I’ve ran OH3 on an RPI for many years and recently migrated over to OH4 on a Synology NAS via a docker container.
OH works fine and is responsive with no immediate issues however the only thing that’s irritating me is heavy hard disk paging every minute. It’s very noticeable in a 4 drive NAS. If I stop the OH docker the paging stops.
I don’t have anything that polls every minute, nothing in the logs and with it operating normally I’m a bit stumped on where to look first. Is there anything that may be default to OH that could be polling or refreshing every minute?
I have a few bindings installed (SQL, MQTT etc) but removing these makes no change.
Any pointers in a direction would be much appreciated
OH inside Docker uses a helluva lot more RAM than if you run direct.
And your NAS probably does not much more (free!) RAM than your RasPi has.
So this is sort of the expectable effect of your decision to change hardware.
I’m not sure if lack of RAM is the issue. It’s not paging all of the time. The container is limited to 2GB and it’s currently sitting at 549mb used. OH is fast and responsive and works well it’s just that exactly every minute it pages for around 5-10seconds hard then stops. I think it’s possibly java. Not really a problem it’s just noisy and distracting more than anything and would like to understand what it’s actually doing.
