This should significantly reduce the MemoryTracking drift, test shows
that there is 0 drift after query storm (100 queries, via http/tcp/tcp
in one session).
TL;DR;
To track memory, clickhouse creates memory tracker object for each
thread **explicitly**, but until it is not created the memory
allocations are not under account.
There should not be lot of allocations w/o memory tracker, since most of
the time it is created early enough, but even this maybe enough to
trigger some problems.
Plus sometimes it is not possible to create it, for example some 3d
party library does not allow to do this explicitly:
- for example before #15740 allocations from librdkafka threads,
- or even worse, poco threads, they don't have any routines to do this.
This won't be a problem for `MemoryTracking` metric if the deallocation
will be done from the same thread w/o memory tracker (or vise versa),
but this is not always true.
NOTE, that this will slow down per-thread allocations w/o memory
tracker, since before this patch there were no memory tracking for them
while now they will be accounted in total_memory_tracker, and for
total_memory_tracker max_untracked_memory is always reached.
But this should not be significant.
This is a first pass at setting an x509 keypair dynamically.
I'm not so sure it should be a singleton (especaially because it depends
on the SSLManager singleton).
Needs more better error handling.
Functionally, it works. Simply change the cert/key path in config, or
touch cert; touch config.
* master: (279 commits)
Convert to python3 (#15007)
update en and ru docs: add 128/256 types (#15515)
Drop unused code for numeric_limits<int128> in MergeTreeDataSelectExecutor (#15519)
Fix comment
More defaults on defaults
Recursive defaults
Fix bash
Update ActionsVisitor.cpp
Don't add memcpy for sanitizers builds
Fix gRPC build scripts.
Neutrialize thinlto's memcpy libcall gen.
Revert "Add fno-builtin-memcpy"
Throw an error when a single parameter is passed to ReplicatedMergeTree instead of ignoring it
Update SECURITY.md
Update tsan_suppressions.txt
Update ActionsVisitor.cpp
remove unrelated changes
Update SECURITY.md
fixes
Update ActionsVisitor.cpp
...
* master: (224 commits)
Update in.md
Add a test for embedded configs
Allow to run without /proc/self/maps
Update adopters.md
More convenient
Enable embedded configs for AArch64
More results
More handy
More handy
Added a comment
Adjustments
Adjustments
Adjustments
Added new results
Step 1: make adding hardware benchmark results more convenient
Revert "Avoid deadlocks in Log/TinyLog"
Fix MSan report in QueryLog
add some disabled tests from arcadia to skip_list.json
Update skip_list.json
Bump CI. [2]
...
# Conflicts:
# docker/test/testflows/runner/Dockerfile
* master: (375 commits)
Update type-conversion-functions.md
Update maxmap.md
Update maxmap.md
Update maxmap.md
Update single_fixed_string_groupby.xml
Alter remove column properties and TTLs (#14742)
better fixed string group by support
Fix incorrect key condition of fixed strings.
constant output order
more tests for #14646
Maybe fix MSan report in base64
Proper exception message for wrong number of arguments of CAST
Added a test
Fix buffer overflow in "bar" function
Update convertMySQLDataType.cpp
Fix clang-tidy
Remove obsolete code from performance test
Slightly better code
Even more
Even more
...
# Conflicts:
# src/Interpreters/Context.cpp
Add inter-server cluster secret, it is used for Distributed queries
inside cluster, you can configure in the configuration file:
<remote_servers>
<logs>
<shard>
<secret>foobar</secret> <!-- empty -- works as before -->
...
</shard>
</logs>
</remote_servers>
And this will allow clickhouse to make sure that the query was not
faked, and was issued from the node that knows the secret. And since
trust appeared it can use initial_user for query execution, this will
apply correct *_for_user (since with inter-server secret enabled, the
query will be executed from the same user on the shards as on initator,
unlike "default" user w/o it).
v2: Change user to the initial_user for Distributed queries if secret match
v3: Add Protocol::Cluster package
v4: Drop Protocol::Cluster and use plain Protocol::Hello + user marker
v5: Do not use user from Hello for cluster-secure (superfluous)
Add distributed_ddl.pool_size to control maximum parallel to handle
distributed DDL.
Also:
- convert Exception constructors to fmt-like
- use sleepFor* over std::this_thread::sleep_for()
Functions to encrypt/decrypt any input data with OpenSSL's ciphers
with custom key, iv, and add (-gcm mode only).
_mysql versions are 100% compatitable with corresponding MySQL functions
Supported modes depend on OpenSSL version, but generally are:
aes-{128,192,56}-{ecb,cbc,cfb1,cfb8,cfb128,ofb,gcm}
Please note that in a -gcm mode a 16-byte tag is appended to the ciphertext
on encryption and is expected to be found at the end of ciphertext on decryption.
Added tests that verify compatibility with MySQL functions,
and test vectors for GCM mode from OpenSSL.
Added masking rules for aes_X funtions
Rules are installed by default to config.d/query_masking_rules.xml
* master: (414 commits)
Add .reference
Update 01460_mark_inclusion_search_crash.sql
Fix bug in mark inclusion search.
cosmetic fixes for performance report
cosmetic fixes in performance report
Add more docs about functional tests
Add ability to specify Default codec for columns (#14049)
better
done
Disable force TTL on optimise
Update CMakeLists.txt
Update custom parts of storage AST only if it has extended definition
DOCSUP-712: Documented the ttl_only_drop_parts setting (#13823)
Small fixes
gix comment and useDefault*(), add tests for nullables
Fix a build for old some OS with old find
Update hdfs.md
Fix 01085_max_distributed_connections flackiness
Fix 00974_distributed_join_on flackiness (by allow retries and hide logs)
Update adopters.md
...
* master: (30 commits)
Documentation improve: Translate [select] section into Chinese to improve readability (#13814)
Update adopters.md
partially disable test with MaterializeMySQL
one more unroll for arerage.cpp
Update adopters.md
Update adopters.md
Update adopters.md
Update adopters.md
fix#13819
DOCSUP-1888: Documented the input_format_avro_allow_missing_fields setting (#13671)
Fix "Arcadia" and "Unbundled" builds
DOCSUP-928: Documented the groupArraySample function (#13791)
Documentation about ReplacingMergeTree extended with type DateTime64 for column (#13498)
doc: update quotas.md (#13400)
Fix 01356_initialize_aggregation in unbundled build (change topKWeighted order)
Fix style
Add normalizedQueryHash function with tests
Fix topK/topKWeighted merge (wtih non-default parameters)
Add test
Add function "normalizeQuery"
...
# Conflicts:
# programs/server/Server.cpp
# src/Access/AccessControlManager.cpp
# src/Access/AccessControlManager.h
* master: (70 commits)
Fix terrible markup due to wrong @blinkov's script.
Update extended-roadmap.md
Update extended-roadmap.md
Fix bug on the website
Suppress Shellcheck
Fix typos
Fix typos
Install codespell
Fix typo
Update gtest_thread_pool_schedule_exception.cpp
Fix flaky unit test
Fix set index with const column pred
Updating healthcheck parameters.
Fix step overflow in range()
* Moving to standard healthcheck parameters. * Adding docker-compose pull and docker-compose down before executing docker-compose up. * Adding docker-compose ps if any of the containers are unhealthy.
Fix topK test
Adding missing certificates and dhparam.pem for openldap4.
Keep "metadata_loading" test only for release build
Fix error in test
Fix typo
...
This PR adds a server level config for overriding the default max number
of threads in global thread pool that is currently allowed (10,000).
This might be useful in scenarios where there are a large number of
distributed queries that are executing concurrently and where the
default number of max threads might not be necessarily be sufficient.
* master: (114 commits)
Update PushingToViewsBlockOutputStream.cpp
Update PushingToViewsBlockOutputStream.cpp
make clang-10 happy
Fix sync_async test (remove timeout)
CLICKHOUSEDOCS-631: temporary_files_codec, join_on_disk_max_files_to_merge settings. (#11242)
Suppress output of cancelled queries in clickhouse-client #9473
Better log messages in ConfigReloader
fix select from StorageJoin
Fix unit tests under MSan
Added test.
Fix build.
Fix arguments for AggregateFunctionQuantile/
Update style.md
Add a guide on error messages.
Report dictionary name on dictionary load errors.
more types in ASOF JOIN (#11301)
Fix part_log test
Update test.
Add perftest.
Parallel processing for PushingToViewsBlockOutputStream::writeSuffix
...
Example of such cases:
- SELECT GROUP BY LIMIT
- SELECT GROUP BY with subsequent MEMORY_LIMIT_EXCEEDED error
And it should be two-level aggregation, since otherwise there will be
only one hashtable which will be cleared correctly, only if you have
two-level GROUP BY some of hashtables will not be cleared since nobody
consume rows.
Before this patch:
09:39.015292 [ 3070801 ] {609a0610-e377-4132-9cf3-f49454cf3c96} <Information> executeQuery: Read 1000000 rows, 7.63 MiB in 0.707 sec., 1413826 rows/sec., 10.79 MiB/sec.
09:39.015348 [ 3070801 ] {609a0610-e377-4132-9cf3-f49454cf3c96} <Debug> MemoryTracker: Peak memory usage (for query): 51.93 MiB.
09:39.015942 [ 3070801 ] {} <Trace> Aggregator: Destroying aggregate states <-- **problem**
09:39.017057 [ 3070801 ] {} <Trace> Aggregator: Destroying aggregate states <--
09:39.017961 [ 3070801 ] {} <Debug> MemoryTracker: Peak memory usage (for query): 51.93 MiB.
09:39.018029 [ 3070801 ] {} <Information> TCPHandler: Processed in 0.711 sec.
After this patch:
16:24.544030 [ 3087333 ] {79da208a-b3c0-48d4-9943-c974a3cbb6ea} <Information> executeQuery: Read 1000000 rows, 7.63 MiB in 0.599 sec., 1670199 rows/sec., 12.74 MiB/sec.
16:24.544084 [ 3087333 ] {79da208a-b3c0-48d4-9943-c974a3cbb6ea} <Debug> MemoryTracker: Peak memory usage (for query): 72.11 MiB.
16:24.544398 [ 3087333 ] {79da208a-b3c0-48d4-9943-c974a3cbb6ea} <Trace> Aggregator: Destroying aggregate states
16:24.545485 [ 3087333 ] {79da208a-b3c0-48d4-9943-c974a3cbb6ea} <Trace> Aggregator: Destroying aggregate states
16:24.547053 [ 3087333 ] {} <Debug> MemoryTracker: Peak memory usage (for query): 72.11 MiB.
16:24.547093 [ 3087333 ] {} <Information> TCPHandler: Processed in 0.603 sec.
1. Moved Volume to separate file
2. Created IVolume interface and implemented current behaviour in implementation of new interface — VolumeJBOD
3. Replaced all old volume usages with new VolumeJBOD. Where it is unnecessary to have JBOD — left just IVolume.
4. Removed old Volume completely
5. Moved StoragePolicy to separated files
6. Moved DiskSelector to separated files
7. Removed DiskSpaceMonitor file
* New metrics provider (Procfs) + Refactored TasksStatsCounters
* Trivial statless test that ProcFS is provided
* Trivial perf test for ProcfsMetricsProvider
Co-authored-by: alexey-milovidov <milovidov@yandex-team.ru>
* Skip the `/* comments */ SELECT @@variables ...` from mysql-connector-java setup for MySQL Handler #9336
mysql-connector setup query:
/* mysql-connector-java-5.1.38 ( Revision: ${revinfo.commit} ) */SELECT @@session.auto_increment_increment AS auto_increment_increment, @@character_set_client AS character_set_client, @@character_set_connection AS character_set_connection, @@character_set_results AS character_set_results, @@character_set_server AS character_set_server, @@init_connect AS init_connect, @@interactive_timeout AS interactive_timeout...
ClickHouse side Error:
{} <Error> executeQuery: Code: 62, e.displayText() = DB::Exception: Syntax error: failed at position 74: @@session.auto_increment_increment AS auto_increment_increment, @@character_set_client AS character_set_client, @@character_set_connection AS character_set_conn. Expected one of: CAST, NULL...
Client side Exception:
java.sql.SQLException: Syntax error: failed at position 74: @@session.auto_increment_increment AS auto_increment_increment, @@character_set_client AS character_set_client, @@character_set_connection AS character_set_conn. Expected one of: CAST...
* add repalce 'SHOW VARIABLES' for mysql-connector-java-5.1.34 #9336
* Add java client(JDBC) integration test to test_mysql_protocol
* shift out java tests from dbms
* Update MySQLHandler.cpp
* Update MySQLHandler.cpp
* test_mysql_protocol: add Test.java exit code 1 when expection
Co-authored-by: alexey-milovidov <milovidov@yandex-team.ru>
