When inserting into a compressed chunk with constraints present,
we need to decompress relevant tuples in order to do speculative
inserting. Usually we used segment by column values to limit the
amount of compressed segments to decompress. This change expands
on that by also using segment metadata to further filter
compressed rows that need to be decompressed.
## 2.10.2 (2023-04-20)
**Bugfixes**
* #5410 Fix file trailer handling in the COPY fetcher
* #5446 Add checks for malloc failure in libpq calls
* #5233 Out of on_proc_exit slots on guc license change
* #5428 Use consistent snapshots when scanning metadata
* #5499 Do not segfault on large histogram() parameters
* #5470 Ensure superuser perms during copy/move chunk
* #5500 Fix when no FROM clause in continuous aggregate definition
* #5433 Fix join rte in CAggs with joins
* #5556 Fix duplicated entries on timescaledb_experimental.policies view
* #5462 Fix segfault after column drop on compressed table
* #5543 Copy scheduled_jobs list before sorting it
* #5497 Allow named time_bucket arguments in Cagg definition
* #5544 Fix refresh from beginning of Continuous Aggregate with variable time bucket
* #5558 Use regrole for job owner
* #5542 Enable indexscan on uncompressed part of partially compressed chunks
**Thanks**
* @nikolaps for reporting an issue with the COPY fetcher
* @S-imo-n for reporting the issue on Background Worker Scheduler crash
* @geezhu for reporting issue on segfault in historgram()
* @mwahlhuetter for reporting the issue with joins in CAggs
* @mwahlhuetter for reporting issue with duplicated entries on timescaledb_experimental.policies view
* @H25E for reporting error refreshing from beginning of a Continuous Aggregate with variable time bucket
This was previously disabled as no data resided on the
uncompressed chunk once it was compressed, but this is not
the case anymore with partially compressed chunks, so we
enable indexscan for the uncompressed chunk again.
Fixes#5432
Co-authored-by: Ante Kresic <ante.kresic@gmail.com>
Instead of using a user name to register the owner of a job, we use
regrole. This allows renames to work properly since the underlying OID
does not change when the owner name changes.
We add a check when calling `DROP ROLE` that there is no job with that
owner and generate an error if there is.
Our cost model should be self-consistent, and the relative values for
the remote tuple and startup costs should reflect their real cost,
relative to costs of other operations like CPU tuple cost.
For example, now remote costs are set even lower than the parallel tuple
and startup cost. Contrary to that, their real world cost is going to be
an order of magnitude higher or more, because parallel tuples are sent
through shared memory, and remote tuples are sent over the network.
Increasing these costs leads to query plan improvements, e.g. we start
to favor the GROUP BY pushdown in some cases.
Commit 3f9cb3c2 introduced new repair tests for broken Continuous
Aggregates with JOIN clause but in the post.repair.sql we not properly
calling the post.repair.cagg_joins.sql because a wrong usage of psql
`if` statement.
Whenever we create a template sql file (*.sql.in) we should add the
respective .gitignore entry for the generated test files.
So added a CI check to check for missing gitignore entries for generated
test files.
The start_scheduled_jobs function mistakenly sorts the scheduled_jobs
list in-place. As a result, when the ts_update_scheduled_jobs_list
function compares the updated list of scheduled jobs with the existing
scheduled jobs list, it is comparing a list that is sorted by job_id to
one that is sorted by next_start time. Fix that by properly copying the
scheduled_jobs list into a new list and use that for sorting.
Fixes#5537
The cmake call to "find_package(OpenSSL)" on windows may set the
variable OPENSSL_LIBRARIES to a non-standard list which might include
non path items such as "debug", "optimized". These non-standard
elements in the list would cause the link failure with error
"LNK1104".
Fix:
Check and retain only valid paths in OPENSSL_LIBRARIES list.
closes#407
With recent changes, we enabled analyze on uncompressed chunk tables
for compressed chunks. This change includes analyzing the compressed
chunks table when analyzing the hypertable and its chunks,
enabling us to remove the generating stats when compressing chunks.
In case of joins in the continuous aggregates, pass the required
structs to the new rte created. These values are required by the
planner to finally query the materialized view.
Fixes#5433
Commit 16fdb6ca5e introduced `timescaledb_experimental.policies` view
to expose the Continuous Aggregate policies but the current JOINS over
our catalog are not accurate.
Fixed it by properly JOIN the underlying catalog tables to expose the
correct information without duplicates about the Continuous Aggregate
policies.
Fixes#5492
When refreshing from the beginning (window_start=NULL) of a
Continuous Aggregate with variable time bucket we were getting a
`timestamp out of range` error.
Fixed it by setting `-Infinity` when passing `window_start=NULL` when
refreshing a Continuous Aggregate with variable time bucket.
Fixes#5474, #5534
Several error messages for continuous aggregates are not following the
error message style guidelines at
https://www.postgresql.org/docs/current/error-style-guide.html
In particular, they do not write the hints and detailed messages as
full sentences.
This is mostly a cosmetic change. When only 1 child is present there
is no need for ordered append. In this situation we might still
benefit from a ChunkAppend node here due to runtime chunk exclusion
when we have non-immutable constraints, so we still add the ChunkAppend
node in that situation even with only 1 child.
This patch drops the following internal SQL functions which were
unused:
_timescaledb_internal.is_main_table(regclass);
_timescaledb_internal.is_main_table(text, text);
_timescaledb_internal.hypertable_from_main_table(regclass);
_timescaledb_internal.main_table_from_hypertable(integer);
_timescaledb_internal.time_literal_sql(bigint, regtype);
While executing compression operations in parallel with
inserting into chunks (both operations which can potentially
change the chunk status), we could get into situations where
the chunk status would end up inconsistent. This change re-reads
the chunk status after locking the chunk to make sure it can
decompress data when handling ON CONFLICT inserts correctly.
Verify that insertion into compressed chunks does not block
each other if the chunks is already partially compressed. Also
check that using the RETURNING clause works the same.
These tests try to verify that changing physical layout
of chunks (either compressed or uncompressed) should
yield consistent results. They also verify index mapping
on compressed chunks is handled correctly.
Commit 8afdddc2da added the first step for deprecating the old format
of Continuous Aggregate but just for PostgreSQL 15 and later versions.
During the extension update we emit a message about the deprecation but
this has being emited even if the user is using PostgreSQL versions
before 15.
Fixed it by emiting the WARNING just when PostgreSQL version is greater
or equal to 15.
This patch moves the support functions for histogram, first and last
into the _timescaledb_functions schema. Since we alter the schema
of the existing functions in upgrade scripts and do not change the
aggregates this should work completely transparently for any user
objects using those aggregates.
Commit 57fde383b3dddd0b52263218e65a0135981c2d34 changed the
messaging but did not format the error hint correctly.
This patch fixes the error hint.
Fixes#5490
Now we look them up again at execution time, which adds up for tables
with a large number of chunks.
This gives about 15% speedup (100 mcs) on a small query on a table from
tests with 50 chunks:
`select id, ts, value from metric_compressed order by id, ts limit 100;`
Decompression produces records which have all the decompressed data
set, but it also retains the fields which are used internally during
decompression.
These didn't cause any problem - unless an operation is being done
with the whole row - in which case all the fields which have ended up
being non-null can be a potential segfault source.
Fixes#5458#5411
Currently internal user objects like chunks and our functions
live in the same schema making locking down that schema hard.
This patch adds a new schema _timescaledb_functions that is meant
to be the schema used for timescaledb internal functions to
allow separation of code and chunks or other user objects.
This patch does following:
1. Executor changes to parse qual ExprState to check if SEGMENTBY
column is specified in WHERE clause.
2. Based on step 1, we build scan keys.
3. Executor changes to do heapscan on compressed chunk based on
scan keys and move only those rows which match the WHERE clause
to staging area aka uncompressed chunk.
4. Mark affected chunk as partially compressed.
5. Perform regular UPDATE/DELETE operations on staging area.
6. Since there is no Custom Scan (HypertableModify) node for
UPDATE/DELETE operations on PG versions < 14, we don't support this
feature on PG12 and PG13.
Refactor the code path that handles remote distributed COPY. The
main changes include:
* Use a hash table to lookup data node connections instead of a list.
* Refactor the per-data node buffer code that accumulates rows into
bigger CopyData messages.
* Reduce the default number of rows in a CopyData message to 100. This
seems to improve throughput, probably striking a better balance
between message overhead and latency.
* The number of rows to send in each CopyData message can now be
changed via a new foreign data wrapper option.
OSM chunks have their own fdw_private which conflicts with checks in
the MergeAppend code path causing segfaults. This commit fixes this by
returning early when there is an OSM chunk in the MergeAppendPath.
Add isolation test case to check that the chunk object created during
chunk copy/move operation on the destination datanode always has
superuser credentials till the end of the operation.
The joins could be between a continuous aggregate and hypertable,
continuous aggregate and a regular Postgres table,
and continuous aggregate and a regular Postgres view.
There is a bug in `width_bucket()` causing an overflow and subsequent
NaN value as a result of dividing with `+inf`. The NaN value is
interpreted as an integer and hence generates an index out of range for
the buckets.
This commit fixes this by generating an error rather than
segfaulting for bucket indexes that are out of range.
During chunk creation, the chunk's dimensional CHECK constraints are
created via an "upcall" to PL/pgSQL code. However, creating
dimensional constraints in PL/pgSQL code sometimes fails, especially
during high-concurrency inserts, because PL/pgSQL code scans metadata
using a snapshot that might not see the same metadata as the C
code. As a result, chunk creation sometimes fail during constraint
creation.
To fix this issue, implement dimensional CHECK-constraint creation in
C code. Other constraints (FK, PK, etc.) are still created via an
upcall, but should probably also be rewritten in C. However, since
these constraints don't depend on recently updated metadata, this is
left to a future change.
Fixes#5456
This patch introduces a C-function to perform the recompression at
a finer granularity instead of decompressing and subsequently
compressing the entire chunk.
This improves performance for the following reasons:
- it needs to sort less data at a time and
- it avoids recreating the decompressed chunk and the heap
inserts associated with that by decompressing each segment
into a tuplesort instead.
If no segmentby is specified when enabling compression or if an
index does not exist on the compressed chunk then the operation is
performed as before, decompressing and subsequently
compressing the entire chunk.
There is a security loophole in current core Postgres, due to which
it's possible for a non-superuser to gain superuser access by attaching
dependencies like expression indexes, triggers, etc. before logical
replication commences.
To avoid this, we now ensure that the chunk objects that get created
for the subscription are done so as a superuser. This avoids malicious
dependencies by regular users.
When calling the `cagg_watermark` function to get the watermark of a
Continuous Aggregate we execute a `SELECT MAX(time_dimension)` query
in the underlying materialization hypertable.
The problem is that a `SELECT MAX(time_dimention)` query can be
expensive because it will scan all hypertable chunks increasing the
planning time for a Realtime Continuous Aggregates.
Improved it by creating a new catalog table to serve as a cache table
to store the current Continous Aggregate watermark in the following
situations:
- Create CAgg: store the minimum value of hypertable time dimension
data type;
- Refresh CAgg: store the last value of the time dimension materialized
in the underlying materialization hypertable (or the minimum value of
materialization hypertable time dimension data type if there's no
data materialized);
- Drop CAgg Chunks: the same as refresh cagg.
Closes#4699, #5307
During the compression autovacuum use to be disabled for uncompressed
chunk and enable after decompression. This leads to postgres
maintainence issue. Let's not disable autovacuum for uncompressed
chunk anymore. Let postgres take care of the stats in its natural way.
Fixes#309
No functional changes, mostly just reshuffles the code to prepare for
batch decompression.
Also removes unneeded repeated column value stores and ExecStoreTuple,
to save 3-5% execution time on some queries.
Invalidate the catalog snapshot in the scanner to ensure that any
lookups into `pg_catalog` uses a snapshot that is consistent with the
snapshot used to scan TimescaleDB metadata.
This fixes an issue where a chunk could be looked up without having a
proper relid filled in, causing an assertion failure
(`ASSERT_IS_VALID_CHUNK`). When a chunk is scanned and found (in
`chunk_tuple_found()`), the Oid of the chunk table is filled in using
`get_relname_relid()`, which could return InvalidOid due to use of a
different snapshot when scanning `pg_class`. Calling
`InvalidateCatalogSnapshot()` before starting the metadata scan in
`Scanner` ensures the pg_catalog snapshot used is refreshed.
Due to the difficulty of reproducing this MVCC issue, no regression or
isolation test is provided, but it is easy to hit this bug when doing
highly concurrent COPY:s into a distributed hypertable.
