Commit Graph

42795 Commits

Author SHA1 Message Date
Alexander Gololobov
a708adecee Pass element type as param to ctor 2023-06-21 14:04:45 +02:00
Alexey Milovidov
ced1ae7287
Update emptyArray.cpp 2023-06-20 21:20:25 +03:00
Alexander Gololobov
29f6ce871d Avoid initializing DateLUT from emptyArray function registration 2023-06-20 19:00:41 +02:00
Kruglov Pavel
0edfbb45ad
Merge pull request #50873 from Avogar/parquet-big-integers
Fallback to parsing big integer from String instead of exception in Parquet format
2023-06-20 16:10:46 +02:00
郭小龙
154d19ce72
The time format (connected_time field) is incorrect (#51026)
* The time format (connected_time field) is incorrect

* Change the description and increase the field list display

* Update 02735_system_zookeeper_connection.reference

* Update 02735_system_zookeeper_connection.sql

* Update 02735_system_zookeeper_connection.sql

* Update 02735_system_zookeeper_connection.reference

* fix test

---------

Co-authored-by: Alexander Tokmakov <tavplubix@gmail.com>
Co-authored-by: Alexander Tokmakov <tavplubix@clickhouse.com>
2023-06-20 16:57:55 +03:00
Alexander Tokmakov
bca0d76cbf
Merge pull request #50107 from ClickHouse/rmt_better_background_tasks_scheduling
Improve scheduling of background tasks in ReplicatedMergeTree
2023-06-20 16:25:19 +03:00
Kseniia Sumarokova
c3227fd44f
Merge pull request #51161 from kssenii/more-metrics-in-cache
More profile events for fs cache
2023-06-20 12:44:00 +02:00
kssenii
7f9cd56f61 One more 2023-06-19 19:22:45 +02:00
Han Fei
959f66eb8a
Merge pull request #50697 from kitaisreal/sort-block-improvements
Sort block improvements
2023-06-19 19:02:03 +02:00
kssenii
ec86028fd8 Change to microseconds 2023-06-19 17:48:56 +02:00
kssenii
0ca2916163 More metrics in cache 2023-06-19 17:22:41 +02:00
Sergei Trifonov
5867f39fc7
Merge pull request #50675 from ClickHouse/unify-priorities-pools
Unify priorities for connection pools
2023-06-19 17:16:15 +02:00
Alexey Milovidov
f9558345e8
Merge pull request #51138 from ClickHouse/print-changed-settings
Print changed settings on fatal errors
2023-06-19 15:50:44 +03:00
Alexander Tokmakov
a00842498e Merge branch 'master' into rmt_better_background_tasks_scheduling 2023-06-19 12:37:17 +02:00
Alexander Tokmakov
caceb7c862
Revert "Added ability to implicitly use file/hdfs/s3 table functions in clickhouse-local" 2023-06-19 13:21:09 +03:00
Maksim Kita
4c72995537 Fixed build 2023-06-19 13:02:35 +03:00
Alexey Milovidov
dcd4cfa5c2 Fix build 2023-06-19 03:28:59 +02:00
Alexey Milovidov
9ae4915655 Code simplification 2023-06-18 21:08:37 +02:00
Alexey Milovidov
b19b87a3e5 Whitespaces 2023-06-18 20:42:27 +02:00
Alexey Milovidov
a6dde6ab62 Print changed settings on fatal errors 2023-06-18 20:38:25 +02:00
Maksim Kita
977cd65cfb Fixed build 2023-06-18 12:52:08 +03:00
Maksim Kita
2f3dce5fe6 Fixed tests 2023-06-18 12:52:08 +03:00
Maksim Kita
7d17a70d17 Fixed tests 2023-06-18 12:52:08 +03:00
Maksim Kita
8d3212363e Fixed tests 2023-06-18 12:52:08 +03:00
Maksim Kita
ca298d66cb Sort block improvements 2023-06-18 12:52:08 +03:00
Alexey Milovidov
ce38d64c5a
Merge pull request #50856 from ClickHouse/relax-too-many-parts
Relax "too many parts" further
2023-06-18 01:49:46 +03:00
Alexander Tokmakov
b4e455430c
Merge branch 'master' into rmt_better_background_tasks_scheduling 2023-06-17 22:19:21 +03:00
Alexey Milovidov
c7a18c640e Slightly better diagnostics 2023-06-17 17:20:44 +02:00
Alexey Milovidov
61e423c5df Fix typo 2023-06-17 17:14:49 +02:00
Alexey Milovidov
98bc124317 Fix typo 2023-06-17 17:13:19 +02:00
Alexey Milovidov
e913a58284
Merge pull request #51049 from ClickHouse/ddl_replication_improvements
An optimiation for ALTERs and Replicated db with one shard
2023-06-17 16:57:59 +03:00
Alexey Milovidov
9fc5b9c2c5
Merge branch 'master' into relax-too-many-parts 2023-06-17 16:14:34 +03:00
Alexey Milovidov
c2e4a7714f
Merge pull request #48821 from alekseygolub/localfilefunction
Added ability to implicitly use file/hdfs/s3 table functions in clickhouse-local
2023-06-17 15:55:13 +03:00
Michael Kolupaev
03baa19e88 Fix ParallelReadBuffer seek 2023-06-16 17:50:51 -07:00
Michael Kolupaev
c5536e4906 Fix ColumnConst::forEachSubcolumn 2023-06-16 17:48:36 -07:00
kssenii
74f2dea57b Fix typo 2023-06-17 02:11:48 +02:00
kssenii
b902f5d045 Small style corrections 2023-06-17 01:54:38 +02:00
Alexey Milovidov
df9abccfaa
Merge branch 'master' into localfilefunction 2023-06-17 00:45:54 +03:00
Alexey Milovidov
a18dc73271 Fix error 2023-06-16 23:12:01 +02:00
Alexey Milovidov
688d5d59af Merge branch 'master' into relax-too-many-parts 2023-06-16 23:08:29 +02:00
Alexey Milovidov
ec952d401d
Merge pull request #51114 from ClickHouse/clang-17-3
Compatibility with clang-17
2023-06-17 00:07:45 +03:00
Alexey Milovidov
d94e47a930 Compatibility with clang-17 2023-06-16 23:07:19 +02:00
Alexey Milovidov
4e45e39e25
Merge pull request #51099 from ClickHouse/fix_use_local_vars_after_exception
Fix for part_names_mutex used after destruction
2023-06-16 23:52:50 +03:00
Alexey Milovidov
782889a051
Merge pull request #51042 from ClickHouse/sys_zookeep_parallel_list
Make multiple parallel list requests when reading from system.zookeeper
2023-06-16 23:46:53 +03:00
Michael Kolupaev
10d597676c Refactor IColumn::forEachSubcolumn to make it slightly harder to implement incorrectly 2023-06-16 11:08:12 -07:00
Michael Kolupaev
c2900177a2 Fix race condition in ColumnLowCardinality caused by subtly broken copy-on-write for the dictionary 2023-06-16 11:06:28 -07:00
Alexander Gololobov
8289d8fefa Fix for part_names_mutex used after destruction 2023-06-16 18:46:07 +02:00
Salvatore Mesoraca
6b49968bc4
Generate safe IVs
IVs should never be re-used when using CTR.
The previous implementation had a 50% probability to generate
an IV that was already used after 65536 generations.
This happens because std::mt19937 is seeded using a 32bit integer
and returns a 32 bits integer.
Because of the birthday problem, collisions have a 50% chance after
only 2^16 IV's generations.
CH uses one IV per file and, for most use cases, 65k files are
not as many as they seem.
Also, the initial entropy is gathered using std::random_device
which is "best-effort" and it's allowed by the standard to
return a fixed sequence of numbers.
2023-06-16 17:09:08 +02:00
Nikolai Kochetov
df1ea0b572
Merge pull request #50489 from ClickHouse/vdimir/alter_moving_garbage
Cleanup moving parts
2023-06-16 15:39:32 +02:00
Vitaly Baranov
f7638de213
Fix crash in BackupCoordinationStageSync::setError(). (#51012) 2023-06-16 15:32:30 +02:00