Unlock a world of possibilities! Login now and discover the exclusive benefits awaiting you.
Dear Experts,
Recently we have added an additional Publisher node on different machine & have clustered it. Problem we are facing is as below:
1. As per observations, the tasks are routed to first QDS as preference. Suppose, I am running 15 tasks at a time, the tasks are randomly routed on the clustered QDS (QDS1 & QDS2). When all engines on QDS1 are occupied & engines on QDS2 are still available for execution, even then the tasks are queued up on QDS1 instead of executing on QDS2.
2. When earlier clustering was performed, the LoadBalancer.xml file in C:\ProgramData\QlikTech\DistributionService showed 2 rows for 2 clustered nodes as ClusterID=1 & ClusterID=2, but since we have done reconfiguration again, only 1 row is seen in LoadBalancer.xml file after clustering.
QV version = 11.20 SR 12
Not able to figure out the root cause. Kindly assist on the same.
Regards,
Yuvraj Killedar
Take a look at the files and formula for loadbalancing mentioned here: https://help.qlik.com/en-US/qlikview/November2017/Subsystems/Server/Content/QlikView-Server/QVSRM_Cl...
Make sure that the QMC displays 2 engines in the Distribution Services section.
Hi Miguel,
Thank you for responding, but have already gone through the document. The calculation mentioned in the document is to override default settings. We want to host our environment with default settings.
Strange thing is that, it worked as expected during testing phase, but now since we are planning for final implementation, it is giving issues..!!
The QMC shows 2 servers as clustered for QDS.
Excuse me for asking the obvious but what then is the difference between testing phase and now? Specifically:
If all the above is properly set up, and there are no differences between the testing and production environment, I would recommend you to check with Qlik Support.