How to Achiev High Availability with Alfresco Community Edition in an Enterprise Setting

[KarekMedAM]

I recently installed the latest version of Alfresco Content Services Community Edition, and it appears to be a strong contender for enterprise-level use. However, unlike the Enterprise Edition, the Community Edition does not natively support clustering, which limits options for achieving high availability directly within the platform.

Given this limitation, I’m interested in exploring the best methods for implementing high availability with Alfresco Community. Could you provide guidance on recommended architectures or approaches that would allow the Community Edition to perform reliably at an enterprise level? Specifically, any insights on third-party tools or configurations that can help manage redundancy, failover, and load balancing effectively would be highly appreciated.

Thnak you

[WilliamSilva-br]

We work with Alfresco Community on several successful projects in Brazil and these issues are not limiting.
 

  

[WilliamSilva-br]

A well - defined project within a good architectural foundation is enough.
 Tell us more about your project so I can help.

[KarekMedAM]

@WilliamSilva-br Thank you for sharing your experience.

In my case, I'm setting the Alfresco Content Community for an enterprise that has a lot of critical documents (around 80 TB) and they want the architecture to be highly available. I thought for clustering, so when one node get down  the system will continue working. (eliminate the single point of frailer) However clustering is only supported in the enterprise edition. Long story short, I want to ensure the enterprise that their documents are secure, redundant, and accessible 24/7.

What is the best architecture to go for in such a scenario? and is there any recommendations on setting ACS community for a large company.

Thank you,

Regards.

[WilliamSilva-br]

1- In my case, I'm setting the Alfresco Content Community for an enterprise that has a lot of critical documents (around 80 TB).:

R-
In this case, you need to know the size of the documents and what the folder structure will be like.
Once the folder structure is created and the documents are ready, the ideal is to have a front end (research portal), using CMIS for document findability.
Note: Tomcat + Solr uses memory in search and indexing.

 

[KarekMedAM]

Thank you,

I have no problem with searching, everything works fine. I just want to eliminate the points of failed or something like clustering in the enterprise edition .

[WilliamSilva-br]

The ideal is to contact Alfresco to evaluate the Enterprise version for 30 days. 
Give me your email and I'll send you my contact at Hyland.

[WilliamSilva-br]

1- However clustering is only supported in the enterprise edition. Long story short, I want to ensure the enterprise that their documents are secure, redundant, and accessible 24/7.

A- We only use the community version in our projects.
In our last migration there were + 40 TB (+ 9 million documents) in 04-Alfresco 7.3 - on premise.
When we talk about a good foundation we have:
1- tuning Lunix, tomcat, solr, etc.
1.1- Monitoring using ELK.
1.2 - Have PostgresQL and ELK in separate containers.
2- Don't forget that Alfresco -Docker is a distributed application and we can have k8s pods clustering together.
In reality, each project is a unique case and never an absolute truth. In my opinion, the ideal is to hire a consultancy that transfers knowledge, to have Alfresco Community customized, adjusted and very well parameterized.
Sds..

  

[KarekMedAM]

Thank you @WilliamSilva-br , In my case I installed Alfresco manually (zip file on tomcat) do you advise to go with docker in order to achieve high availability? if so, do you any recommendations for that!

In your expertise with alfresco did you considered applying high availability in postgres like using jpool or is not not necessary!?

[WilliamSilva-br]

Hello,

1-advise to go with docker in order to achieve high availability? if so, do you any recommendations for that!

R- In our projects we use Docker + Elk or Grafana and we also do monitoring integrated with Zabix.

2- In your expertise with alfresco did you considered applying high availability in postgres like using jpool or is not not necessary!?
R- The ideal is to understand the architecture of Alfresco and its structure. In our deployments, we always install PostgresSQL in a separate instance and yes, we also tune it.