We are having performance issues with 1.8 version.
Influxdb service is filling and consuming all the memory of the server . It was 64GB memory.
We increased the memory to 96GB. But 2 days later it filled that 96GB memory too.
So this effects every application on this server. Grafana and influxdb integration also.
Every 4-5 days i need to restart the influxdb service to relax the influxdb server and decrease memory usage. It decrease to 16-18GB influxdb service usage. But then it fulls and consumes all 96GB. How can i identify the problem?
Hello @cslm,
yes unfortunately this is a common issue with 1.x. It’s a large reason why the storage engine was rewritten for 3.x:
Unfortunately there aren’t a lot of ways to diagnose this problem. 1.x consumes all available memory and there isn’t operator control on limiting it. I wish I had a better answer for you.
Thanks Anaisdg.
Is restarting service every 12 hours automatically can be a workaround solution?
What can be a workaround for this? We cannot upgrade influxdb for some internal reasons.
@cslm did you find any work around for this?
Hello chakrag,
Only solution i found is restarting service before it fills all memory. I wrote a script thay is checking memory every 2 minutes. İf memory is filled more than %45 the script is restarting influxdb service.