vip-manager: cluster ip confusion (AWS)

[prensgold]

Hi Vitaliy ,

When I ran your scripts on the servers in my local, I could only see the cluster ip on one server.

In the screenshot below, the cluster ip output from the servers I set up in my own locale.

When I want to install and test your current script on amazon, I can see the cluster ip on all 3 servers.
and when I tried to connect to the database with port 5000 from each server, I was able to connect.

In the screenshot below, the cluster ip output from the servers I set up on amazon.

Cluster ip status confused me in two installations.
What is the correct output here? The operating systems are in the same version.
How can we interpret this situation?

[vitabaks]

Hi @prensgold

Thanks!

I'm not sure, but maybe it's a bug in the vip-manager version 2.1.0

Have you deployed a new cluster on AWS? using etcd version 3.5.7 or do you have a different version?

Can you attach logs from all servers?

sudo journalctl -u vip-manager

[prensgold]

yes , i installed version 3.5.7

[vitabaks]

an issue was created on the vip-manager project repository

cybertec-postgresql/vip-manager#147

[vitabaks]

Is there any news on this issue?

[vitabaks]

AWS doesn't support sending of gratuitous ARP messages (none of the public hosters do, as you could try to impersonate another server easily), so vip-manager would not work with AWS.

cybertec-postgresql/vip-manager#147 (comment)

[vitabaks]

Recommendations for cloud environments:

• For the "Type A" scheme: Use DNS records (round-robin) listing all HAProxy load balancing servers instead of relying on cluster_vip.
• For the "Type B" scheme: Use libpq target_session_attrs, ensuring read/write connections go to the primary database, as an alternative to cluster_vip.