Description of problem: [RHEL 8.8] Kdump fails over NFS when using FQDN name if the system has two interfaces(NIC1 used to reach the NFS server and NIC2 used to reach DNS server).

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

kexec-tools-2.0.25-5.el8_8.1.x86_64

How reproducible: Always in customer system. 


Steps to Reproduce:
1. On RHEL 8.8 system which is having 2 NIC cards, and both should be in two different network range. 
2. For example, NIC1 IP should be within the range of DNS server's IP, NIC2 IP should be within the range of NFS server's IP. 
3. install kexec-tools-2.0.25-5.el8_8.1.x86_64
4. Mount the NFS share with FQDN name, also add the entry in /etc/fstab file with FQDN name.
5. Add the nfs entry in /etc/kdump.conf file with FQDN name. 

~~~
default reboot
nfs <FQDN-name>:/nfsdump  
path /nfsdump
core_collector makedumpfile -c --message-level 1 -d 1
~~~


Actual results:

[   70.737299] kdump[548]: dump path "/kdumproot/nfsdump" is not mounted, trying to mount...
[   70.765322] kdump.sh[550]: mount.nfs: Failed to resolve server FQDN-name-of-nfs-server: Name or service not known    <<<-------
[   70.810298] kdump[552]: failed to dump to "/kdumproot/nfsdump", it's not a mount point!
[   70.852691] kdump[554]: saving vmcore failed


Expected results:

Kdump should work as expected upon using the FQDN name. 


Additional info: 

- After downgrading the kexec-tools package to kexec-tools-2.0.24-6.el8.x86_64 the issue is not reproducible. 
- Upon looking further, I have noticed that the issue introduced in the following kexec-tools-2.0.25-3.el8.x86_64.rpm  package. 
  The BZ is https://bugzilla.redhat.com/show_bug.cgi?id=1950282