We have 2 Network Adapters (each vm) assigned with the same subnet IP addresses on Azure SLES 12.x on two VMs, as this is requriement by client application team. However, we were facing some network communications issues (unable to reach) between two servers. Communication is fine if we disabled one of the network adpater on both servers.

So, we have kept some routing entries in /etc/init.d/after.local file to resolve this.



ip route add default via 170.162.92.1 dev eth0 table 101
ip route add default via 170.162.92.1 dev eth1 table 102
ip rule add from 170.162.92.71 lookup 101 pref 10100
ip rule add from 170.162.92.107 lookup 102 pref 10101



But after this when we see the routing table, we are seeing eth1 is the first entry in the list. Just I would like know community members opinion on this if we face any issues in future because of this as we are going deploy pacemaker cluster on this for SAP HANA.




#ip route list
default via 170.162.92.1 dev eth0 proto dhcp
168.63.xxx.xxx via 170.162.92.1 dev eth0 proto dhcp
169.254.xxx.xxx via 170.162.92.1 dev eth0 proto dhcp
170.162.92.0/24 dev eth1 proto kernel scope link src 170.162.92.107
170.162.92.0/24 dev eth0 proto kernel scope link src 170.162.92.71




#route -n

Kernel IP routing table
Destination Gateway Genmask Flags Metric Ref Use Iface
0.0.0.0 170.162.92.1 0.0.0.0 UG 0 0 0 eth0
168.63.xxx.xxx 170.162.92.1 255.255.255.255 UGH 0 0 0 eth0
169.254.xxx.xxx 170.162.92.1 255.255.255.255 UGH 0 0 0 eth0
170.162.92.0 0.0.0.0 255.255.255.0 U 0 0 0 eth1
170.162.92.0 0.0.0.0 255.255.255.0 U 0 0 0 eth0

I can't remember exactly where I found the information but you'll likely want to set your NICs up so that the server replies on the same NIC that the request came in.

1 members found this post helpful.

I am not really sure whether you are looking for sub-interface configuration.

Check out this link: http://cgi.linuxfocus.org/~guido/scr...fig-howto.html

From site above: