Sven Klemm

I think it would be useful to track hypertable vs chunk runtime exclusion separately (in EXPLAIN output).

So this bug did not happen in 2.5.0 but started with 2.5.2? Did you try 2.5.1 too. Can you show the configuration of your distributed hypertable(s). What triggers this behaviour?...

So versions before 2.5.0 did not show this behaviour? Which version did you use before 2.5.0? There is indeed a better way to delete data in hypertables with drop_chunks but...

Duplicate of #3868

Did you have a look at GapFill you can do queries to get results like you are suggesting. https://docs.timescale.com/api/latest/hyperfunctions/gapfilling-interpolation/time_bucket_gapfill/#time-bucket-gapfill

No, gapfill can look beyond if you tell it to. But it is not an exact match what you are asking for but the functionality is similar. One limitation of...

You can reference channelname in subquery like so: ``` SELECT time_bucket_gapfill('1h',sampletime), channelname, interpolate( avg(value), (SELECT (sampletime,value) FROM t t_sub WHERE t_sub.channelname = t_outer.channelname AND sampletime = '2022-01-01 11:30' ORDER BY...

I think you want lower inclusive and upper exclusive. Upper inclusive and lower exclusive would work too but might be a little weird since time_bucket returns lower. Currently it does...

Hmm installing timescaledb should have pulled in postgresql as dependency was that not the case?

Do you have postgres 14 installed too, it consults pg_config for the version to look for?