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 | 859,641,850 | 24,885 | vdimir | Let's merge master and trigger ci again | 2021-06-11T14:58:34 | 2021-06-11T14:58:34 | {} | MEMBER |
ClickHouse | ClickHouse | 859,947,199 | 25,146 | bnaecker | > > Is there any more detail or context for this error? I'm not sure which GitHub check (if any) that comes from.
>
> `ClickHouse special build check — 3/5 builds are OK`, `clang-11-aarch64`, `Build log`.
Thanks, I see it now. However, I'm not able to run the Docker build container as that action describes. Runni... | 2021-06-11T22:55:40 | 2021-06-11T22:55:40 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 858,284,724 | 24,075 | l1t1 | version 21.7.1.7113 still has this problem.
@filimonov | 2021-06-10T04:12:46 | 2021-06-10T10:18:40 | {} | NONE |
ClickHouse | ClickHouse | 858,632,539 | 25,139 | abyss7 | Can you please provide any data about your HTTP request? Especially the field values in your POST headers. | 2021-06-10T13:40:22 | 2021-06-10T13:40:22 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 858,691,170 | 25,139 | lopn | POST /?wait_end_of_query=1&database=default&user=xxxx&password=xxxxx HTTP/1.1
Host: 172.31.32.7:8123
Content-Length: 372086
Content-Type: multipart/form-data; boundary=----WebKitFormBoundary7MA4YWxkTrZu0gW
----WebKitFormBoundary7MA4YWxkTrZu0gW
Content-Disposition: form-data; name="query"
with ps as (SELECT
... | 2021-06-10T14:52:07 | 2021-06-10T14:54:14 | {} | NONE |
ClickHouse | ClickHouse | 858,753,569 | 25,126 | alesapin | @mergifyio update | 2021-06-10T16:09:43 | 2021-06-10T16:09:43 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 859,347,955 | 25,172 | kssenii | Failure `Yandex synchronization check (only for Yandex employees)` - not related to changes of this PR.
| 2021-06-11T07:27:26 | 2021-06-11T07:27:26 | {} | MEMBER |
ClickHouse | ClickHouse | 859,356,876 | 21,451 | mezhekov | I have a similar problem, the point is that replication does not work, it grows delay | 2021-06-11T07:42:56 | 2021-06-11T07:42:56 | {} | NONE |
ClickHouse | ClickHouse | 859,729,703 | 25,220 | filimonov | Also in context of https://github.com/ClickHouse/ClickHouse/issues/15552
Security scanner can came and set up the password :) | 2021-06-11T17:22:38 | 2021-06-11T17:22:38 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 859,779,593 | 25,149 | alexey-milovidov | Performance test has shown a slowdown. | 2021-06-11T18:59:21 | 2021-06-11T18:59:21 | {} | MEMBER |
ClickHouse | ClickHouse | 859,782,916 | 25,146 | alexey-milovidov | > Is there any more detail or context for this error? I'm not sure which GitHub check (if any) that comes from.
`ClickHouse special build check — 3/5 builds are OK`, `clang-11-aarch64`, `Build log`.
| 2021-06-11T19:04:56 | 2021-06-11T19:04:56 | {} | MEMBER |
ClickHouse | ClickHouse | 859,961,342 | 25,166 | alexey-milovidov | That's alright. | 2021-06-11T23:53:23 | 2021-06-11T23:53:23 | {} | MEMBER |
ClickHouse | ClickHouse | 858,487,635 | 17,623 | zouyonghao | This is a great document.
I'm interesting about the fuzzing jobs:
Fuzzing of cluster configurations
Fuzzing of ClickHouse versions for tests with distributed queries for compatibility
Could you introduce about them or provide some details?
Thanks | 2021-06-10T10:00:13 | 2021-06-10T10:00:13 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 858,732,998 | 25,163 | KochetovNicolai | In integration tests, it is a timeout to start cluster | 2021-06-10T15:44:19 | 2021-06-10T15:44:19 | {} | MEMBER |
ClickHouse | ClickHouse | 859,371,623 | 25,166 | tavplubix | ClickHouse special build check is relevant:
```
2021-06-10 14:37:35 /build/obj-x86_64-linux-gnu/../src/Storages/MergeTree/MergeTreeData.cpp:915:5: error: use range-based for loop instead [modernize-loop-convert,-warnings-as-errors]
2021-06-10 14:37:35 for (size_t i = 0; i < part_names_with_disks.size(); ++i)
20... | 2021-06-11T08:06:26 | 2021-06-11T08:06:26 | {} | MEMBER |
ClickHouse | ClickHouse | 859,641,944 | 24,885 | vdimir | @mergifyio update | 2021-06-11T14:58:41 | 2021-06-11T14:58:41 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 859,634,225 | 25,200 | tavplubix | Stress test (thread) - infinite recursion in `StorageMerge::getVirtuals` - #25225
Other stress tests passed. Let's run it once again to find new failures | 2021-06-11T14:47:21 | 2021-06-11T21:02:45 | {} | MEMBER |
ClickHouse | ClickHouse | 858,549,155 | 25,107 | alesapin | Unrelated failures. | 2021-06-10T11:41:30 | 2021-06-10T11:41:30 | {} | MEMBER |
ClickHouse | ClickHouse | 858,658,262 | 25,139 | lopn | https://github.com/ClickHouse/ClickHouse/blob/3e63b033e87d7b183a4024fcf8d4dadbeb471d4b/src/Server/HTTP/HTMLForm.cpp#L253
I want to know how to config the value_length_limit | 2021-06-10T14:12:47 | 2021-06-10T14:14:48 | {} | NONE |
ClickHouse | ClickHouse | 859,535,860 | 25,069 | Jokser | S3 object metadata is just a `map<string, string>`. We can add any new field to it. | 2021-06-11T12:09:12 | 2021-06-11T12:09:12 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 859,967,094 | 24,870 | alexey-milovidov | @vdimir You may found it interesting for review. | 2021-06-12T00:20:25 | 2021-06-12T00:20:25 | {
"eyes": 1
} | MEMBER |
ClickHouse | ClickHouse | 858,692,567 | 25,139 | lopn | POST /?wait_end_of_query=1&database=default&user=xxxx&password=xxxx HTTP/1.1
Host: 172.31.32.7:8123
Content-Length: 185809
Content-Type: multipart/form-data; boundary=----WebKitFormBoundary7MA4YWxkTrZu0gW
----WebKitFormBoundary7MA4YWxkTrZu0gW
Content-Disposition: form-data; name="query"
with ps as (SELECT
r... | 2021-06-10T14:53:45 | 2021-06-10T14:54:58 | {} | NONE |
ClickHouse | ClickHouse | 858,842,363 | 25,131 | volfco | Setting `<query_plan_max_optimizations_to_apply>0</query_plan_max_optimizations_to_apply>` still causes mutations to back up, take a lot longer then I'd expect, and consume all system resources. But they are progressing slowly.
I think the idea of not trying to do optimizations for mutations makes sense. The documen... | 2021-06-10T17:55:52 | 2021-06-10T17:55:52 | {} | NONE |
ClickHouse | ClickHouse | 859,360,265 | 21,394 | kasnet | How could I solve this problem? | 2021-06-11T07:48:37 | 2021-06-11T07:48:37 | {} | NONE |
ClickHouse | ClickHouse | 859,503,043 | 25,049 | alesapin | Failures unrelated to changes. | 2021-06-11T11:05:23 | 2021-06-11T11:05:23 | {} | MEMBER |
ClickHouse | ClickHouse | 859,643,424 | 24,885 | mergify[bot] | **Command `update`: success**
> **Branch has been successfully updated**
>
| 2021-06-11T15:00:44 | 2021-06-11T15:00:44 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 859,697,654 | 25,210 | tavplubix | Functional stateless tests (release, DatabaseReplicated) — fail: 0, passed: 3000, skipped: 109
Exec. time 1h:21m
Let's restart the task | 2021-06-11T16:24:25 | 2021-06-11T16:40:04 | {} | MEMBER |
ClickHouse | ClickHouse | 858,682,224 | 25,139 | abyss7 | If you are sure that the problem is exacly with this limitation, then I will just add a server setting for it. | 2021-06-10T14:41:42 | 2021-06-10T14:42:08 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 858,709,570 | 25,139 | lopn | @abyss7 | 2021-06-10T15:14:19 | 2021-06-10T15:14:19 | {} | NONE |
ClickHouse | ClickHouse | 859,197,887 | 25,146 | alexey-milovidov | PS. Just writing `const char *` is fine. | 2021-06-11T01:39:43 | 2021-06-11T01:39:43 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 859,341,912 | 25,172 | meoww-bot | @alexey-milovidov Hi, can you give me the details of Yandex synchronization check and Yandex third-party checks? I don't know why these checks fail. Thanks. | 2021-06-11T07:18:02 | 2021-06-11T07:18:02 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 859,395,321 | 23,444 | vpanfilov | Seems to be connected with https://github.com/ClickHouse/ClickHouse/issues/21992 | 2021-06-11T08:44:52 | 2021-06-11T08:44:52 | {} | NONE |
ClickHouse | ClickHouse | 859,563,201 | 25,202 | KochetovNicolai | Version 20.12.3.3 is not supported already. Could you try 21.3 at least? | 2021-06-11T12:57:57 | 2021-06-11T12:57:57 | {} | MEMBER |
ClickHouse | ClickHouse | 859,960,799 | 20,495 | mergify[bot] | **Command `update`: success**
> **Branch has been successfully updated**
>
| 2021-06-11T23:50:48 | 2021-06-11T23:50:48 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 859,962,166 | 25,166 | alexey-milovidov | Integration test has failed:
> test_backward_compatibility/test_detach_part_wrong_partition_id.py::test_detach_part_wrong_partition_id | 2021-06-11T23:57:23 | 2021-06-11T23:57:23 | {} | MEMBER |
ClickHouse | ClickHouse | 859,966,759 | 24,875 | alexey-milovidov | Docker image was not rebuilt, so we cannot see changes. Let's try in master. | 2021-06-12T00:18:24 | 2021-06-12T00:18:24 | {} | MEMBER |
ClickHouse | ClickHouse | 858,647,966 | 25,169 | nikitamikhaylov | Need to check the same scenario with TSVWithNames format. | 2021-06-10T14:00:06 | 2021-06-10T14:00:06 | {} | MEMBER |
ClickHouse | ClickHouse | 858,660,269 | 20,243 | roman-vynar | @tavplubix Thanks a ton for your comment.
This DROP NO DELAY|SYNC or database_atomic_wait_for_drop_and_detach_synchronously is undocumented or poorly documented. | 2021-06-10T14:15:20 | 2021-06-10T14:15:20 | {
"+1": 9
} | NONE |
ClickHouse | ClickHouse | 859,607,821 | 24,573 | gyuton | Internal documentation ticket: DOCSUP-10285. | 2021-06-11T14:07:40 | 2021-06-11T14:07:40 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 859,618,743 | 25,171 | alexey-milovidov | @bharatnc It will be wonderful! | 2021-06-11T14:24:22 | 2021-06-11T14:24:22 | {} | MEMBER |
ClickHouse | ClickHouse | 859,657,639 | 25,218 | MaxWk | the second commit fix the issue:https://github.com/ClickHouse/ClickHouse/issues/25219 | 2021-06-11T15:23:22 | 2021-06-11T15:23:22 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 859,968,120 | 24,920 | alexey-milovidov | @mergifyio update | 2021-06-12T00:26:22 | 2021-06-12T00:26:22 | {} | MEMBER |
ClickHouse | ClickHouse | 858,561,489 | 24,041 | l1t1 | it returns a value with type float64 and overflow when running
select arrayProduct(range(1,22)) a, toTypeName(a)
14197454024290337000 | Float64
-- | --
the right answer should be
51090942171709440000 | 2021-06-10T12:02:53 | 2021-06-10T12:02:53 | {} | NONE |
ClickHouse | ClickHouse | 858,631,388 | 10,368 | nvartolomei | Related #25168 (posting here to have cross-references between these related issues) | 2021-06-10T13:38:49 | 2021-06-10T13:38:49 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 859,349,466 | 25,079 | alesapin | Ok, let's merge and check in master. | 2021-06-11T07:30:10 | 2021-06-11T07:30:10 | {} | MEMBER |
ClickHouse | ClickHouse | 859,534,619 | 24,885 | Jokser | @vdimir Is there any CI issues? Stateless tests are timed-out too fast. Can we re-trigger CI? | 2021-06-11T12:07:08 | 2021-06-11T12:07:08 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 859,543,650 | 23,311 | mergify[bot] | **Command `update`: success**
> **Branch has been successfully updated**
>
| 2021-06-11T12:23:10 | 2021-06-11T12:23:10 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 859,959,169 | 24,822 | gyuton | Internal documentation ticket: DOCSUP-10286. | 2021-06-11T23:42:55 | 2021-06-11T23:42:55 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 858,476,334 | 25,161 | KochetovNicolai | Looks fine.
But looks like we heed a test. | 2021-06-10T09:44:28 | 2021-06-10T09:44:28 | {} | MEMBER |
ClickHouse | ClickHouse | 858,598,887 | 24,041 | den-crane | right answer 1.1240007277776077e21
https://www.wolframalpha.com/input/?i=factorial+22
1124000727777607680000
| 2021-06-10T12:57:29 | 2021-06-10T12:57:29 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 858,714,753 | 25,130 | gyuton | @akuzm Hi! Can you review this, please? | 2021-06-10T15:20:48 | 2021-06-10T15:20:48 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 858,787,124 | 25,168 | nvartolomei | Yes, it was the source replica which got lost. I think I know how to write
a test for this and other similar cases.
Create 2 replicas, make an insert then network partition the replicas. Make
1 more insert and run an optimize final.
A: Stop the replica, delete the part created by second insert and the
merged part, st... | 2021-06-10T16:58:17 | 2021-06-10T16:58:17 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 859,197,693 | 25,146 | alexey-milovidov | clang-tidy:
```
2021-06-09 23:42:23 /build/obj-x86_64-linux-gnu/../base/common/getResource.cpp:36:5: error: 'auto sym_start' can be declared as 'const auto *sym_start' [readability-qualified-auto,-warnings-as-errors]
2021-06-09 23:42:23 auto sym_start = reinterpret_cast<const char*>(dlsym(RTLD_DEFAULT, symbol_na... | 2021-06-11T01:39:03 | 2021-06-11T01:39:03 | {} | MEMBER |
ClickHouse | ClickHouse | 859,316,346 | 23,780 | SaintBacchus | @amosbird has any design for this? | 2021-06-11T06:36:00 | 2021-06-11T06:36:00 | {} | NONE |
ClickHouse | ClickHouse | 859,953,898 | 25,146 | alexey-milovidov | > `docker run --network=host --volume=/place/sandbox-data/ramdrive/991451814/ClickHouse/build_output_folder:/output --volume=/place/sandbox-data/ramdrive/991451814/ClickHouse/docker/packager/../..:/build --volume=/place/sandbox-data/ramdrive/991451814/local_ccache:/ccache -e CC=clang-11 -e CXX=clang++-11 -e CCACHE_DIR... | 2021-06-11T23:20:54 | 2021-06-11T23:20:54 | {} | MEMBER |
ClickHouse | ClickHouse | 859,960,559 | 20,495 | alexey-milovidov | @mergifyio update | 2021-06-11T23:49:36 | 2021-06-11T23:49:36 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 859,152,420 | 25,187 | alexey-milovidov | 21.6 and 21.7 cannot execute the following query on GitHub dataset:
```
SELECT
actor_login IN ('alexey-milovidov', 'alesapin', 'blinkov', 'KochetovNicolai', 'akuzm', 'abyss7', '4ertus2', 'vitlibar', 'CurtizJ', 'tavplubix', 'nikitamikhaylov', 'qoega', 'Akazz', 'Jokser', 'millb', 'stavrolia', 'zlobober', 'prolle... | 2021-06-10T23:29:02 | 2021-06-10T23:29:02 | {} | MEMBER |
ClickHouse | ClickHouse | 859,543,495 | 23,311 | Avogar | @Mergifyio update | 2021-06-11T12:22:54 | 2021-06-11T12:22:54 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 859,960,313 | 25,212 | alexey-milovidov | Only `test_cluster_copier` remains. | 2021-06-11T23:48:29 | 2021-06-11T23:48:29 | {} | MEMBER |
ClickHouse | ClickHouse | 858,631,274 | 24,041 | l1t1 | range(1,22) is an array from 1 to 21 | 2021-06-10T13:38:38 | 2021-06-10T13:38:38 | {} | NONE |
ClickHouse | ClickHouse | 858,633,786 | 25,167 | den-crane | ```
21.3.2.5
┌───────date─┬──────────────url_id─┬─event_type─┬─count─┐
│ 1970-01-01 │ 0 │ 0 │ 69 │
│ 1970-01-01 │ 0 │ 10 │ 370 │
│ 1970-01-01 │ 100825704 │ 0 │ 13 │
│ 1970-01-01 │ 1225804048 │ 10 │ 135 │
│ 1971-1... | 2021-06-10T13:42:01 | 2021-06-10T13:42:01 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 858,866,094 | 10,892 | alexey-milovidov | @yingshikong186 What issues do you have with this version? | 2021-06-10T18:19:16 | 2021-06-10T18:19:16 | {} | MEMBER |
ClickHouse | ClickHouse | 859,018,562 | 10,892 | den-crane | @yingshikong186 `alter table drop column` can be blocked for hours by a running merge. | 2021-06-10T20:18:49 | 2021-06-10T20:18:49 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 859,075,364 | 10,892 | alexey-milovidov | @den-crane It always works instantly in recent ClickHouse releases, no locking is performed. | 2021-06-10T21:15:21 | 2021-06-10T21:15:40 | {} | MEMBER |
ClickHouse | ClickHouse | 859,186,984 | 24,467 | zhicwu | Link to #25047. | 2021-06-11T01:07:24 | 2021-06-11T01:07:24 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 859,279,346 | 25,171 | bharatnc | I can take this issue if no one is already working on this ? | 2021-06-11T05:31:41 | 2021-06-11T05:31:41 | {
"heart": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 859,620,883 | 25,146 | bnaecker | > clang-tidy:
>
> ```
> 2021-06-09 23:42:23 /build/obj-x86_64-linux-gnu/../base/common/getResource.cpp:36:5: error: 'auto sym_start' can be declared as 'const auto *sym_start' [readability-qualified-auto,-warnings-as-errors]
> 2021-06-09 23:42:23 auto sym_start = reinterpret_cast<const char*>(dlsym(RTLD_DEFAUL... | 2021-06-11T14:27:34 | 2021-06-11T14:27:34 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 859,734,835 | 25,195 | filimonov | Some types of inactive parts are almost useless (_tmp), automatic cleaning for them after some retention period should be good.
Users almost never check them and sometimes a lot of disk space is wasted by them.
Extra thoughts:
May be we can add a warning at server start about the size occupied by detached par... | 2021-06-11T17:32:43 | 2021-06-11T17:32:43 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 859,909,894 | 25,200 | tavplubix | All stress tests passed | 2021-06-11T21:03:36 | 2021-06-11T21:03:36 | {} | MEMBER |
ClickHouse | ClickHouse | 858,588,899 | 25,159 | den-crane | Mention that it works slower than sum and has sense only for FLOAT types. Thanks. | 2021-06-10T12:43:20 | 2021-06-10T12:43:20 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 858,883,541 | 24,130 | AnaUvarova | > Файл docs/clean случайно остался?
Да) | 2021-06-10T18:34:04 | 2021-06-10T18:34:04 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 859,367,121 | 25,166 | tavplubix | Automatic cleanup may be useful after this PR and #15701: #25195 | 2021-06-11T07:59:39 | 2021-06-11T07:59:39 | {} | MEMBER |
ClickHouse | ClickHouse | 859,914,042 | 25,200 | tavplubix | Functional stateless tests (ANTLR debug) - `Logical error: 'Can't find queue-0000000633 in the memory queue. It is a bug. Entry: format version: 4` -
It's related to intersecting parts and merge cancellation.
Fwd from Telegram chat:
```
Alexander Sapin, [08.06.21 18:30]
https://clickhouse-test-reports.s3.yandex... | 2021-06-11T21:13:46 | 2021-06-11T21:13:46 | {} | MEMBER |
ClickHouse | ClickHouse | 859,953,520 | 25,146 | alexey-milovidov | > As an aside, it seems this error arises because we're cross compiling. Presumably this is being build on an x86 system, but we're generating code for ARM. Is that correct?
Yes, that's correct.
> I'm just trying to understand how the target architecture is being set by the build scripts.
It is being set in `c... | 2021-06-11T23:19:18 | 2021-06-11T23:19:18 | {} | MEMBER |
ClickHouse | ClickHouse | 859,153,569 | 25,028 | alexey-milovidov | Reverted to 21.5 due to bug https://github.com/ClickHouse/ClickHouse/issues/25187 | 2021-06-10T23:32:20 | 2021-06-10T23:32:28 | {} | MEMBER |
ClickHouse | ClickHouse | 859,625,902 | 25,146 | bnaecker | > AArch64 build:
>
> > Jun 09 23:42:47 ld.lld: error: /build/build_docker/contrib/cctz-cmake/libtzdata.a(Abidjan.o) is incompatible with elf64-littleaarch64
Is there any more detail or context for this error? I'm not sure which GitHub check (if any) that comes from. | 2021-06-11T14:35:06 | 2021-06-11T14:35:06 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 858,620,951 | 24,883 | mjh568412187 | > That is related to Atomic database behaviour.
>
> See https://kb.altinity.com/engines/altinity-kb-atomic-database-engine#q-i-cannot-reuse-zookeeper-path-after-dropping-the-table
Thank you | 2021-06-10T13:25:13 | 2021-06-10T13:25:13 | {} | NONE |
ClickHouse | ClickHouse | 858,757,758 | 25,126 | mergify[bot] | **Command `update`: success**
> **Branch has been successfully updated**
>
| 2021-06-10T16:15:44 | 2021-06-10T16:15:44 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 858,763,731 | 25,131 | volfco | @KochetovNicolai Making that config change did the trick on my testing dataset.
The mutations I used in my test database were all single row updates (`ALTER TABLE media UPDATE ... WHERE uuid = ...`). That might be a problem for the optimizer if it's trying to group multiple single row operations into one larger opt... | 2021-06-10T16:23:07 | 2021-06-10T16:23:07 | {} | NONE |
ClickHouse | ClickHouse | 858,775,980 | 25,168 | tavplubix | Replication queue deadlock (which is actually livelock) may happen due to #10368, because it's usually impossible to cancel scheduled merge even if some source part is lost. But there is another way to deal with such merges, see discussion in #24777.
> The problem is caused by the fact that source replica crashes (... | 2021-06-10T16:41:35 | 2021-06-10T16:41:35 | {} | MEMBER |
ClickHouse | ClickHouse | 859,388,262 | 25,166 | KochetovNicolai | @alexey-milovidov , this pr disable removing of parts at start completely.
There was some complex logic - (broken parts with level = 0) or (broken parts which cover at leas two existing parts) are were removed.
We have decided that we could always move parts to detached. | 2021-06-11T08:33:13 | 2021-06-11T08:33:13 | {} | MEMBER |
ClickHouse | ClickHouse | 859,910,800 | 25,210 | tavplubix | Functional stateless tests (release, DatabaseReplicated) — fail: 0, passed: 3000, skipped: 109
Exec. time 1h:16m | 2021-06-11T21:05:41 | 2021-06-11T21:05:41 | {} | MEMBER |
ClickHouse | ClickHouse | 859,964,107 | 25,101 | alexey-milovidov | @mergifyio update | 2021-06-12T00:05:40 | 2021-06-12T00:05:40 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 858,658,814 | 25,167 | adiletkabylbekov | yes, it's okay on 21.3.12.2! thank you!
but what was the issue with 21.3.2.5, is it related to datatypes? | 2021-06-10T14:13:28 | 2021-06-10T14:13:28 | {} | NONE |
ClickHouse | ClickHouse | 858,691,107 | 25,167 | den-crane | duplicate of https://github.com/ClickHouse/ClickHouse/issues/21773
fixed https://github.com/ClickHouse/ClickHouse/pull/21841
fixed in 21.3.3.14 | 2021-06-10T14:52:02 | 2021-06-10T14:52:23 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 858,696,159 | 25,139 | lopn | DFL_MAX_VALUE_LENGTH = 256 * 1024
https://github.com/ClickHouse/ClickHouse/blob/3e63b033e87d7b183a4024fcf8d4dadbeb471d4b/src/Server/HTTP/HTMLForm.h#L132
this default is 256KB， how can i config it | 2021-06-10T14:58:04 | 2021-06-10T15:05:47 | {} | NONE |
ClickHouse | ClickHouse | 859,197,498 | 25,146 | alexey-milovidov | AArch64 build:
> Jun 09 23:42:47 ld.lld: error: /build/build_docker/contrib/cctz-cmake/libtzdata.a(Abidjan.o) is incompatible with elf64-littleaarch64 | 2021-06-11T01:38:28 | 2021-06-11T01:38:28 | {} | MEMBER |
ClickHouse | ClickHouse | 859,200,360 | 25,158 | yanye666 | Even if ‘detail_id’ is not an empty string, but the same, there will be this problem | 2021-06-11T01:47:10 | 2021-06-11T01:47:10 | {} | NONE |
ClickHouse | ClickHouse | 858,759,887 | 25,172 | meoww-bot | Should I remove the machine_translated label? | 2021-06-10T16:17:29 | 2021-06-10T16:17:29 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 858,812,482 | 25,168 | tavplubix | Most likely it will not cover complicated corner-cases and distributed race conditions like this: https://github.com/ClickHouse/ClickHouse/pull/24777#issuecomment-851525932
Such test will check only the simplest case. | 2021-06-10T17:26:25 | 2021-06-10T17:26:25 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 858,861,376 | 25,172 | alexey-milovidov | > Should I remove the machine_translated label?
Yes. | 2021-06-10T18:14:35 | 2021-06-10T18:14:35 | {} | MEMBER |
ClickHouse | ClickHouse | 859,153,787 | 25,187 | alexey-milovidov | 21.5 is Ok. | 2021-06-10T23:32:57 | 2021-06-10T23:32:57 | {} | MEMBER |
ClickHouse | ClickHouse | 859,558,023 | 25,158 | KochetovNicolai | Hi!
It is expected behaviour (or implementation detail, if you wish). It would be very inexpensive to merge all the rows in case you have, like 1 billion same rows with +1 sign, and then 1 billion same rows with -1 sign. Implementation uses a window, it's size about 8192 (minus 3, probably). Only rows are in the same ... | 2021-06-11T12:49:12 | 2021-06-11T12:49:12 | {} | MEMBER |
ClickHouse | ClickHouse | 859,608,997 | 25,158 | KochetovNicolai | How does this window work?
Let's N = 8189
We keep N consecutive rows in memory. We can collapse only those rows. When we collapse any two rows, we read two more new. When there are no rows which can be collapsed, we write first row in window and read another one.
Implementation is here:
https://github.com/ClickH... | 2021-06-11T14:09:21 | 2021-06-11T14:09:21 | {} | MEMBER |
ClickHouse | ClickHouse | 859,728,536 | 25,220 | filimonov | Maybe it should require some grant? | 2021-06-11T17:20:16 | 2021-06-11T17:20:26 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 859,774,185 | 25,224 | alexey-milovidov | #25187 | 2021-06-11T18:48:35 | 2021-06-11T18:48:35 | {} | MEMBER |
ClickHouse | ClickHouse | 846,646,266 | 20,202 | alexey-milovidov | @KochetovNicolai There is no comment about `MergeTreeDataSelectCache`.
You've missed the case when PVS-Studio argued about uncomprehensible code:
> PVS check — Have 7 new errors, 21 total errors
https://clickhouse-test-reports.s3.yandex.net/20202/817bc1377c131e83d13bbd90260f8f2d97856bfd/pvs_studio_report/pvs-st... | 2021-05-23T23:53:35 | 2021-05-23T23:53:35 | {} | MEMBER |
ClickHouse | ClickHouse | 846,976,905 | 21,701 | theonedemon | @kssenii any news ? | 2021-05-24T11:23:08 | 2021-05-24T11:23:08 | {} | NONE |
ClickHouse | ClickHouse | 847,080,871 | 23,902 | vzakaznikov | These issues seem to be regression. On the latest stable 21.5.5.12 it works as expected. | 2021-05-24T14:28:22 | 2021-05-24T14:28:22 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 847,292,048 | 24,457 | mergify[bot] | **Command `update`: success**
> **Branch has been successfully updated**
>
| 2021-05-24T19:49:27 | 2021-05-24T19:49:27 | {} | CONTRIBUTOR |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.