Skip to content
Merged
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
38 changes: 34 additions & 4 deletions diagnostic.sql
Original file line number Diff line number Diff line change
Expand Up @@ -42,10 +42,10 @@ $$ SET search_path = pg_catalog, pg_temp;
CREATE OR REPLACE FUNCTION pg_temp.check_cagg_large_materialization_range() RETURNS void LANGUAGE plpgsql AS
$$
DECLARE
cagg regclass;
range text;
v_cagg regclass;
v_range text;
BEGIN
FOR cagg, range IN
FOR v_cagg, v_range IN
SELECT
format('%I.%I', c.user_view_schema, c.user_view_name)::regclass AS continuous_aggregate,
CASE
Expand All @@ -62,7 +62,36 @@ BEGIN
WHEN d.column_type =ANY('{timestamp,timestamptz}'::regtype[]) THEN _timescaledb_functions.to_interval(r.greatest_modified_value - r.lowest_modified_value) > 250 * f.bucket_width::interval
END
LOOP
RAISE WARNING 'Continuous aggregate % has large materialization range ''%''.', cagg, range;
RAISE WARNING 'Continuous aggregate % has large materialization range ''%''.', v_cagg, v_range;
END LOOP;
END
$$ SET search_path = pg_catalog, pg_temp;

CREATE OR REPLACE FUNCTION pg_temp.check_cagg_chunk_interval() RETURNS void LANGUAGE plpgsql AS
$$
DECLARE
v_cagg regclass;
v_cagg_width text;
v_chunk_width text;
BEGIN
FOR v_cagg, v_chunk_width, v_cagg_width IN
SELECT
format('%I.%I', c.user_view_schema, c.user_view_name)::regclass AS continuous_aggregate,
CASE
WHEN d.column_type =ANY('{int4,int8}'::regtype[]) THEN d.interval_length::text
WHEN d.column_type =ANY('{timestamp,timestamptz}'::regtype[]) THEN _timescaledb_functions.to_interval(d.interval_length)::text
END AS chunk_width,
f.bucket_width cagg_width
FROM _timescaledb_catalog.continuous_agg c
JOIN _timescaledb_catalog.continuous_aggs_bucket_function f ON f.mat_hypertable_id=c.mat_hypertable_id
JOIN _timescaledb_catalog.dimension d ON d.hypertable_id=c.mat_hypertable_id
WHERE
CASE
WHEN d.column_type =ANY('{int4,int8}'::regtype[]) THEN d.interval_length <= f.bucket_width::int
WHEN d.column_type =ANY('{timestamp,timestamptz}'::regtype[]) THEN _timescaledb_functions.to_interval(d.interval_length) <= f.bucket_width::interval
END
LOOP
RAISE WARNING 'Continuous aggregate % has chunk width smaller than bucket width % <= %', v_cagg, v_chunk_width, v_cagg_width;
END LOOP;
END
$$ SET search_path = pg_catalog, pg_temp;
Expand Down Expand Up @@ -169,6 +198,7 @@ BEGIN
PERFORM pg_temp.check_job_failures();
PERFORM pg_temp.check_compressed_chunk_batch_sizes();
PERFORM pg_temp.check_cagg_large_materialization_range();
PERFORM pg_temp.check_cagg_chunk_interval();
END
$$ SET search_path = pg_catalog, pg_temp;

Expand Down