Hi community,
Looking at the following post, it seems FGT-1 sends the SYN to FGT-2 through the LAN switch.
In the following situation, what happens if link 1 goes down? The SYN cannot be distributed from FGT-1 to FGT-2, then does the active-active configuration stop working? If not, what happens?
Regards,
Julián
Solved! Go to Solution.
Hi Julian
If link 1 goes down even syn from client will not reach the primary FW, the default gateway of the client will simply not be reachable anymore. In that case the active-active HA still works but is useless, since your client network is just isolated from the rest of the world.
I think the best to do is to set link 1 as monitored interface in your HA config, so the primary will fail-over when link 1 goes down.
Hi Julian
If link 1 goes down even syn from client will not reach the primary FW, the default gateway of the client will simply not be reachable anymore. In that case the active-active HA still works but is useless, since your client network is just isolated from the rest of the world.
I think the best to do is to set link 1 as monitored interface in your HA config, so the primary will fail-over when link 1 goes down.
Hi AEK,
If link 1 goes down even syn from client will not reach the primary FW, the default gateway of the client will simply not be reachable anymore. In that case the active-active HA still works but is useless, since your client network is just isolated from the rest of the world.
But if my client network is isolated from the rest of the world is like that active-active HA doesn't work.
I think the best to do is to set link 1 as monitored interface in your HA config, so the primary will fail-over when link 1 goes down.
Good point, if primary fails-over when link 1 goes down, now the default gateway will be reachable. In that case the clients will not be isolated, but there will not be load balance because the new primary will not be able to forward packets to the new secondary through the switch because link 1 is down. Is that right?
Regards,
Julián
FortiGate's active-active is not like Forcepoint's.
FGT's active-active is not a true active-active, here only one FGT (the primary) will receive the packets from client, while the secondary active will receive some offloading from primary only, but nothing from the client.
In case your link 1 is down, if the primary doesn't fail over then your network is isolated. Once it fails over then the clients will see their gateway again through link 2.
Hi AEK,
Yes, I understand. But in case we monitor link 1 and primary fails over, there will not be load balance because primary will not be able to forward packets to secondary through switch, is that right?
Regards,
Julián
Hi Julian
According to the tech tip you shared I fully agree with you.
You can also check this page to understand more how A-A HA works.
https://docs.fortinet.com/document/fortigate/7.6.3/administration-guide/966077/ha-and-load-balancing
On the other hand, personally in all my integrations I never used FGT in A-A HA mode since I'm still not convinced of its added value. I always set it up in A-P HA, and sometimes with virtual clustering when I have VDOMs in order to avoid having unemployed nodes, and because I think virtual clustering can be more efficient than A-A HA.
Hope it helps.
User | Count |
---|---|
2403 | |
1290 | |
778 | |
528 | |
454 |
The Fortinet Security Fabric brings together the concepts of convergence and consolidation to provide comprehensive cybersecurity protection for all users, devices, and applications and across all network edges.
Copyright 2025 Fortinet, Inc. All Rights Reserved.