Hello team,

We’re currently running version 1.7.0 of your SDK without any issues. We tried to upgrade to the newest version this week, however after deploying the version we could see immediately a high increase on opened threads & increased memory usage for each running pod. Here you can see the reported avg thread count for the deployed period (14:55 we started rolling out the version 1.9.0, at around 15:12 we started rolling out the previous version).

At the same time memory consumption of each pod went approx. 50% up (I mean, needs to handle all the active threads etc), which leads me to an assumption that we might be facing some sort of resource leak in the library. Or is that something expected from your end? It’s hard to correlate any changelog items to this.

Also worth mentioning: We also tested other versions and it seems to have the same issues, so the issue was probably introduced in the version 1.8.0.