I am trying to create a two instances Alfresco Community point that points to One database (Postgres) and one filesystem. if any one has this solution, please guide me with the instructions to do so.
The steps are similar to setting up a single instance as long as each instance is configured to access it's own dedicated resources, including DB schema user & root folder (alf_data).
Each instance should run on a dedicated machine (could be virtual or even a container). This will be easier to administer and will avoid port conflicts.
These machines can mount the same file system volume as long as they reference different root folders. The mounted filesystem will need to have two separate Alfresco "dir.root" folders, eg. alf_data1 & alf_data2.
The DB should run on a separate machine that is accessible on the same local network. You will create two separate DB schemas, eg. alfresco1 & alfresco2
Make sure you alfresco-global.properties is correctly configured for each instance (dir.root & db.* properties).
- each instance must point to it's own DB schema user (eg. alfresco1, alfresco2)
- each instance must point to it's own dir.root (alf_data) folder on the filesystem
- if you run the two instances on the same machine (not recommended) then you will also need to reconfigure all conflicting ports
I am confused here, if each instance has its one folder structure and database, the user see's two different sites depending on the load balance. My aim is to deploy alfresco into two different machines, depending on the load the load balance will rout to machine 1 or machine 2, and user should see the same data irrespective of the machine they are accessing.
Ah, ok. That is completely different :-)
Sorry, ACS Community does not support clustering - that is an Enterprise-specific feature.
If you would like to setup a single instance of Alfresco that clusters (load balances) across multiple repository nodes then you would need to consider ACS Enterprise, please refer to:
Setting up clustering | Alfresco Documentation
Some related links:
What's going on with Alfresco clustering? | ECM Architect
Thank you for your response. We have successfully created clustering on the Community version of the alfresco. This is the following I did for achieving this.
1, installed alfresco on two different servers.
2, pointed both alfresco's to a postgres database
3, copied one of the alf_data folder from one of the server to NFS location.
4, mounted the NFS location in both the servers.
5, change dir.root line in alfresco-global.properties to the mount location of NFS.
6, everything works like a charm.
I any one need more info, please ask me the question. I will see if I can respond to your question about clustering.
Retrieving data ...