Hi! I’ve got this kind of table:
CREATE TABLE raw_m1 (
datetime timestamp NOT NULL,
symbol varchar NOT NULL,
open double precision NOT NULL,
high double precision NOT NULL,
low double precision NOT NULL,
close double precision NOT NULL,
volume double precision NOT NULL
) WITH (
tsdb.hypertable,
tsdb.partition_column='datetime',
tsdb.segmentby='symbol',
tsdb.orderby='datetime DESC'
);
And when I’m creating a view on it like this:
CREATE MATERIALIZED VIEW m15(datetime, symbol, cs)
WITH (tsdb.continuous,
tsdb.materialized_only = FALSE) AS
SELECT time_bucket('15 min', datetime) AS bucket,
symbol,
ROLLUP(candlestick((datetime)::timestamp without time zone, open, high, low, close, volume))
FROM raw_m1
GROUP BY bucket, symbol
ORDER BY bucket DESC;
I’ve got warning
using non-immutable functions in continuous aggregate view may lead to inconsistent results on rematerialization
I’m aware of the fact that this warning is due to timestamp type of datetime field instead of timestamptz, but I don’t really understand the reason for this. The underlying types in both timestamps are the same and all calculations with them should be equivalent.
I’d like to know, does this warning have serious meaning or may I use timestamp without caring for this?