High Availability for BO Enterprise XI 3.1 how does it work?

[Former Member]

Dear colleagues,

we are planning a HA infrastructure for BO with a cluster of 2 nodes running on Windows Server. From documentation I see that HA is implemented using several CMS servers. How does the client's access to the "correct" CMS server work? Is there any kind of relocatible name available? How is the client informed which instance is up? Many thanks in advance.

Best regards,

Detlef Kreiss (HP Outsourcing services)

[Former Member · Accepted solution]

Implementing High Availability only for BO Enterprise XI 3.1 can be achieved by two servers. One of your server acts as primary but it does not mean that all the requests from application server are routed through the primary server.

Application Server takes care of routing the request to a CMS.

If you are really looking at deploying High Availability in all the layers,

You may need a LOAD BALCNING ROUTER to provide one single URL to the end user

Application server to be deployed in more than one server and clustered

BOE services to be installed in more than one server and CMS across these servers to be clustered

File Repository location to be on clustered network file share or a SAN with raid -1

And your repository database has to be a highly available one too

Apart from all these, you need to do proper sizing and collect requirements to deploy only the necessary services either vertically or horizontally to increase the scalability

Mohanraj CP

Hewlett Packard

Bangalore, India

[former_member136842]

Hello,

f.e. if a user logs into InfoView the Tomcat AS decides which CMS the Session will be allocated. The first CMS who response in the Cluster gets the session.

Regards

-Seb.