Commit Graph

117045 Commits

Author SHA1 Message Date
Dmitry Kardymon
c032dee3b4 Merge remote-tracking branch 'origin/master' into ADQM-867 2023-06-04 18:45:47 +00:00
Amos Bird
a66f68e5df
fix again 2023-06-05 01:48:13 +08:00
alesapin
8d8d06285f Add integration test 2023-06-04 18:59:22 +02:00
cmsxbc
4bb44c7c72
1. skip extract darwin toolchain in builder when uncessary
2. update MacOSX SDK version in toolchain readme to match in builder
2023-06-04 23:06:21 +08:00
alesapin
02923138a3 Merge branch 'master' into azure 2023-06-04 16:43:28 +02:00
alesapin
0f4dd26ceb Add async iteration to object storage 2023-06-04 16:03:44 +02:00
Sergey Kazmin
a3a1283467 remove try-catch from the impl of typeid_cast for refs 2023-06-04 16:22:58 +03:00
Sergey Kazmin
63eab2783e Make typeid_cast for pointers noexcept 2023-06-04 16:22:58 +03:00
robot-ch-test-poll
b8f9a57797
Merge pull request #50539 from ClickHouse/skim-msan
Disable skim (Rust library) under memory sanitizer
2023-06-04 15:03:14 +02:00
Amos Bird
e24c9267bc
fix 2023-06-04 20:06:27 +08:00
Maksim Kita
46cbdeeb7e Fixed tests 2023-06-04 15:02:46 +03:00
kssenii
24e015b961 Fix 2023-06-04 13:34:20 +02:00
kssenii
411cf84db2 Merge remote-tracking branch 'upstream/master' into add-some-assertions-2 2023-06-04 13:32:57 +02:00
Amos Bird
92b2200c55
mutation stages can be empty 2023-06-04 18:32:12 +08:00
Amos Bird
97bd3f0483
Proper mutation of skip indices and projections 2023-06-04 18:32:08 +08:00
alesapin
4fd674a3b2
Merge pull request #50521 from ClickHouse/get_rid_of_strange_object_storage
Remove strange object storage methods
2023-06-04 12:09:09 +02:00
auxten
db806bd394
Resize underlying vector only pos_offset == vector.size() 2023-06-04 17:44:29 +08:00
Azat Khuzhin
60c2245da4 Make 01565_query_loop_after_client_error slightly more robust
CI founds one failure of this test [1], which is quite interesting, the
test itself should not takes too long:

    2023.06.04 05:01:36.889144 [ 42287 ] {} <Trace> DynamicQueryHandler: Request URI: /?query=SELECT+%27Running+test+stateless%2F01565_query_loop_after_client_error.expect+from+pid%3D1053%27&database=system&connect_timeout=30&receive_timeout=30&send_timeout=30&http_connection_timeout=30&http_receive_timeout=30&http_send_timeout=30&output_format_parallel_formatting=0
    ...
    2023.06.04 05:01:37.364595 [ 42844 ] {fa79939f-1fa0-4f3b-b599-fd2269122d6a} <Debug> executeQuery: (from [::1]:40264) (comment: 01565_query_loop_after_client_error.expect) INSERT INTO t01565(c0, c1) VALUES  (stage: Complete)
    2023.06.04 05:01:37.366485 [ 42844 ] {fa79939f-1fa0-4f3b-b599-fd2269122d6a} <Information> TCPHandler: Change cancellation status from NOT_CANCELLED to FULLY_CANCELLED. Log message: Received 'Cancel' packet from the client, canceling the query.
    2023.06.04 05:01:37.366810 [ 42844 ] {fa79939f-1fa0-4f3b-b599-fd2269122d6a} <Debug> TCPHandler: Processed in 0.002539914 sec.

But for the last INSERT the client itself works **very slow**, which
seems was the reason why "\r" had been sent too early and was never
interpreted:

    send: sending "INSERT INTO t01565(c0, c1) VALUES ('1', 1) ;\n" to { exp4 }

    expect: does " " (spawn_id exp4) match glob pattern "INSERT"? no
    I
    expect: does " \u001b[21GI\u001b[0m\u001b[J\u001b[22G" (spawn_id exp4) match glob pattern "INSERT"? no
    
    expect: does " \u001b[21GI\u001b[0m\u001b[J\u001b[22G\u001b[21G" (spawn_id exp4) match glob pattern "INSERT"? no
    INSERT INTO t
    expect: does " \u001b[21GI\u001b[0m\u001b[J\u001b[22G\u001b[21GINSERT INTO t\u001b[0m" (spawn_id exp4) match glob pattern "INSERT"? yes
    expect: set expect_out(0,string) "INSERT"
    expect: set expect_out(spawn_id) "exp4"
    expect: set expect_out(buffer) " \u001b[21GI\u001b[0m\u001b[J\u001b[22G\u001b[21GINSERT"
    send: sending "\r" to { exp4 }

    expect: does " INTO t\u001b[0m" (spawn_id exp4) match glob pattern "Ok."? no
    
    expect: does " INTO t\u001b[0m\u001b[J" (spawn_id exp4) match glob pattern "Ok."? no
    
    expect: does " INTO t\u001b[0m\u001b[J\u001b[34G" (spawn_id exp4) match glob pattern "Ok."? no
    
    expect: does " INTO t\u001b[0m\u001b[J\u001b[34G\u001b[21G" (spawn_id exp4) match glob pattern "Ok."? no
    INSERT INTO t01565(c0, c1) VALUES ('1', 1) ;

    
    expect: does " INTO t\u001b[0m\u001b[J\u001b[34G\u001b[21G\u001b[JINSERT INTO t01565\u001b[0;22;33m(\u001b[0mc0\u001b[0;1m,\u001b[0m c1\u001b[0;22;33m)\u001b[0m VALUES \u001b[0;22;33m(\u001b[0;22;36m'1'\u001b[0;1m,\u001b[0m \u001b[0;22;32m1\u001b[0;22;33m)\u001b[0m \u001b[0;1m;\u001b[0m\r\n\r\n\u001b[0m\u001b[1G" (spawn_id exp4) match glob pattern "Ok."? no
    expect: timed out

Here you can see that it matched "INSERT" and then expect receive data
from the client almost byte by byte.

So I hope that expecting the last part of the query should fix the
problem.

  [1]: https://s3.amazonaws.com/clickhouse-test-reports/50429/228ebab86db95dca1e29967061d245985bc86a0f/stateless_tests__release__s3_storage__[2_2].html
Signed-off-by: Azat Khuzhin <a.khuzhin@semrush.com>
2023-06-04 11:37:48 +02:00
Alexey Milovidov
07a6e49941
Merge pull request #50527 from ClickHouse/fix-bad-code-rocksdb
Fix low quality code around metadata in RocksDB (experimental feature never used in production)
2023-06-04 06:48:19 +03:00
Alexey Milovidov
640b758ab0
Merge pull request #50533 from kssenii/add-some-metrics
Add profile events for fs cache eviction
2023-06-04 06:03:29 +03:00
Alexey Milovidov
53ec091c8d Disable skim (Rust library) under memory sanitizer 2023-06-04 05:00:29 +02:00
Alexey Milovidov
7ceedbd180 Update tests 2023-06-04 04:45:55 +02:00
Alexey Milovidov
c62558f982 Control memory usage in generateRandom 2023-06-04 04:44:51 +02:00
Alexey Milovidov
8c1091dc58
Merge branch 'master' into fix-secure-async-read-write 2023-06-04 05:06:59 +03:00
Alexey Milovidov
c84c0be1a7
Merge branch 'master' into try-fix-flaky-test 2023-06-04 05:06:40 +03:00
pufit
e548e4b38c
Merge pull request #50278 from ClickHouse/roles_in_config
Specify roles in users.xml
2023-06-03 21:10:26 -04:00
Alexey Milovidov
28862129a5
Merge pull request #50385 from marising/add_boost_link
Link boost::context library to clickhouse_common_io
2023-06-04 04:02:31 +03:00
Alexey Milovidov
6f01e2ad9f Fix incorrect constant folding 2023-06-04 02:49:45 +02:00
Nikolay Degterinsky
aa19b2fc40
Merge pull request #50044 from azat/optimize_use_projections-alias
Preserve backward incompatibility for renamed settings by using aliases
2023-06-04 02:33:38 +02:00
robot-ch-test-poll4
b1de1d3ada
Merge pull request #50535 from kitaisreal/query-node-small-fix
QueryNode small fix
2023-06-04 02:20:15 +02:00
alesapin
57070227a8 Fxi 2023-06-03 23:44:19 +02:00
alesapin
a0df856605 Able to insert 2023-06-03 23:11:03 +02:00
kssenii
009fe3d25e Add profile events for eviction 2023-06-03 21:35:27 +02:00
Azat Khuzhin
69aec7af9b Add new metrics BrokenDistributedBytesToInsert/DistributedBytesToInsert
Useful to see at the server status overall.

Signed-off-by: Azat Khuzhin <a.khuzhin@semrush.com>
2023-06-03 20:49:19 +02:00
Azat Khuzhin
0f37be5492 Fix description for BrokenDistributedFilesToInsert
Signed-off-by: Azat Khuzhin <a.khuzhin@semrush.com>
2023-06-03 20:49:19 +02:00
Maksim Kita
e61131c4bb QueryNode small fix 2023-06-03 21:45:53 +03:00
Maksim Kita
e28dfb7ea8 Updated tests 2023-06-03 21:44:31 +03:00
pufit
1aaac69e13
Merge branch 'master' into roles_in_config 2023-06-03 14:42:10 -04:00
Azat Khuzhin
13a1226971 Fix nested distributed SELECT in WITH clause
For the CSE (common scalar expression, form of CTE) form of WITH it will
set max_result_rows to 1, since there should not be more rows, but this
will be applied for the DESC TABLE as well (service distributed query
that required to obtain the structure).

Note, that it is a problem only for nested distributed queries because
getSubqueryContext() from InterpreterSelectQuery.cpp resets those limits
as well, but this does not helps, for the nested DESC since it is
executed on the remote node that has max_result_rows=1.

Signed-off-by: Azat Khuzhin <a.khuzhin@semrush.com>
2023-06-03 20:22:51 +02:00
kssenii
05fbebbdac Merge remote-tracking branch 'upstream/master' into add-some-assertions-2 2023-06-03 20:19:14 +02:00
alesapin
894457d6e1 Style fxi 2023-06-03 14:36:52 +02:00
Maksim Kita
7a7e03a2ff Function if constant folding 2023-06-03 14:16:59 +03:00
Sergei Trifonov
f1eba08eaf
Merge pull request #50379 from alekar/minor-cgroup-improvements
Minor improvements in CGroup awareness.
2023-06-03 12:36:39 +02:00
Sema Checherinda
2177e562b8
Merge pull request #50518 from CheSema/lesser-traces
less traces in logs
2023-06-03 10:57:45 +02:00
Alexey Milovidov
8018e58b74
Merge pull request #49894 from evillique/postgresql-uuid
Add UUID data type to PostgreSQL
2023-06-03 07:13:41 +03:00
Alexey Milovidov
ccba3500dd
Update StoragePostgreSQL.cpp 2023-06-03 07:13:26 +03:00
Alexey Milovidov
054b908b9c
Merge pull request #50416 from ClickHouse/revert-49656-rs/entropy-learned-hashing
Revert "Add SQL functions for Entropy Learned Hashing"
2023-06-03 06:50:48 +03:00
Alexey Milovidov
129fe9158f
Merge pull request #50486 from Algunenano/system_on_cluster
Mention ON CLUSTER options for SYSTEM queries
2023-06-03 06:48:54 +03:00
Alexey Milovidov
dbfb3b810b
Merge branch 'master' into nulllogger 2023-06-03 06:47:42 +03:00
Alexey Milovidov
30be723a9a Fix bad code around metadata in RocksDB 2023-06-03 05:11:02 +02:00