Release 2.6.0

This release is medium priority for upgrade. We recommend that you upgrade at the next available opportunity. This release adds major new features since the 2.5.2 release, including: Compression in continuous aggregates Experimental support for timezones in continuous aggregates Experimental support for monthly buckets in continuous aggregates It also includes several bug fixes. Telemetry reports are switched to a new format, and now include more detailed statistics on compression, distributed hypertables and indexes. **Features** * #3768 Allow ALTER TABLE ADD COLUMN with DEFAULT on compressed hypertable * #3769 Allow ALTER TABLE DROP COLUMN on compressed hypertable * #3943 Optimize first/last * #3945 Add support for ALTER SCHEMA on multi-node * #3949 Add support for DROP SCHEMA on multi-node **Bugfixes** * #3808 Properly handle max_retries option * #3863 Fix remote transaction heal logic * #3869 Fix ALTER SET/DROP NULL contstraint on distributed hypertable * #3944 Fix segfault in add_compression_policy * #3961 Fix crash in EXPLAIN VERBOSE on distributed hypertable * #4015 Eliminate float rounding instabilities in interpolate * #4019 Update ts_extension_oid in transitioning state * #4073 Fix buffer overflow in partition scheme **Improvements** Query planning performance is improved for hypertables with a large number of chunks. **Thanks** * @fvannee for reporting a first/last memory leak * @mmouterde for reporting an issue with floats and interpolate
timescale · Feb 17, 2022 · 23962c8 · 23962c8
1 parent fc5341c
commit 23962c8
Show file tree

Hide file tree

Showing 8 changed files with 179 additions and 166 deletions.
diff --git a/.github/workflows/abi.yaml b/.github/workflows/abi.yaml
@@ -68,7 +68,7 @@ jobs:
           cp build_abi/install_lib/* `pg_config --pkglibdir`
           chown -R postgres /mnt
           set -o pipefail
-          sudo -u postgres make -C build_abi -k regresscheck regresscheck-t regresscheck-shared | tee installcheck.log
+          sudo -u postgres make -C build_abi -k regresscheck regresscheck-t regresscheck-shared IGNORES="memoize" | tee installcheck.log
         EOF
 
     - name: Show regression diffs

diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -4,7 +4,16 @@
 `psql` with the `-X` flag to prevent any `.psqlrc` commands from
 accidentally triggering the load of a previous DB version.**
 
-## Unreleased
+## 2.6.0 (2022-02-16)
+This release is medium priority for upgrade. We recommend that you upgrade at the next available opportunity.
+
+This release adds major new features since the 2.5.2 release, including:
+
+* Compression in continuous aggregates
+* Experimental support for timezones in continuous aggregates
+* Experimental support for monthly buckets in continuous aggregates
+
+The release also includes several bug fixes. Telemetry reports now include new and more detailed statistics on regular tables and views, compression, distributed hypertables, and continuous aggregates, which will help us improve TimescaleDB.  
 
 **Features**
 * #3768 Allow ALTER TABLE ADD COLUMN with DEFAULT on compressed hypertable
@@ -23,6 +32,9 @@ accidentally triggering the load of a previous DB version.**
 * #4019 Update ts_extension_oid in transitioning state
 * #4073 Fix buffer overflow in partition scheme
 
+**Improvements**
+* Query planning performance is improved for hypertables with a large number of chunks.
+
 **Thanks**
 * @fvannee for reporting a first/last memory leak
 * @mmouterde for reporting an issue with floats and interpolate

diff --git a/sql/CMakeLists.txt b/sql/CMakeLists.txt
@@ -32,11 +32,12 @@ set(MOD_FILES
     updates/2.4.1--2.4.2.sql
     updates/2.4.2--2.5.0.sql
     updates/2.5.0--2.5.1.sql
-    updates/2.5.1--2.5.2.sql)
+    updates/2.5.1--2.5.2.sql
+    updates/2.5.2--2.6.0.sql)
 
 # The downgrade file to generate a downgrade script for the current version, as
 # specified in version.config
-set(CURRENT_REV_FILE reverse-dev.sql)
+set(CURRENT_REV_FILE 2.6.0--2.5.2.sql)
 # Files for generating old downgrade scripts. This should only include files for
 # downgrade to from one version to its previous version since we do not support
 # skipping versions when downgrading.

diff --git a/sql/updates/2.5.2--2.6.0.sql b/sql/updates/2.5.2--2.6.0.sql
@@ -0,0 +1,42 @@
+DROP FUNCTION IF EXISTS @extschema@.recompress_chunk;
+DROP FUNCTION IF EXISTS @extschema@.delete_data_node;
+DROP FUNCTION IF EXISTS @extschema@.get_telemetry_report;
+
+-- Also see the comments for ContinuousAggsBucketFunction structure.
+CREATE TABLE _timescaledb_catalog.continuous_aggs_bucket_function(
+  mat_hypertable_id integer PRIMARY KEY REFERENCES _timescaledb_catalog.hypertable (id) ON DELETE CASCADE,
+  -- The schema of the function. Equals TRUE for "timescaledb_experimental", FALSE otherwise.
+  experimental bool NOT NULL,
+  -- Name of the bucketing function, e.g. "time_bucket" or "time_bucket_ng"
+  name text NOT NULL,
+  -- `bucket_width` argument of the function, e.g. "1 month"
+  bucket_width text NOT NULL,
+  -- `origin` argument of the function provided by the user
+  origin text NOT NULL,
+  -- `timezone` argument of the function provided by the user
+  timezone text NOT NULL
+);
+
+-- in tables.sql the same is done with GRANT SELECT ON ALL TABLES IN SCHEMA
+GRANT SELECT ON _timescaledb_catalog.continuous_aggs_bucket_function TO PUBLIC;
+
+SELECT pg_catalog.pg_extension_config_dump('_timescaledb_catalog.continuous_aggs_bucket_function', '');
+
+-- Adding overloaded versions of invalidation_process_hypertable_log() and invalidation_process_cagg_log()
+-- with bucket_functions argument is done in cagg_utils.sql. Note that this file is included when building
+-- the update scripts, so we don't have to do it here.
+
+DROP VIEW IF EXISTS timescaledb_information.continuous_aggregates;
+
+CREATE FUNCTION @extschema@.delete_data_node(
+    node_name              NAME,
+    if_exists              BOOLEAN = FALSE,
+    force                  BOOLEAN = FALSE,
+    repartition            BOOLEAN = TRUE,
+	drop_database          BOOLEAN = FALSE
+) RETURNS BOOLEAN AS '@MODULE_PATHNAME@', 'ts_data_node_delete' LANGUAGE C VOLATILE;
+
+CREATE FUNCTION @extschema@.get_telemetry_report() RETURNS jsonb
+    AS '@MODULE_PATHNAME@', 'ts_telemetry_get_report_jsonb'
+LANGUAGE C STABLE PARALLEL SAFE;
+
diff --git a/sql/updates/2.6.0--2.5.2.sql b/sql/updates/2.6.0--2.5.2.sql
@@ -0,0 +1,119 @@
+
+DROP PROCEDURE IF EXISTS @extschema@.recompress_chunk;
+DROP FUNCTION IF EXISTS _timescaledb_internal.chunk_status;
+DROP FUNCTION IF EXISTS @extschema@.delete_data_node;
+DROP FUNCTION IF EXISTS @extschema@.get_telemetry_report;
+
+CREATE FUNCTION @extschema@.delete_data_node(
+  node_name              NAME,
+  if_exists              BOOLEAN = FALSE,
+  force                  BOOLEAN = FALSE,
+  repartition            BOOLEAN = TRUE
+) RETURNS BOOLEAN AS '@MODULE_PATHNAME@', 'ts_data_node_delete' LANGUAGE C VOLATILE;
+
+CREATE FUNCTION @extschema@.get_telemetry_report(always_display_report boolean DEFAULT false) RETURNS TEXT
+  AS '@MODULE_PATHNAME@', 'ts_get_telemetry_report' LANGUAGE C STABLE PARALLEL SAFE;
+
+DO $$
+DECLARE
+    caggs text[];
+    caggs_nr int;
+BEGIN
+    SELECT array_agg(format('%I.%I', user_view_schema, user_view_name)) FROM _timescaledb_catalog.continuous_agg WHERE bucket_width < 0 INTO caggs;
+    SELECT array_length(caggs, 1) INTO caggs_nr;
+    IF caggs_nr > 0 THEN
+        RAISE EXCEPTION 'Downgrade is impossible since % continuous aggregates exist which use variable buckets: %', caggs_nr, caggs
+              USING HINT = 'Remove the corresponding continuous aggregates manually before downgrading';
+    END IF;
+END
+$$ LANGUAGE 'plpgsql';
+
+-- It's safe to drop the table.
+-- ALTER EXTENSION is required to revert the effect of pg_extension_config_dump()
+-- See "The list of tables configured to be dumped" test in test/sql/updates/post.catalog.sql
+ALTER EXTENSION timescaledb DROP TABLE _timescaledb_catalog.continuous_aggs_bucket_function;
+
+-- Actually drop the table.
+-- ALTER EXTENSION only removes the table from the extension but doesn't drop it.
+DROP TABLE IF EXISTS _timescaledb_catalog.continuous_aggs_bucket_function;
+
+-- Drop overloaded versions of invalidation_process_hypertable_log() and invalidation_process_cagg_log()
+-- with bucket_functions argument.
+
+ALTER EXTENSION timescaledb DROP FUNCTION _timescaledb_internal.invalidation_process_hypertable_log(
+    mat_hypertable_id INTEGER,
+    raw_hypertable_id INTEGER,
+    dimtype REGTYPE,
+    mat_hypertable_ids INTEGER[],
+    bucket_widths BIGINT[],
+    max_bucket_widths BIGINT[],
+    bucket_functions TEXT[]
+);
+
+DROP FUNCTION IF EXISTS _timescaledb_internal.invalidation_process_hypertable_log(
+    mat_hypertable_id INTEGER,
+    raw_hypertable_id INTEGER,
+    dimtype REGTYPE,
+    mat_hypertable_ids INTEGER[],
+    bucket_widths BIGINT[],
+    max_bucket_widths BIGINT[],
+    bucket_functions TEXT[]
+);
+
+ALTER EXTENSION timescaledb DROP FUNCTION _timescaledb_internal.invalidation_process_cagg_log(
+    mat_hypertable_id INTEGER,
+    raw_hypertable_id INTEGER,
+    dimtype REGTYPE,
+    window_start BIGINT,
+    window_end BIGINT,
+    mat_hypertable_ids INTEGER[],
+    bucket_widths BIGINT[],
+    max_bucket_widths BIGINT[],
+    bucket_functions TEXT[],
+    OUT ret_window_start BIGINT,
+    OUT ret_window_end BIGINT
+);
+
+DROP FUNCTION IF EXISTS _timescaledb_internal.invalidation_process_cagg_log(
+    mat_hypertable_id INTEGER,
+    raw_hypertable_id INTEGER,
+    dimtype REGTYPE,
+    window_start BIGINT,
+    window_end BIGINT,
+    mat_hypertable_ids INTEGER[],
+    bucket_widths BIGINT[],
+    max_bucket_widths BIGINT[],
+    bucket_functions TEXT[],
+    OUT ret_window_start BIGINT,
+    OUT ret_window_end BIGINT
+);
+
+--undo compression feature for caggs
+--check that all continuous aggregates have compression disabled
+--this check is sufficient as we cannot have compressed chunks, if
+-- compression is disabled
+DO $$
+DECLARE
+  cagg_name NAME;
+  cnt INTEGER := 0;
+BEGIN
+    FOR cagg_name IN
+        SELECT view_name,
+               materialization_hypertable_schema,
+               materialization_hypertable_name
+        FROM timescaledb_information.continuous_aggregates
+        WHERE compression_enabled is TRUE
+    LOOP
+        RAISE NOTICE 'compression is enabled for continuous aggregate: %', cagg_name;
+        cnt := cnt + 1;
+    END LOOP;
+    IF cnt > 0 THEN
+       RAISE EXCEPTION 'cannot downgrade as compression is enabled for continuous aggregates' 
+            USING DETAIL = 'Please disable compression on all continuous aggregates before downgrading.',
+            HINT = 'To disable compression, call decompress_chunk to decompress chunks, then drop any existing compression policy on the continuous aggregate, and finally run ALTER MATERIALIZED VIEW % SET timescaledb.compress = ''false''. ';
+    END IF;
+END $$;
+
+-- revert changes to continuous aggregates view definition
+DROP VIEW IF EXISTS timescaledb_information.continuous_aggregates;
+
diff --git a/sql/updates/latest-dev.sql b/sql/updates/latest-dev.sql
@@ -1,42 +0,0 @@
-DROP FUNCTION IF EXISTS @extschema@.recompress_chunk;
-DROP FUNCTION IF EXISTS @extschema@.delete_data_node;
-DROP FUNCTION IF EXISTS @extschema@.get_telemetry_report;
-
--- Also see the comments for ContinuousAggsBucketFunction structure.
-CREATE TABLE _timescaledb_catalog.continuous_aggs_bucket_function(
-  mat_hypertable_id integer PRIMARY KEY REFERENCES _timescaledb_catalog.hypertable (id) ON DELETE CASCADE,
-  -- The schema of the function. Equals TRUE for "timescaledb_experimental", FALSE otherwise.
-  experimental bool NOT NULL,
-  -- Name of the bucketing function, e.g. "time_bucket" or "time_bucket_ng"
-  name text NOT NULL,
-  -- `bucket_width` argument of the function, e.g. "1 month"
-  bucket_width text NOT NULL,
-  -- `origin` argument of the function provided by the user
-  origin text NOT NULL,
-  -- `timezone` argument of the function provided by the user
-  timezone text NOT NULL
-);
-
--- in tables.sql the same is done with GRANT SELECT ON ALL TABLES IN SCHEMA
-GRANT SELECT ON _timescaledb_catalog.continuous_aggs_bucket_function TO PUBLIC;
-
-SELECT pg_catalog.pg_extension_config_dump('_timescaledb_catalog.continuous_aggs_bucket_function', '');
-
--- Adding overloaded versions of invalidation_process_hypertable_log() and invalidation_process_cagg_log()
--- with bucket_functions argument is done in cagg_utils.sql. Note that this file is included when building
--- the update scripts, so we don't have to do it here.
-
-DROP VIEW IF EXISTS timescaledb_information.continuous_aggregates;
-
-CREATE FUNCTION @extschema@.delete_data_node(
-    node_name              NAME,
-    if_exists              BOOLEAN = FALSE,
-    force                  BOOLEAN = FALSE,
-    repartition            BOOLEAN = TRUE,
-	drop_database          BOOLEAN = FALSE
-) RETURNS BOOLEAN AS '@MODULE_PATHNAME@', 'ts_data_node_delete' LANGUAGE C VOLATILE;
-
-CREATE FUNCTION @extschema@.get_telemetry_report() RETURNS jsonb
-    AS '@MODULE_PATHNAME@', 'ts_telemetry_get_report_jsonb'
-LANGUAGE C STABLE PARALLEL SAFE;
-

diff --git a/sql/updates/reverse-dev.sql b/sql/updates/reverse-dev.sql
@@ -1,119 +0,0 @@
-
-DROP PROCEDURE IF EXISTS @extschema@.recompress_chunk;
-DROP FUNCTION IF EXISTS _timescaledb_internal.chunk_status;
-DROP FUNCTION IF EXISTS @extschema@.delete_data_node;
-DROP FUNCTION IF EXISTS @extschema@.get_telemetry_report;
-
-CREATE FUNCTION @extschema@.delete_data_node(
-  node_name              NAME,
-  if_exists              BOOLEAN = FALSE,
-  force                  BOOLEAN = FALSE,
-  repartition            BOOLEAN = TRUE
-) RETURNS BOOLEAN AS '@MODULE_PATHNAME@', 'ts_data_node_delete' LANGUAGE C VOLATILE;
-
-CREATE FUNCTION @extschema@.get_telemetry_report(always_display_report boolean DEFAULT false) RETURNS TEXT
-  AS '@MODULE_PATHNAME@', 'ts_get_telemetry_report' LANGUAGE C STABLE PARALLEL SAFE;
-
-DO $$
-DECLARE
-    caggs text[];
-    caggs_nr int;
-BEGIN
-    SELECT array_agg(format('%I.%I', user_view_schema, user_view_name)) FROM _timescaledb_catalog.continuous_agg WHERE bucket_width < 0 INTO caggs;
-    SELECT array_length(caggs, 1) INTO caggs_nr;
-    IF caggs_nr > 0 THEN
-        RAISE EXCEPTION 'Downgrade is impossible since % continuous aggregates exist which use variable buckets: %', caggs_nr, caggs
-              USING HINT = 'Remove the corresponding continuous aggregates manually before downgrading';
-    END IF;
-END
-$$ LANGUAGE 'plpgsql';
-
--- It's safe to drop the table.
--- ALTER EXTENSION is required to revert the effect of pg_extension_config_dump()
--- See "The list of tables configured to be dumped" test in test/sql/updates/post.catalog.sql
-ALTER EXTENSION timescaledb DROP TABLE _timescaledb_catalog.continuous_aggs_bucket_function;
-
--- Actually drop the table.
--- ALTER EXTENSION only removes the table from the extension but doesn't drop it.
-DROP TABLE IF EXISTS _timescaledb_catalog.continuous_aggs_bucket_function;
-
--- Drop overloaded versions of invalidation_process_hypertable_log() and invalidation_process_cagg_log()
--- with bucket_functions argument.
-
-ALTER EXTENSION timescaledb DROP FUNCTION _timescaledb_internal.invalidation_process_hypertable_log(
-    mat_hypertable_id INTEGER,
-    raw_hypertable_id INTEGER,
-    dimtype REGTYPE,
-    mat_hypertable_ids INTEGER[],
-    bucket_widths BIGINT[],
-    max_bucket_widths BIGINT[],
-    bucket_functions TEXT[]
-);
-
-DROP FUNCTION IF EXISTS _timescaledb_internal.invalidation_process_hypertable_log(
-    mat_hypertable_id INTEGER,
-    raw_hypertable_id INTEGER,
-    dimtype REGTYPE,
-    mat_hypertable_ids INTEGER[],
-    bucket_widths BIGINT[],
-    max_bucket_widths BIGINT[],
-    bucket_functions TEXT[]
-);
-
-ALTER EXTENSION timescaledb DROP FUNCTION _timescaledb_internal.invalidation_process_cagg_log(
-    mat_hypertable_id INTEGER,
-    raw_hypertable_id INTEGER,
-    dimtype REGTYPE,
-    window_start BIGINT,
-    window_end BIGINT,
-    mat_hypertable_ids INTEGER[],
-    bucket_widths BIGINT[],
-    max_bucket_widths BIGINT[],
-    bucket_functions TEXT[],
-    OUT ret_window_start BIGINT,
-    OUT ret_window_end BIGINT
-);
-
-DROP FUNCTION IF EXISTS _timescaledb_internal.invalidation_process_cagg_log(
-    mat_hypertable_id INTEGER,
-    raw_hypertable_id INTEGER,
-    dimtype REGTYPE,
-    window_start BIGINT,
-    window_end BIGINT,
-    mat_hypertable_ids INTEGER[],
-    bucket_widths BIGINT[],
-    max_bucket_widths BIGINT[],
-    bucket_functions TEXT[],
-    OUT ret_window_start BIGINT,
-    OUT ret_window_end BIGINT
-);
-
---undo compression feature for caggs
---check that all continuous aggregates have compression disabled
---this check is sufficient as we cannot have compressed chunks, if
--- compression is disabled
-DO $$
-DECLARE
-  cagg_name NAME;
-  cnt INTEGER := 0;
-BEGIN
-    FOR cagg_name IN
-        SELECT view_name,
-               materialization_hypertable_schema,
-               materialization_hypertable_name
-        FROM timescaledb_information.continuous_aggregates
-        WHERE compression_enabled is TRUE
-    LOOP
-        RAISE NOTICE 'Compression is enabled for continuous aggregate: %', cagg_name;
-        cnt := cnt + 1;
-    END LOOP;
-    IF cnt > 0 THEN
-       RAISE EXCEPTION 'Cannot downgrade as compression is enabled for continuous aggregates' 
-            USING DETAIL = 'Please disable compression on all continuous aggregates before downgrading',
-            HINT = 'To disable compression, call decompress_chunk to decompress chunks, then drop any existing compression policy on the continuous aggregate, and finally run ALTER MATERIALIZED VIEW % SET timescaledb.compress = ''false'' ';
-    END IF;
-END $$;
-
--- revert changes to continuous aggregates view definition
-DROP VIEW IF EXISTS timescaledb_information.continuous_aggregates;
-

diff --git a/version.config b/version.config
@@ -1,3 +1,3 @@
-version = 2.6.0-dev
+version = 2.6.0
 update_from_version = 2.5.2
 downgrade_to_version = 2.5.2