Markos Fountoulakis

This is related to https://github.com/netdata/netdata/issues/7587 , and could be useful together with it.

@maciejasembler please configure `dbengine` as described in the [docs](https://learn.netdata.cloud/docs/agent/database/engine#configuration) since `history` does not apply for `dbengine`. After that, let's see if the memory leak persists.

Can you share a screenshot of the `NetData DB engine RAM usage (netdata.dbengine_ram)` chart, which is in the "Netdata Monitoring->dbengine" section of the dashboard on the right?

A lot of memory assigned to collectors seems to be growing. This could explain some of the problems. After looking at your screenshot with the database files I'm not sure...

> > > Today's stats: > > ``` > bash-5.0# du -sh > 348.0M . > ``` > > Still - the memleak is present despite setting those explicitly. >...

> > > Today's stats: > > ``` > bash-5.0# du -sh > 348.0M . > ``` > > Still - the memleak is present despite setting those explicitly. >...

@maciejasembler can you share your live netdata configuration with me? You can get it e.g. like this `wget 127.0.0.1:19999/netdata.conf` If it's too big or has sensitive information you can share...

Ah, yes. `default memory mode = dbengine` does not belong in `netdata.conf` but in `stream.conf`. Please double check the documentation for streaming.

> > > Ok, let me try that. > > Do you know if the same applies for dbengine multihost disk space and page cache size ? (like if this...

Hi @maciejasembler , FYI I am no longer affiliated with the netdata organization so I can only keep an eye on this on a best effort basis. We should probably...