Issues when Deliberately Shutting Down Nodes in Cluster
We are testing a cluster environment with three nodes. The purpose is to shut down nodes during low utilization times. The thinking behind this was that since one of the benefits of a clustered QVS is the failover aspect, so we can potentially shut down nodes at will and this should be no problem as long as the other nodes have resources. However we are seeing another set of quirks.
Among them are documents not showing up in accesspoint. Intermittently apps are showing and disappearing. Sometimes when opening an app we are getting a File not Found error message, when the qvw is clearly published in the root folder. And also QMC is stalling/freezing when trying to access list of published documents or seeing which documents are open on which node.
I am leaning towards these being caused by QMC having trouble syncing PGO files between each node with one node down. Anyone have any insight? Is there something we are doing wrong with the clustering? Any insight is appreciated.
Re: Issues when Deliberately Shutting Down Nodes in Cluster
Just a side note as well regarding QMC and pgo files, the QMC/QMS has no involvement in pgo files, the pgo files are strictly handled by the QlikView Server services only, just wanted to make this point for any future reviewers. The other poster is correct as well in their explanation of things, I do not recall when this got addressed, but I would expect 12.10 should be good, as I do not recall having run into this of late.
To help users find verified answers, please don't forget to use the "Accept as Solution" button on any posts that helped you resolve your problem or question.