I am looking at deploying the load balance in my NSX environment to load balance my SMTP senders. In reading the instructions it talks about the IP pools for the service engines. What I cannot tell for certain is do I need enough IP's to deploy the SE's on each host, or maybe just 2 or 3. There will not be a large load on the load balancer. Wondering if anyone has any experience in a deployment on this level.
Hello,
As far as I understood your requirement, you can have a static IP pool of 4-6 IP addresses for the Service Engine(SE) VMs, if the need is to have a max of 4-6 SE VMs in your setup. But if you need dedicated SE groups for different components then you may need to have a larger IP pool of 8-12 IPs in it again depending on your actual requirements. Hope this would help,
Regards
Sriram
Hello,
As far as I understood your requirement, you can have a static IP pool of 4-6 IP addresses for the Service Engine(SE) VMs, if the need is to have a max of 4-6 SE VMs in your setup. But if you need dedicated SE groups for different components then you may need to have a larger IP pool of 8-12 IPs in it again depending on your actual requirements. Hope this would help,
Regards
Sriram
I just wanted to make sure I was not required to have one on every host before I tried to deploy it. I am only looking at using it to load balance two servers we are using for SMTP relaying to M365.
Thank you
Yes that should be totally fine, since there is no such requirement. It should be fine for load balancing your two backend SMTP servers.
Thanks
Sriram
Hi
Each Service Engine will require an IP address for Management and an additional IP Address (Self-IP) that will be used for the actual LB Traffic that it needs to service.
You will thus need two IP Pools, one for Management and the other for the Data traffic.
Sulaiman