Bug 2064342

+++ This bug was initially created as a clone of Bug #1654862 +++

Description of problem:
Forward port final fixes to 8.7.

A customer recently encountered one of the limitations of the resource-agent IPsrcaddr and the resource agent failed to start. It was not clear why it would not start.

After recreating and debugging the issue it was found that the resource agent failed to start because they encountered one of the limitations of IPsrcaddr.
- https://github.com/ClusterLabs/resource-agents/blob/master/heartbeat/IPsrcaddr#L38


The limitation they encountered was that the IP for ipaddress of the IPsrcaddr resource agent was for an IP that was NOT in the same network as the default gateway. 

If we are going to continue with limitations then we should have some debugging that that will log a reason for the resource agent not starting if they are encountering one of the limitations. 

This example would be pretty simple to do:
  srca_start() {
     [.....]
     else:
        if [ -z $Network }; then
           errorexit "You encountered a limitation of this agent. The $ipaddress is not in same network as the default route."
       fi  
       ip route replace  .....
       [.....]
      fi
      return $rc
     }

We should provide more informative error message for any time we hit a limitation that we can determine why it failed to start. 

Version-Release number of selected component (if applicable):
resource-agents-4.1.1-12.el7_6.4.x86_64

How reproducible:
Every time

Steps to Reproduce:
Create a VIP where the ip is on different network than that of the default route. 
# pcs resource create ClusterIP ocf:heartbeat:IPaddr2 ip=192.168.122.190 cidr_netmask=24 op monitor interval=30s --group RG_VIP

# ip -o address
1: lo    inet 127.0.0.1/8 scope host lo\       valid_lft forever preferred_lft forever
2: ens3    inet 192.168.100.172/24 brd 192.168.100.255 scope global noprefixroute dynamic ens3\       valid_lft 86100sec preferred_lft 86100sec
3: ens8    inet 192.168.122.172/24 brd 192.168.122.255 scope global noprefixroute dynamic ens8\       valid_lft 3300sec preferred_lft 3300sec
3: ens8    inet 192.168.122.190/24 brd 192.168.122.255 scope global secondary ens8\       valid_lft forever preferred_lft forever

# ip ro
  default via 192.168.100.1 dev ens3 proto dhcp metric 100
  192.168.100.0/24 dev ens3 proto kernel scope link src 192.168.100.172 metric 100
  192.168.122.0/24 dev ens8 proto kernel scope link src 192.168.122.172 metric 101

Then the IPsrcaddr is added.
# pcs resource create ip_src ocf:heartbeat:IPsrcaddr ipaddress=192.168.122.190 cidr_netmask=24 trace_ra=1 --group RG_VIP

Actual results:
The resource fails to start, but the reason for the resource not starting is not provided. The end user would be left wondering why the resource did not start.

Expected results:
If we are going to continue to have limitations on how IPsrcaddr is used then we should provide bettter error messages when the resource fails to start because of one of those limitations.

Additional info: