LuciferInLove opened a new issue #2859: URL: https://github.com/apache/rocketmq/issues/2859
Hello. We had the rocketmq cluster installed with master-slave architecture, and then we moved it to the new dledger architecture. Brokers are running on small ec2 instances with such java options: ``` -Xms128m -Xmx128m -Xmn128m -XX:G1HeapRegionSize=1m -XX:MaxDirectMemorySize=256m ``` When we've used the master-slave architecture, everything worked correctly, but when we switched to the dledger architecture, we started getting Out of memory errors. We see many threads that allocate 4mb buffer. Is it the right behavior? The log file with errors is attached. [hs_err_pid_1.log](https://github.com/apache/rocketmq/files/6433664/hs_err_pid_1.log) -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected]
