Dear community
I have an interesting case where I would need some help.
I have two systems OH3 running, both with Debian Buster, Java 11, openhabian and so on.
System 1
AMD system which comes to its hardware limit with OH3
This is the main system with persistence database MariaDB, deconz, KNX, several things, items and rules configured with VSCode
It was running on OH2.5 with relative good performance, but I did a mistake and updated to OH3. (It is running all the stuff, but, as I mentioned somehow slow)
System 2
Intel system, was installed with fresh installation as described in manual. No persistence until now, but I want to run rr4j
On this system my son integrated already many rules, things, equipment’s, models so that I do not want him to start from the scratch again.
My questions
How do I get from System 1 all the stuff transferred to System 2 without destroying already existing data
What can you recommend to me as best? Just as a hint, I have an older desktop PC which could be used temporarily, then as System 3 (also Debian) if it would help to merge.
MariaDB does get the persistence data defined by the Groups, can I define it in the same way with rr4j?
Unless your AMD system has less than 1GB of RAM and/or was built in the late 2000s it should be more than powerful enough to run OH 3. Most people report that OH 3 runs better in more respects than OH 2.5 so there is most likely something else going wrong. Remember, OH can run quite nicely on an RPi 2. Your AMD system is almost certainly more powerful than an RPi 2.
In terms of your question, you’ve not declared how you’ve defined all your stuff. Config files or through the UI?
If through config files, you should be able to just copy those files over to the new OH and they will be loaded along side what ever the System 2 already has. I’d recommend installing the bindings and rediscovering the Things if those are not defined through config files.
If you’ve a lot of stuff defined through the UI, you can query for and save the JSON for those through the REST API Docs. Copy that JSON over to the REST API docs on System 2 and essentially recreate them on System 2.
Yes, all the persistence stuff works the same at that level. Though in OH 3 keep in mind that rrd4j comes with a default strategy of saving all supported Items every minute and on every change.
What’s nice about that is no matter where you are in MainUI, you can click the “Analyze” button and you will get a chart even if you didn’t think you would want a chart of that Item when you first set up that Item. It’s rrd4j so the size of the database remains fixed no matter how long you keep adding data to it so you won’t run out of disk space.
So you may not want to limit rrd4j with Group based Persistence. I let it save everything.
