We have one CentOS7 or RedHat 7 "BIG" HP Proliant DL360 Gen 9 server with 4 NICs, two external switches Switch 1 and Switch 2 and many "small" servers.
Those small servers have two NICs: those NICs send exact the same traffic through each NIC to different switches. NIC1 is connected to Switch 1 and NIC2 to Switch 2. So we get redundancy with small servers: each server sends the same traffic through two switches towards one "BIG" server....
I tried to understand if I can by using teaming on BIG server create such configuration, for example:
NIC1 and NIC2 are team1, two NICs need for 2Gbps bandwidth.. NIC3 and NIC4 are team2 here is exactly the same mirrored 2 Gbps traffic...
Each team has the same traffic - and somehow (how?) we need to "know" or maybe define initial ACTIVE team. WE have to switch to another team if, for example Switch 1 dies, or even if on cable/NIC stops work, failed even if cable is taken out of connector..
Application on BIG server should not notice it because it wants to see just one static IP all the time, independently which team is ACTIVE and which is passive..
My question: is it possible to configure our BIG by using "teaming" in the way I described above or we need to buy some High Availability 3rd Party SW?
Thank you for your attention :-)
