The first is the MapDB database. It only saves your most recent states of your Items so:
you have a lot of Items
you have a number of Items with really big states (e.g. Images)
something has gone wrong.
It this were running on an SD card I’d suspect the card is wearing out and failing to clear old data when new data is written.
If the problem is 1 or 2, create a custom mapdb config to only save those Items you need it to instead of all Items. If the problem is 3 then there isn’t enough information to guess at the cause or a solution.
The second one is InfluxDB. It;'s going to suffer the same problems if 1 or 2 are the root cause only more so since InfluxDB saves a history of Item states, not just the latest state.
I have deleted influxdb and just swapped too mapdb and rrd4j
Mapdb is still behaving strangely it seems too use all the available space available with a mapdb.t file over 28gb
The cpu then starts spiking too max and the system becomes pretty unstable restarting the system usually deletes the file frees the space and then drops cpu back too normal
This is happening over the space of around 2 days
1056 items
Maby a few items containing pictures nothing too special
There is something else going on then. You don’t have enough Items nor enough large Items to account for 28 GB of storage in MapDB which only saves the most recent state of the Item.
I can’t imagine what could be the problem but it could be anything from a corrupt file system to malware, to some latent bug in MapDB that you are the first to uncover.
It would be really odd for you to be the only person experiencing this problem with MapDB. This is probably the second most use add-on of all.
Try querying for all the records in MapDB through the API explorer. Maybe that will point to a specific set of Items that are growing or something like that. Unfortunately there doesn’t seem to be a DB viewer app to look at the file directly.
Obviously don’t wait for the file to be many gig before running the query or else it will probably time out. But if it gets to 30 GB in a couple of days, you should be able to see the number of records or the size of certain records growing by comparing queries from just a couple of minutes apart.
Deleted mapdb and all mapdb files and started using default rrd4j storage been fine since.
Reinstalled mapdb and allowed it too recreate the db files
Seems too be working fine right now only just done it
Mapdb.t file is transaction log i think something was wrong with the actual storing of the data into the actual db thus the transaction log filled up waiting too write too the db
I think there was corruption or user permission errors with the old db that got pulled over with the openhab-cli restore