Qlik Sense High Memory usage on Failover Node (Related with Dispatcher service)
Hello guys, im having this trouble like more than 1years.
In our environment, we have a failover node which does not have any purpose other than being a failover. So there are no task/manuel load on this server.
Somehow server is going to use all of the memory when it reached +%80 server becomes unresponsive. It happens like every month, still i dont know it happens. When the problem occurs, i login to server, i can see there are +1million handle counts on the CPU and +1million log entries on the EventViewer which are related to QlikSenseServiceDispatcher. Also with rammap i checked memory consumption, there are +1million process belong to NLAppsearch.exe (even though i disabled this process from config files).
Are there anyone who experienced something like this? I am using QS FEB 2023 release.
thanks for your reply but i dont have any problem with disk space, its about Memory usage somehow goes %100 and server becomes unresponsive and only thing we can do is reboot the server.
We also have a open ticket on Microsoft and their support engineer told us to disable QlikSenseService Dispatcher due to it has so many handle counts on CPU (+1m).
I had a ticket on Qlik too and its closed with response increase maximum RAM of server, but its already have 64gb of ram and its enough for a failover server which waits without any load(untill promoto as a central). Neither SupportQlik nor we couldnt find root cause of this memory leak :(. Only thing we can do is reboot server everytime it happens.
Hello there Alan, we already have 3TB of memory for engine operations on other nodes. We have this problem only on a failovernode and it does NOT have any Engine load. How come it needs 128GB for just operating dispatcher&Repository service?
