I observe long running nodes have high non heap memory footprint then a recently started node. That is the reason I am interested in find non heap memory usage by Cassandra node. What could be the reason of high non heap memory footprint in long running cluster?
Regards Himanshu On Sat, Apr 11, 2020 at 11:56 PM Pekka Enberg <penb...@scylladb.com> wrote: > Hi Himanshu, > > On Sat, Apr 11, 2020 at 8:47 PM HImanshu Sharma > <himanshusharma0...@gmail.com> wrote: > > I am not worried about page cache. I want to monitor memory pressure, I > want to check that if heap+non heap usage goes above certain level then I > can take certain action. But due to this page cache thing, I am not sure > how to find actual memory usage( heap and off heap). Heap will not be more > than 8G. Off heap increases with data on disk due to off heap storage of > bloom filters and index files. I want to know the exact amount of memory > occupied by off heap components. Nodetool info output does not look > appropriate to me or I am missing something. > > There's "nodetool tablestats", which shows off-heap size for tables, > but I don't know if that's sufficient for your purposes. > > Out of curiosity, why are you not interested in how much memory your > process allocated in the page cache? As you said earlier, the resident > set size (RSS) in Linux indeed includes memory allocated in the page > cache: > > https://stackoverflow.com/a/21049737/153224 > > But it is as much part of the "memory pressure" as anything else. > > - Pekka >
