("/opt/CA/CAlib/libnss_dns.so.2", O_RDONLY) = -1 ENOENT (No such file or directory)


From above red part we find out why.
[oracle@node1 ~]$ ls -ltr /etc/nsswitch.conf
-rw------- 1 root root 1730 Feb 23 09:07 /etc/nsswitch.conf

Oracle don't have access to above file.
After grant permission to oracle, the ping succeed for oracle, so is the netbackup.