* impl
* fix style
* make executeQueryWithParallelReplicas similar to executeQuery
* impl for parallel replicas
* cleaner code for remote sorting properties
* update test
* fix
* handle when nodes of old versions participate
* small fixes
* temporary enable for testing
* fix after merge
* Revert "temporary enable for testing"
This reverts commit cce7f8884c.
* review fixes
* add bc test
* Update src/Core/Settings.h
Can be useful if the table constantly recreated, i.e. in some tests.
v2: fix tests (disable for ordinary database)
v3: adjust test after merging replacing with empty parts pr
Signed-off-by: Azat Khuzhin <a.khuzhin@semrush.com>
(cherry picked from commit d0ce500f89)
In case of merges/mutations, separate MemoryTracker object is used, to
track memory separately, to report this metrics.
However, the tail of those allocations hadn't been taken into account,
and in case of lots of mutations/merges it can be significant.
100% reproducible with the following example:
Start the server w/o frequent sync with RSS:
$ clickhouse-server -C /src/programs/server/config.xml -- --asynchronous_metrics_update_period_s=86400
Run the following snippet:
create table data (key Int, value Nullable(Decimal(38, 0))) engine=ReplicatedMergeTree('/ch/data', '1') order by key
insert into data values (1, -13028236692093846346337460743176821145)
alter table data modify column value UInt256
Code: 341. DB::Exception: Received from localhost:9000. DB::Exception: Exception happened during execution of mutation 'mutation_2.txt' with part 'all_1_1_0' reason: 'Code: 407. DB::Exception: Convert overflow: while executing 'FUNCTION _CAST(value :: 1, 'UInt256' :: 2) -> _CAST(value, 'UInt256') UInt256 : 3': (while reading from part ./store/dd9/dd9ef183-1270-4877-af22-8a000bdcc1b8/all_1_1_0/): While executing MergeTreeInOrder. (DECIMAL_OVERFLOW) (version 22.12.1.1)'. This error maybe retryable or not. In case of unretryable error, mutation can be killed with KILL MUTATION query. (UNFINISHED)
And track the memory:
select value/1e9 from system.metrics where metric like 'MemoryTracking'
Reported-by: @filimonov
Signed-off-by: Azat Khuzhin <a.khuzhin@semrush.com>
