Lukas Crepaz

Same here, panning with mouse is almost unusable, although mine is stuck mostly and just moves very little I'm using latest Manjaro Linux Same problem was described here (in german):...

@davidkohn88 Thanks for the workaround, it's working good. We also experimented with using the simple interval/average functions like this ``` SELECT time_bucket('1 month', time) as dt, toolkit_experimental.integral(time_weight('LOCF', time, value), 'hours')...

@Starnick4444 sadly this issue still persists. tested with latest version `1.5.4` and `SURREAL_ROCKSDB_WRITE_BUFFER_SIZE=67108864` and `SURREAL_ROCKSDB_MAX_WRITE_BUFFER_NUMBER=24` as well as with `SURREAL_ROCKSDB_MAX_WRITE_BUFFER_NUMBER=4`

no we do not have any indexes defined

@sgirones @Starnick4444 maybe some additional info: this is a typical debug message of our ingestion from vector: ``` 2024-07-25T12:11:44.519528Z DEBUG request:import:execute:process:executor: surrealdb_core::dbs::executor: Executing: CREATE metrics_buffer_send_duration_seconds:['2024-07-25T12:11:43.483241190Z'] CONTENT { host: '426680319cde', kind:...

we tried different setups on different clusters for the last week and what we've noticed is, that with a low `SURREAL_ROCKSDB_MAX_WRITE_BUFFER_NUMBER`, e.g. 4 or 2, the memory consumption increases at...

Are there news about this feature? We're currently exploring river and might have a use case, where we need to create a batch of multiple jobs for a specific customer...

@bgentry thx for your quick response! regarding your options: i think both of them would work for us: for the first one we could create a timer per customer where...