'Cluster WSO2 AM All-in-One Active-Active Deployment

I have deployed API manager 4.0.0 All-in-one on 2 VMs. I am using Mysql as DB which is on separate VM and I am sharing the databases as mentioned in the document. Now I am trying to cluster these 2 nodes as mentioned in this document. There are few things which are not clear to me from this document.

  1. Which node is manager and which is worker or they are both manager or worker? What is basic difference between manager and worker?

  2. If I use nfs to share the resources between all the nodes, which node do we setup nfs? (I setup nfs in a different vm , and both nodes are mounted to nfs server, is that right?)

  3. What happens under the hood when you publish an API in version 4.0.0. I understand that when an API is published, it gets deployed on the API Gateway, and the API lifecycle state will be changed to PUBLISHED. What artifacts are persisted in the DB (and where) and what artifacts are persisted to the filesystem (my understanding is that they are located at <APIM_HOME>/repository/deployment/server/synapse-configs/default directory as XMLs, but i don't think i notice something changed in the directory, where are they?).

  4. What does step1 and step9 mean? why we need this?

wso2-am


Solution 1:[1]

  1. We had this manager-worker concept in the older versions. But in APIM v4, we don't have that concept. Both nodes are accepting requests.
  2. In APIM v4 we have inbuilt artifact synchroniser by default and you don't need a NFS for API artifacts and rate limiting policies. But if you are using tenants and userstores, then you need to have the NFS. In that case, you can mount both the nodes to the NFS server.
  3. Before APIM v4, we had this file system based artifact approach. But in the latest it is loaded to the memory. When you create an API, it publishes an event to itself and to the other node. Then both the nodes load the API from the database to the memory.
  4. Step 1: You can't use the default shipped keystores in production. You have to change those. Step 9: This is for distributed caching.

Sources

This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.

Source: Stack Overflow

Solution Source
Solution 1 Pubci

Related Questions