-
Notifications
You must be signed in to change notification settings - Fork 101
I have two Kubernetes clusters. I have created hazelcast-cluster with members A,B,C in East AKS cluster and hazelcast-cluster with members D,E,F in West AKS cluster. Now I want A,B,C to discover D,E,F. How can I do that? #285
Comments
So you'd like to have one Hazelcast cluster stretched across multiple Kubernetes clusters, right? Technically, you could do it by using host network or host port (but it's not a recommended approach). Also, if you have one Hazelcast cluster across multiple regions, the performance will be terrible, so you shouldn't do it at all. The proper way to do it is to have 2 separate Hazelcast cluster with the WAN replication in between. WAN, however is an enterprise feature. If you're interested more in WAN, please check this guide: https://guides.hazelcast.org/kubernetes-wan/. |
@leszko I am pretty much naive with Hazelcast. Going for enterprise feature is not an option for me. |
I also have the same problem please let us know the solution if its possible |
@AayushiKhandelwal123 and @leszko , please let me know if you have any solution . |
Hi,
No didn't get the solution in free hazelcast version. Even hazelcast team
told me to go for enterprise hazelcast WAN feature.
…On Mon, Feb 8, 2021, 11:05 PM paritalagt ***@***.***> wrote:
@AayushiKhandelwal123 <https://github.com/AayushiKhandelwal123> , please
let me know if you have any solution .
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#285 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AIJSF5RVJAVJM3ZLC6WCF5LS6AOHVANCNFSM4U7VPGSA>
.
|
Updating old issue. as we have similar situation using enterprise HZ. Wondering whether we can use stretched cluster using TCP proxy thru HTTPROXY kubernetes definition. Basically exposing public addressable ingress endpoint for respective cluster and route the traffic to respective pod by using custom POD selector in the respective K8s cluster. |
Both Kubernetes clusters are in different network. I saw this #141 but it has been closed. So I was looking for a solution other than using hazelcast enterprise feature or having one AKS cluster in two regions.
The text was updated successfully, but these errors were encountered: