owner stringclasses 14
values | repo stringclasses 14
values | id int64 116k 4.61B | issue_number int32 1 181k | author stringlengths 1 39 | body stringlengths 1 262k | created_at timestamp[us]date 2000-06-06 02:40:44 2026-06-02 20:16:18 | updated_at timestamp[us]date 2000-06-06 02:40:44 2026-06-02 20:22:56 | reactions unknown | author_association stringclasses 5
values |
|---|---|---|---|---|---|---|---|---|---|
ClickHouse | ClickHouse | 866,341,127 | 25,549 | tavplubix | But there is no `ALTER PARTITION` queries in the test... | 2021-06-22T21:18:51 | 2021-06-22T21:18:51 | {} | MEMBER |
ClickHouse | ClickHouse | 865,812,797 | 24,420 | PHaroZ | many thanks for your work ! | 2021-06-22T09:27:32 | 2021-06-22T09:27:32 | {} | NONE |
ClickHouse | ClickHouse | 865,893,523 | 25,455 | KochetovNicolai | @vitlibar you have merge it too early
```
2021-06-22 13:11:00 01913_join_push_down_bug: [ FAIL ] 0.10 sec. - return code 115
2021-06-22 13:11:00 [7d1627c41c66] 2021.06.22 13:11:00.258127 [ 160 ] {ed22d544-e4cc-4f6e-ab47-ffdd9b9f24a1} <Error> executeQuery: Code: 115, e.di... | 2021-06-22T11:14:21 | 2021-06-22T11:14:21 | {} | MEMBER |
ClickHouse | ClickHouse | 866,280,022 | 22,068 | alexey-milovidov | No info provided. | 2021-06-22T19:41:39 | 2021-06-22T19:41:39 | {} | MEMBER |
ClickHouse | ClickHouse | 866,335,534 | 22,841 | alexey-milovidov | https://github.com/ClickHouse/ClickHouse/commit/133443bd144de6bec11894099871fa8ec64112c7 | 2021-06-22T21:09:26 | 2021-06-22T21:09:26 | {} | MEMBER |
ClickHouse | ClickHouse | 865,737,883 | 25,544 | excitoon | 21.1 is not maintained. | 2021-06-22T08:37:12 | 2021-06-22T08:37:33 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 866,321,272 | 10,409 | alexey-milovidov | I don't see how perfect hashing can be applied.
It is for preparation of static hash tables with data known in advance. | 2021-06-22T20:45:38 | 2021-06-22T20:46:06 | {} | MEMBER |
ClickHouse | ClickHouse | 866,351,466 | 23,850 | alexey-milovidov | Also you may ask why when changing from two to three values, the time changes from 1s to just 1.2s - less that 50%.
Most likely this is because the dependency between time and the amount of data is not strictly linear.
Roaring bitmap is not a simple data structure. There can be some steps in query execution time du... | 2021-06-22T21:36:50 | 2021-06-22T21:36:50 | {} | MEMBER |
ClickHouse | ClickHouse | 866,386,258 | 25,565 | mergify[bot] | **Command `update`: success**
> **Branch has been successfully updated**
>
| 2021-06-22T22:46:40 | 2021-06-22T22:46:40 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 866,388,863 | 14,441 | alexey-milovidov | @tavplubix We can do lazy loading of all or a subset of databases during server startup.
We can make make a wrapper `DatabaseLoader` that will load the underlying database in a thread.
All the methods will wait for loading to finish and redirect call to the method of the underlying database, or if it was loaded with ... | 2021-06-22T22:53:27 | 2021-06-22T22:54:16 | {} | MEMBER |
ClickHouse | ClickHouse | 866,390,176 | 588 | alexey-milovidov | Fixed in #22976:
```
milovidov@github-explorer:~$ sudo ls -l /etc/clickhouse-server/
total 68
dr-x------ 2 clickhouse clickhouse 4096 Dec 1 2020 config.d
-r-------- 1 clickhouse clickhouse 38445 Dec 7 2020 config.xml
dr-x------ 2 clickhouse clickhouse 4096 Dec 1 2020 users.d
-r-------- 1 clickhouse cli... | 2021-06-22T22:56:47 | 2021-06-22T22:57:21 | {} | MEMBER |
ClickHouse | ClickHouse | 865,866,597 | 25,552 | vitlibar | Depends on https://github.com/ClickHouse/ClickHouse/pull/23237 which doesn't exist in `21.5`. | 2021-06-22T10:31:39 | 2021-06-22T10:31:39 | {} | MEMBER |
ClickHouse | ClickHouse | 865,931,172 | 25,551 | tavplubix | > The only failure - `Functional stateless tests (ubsan) — fail: 1` - Timeout on wating for prompt because of `Cannot create file` exception...
It fails because of TOCTOU issue here:
https://github.com/ClickHouse/ClickHouse/blob/bdc8fe1ca08d707099d1f0c36fa0d1aa5bbfeca9/programs/client/Client.cpp#L578-L579
Could ... | 2021-06-22T12:13:19 | 2021-06-22T12:13:19 | {} | MEMBER |
ClickHouse | ClickHouse | 865,961,930 | 24,980 | nikitamikhaylov | @Andr0901 Would you like to bring this PR up to the mergeable state? | 2021-06-22T13:00:48 | 2021-06-22T13:00:48 | {} | MEMBER |
ClickHouse | ClickHouse | 865,990,424 | 25,457 | morozovsk | when you create/recreate materialized view just use TO to existing table. so when you need to recreate materialized view (for example if you need to add new column) you can just drop materialized view and create new one instead and your data will be saved at existing table. | 2021-06-22T13:36:16 | 2021-06-22T13:36:16 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 866,044,246 | 25,483 | qoega | FYI @bharatnc | 2021-06-22T14:41:47 | 2021-06-22T14:41:47 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 866,391,772 | 18,923 | alexey-milovidov | Does not reproduce:
```
milovidov@milovidov-desktop:~/work/ClickHouse/programs/server$ docker run -it --rm --network host yandex/clickhouse-server:21.6
Processing configuration file '/etc/clickhouse-server/config.xml'.
Merging configuration file '/etc/clickhouse-server/config.d/docker_related_config.xml'.
Logging ... | 2021-06-22T23:01:12 | 2021-06-22T23:01:12 | {} | MEMBER |
ClickHouse | ClickHouse | 865,997,681 | 21,363 | abyss7 | @MergifyIo update | 2021-06-22T13:45:33 | 2021-06-22T13:45:33 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 866,034,070 | 25,428 | mergify[bot] | **Command `update`: success**
> **Branch has been successfully updated**
>
| 2021-06-22T14:29:37 | 2021-06-22T14:29:37 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 866,039,604 | 25,595 | mergify[bot] | **Command `update`: success**
> **Branch already up to date**
>
| 2021-06-22T14:36:20 | 2021-06-22T14:36:20 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 866,327,787 | 23,199 | alexey-milovidov | `MergeTreeReaderWide::readRows` - make a loop over columns to request reading more than one column at once. | 2021-06-22T20:56:44 | 2021-06-22T20:56:44 | {} | MEMBER |
ClickHouse | ClickHouse | 865,860,881 | 25,452 | Avogar | @filimonov maybe you have an idea how to solve this ambiguity? One of the possible solution is to add new data type RawString, but this is overkill I think. | 2021-06-22T10:22:41 | 2021-06-22T10:22:41 | {} | MEMBER |
ClickHouse | ClickHouse | 865,873,764 | 25,452 | amosbird | > @filimonov maybe you have an idea how to solve this ambiguity? One of the possible solution is to add new data type RawString, but this is overkill I think.
It's also possible to give a different name like `JSONExtractWithRawString`. However, I don't understand the impact of this ambiguity. What would break? | 2021-06-22T10:43:40 | 2021-06-22T10:44:24 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 865,895,442 | 25,455 | KochetovNicolai | Will try to fix it in 21.3 branch | 2021-06-22T11:16:50 | 2021-06-22T11:16:50 | {} | MEMBER |
ClickHouse | ClickHouse | 866,022,425 | 25,028 | l1t1 | is this issue confirmed?
https://github.com/ClickHouse/ClickHouse/issues/24075 | 2021-06-22T14:15:32 | 2021-06-22T14:15:32 | {} | NONE |
ClickHouse | ClickHouse | 866,318,127 | 24,209 | alexey-milovidov | I checked the code and can qualify it as a bug.
`KeyGetterForDict::findKey` is reading the whole dictionary for every record.
This code simply should not exist. | 2021-06-22T20:40:21 | 2021-06-22T20:40:21 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 866,341,705 | 22,108 | alexey-milovidov | The issue is fixed.
20.9:
```
SELECT bitmapMax(argMax(x, y))
FROM remote('127.0.0.{2,3}', view(
SELECT
groupBitmapState(toUInt32(number)) AS x,
number AS y
FROM numbers(10)
GROUP BY number
))
Query id: 3271c235-c0d8-42c5-ba59-1922786c7f34
Connecting to localhost:9000 as use... | 2021-06-22T21:19:44 | 2021-06-22T21:19:44 | {} | MEMBER |
ClickHouse | ClickHouse | 866,352,897 | 7,862 | alexey-milovidov | Most likely this issue is fixed in https://github.com/ClickHouse/ClickHouse/pull/25015 (version 21.7).
There is very similar issue that has been fixed by this PR: https://github.com/ClickHouse/ClickHouse/issues/22108 | 2021-06-22T21:39:37 | 2021-06-22T21:39:52 | {} | MEMBER |
ClickHouse | ClickHouse | 865,912,813 | 24,887 | filimonov | see also https://github.com/ClickHouse/ClickHouse/issues/18110 | 2021-06-22T11:44:50 | 2021-06-22T11:44:50 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 865,916,864 | 25,590 | sunbc996 | CREATE TABLE XXX (
`XXX` Date,
`XXX` String,
`XXX` String,
`roaring_bitmap` AggregateFunction(groupBitmap, UInt64) MATERIALIZED base64Decode(XXX)
) ENGINE = AggregatingMergeTree() PARTITION BY toYYYYMMDD(XXX)
ORDER BY
sdt SETTINGS index_granularity = 8192 | 2021-06-22T11:51:55 | 2021-06-22T11:51:55 | {} | NONE |
ClickHouse | ClickHouse | 865,943,085 | 25,588 | sunzaizhi11 | This problem does not exist in newest versions? | 2021-06-22T12:32:02 | 2021-06-22T12:32:02 | {} | NONE |
ClickHouse | ClickHouse | 865,997,946 | 21,363 | mergify[bot] | **Command `update`: failure**
> **Pull request can't be updated with latest base branch changes**
> Mergify needs the author permission to update the base branch of the pull request.
> amosbird needs to [authorize modification on its head branch](https://help.github.com/articles/allowing-changes-to-a-pull-request-branc... | 2021-06-22T13:45:52 | 2021-06-22T13:45:52 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 865,998,354 | 25,028 | alexey-milovidov | All bugs are fixed.
It is updated back to master 21.7. | 2021-06-22T13:46:24 | 2021-06-22T13:46:33 | {} | MEMBER |
ClickHouse | ClickHouse | 866,033,106 | 25,428 | alesapin | @Mergifyio update | 2021-06-22T14:28:27 | 2021-06-22T14:28:27 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 866,146,362 | 19,970 | r-zenine | @etah000 Any update ?
The company I work for would be delighted to see this capability built in `Clickhouse`. If you think this is the way to go, my team could help improve that fork until it's production ready :) | 2021-06-22T16:34:36 | 2021-06-22T16:34:36 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 866,290,350 | 24,209 | vkosh | @alexey-milovidov,
In real queries I use GROUP BY to aggregate some data from MergeTree table, grouped by dictionary key field, e.g.:
```
select count(), user_dict.email
from click join user_dict on click.user_id = user_dict.id
group by click.user_id;
```
So I actually use workaround from 2.2 case.
Anyway thank... | 2021-06-22T19:57:19 | 2021-06-22T19:57:19 | {} | NONE |
ClickHouse | ClickHouse | 865,671,598 | 25,404 | alexey-milovidov | `std::make_shared<ASTLiteral>` should be created with value 1. | 2021-06-22T07:31:33 | 2021-06-22T07:31:33 | {} | MEMBER |
ClickHouse | ClickHouse | 866,177,128 | 25,529 | alexey-milovidov | `SET max_bytes_before_external_sort = '50G', max_bytes_before_external_group_by = '50G', distributed_aggregation_memory_efficient = 1;` | 2021-06-22T17:16:27 | 2021-06-22T17:16:27 | {} | MEMBER |
ClickHouse | ClickHouse | 866,282,785 | 24,209 | alexey-milovidov | @vkosh You have `LIMIT 1` in queries without GROUP BY, they return first record as soon as it has been read.
In contrast, the queries with GROUP BY have to finish reading and aggregating all data before returning result.
Workaround: remove GROUP BY and replace it with DISTINCT. | 2021-06-22T19:45:56 | 2021-06-22T19:45:56 | {} | MEMBER |
ClickHouse | ClickHouse | 866,287,546 | 3,708 | alexey-milovidov | `agg(x, y ORDER BY a, b LIMIT 10)` will be nice to have. | 2021-06-22T19:53:57 | 2021-06-22T19:53:57 | {} | MEMBER |
ClickHouse | ClickHouse | 866,296,858 | 9,958 | alexey-milovidov | `system.trace_log` helps:
```
milovidov@milovidov-desktop:~/work/ClickHouse$ seq 1 1000 | pv --rate-limit 10 | clickhouse-client --query_profiler_real_time_period_ns 1000000 --query "INSERT INTO t FORMAT TSV"
^C00 B 0:00:20 [10,1 B/s] [ ... | 2021-06-22T20:06:55 | 2021-06-22T20:06:55 | {} | MEMBER |
ClickHouse | ClickHouse | 866,390,998 | 6,921 | alexey-milovidov | No info provided. | 2021-06-22T22:59:11 | 2021-06-22T22:59:11 | {} | MEMBER |
ClickHouse | ClickHouse | 866,400,929 | 25,513 | azat | > [Functional stateless tests (memory) — fail: 1, passed: 3132, skipped: 14](https://clickhouse-test-reports.s3.yandex.net/25513/b96df135aee40b1a54a4fc3f0d4db89e11385564/functional_stateless_tests_(memory).html#fail1)
01442_merge_detach_attach - #25605 | 2021-06-22T23:22:40 | 2021-06-22T23:48:16 | {} | MEMBER |
ClickHouse | ClickHouse | 865,738,133 | 25,543 | excitoon | 21.2 is not maintained. | 2021-06-22T08:37:19 | 2021-06-22T08:37:19 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 865,803,853 | 25,401 | MaxWk | > @MaxWk this is about exporting internal CH metrics to some prometeus, right? Same as exporter.
>
> What I mean, is using CH as a Storage for Prometeus, which a called [remote write](https://prometheus.io/docs/operating/configuration/#remote_write) and [remote read](https://prometheus.io/docs/operating/configuratio... | 2021-06-22T09:21:16 | 2021-06-22T09:21:16 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 865,927,320 | 25,549 | alesapin | ```
2021.06.21 21:04:28.827229 [ 1455 ] {} <Error> test_vofhux.ttl_table2 (1de274a3-4e40-4ffd-9de2-74a34e401ffd): auto DB::StorageReplicatedMergeTree::processQueueEntry(ReplicatedMergeTreeQueue::SelectedEntryPtr)::(anonymous class)::operator()(DB::StorageReplicatedMergeTree::LogEntryPtr &) const: Code: 49, e.displayTe... | 2021-06-22T12:07:48 | 2021-06-22T12:07:48 | {} | MEMBER |
ClickHouse | ClickHouse | 866,096,395 | 25,529 | IgorLelyakov | Добрый день!
Выставил max_bytes_before_external_sort= 85 899 345 92 - тоже самое сообщение об ошибке,
уменьшил выставил max_bytes_before_external_sort= 42 949 672 960 - то же самое сообщение об ошибке о превышении лимита памяти, что порекомендуете еще уменьшить или наоборот увеличить?
max_memory_usage=128 849 018 88... | 2021-06-22T15:39:19 | 2021-06-22T15:39:19 | {} | NONE |
ClickHouse | ClickHouse | 865,850,566 | 25,582 | tavplubix | > > What is the scenario? Why this feature is a candidate for backporting?
>
> `system` database became `ATOMIC` by default (20.8+, #13680), so it will remove all existing data (since it is loaded with force -
>
> https://github.com/ClickHouse/ClickHouse/blob/d6bdde54d94a4eaf337c4e62842d6dd9b90f21ea/src/Interpret... | 2021-06-22T10:07:41 | 2021-06-22T10:07:41 | {} | MEMBER |
ClickHouse | ClickHouse | 865,870,007 | 25,555 | vitlibar | Depends on https://github.com/ClickHouse/ClickHouse/pull/23237 which doesn't exist in `21.3`. | 2021-06-22T10:37:13 | 2021-06-22T10:37:13 | {} | MEMBER |
ClickHouse | ClickHouse | 866,218,955 | 21,363 | tavplubix | Could you please also add a stress-test that concurrently runs `REPLACE PARTITION ... FROM PARTITION ...` and other queries (such as `OPTIMIZE` and `ALTER UPDATE`)? See `00993_system_parts_race_condition_drop_zookeeper` and `01154_move_partition_long` for examples. | 2021-06-22T18:12:16 | 2021-06-22T18:12:16 | {} | MEMBER |
ClickHouse | ClickHouse | 866,385,477 | 12,990 | alexey-milovidov | @qoega It is a different issue. Create another task. | 2021-06-22T22:44:39 | 2021-06-22T22:44:39 | {} | MEMBER |
ClickHouse | ClickHouse | 865,713,173 | 24,705 | vitlibar | Looks like the test `01866_datetime64_cmp_with_constant` has to be adjusted for `20.8`.
@kitaisreal Can you help with that? | 2021-06-22T08:26:40 | 2021-06-22T08:35:16 | {} | MEMBER |
ClickHouse | ClickHouse | 865,855,073 | 25,533 | qoega | similar
https://clickhouse-test-reports.s3.yandex.net/25570/50dc48e99f8a7c39b6a6d6570a0c75f9575252fd/fuzzer_ubsan/report.html#fail1
``` sql
SELECT NULL AND (21
47483648 AND NULL) AND -2147483647, toUUID(((1048576 AND NULL) AND (2147483647 AND 257 AND NULL AND -2147483649) AND NULL) IN (test_01103.t1_distr.id), '000... | 2021-06-22T10:14:19 | 2021-06-22T10:14:19 | {} | MEMBER |
ClickHouse | ClickHouse | 866,104,536 | 25,549 | alesapin | @Mergifyio update | 2021-06-22T15:47:13 | 2021-06-22T15:47:13 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 865,671,453 | 25,455 | vitlibar | @KochetovNicolai Please help to fix the compilation. | 2021-06-22T07:31:18 | 2021-06-22T07:31:18 | {} | MEMBER |
ClickHouse | ClickHouse | 865,922,164 | 25,588 | tavplubix | Fixed in #23433
Version 21.4.4.30 is outdated, please consider upgrading | 2021-06-22T12:00:29 | 2021-06-22T12:00:29 | {} | MEMBER |
ClickHouse | ClickHouse | 866,183,271 | 25,028 | alexey-milovidov | @l1t1 Playground is using x86_64. | 2021-06-22T17:24:55 | 2021-06-22T17:24:55 | {} | MEMBER |
ClickHouse | ClickHouse | 866,203,814 | 21,363 | tavplubix | > @tavplubix from where does the flakiness come in?
I just see that `01744_alter_replace_partition_update` is added to `skip_list.json`. I supposed that it's because this test was failing in flaky check due to parallel run. | 2021-06-22T17:55:54 | 2021-06-22T17:55:54 | {} | MEMBER |
ClickHouse | ClickHouse | 866,325,440 | 23,199 | alexey-milovidov | > Could you please explain, why CH have used only 1 thread for
Due to very low amount of records in a table (just 20 000 records is about nothing).
> Can it process such queries using more threads?
It will make the query processing time even worse. As it will do two times more requests. | 2021-06-22T20:52:53 | 2021-06-22T20:52:53 | {} | MEMBER |
ClickHouse | ClickHouse | 866,326,204 | 23,199 | alexey-milovidov | What do you need is to request more than one column at once from single thread to mitigate obnoxiously high s3 latency. | 2021-06-22T20:54:13 | 2021-06-22T20:54:34 | {} | MEMBER |
ClickHouse | ClickHouse | 866,346,529 | 25,549 | tavplubix | Yes, the following code is incorrect:
https://github.com/ClickHouse/ClickHouse/blob/67e9b85951d47ba2c056f3cb8daaf1cf3459d869/src/Storages/StorageReplicatedMergeTree.cpp#L6865-L6872
https://github.com/ClickHouse/ClickHouse/blob/67e9b85951d47ba2c056f3cb8daaf1cf3459d869/src/Storages/StorageReplicatedMergeTree.cpp#L3163-... | 2021-06-22T21:27:58 | 2021-06-22T21:27:58 | {} | MEMBER |
ClickHouse | ClickHouse | 866,382,653 | 25,565 | alexey-milovidov | @mergifyio update | 2021-06-22T22:37:24 | 2021-06-22T22:37:24 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 866,386,699 | 16,132 | alexey-milovidov | Looks like this PR is abandoned. | 2021-06-22T22:47:56 | 2021-06-22T22:47:56 | {} | MEMBER |
ClickHouse | ClickHouse | 865,690,680 | 25,525 | qoega | It was the same picture here #25212
<img width="823" alt="Снимок экрана 2021-06-22 в 10 59 23" src="https://user-images.githubusercontent.com/2159081/122886833-f9fac780-d348-11eb-89b3-22769efff4e4.png">
Probably we are just lucky :) | 2021-06-22T08:00:29 | 2021-06-22T08:00:29 | {} | MEMBER |
ClickHouse | ClickHouse | 866,034,758 | 25,595 | nikitamikhaylov | @Mergifyio update | 2021-06-22T14:30:29 | 2021-06-22T14:30:29 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 866,227,088 | 25,529 | IgorLelyakov | спасибо, попробую | 2021-06-22T18:23:43 | 2021-06-22T18:23:43 | {} | NONE |
ClickHouse | ClickHouse | 866,281,171 | 24,479 | alexey-milovidov | No info provided. | 2021-06-22T19:43:22 | 2021-06-22T19:43:22 | {} | MEMBER |
ClickHouse | ClickHouse | 866,321,764 | 10,409 | alexey-milovidov | Duplicate of #21047. | 2021-06-22T20:46:31 | 2021-06-22T20:46:31 | {} | MEMBER |
ClickHouse | ClickHouse | 866,392,579 | 18,923 | alexey-milovidov | > Also, we need a cmd line options (with corresponding config settings?) to enforce watchdog installation even for TTY mode and disable it even for non-TTY mode.
There is. | 2021-06-22T23:03:16 | 2021-06-22T23:03:16 | {} | MEMBER |
ClickHouse | ClickHouse | 866,401,364 | 25,513 | azat | @alexey-milovidov should it be rebased or existing passed checks is enough? | 2021-06-22T23:23:51 | 2021-06-22T23:23:51 | {} | MEMBER |
ClickHouse | ClickHouse | 866,469,733 | 25,523 | feihengye | @filimonov
SETTINGS kafka_broker_list = '192.168.1.26:9092', kafka_topic_list = 'data_kv', kafka_group_name = 'clickhouse_new', kafka_format = 'JSONEachRow', kafka_num_consumers = 16, kafka_max_block_size = 1048576, kafka_row_delimiter = '\n', kafka_thread_per_consumer = 5, kafka_skip_broken_messages = 100000; | 2021-06-23T02:19:34 | 2021-06-23T02:19:34 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 865,639,213 | 25,551 | mergify[bot] | **Command `update`: success**
> **Branch has been successfully updated**
>
| 2021-06-22T06:35:57 | 2021-06-22T06:35:57 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 865,855,234 | 25,570 | qoega | Fuzzer ubsan #25533 | 2021-06-22T10:14:33 | 2021-06-22T10:14:33 | {} | MEMBER |
ClickHouse | ClickHouse | 865,914,860 | 25,549 | alesapin | Wow! Flaky check possibly reproduced the bug. I'll try with thread fuzzer. | 2021-06-22T11:48:24 | 2021-06-22T11:48:24 | {} | MEMBER |
ClickHouse | ClickHouse | 866,279,522 | 24,075 | alexey-milovidov | @l1t1 We need a reproducible example. Please provide the dataset to reproduce or a query that reproduces the issue without dataset. | 2021-06-22T19:40:51 | 2021-06-22T19:40:51 | {} | MEMBER |
ClickHouse | ClickHouse | 865,682,300 | 25,582 | azat | >What is the scenario? Why this feature is a candidate for backporting?
`system` database became `ATOMIC` by default (20.8+, #13680), so it will remove all existing data (since it is loaded with force - https://github.com/ClickHouse/ClickHouse/blob/d6bdde54d94a4eaf337c4e62842d6dd9b90f21ea/src/Interpreters/loadMetada... | 2021-06-22T07:47:34 | 2021-06-22T07:47:34 | {} | MEMBER |
ClickHouse | ClickHouse | 865,687,093 | 25,421 | alesapin | For some reason sometimes we have eth1 instead of eth0 in the fetches test. I'll just add it as a fallback.
`eth1: 41478485268 3009510 0 0 0 0 0 2445 188022557 2105260 0 0 0 0 0 ` | 2021-06-22T07:55:03 | 2021-06-22T07:55:03 | {} | MEMBER |
ClickHouse | ClickHouse | 865,704,738 | 25,579 | alexey-milovidov | `.sh` test will be nice to have. | 2021-06-22T08:20:03 | 2021-06-22T08:20:03 | {} | MEMBER |
ClickHouse | ClickHouse | 865,642,736 | 25,552 | vitlibar | @vdimir Please help to adjust the test `01049_join_low_card_bug_long` for `21.5`. | 2021-06-22T06:43:01 | 2021-06-22T08:36:20 | {
"eyes": 1
} | MEMBER |
ClickHouse | ClickHouse | 865,349,535 | 25,472 | vitlibar | Looks like the test `01240_join_get_or_null` has to be adjusted for `20.8`.
@vdimir Can you help with that? | 2021-06-21T21:16:05 | 2021-06-22T08:37:04 | {} | MEMBER |
ClickHouse | ClickHouse | 863,473,352 | 25,428 | alesapin | TODO:
- [x] Integration tests with some data.
- [x] Implement setACL while parsing data | 2021-06-17T18:38:56 | 2021-06-22T11:46:53 | {} | MEMBER |
ClickHouse | ClickHouse | 866,037,255 | 25,165 | nikitamikhaylov | @Mergifyio update | 2021-06-22T14:33:31 | 2021-06-22T14:33:31 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 866,040,670 | 25,583 | alesapin | Ok, ready. | 2021-06-22T14:37:33 | 2021-06-22T14:37:33 | {} | MEMBER |
ClickHouse | ClickHouse | 866,198,460 | 21,363 | abyss7 | @tavplubix from where does the flakiness come in? | 2021-06-22T17:47:46 | 2021-06-22T17:47:46 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 866,296,275 | 9,958 | alexey-milovidov | I confirm, the value of the `NetworkReceiveElapsedMicroseconds` does not include the time spent waiting data from the client. | 2021-06-22T20:05:59 | 2021-06-22T20:08:01 | {} | MEMBER |
ClickHouse | ClickHouse | 866,310,647 | 25,549 | alesapin | Is it my fix or the test show this error?
```
2021.06.22 20:24:21.778684 [ 647 ] {} <Error> test_kd1w0m.ttl_table4 (4d173b4f-7696-4528-8d17-3b4f7696e528): auto DB::StorageReplicatedMergeTree::processQueueEntry(ReplicatedMergeTreeQueue::SelectedEntryPtr)::(anonymous class)::operator()(
DB::StorageReplicatedMergeTree:... | 2021-06-22T20:28:04 | 2021-06-22T20:28:04 | {
"eyes": 1
} | MEMBER |
ClickHouse | ClickHouse | 866,344,206 | 25,549 | tavplubix | It can be related to background removal of empty parts | 2021-06-22T21:23:42 | 2021-06-22T21:23:42 | {} | MEMBER |
ClickHouse | ClickHouse | 866,346,838 | 15,271 | alexey-milovidov | > when current item equal last item, we should not access the RBM in memory or cache
Yes, it is possible and this optimization is easy to implement.
Just track the last value in `RoaringBitmapWithSmallSet` or make this optimization in `SmallSet`.
Compare with the last value on insertion. | 2021-06-22T21:28:32 | 2021-06-22T21:28:32 | {} | MEMBER |
ClickHouse | ClickHouse | 866,397,498 | 19,689 | alexey-milovidov | @vitlibar This is still relevant. | 2021-06-22T23:14:33 | 2021-06-22T23:14:33 | {} | MEMBER |
ClickHouse | ClickHouse | 865,646,709 | 25,536 | vitlibar | Depends on https://github.com/ClickHouse/ClickHouse/pull/23429, can't fix compilation. | 2021-06-22T06:50:43 | 2021-06-22T06:50:43 | {} | MEMBER |
ClickHouse | ClickHouse | 866,039,683 | 25,165 | mergify[bot] | **Command `update`: success**
> **Branch has been successfully updated**
>
| 2021-06-22T14:36:26 | 2021-06-22T14:36:26 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 866,058,491 | 21,866 | Avogar | Should be fixed in https://github.com/ClickHouse/ClickHouse/pull/24341 | 2021-06-22T14:58:26 | 2021-06-22T14:58:26 | {} | MEMBER |
ClickHouse | ClickHouse | 866,101,594 | 19,589 | vdimir | Ref https://github.com/ClickHouse/ClickHouse/issues/20870 | 2021-06-22T15:44:01 | 2021-06-22T15:44:01 | {} | MEMBER |
ClickHouse | ClickHouse | 866,106,567 | 25,549 | mergify[bot] | **Command `update`: success**
> **Branch has been successfully updated**
>
| 2021-06-22T15:49:36 | 2021-06-22T15:49:36 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 866,310,354 | 24,209 | alexey-milovidov | I cannot explain the difference between 2.1 and 2.2, it is strange indeed. | 2021-06-22T20:27:39 | 2021-06-22T20:27:39 | {} | MEMBER |
ClickHouse | ClickHouse | 866,412,299 | 25,579 | azat | >.sh test will be nice to have.
Ok, added (although this looks a little bit superfluous, since this is just a NULL dereference there, and affects only incorrect usage, and covering all such cases may be too much) | 2021-06-22T23:54:28 | 2021-06-22T23:54:28 | {} | MEMBER |
ClickHouse | ClickHouse | 860,284,849 | 7,778 | alexey-milovidov | I'm happy to say that `Set`, `Join` and `Buffer` tables have been instrumented and their memory usage info is available in `system.tables` table.
For other parts this task is too generic - not actionable.
| 2021-06-13T23:23:19 | 2021-06-13T23:23:34 | {} | MEMBER |
ClickHouse | ClickHouse | 860,287,330 | 21,748 | alexey-milovidov | So, the issue should be technically solvable by load balancer. | 2021-06-13T23:42:56 | 2021-06-13T23:42:56 | {} | MEMBER |
ClickHouse | ClickHouse | 860,288,094 | 9,932 | alexey-milovidov | Ok. Looks like it is no longer relevant. | 2021-06-13T23:48:35 | 2021-06-13T23:48:35 | {} | MEMBER |
ClickHouse | ClickHouse | 860,352,899 | 23,964 | alexey-milovidov | +1
Allowing to write
`clickhouse-client --max_threads` is confusing. | 2021-06-14T04:25:28 | 2021-06-14T04:25:28 | {} | MEMBER |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.