mirror of
https://github.com/ClickHouse/ClickHouse.git
synced 2024-12-11 00:42:29 +00:00
6.9 KiB
6.9 KiB
sidebar_position | sidebar_label |
---|---|
1 | 2023 |
2023 Changelog
ClickHouse release v22.8.14.53-lts (4ea67c4007
) FIXME as compared to v22.8.13.20-lts (e4817946d1
)
Performance Improvement
- Backported in #45845: Fixed performance of short
SELECT
queries that read from tables with large number ofArray
/Map
/Nested
columns. #45630 (Anton Popov). - Backported in #46374: Fix too big memory usage for vertical merges on non-remote disk. Respect
max_insert_delayed_streams_for_parallel_write
for the remote disk. #46275 (Nikolai Kochetov). - Backported in #46358: Allow using Vertical merge algorithm with parts in Compact format. This will allow ClickHouse server to use much less memory for background operations. This closes #46084. #46282 (Anton Popov).
Build/Testing/Packaging Improvement
- Backported in #46112: Remove the dependency on the
adduser
tool from the packages, because we don't use it. This fixes #44934. #45011 (Alexey Milovidov). - Backported in #46482: Get rid of unnecessary build for standalone clickhouse-keeper. #46367 (Mikhail f. Shiryaev).
- Backported in #46505: Some time ago the ccache compression was changed to
zst
, butgz
archives are downloaded by default. It fixes it by prioritizing zst archive. #46490 (Mikhail f. Shiryaev).
Bug Fix (user-visible misbehavior in official stable release)
- Backported in #45908: Fixed bug with non-parsable default value for EPHEMERAL column in table metadata. #44026 (Yakov Olkhovskiy).
- Backported in #46238: A couple of seg faults have been reported around
c-ares
. All of the recent stack traces observed fail on inserting intostd::unodered_set<>
. I believe I have found the root cause of this, it seems to be unprocessed queries. Prior to this PR, CH callspoll
to wait on the file descriptors in thec-ares
channel. According to the poll docs, a negative return value means an error has ocurred. Because of this, we would abort the execution and return failure. The problem is thatpoll
will also return a negative value if a system interrupt occurs. A system interrupt does not mean the processing has failed or ended, but we would abort it anyways because we were checking for negative values. Once the execution is aborted, the whole stack is destroyed, which includes thestd::unordered_set<std::string>
passed to thevoid *
parameter of the c-ares callback. Once c-ares completed the request, the callback would be invoked and would access an invalid memory address causing a segfault. #45629 (Arthur Passos). - Backported in #45727: Fix key description when encountering duplicate primary keys. This can happen in projections. See #45590 for details. #45686 (Amos Bird).
- Backported in #46394: Fix
SYSTEM UNFREEZE
queries failing with the exceptionCANNOT_PARSE_INPUT_ASSERTION_FAILED
. #46325 (Aleksei Filatov). - Backported in #46442: Fix possible
LOGICAL_ERROR
in asynchronous inserts with invalid data sent in formatVALUES
. #46350 (Anton Popov). - Backported in #46674: Fix an invalid processing of constant
LowCardinality
argument in functionarrayMap
. This bug could lead to a segfault in release, and logical errorBad cast
in debug build. #46569 (Alexey Milovidov). - Backported in #46879: Fix MSan report in the
maxIntersections
function. This closes #43126. #46847 (Alexey Milovidov). - Backported in #46871: Fix a bug in the
Map
data type. This closes #46855. #46856 (Alexey Milovidov).
NOT FOR CHANGELOG / INSIGNIFICANT
- Another attempt to fix automerge, or at least to have debug footprint #45476 (Mikhail f. Shiryaev).
- Add check for running workflows to merge_pr.py #45803 (Mikhail f. Shiryaev).
- Get rid of progress timestamps in release publishing #45818 (Mikhail f. Shiryaev).
- Add necessary dependency for sanitizers #45959 (Mikhail f. Shiryaev).
- Add helping logging to auto-merge script #46080 (Mikhail f. Shiryaev).
- Fix write buffer destruction order for vertical merge. #46205 (Nikolai Kochetov).
- Get rid of legacy DocsReleaseChecks #46665 (Mikhail f. Shiryaev).