Understanding Clustering Architecture

[#1]

Hey,
So my current environment has two server which is persumably cluster together. The reason I believe that the servers are cluster is because the cluster created PGO's. So when a user go on to either server via the access point, they have access to all the documents. Now the issue and this might be because I don't have a full understanding of the clustering architecture of Qlikview. This morning, QVS service from Server A failed and only the QVS service from Server B was running. When a user went in via the Access Portal from server A, the user recieved an error back saying something alone the line: 'Cannot contact the service'.

I was under the impression that if the QVS from server A goes down and a user access the Access Point from server A, then the Access Point will first try and connect to Server A's QVS and if it can't be contacted then it will go to Server B QVS? If this is not the case, from anyone's experience how do you deal with having high availability in your environment?

Also, is there a way in the QlikView management console to send out alerts to a user when one of Qlikview services goes down?

Many Thanks in advance for everyone's help,
Samuel Pena

[rharmsny]

this should work as you describe it. When the user gets the fail message, is the document loaded in the active server?

[lhr]

if the qvses are clustered then it should look like one qvs in the accesspoint and it should not send users to a node that is offline.

the qvses are clustered if they have the same license (that supports clustering), are the same version and hae the same document folder set.