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 | 787,872,834 | 21,184 | superceix | > It's because ON CLUSTER queries return table with query execution status. Seems like mysql client does not expect that DDL query may return some data.
Will you support this in the future? | 2021-03-01T11:22:08 | 2021-03-01T11:22:08 | {} | NONE |
ClickHouse | ClickHouse | 787,927,406 | 20,182 | filimonov | Also, there are still few unfixed regressions: https://github.com/ClickHouse/ClickHouse/issues?q=is%3Aissue+label%3Aprio-major+is%3Aopen
It would be cool to clean the list before 21.3. | 2021-03-01T12:54:30 | 2021-03-01T12:54:30 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 787,080,299 | 21,220 | Slach | @alexey-milovidov
I also run
`docs/tools/build.py --lang ru --verbose`
and doesn't see any "nowhere" errors inside my branch | 2021-02-27T14:23:01 | 2021-02-27T14:23:01 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 787,106,299 | 21,268 | olgarev | > FYI there is another PR with PostgreSQL doc #21220. I believe it should be merged first. May be after you will just correct and combine the two docs together? (There is no doc for database engine in that pr, which is added here, and other corrections and usage examples). May be you can help figure out what is wrong w... | 2021-02-27T17:26:32 | 2021-02-27T20:50:29 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 787,190,681 | 21,292 | alexey-milovidov | Advantages:
It no longer requires Hyperscan.
Performance is more than 10 times better: 640 MB/sec per core (18 GB/sec on 40 cores) vs. 50 MB/sec per core.
The amount of code is less: 308 vs 582 lines.
I've also added more tests and documentation. | 2021-02-27T22:04:05 | 2021-02-28T01:36:46 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 787,106,843 | 21,220 | kssenii | @Slach, look https://github.com/ClickHouse/ClickHouse/pull/21268#issuecomment-787106299 or here https://github.com/ClickHouse/ClickHouse/commit/01ef06a42c4c213a6ec0d0c9188ab10e5a2f07b9 | 2021-02-27T17:30:22 | 2021-02-28T07:22:11 | {} | MEMBER |
ClickHouse | ClickHouse | 787,435,431 | 21,310 | amosbird | It works in 21+ | 2021-02-28T11:14:09 | 2021-02-28T11:14:09 | {
"+1": 1
} | COLLABORATOR |
ClickHouse | ClickHouse | 787,538,207 | 21,318 | alexey-milovidov | If test cannot run in parallel with itself, it should be added to skip_list.json in "parallel" section. | 2021-02-28T22:43:59 | 2021-02-28T22:44:08 | {} | MEMBER |
ClickHouse | ClickHouse | 787,542,257 | 8,873 | alexey-milovidov | Performance of array operations was improved.
Performance of GROUP BY multiple fields, with LowCardinality in one of them is still slower than Enum, because the dictionary is dynamic and can be different in different blocks of data. | 2021-02-28T23:10:27 | 2021-02-28T23:10:33 | {} | MEMBER |
ClickHouse | ClickHouse | 787,551,033 | 18,583 | alexey-milovidov | We may want to add
`.cfi_startproc`
`.cfi_endproc`
directives, so the function will be visible in profiler.
| 2021-03-01T00:03:17 | 2021-03-01T00:03:17 | {} | MEMBER |
ClickHouse | ClickHouse | 787,591,243 | 21,182 | ikarishinjieva | @kitaisreal @ucasFL Thank you for advise, functional test is added | 2021-03-01T02:26:26 | 2021-03-01T02:26:26 | {} | NONE |
ClickHouse | ClickHouse | 786,549,576 | 20,347 | nikitamikhaylov | > @nikitamikhaylov Hi, is everything OK here? Anything left? The PR was not merged.
I was a bit confused about the translation of Kafka settings and the presence of some settings at all. So, there will be a bit of extra work from my side.
UPD: Will continue after merge. | 2021-02-26T10:10:53 | 2021-03-01T13:49:02 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 788,034,027 | 21,350 | den-crane | Clickhouse version?
Excerpts from /var/log/clickhouse-server/clickhouse-server.log for that period of time? | 2021-03-01T15:26:22 | 2021-03-01T15:28:39 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 787,196,718 | 18,842 | alexey-milovidov | Example: find all websites about ClickHouse and highlight results in their text content:
```
SELECT replaceRegexpAll(extractTextFromHTML(content) AS text, '(?i)(clickhouse)', '\x1b[1;31m\1\x1b[0m')
FROM minicrawl WHERE text ILIKE '%clickhouse%'
``` | 2021-02-27T22:23:00 | 2021-02-27T22:23:28 | {} | MEMBER |
ClickHouse | ClickHouse | 787,400,476 | 21,302 | keen-wolf | Sorry for update so late. It's a long new year holiday~
Three situations for File-like related methods have been carefully analyzed here: the File-Engine for Creating Table, the TableFunction.File and the TableFunction.Input. The first two have been tested to be improved by this feature, It has no effects on the las... | 2021-02-28T05:46:31 | 2021-02-28T05:57:34 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 787,543,428 | 5,751 | alexey-milovidov | The issue has been fixed by yourself, I see your comment in code from 28th June 2019. | 2021-02-28T23:16:18 | 2021-02-28T23:16:18 | {} | MEMBER |
ClickHouse | ClickHouse | 787,564,076 | 18,583 | alexey-milovidov | Do I understand correctly that the asm code assumes specific kind of relocations: `kCpuids(%rip)` and have to be built or linked with specific options? | 2021-03-01T00:46:26 | 2021-03-01T00:47:55 | {} | MEMBER |
ClickHouse | ClickHouse | 787,648,154 | 21,325 | azat | @excitoon by the way, is it possible to write a test to reproduce the problem that you are trying to fix? | 2021-03-01T05:25:08 | 2021-03-01T05:25:08 | {} | MEMBER |
ClickHouse | ClickHouse | 787,869,795 | 21,330 | taiyang-li | I got what you mean, thanks | 2021-03-01T11:16:59 | 2021-03-01T11:16:59 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 788,044,024 | 21,350 | joein | @den-crane Clickhouse version is 20.11.4.13 and I can send you text of exceptions (or part of the log file) in 2 hours | 2021-03-01T15:39:59 | 2021-03-01T15:42:05 | {} | NONE |
ClickHouse | ClickHouse | 788,046,496 | 21,340 | filimonov | It may be caused by merges currently running.
Check `metrics.DiskSpaceReservedForMerge` | 2021-03-01T15:42:56 | 2021-03-01T15:42:56 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 787,074,073 | 20,991 | alexey-milovidov | > AST fuzzer (debug) — Logical error: 'Logical error: no information about file statusMap%2Egoal_id.size0 in StorageLog'.
An issue with nested columns, @CurtizJ.
Fixed here: https://github.com/ClickHouse/ClickHouse/pull/21208 | 2021-02-27T13:38:29 | 2021-02-27T13:40:52 | {} | MEMBER |
ClickHouse | ClickHouse | 787,198,120 | 21,279 | alexey-milovidov | > Stress test (thread) — Killed by signal (in clickhouse-server.log)
FYI @azat | 2021-02-27T22:33:17 | 2021-02-27T22:33:17 | {} | MEMBER |
ClickHouse | ClickHouse | 787,505,535 | 21,079 | azat | >Stress test (thread) — Killed by signal (in clickhouse-server.log)
#21305 (`ZstdDeflatingWriteBuffer`)
>Integration tests (release) — fail: 0, passed: 460, error: 26
```
E ERROR: for hdfs1 Cannot start service hdfs1: driver failed programming external connectivity on endpoint rootteststoragehdfs_h... | 2021-02-28T19:12:16 | 2021-02-28T19:12:16 | {} | MEMBER |
ClickHouse | ClickHouse | 787,541,841 | 15,455 | alexey-milovidov | Performance get much better (20+ GB/sec) if setting `min_bytes_to_use_mmap_io = 1`.
Although this setting is not good in general for production. | 2021-02-28T23:07:48 | 2021-02-28T23:07:48 | {} | MEMBER |
ClickHouse | ClickHouse | 787,854,293 | 21,189 | filimonov | @OmarBazaraa why not just sending it as PR?
Fix looks ok, and follows the Poco recommended/documented way of doing it: "ssl: If ssl=true is specified, a custom SocketFactory subclass creating a SecureStreamSocket must be supplied."
See https://pocoproject.org/docs/Poco.MongoDB.Connection.html#21940 | 2021-03-01T10:52:49 | 2021-03-01T10:52:49 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 787,929,624 | 21,167 | filimonov | will be released in 21.2.5 and 21.1.6 | 2021-03-01T12:58:23 | 2021-03-01T12:58:23 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 787,949,381 | 21,332 | den-crane | Database Atomic?
https://github.com/ClickHouse/ClickHouse/issues/12135 | 2021-03-01T13:29:47 | 2021-03-01T13:29:47 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 787,973,129 | 21,332 | den-crane | yes, right | 2021-03-01T14:04:04 | 2021-03-01T14:04:04 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 787,207,346 | 18,012 | alexey-milovidov | @simPod I cannot reproduce the issue in master:
```
milovidov-desktop :) USE simpod
USE simpod
Query id: 1da63534-dea0-4776-b8b6-03f4de1283f7
Ok.
0 rows in set. Elapsed: 0.001 sec.
milovidov-desktop :) CREATE TABLE t_a ( `fetched_at` DateTime, `a` UUID, `b` UUID, `in` UInt64, `ou... | 2021-02-28T00:06:12 | 2021-02-28T00:06:12 | {} | MEMBER |
ClickHouse | ClickHouse | 787,489,776 | 21,312 | vdimir | Checked that CI with merged master passed. I'm closing this and merge https://github.com/ClickHouse/ClickHouse/pull/19591 | 2021-02-28T17:41:00 | 2021-02-28T17:41:00 | {} | MEMBER |
ClickHouse | ClickHouse | 787,522,139 | 21,320 | ilejn | This is Work In Progress, not ready to merge, too early to run tests.
Implements #17612
Current state
* plenty of debug messages/comment/other nasty things, it is too early to get rid of them, sorry.
* stateless tests are not OK after merge with recent master (LowCardinality things are problematic), worked befor... | 2021-02-28T21:00:11 | 2021-02-28T21:00:11 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 787,642,667 | 21,325 | excitoon | @azat Can you please check if I understood your changes here https://github.com/ClickHouse/ClickHouse/pull/19451/files#diff-bf34afb97d367a2be24088c4256e7bf39a62e9cb30d2b5ac2ff50148fead2e4f correctly.
| 2021-03-01T05:13:57 | 2021-03-01T05:13:57 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 787,825,463 | 21,330 | taiyang-li | OK. I will add this feature to `accurateCastOrNull ` | 2021-03-01T10:06:39 | 2021-03-01T10:06:39 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 787,934,488 | 20,876 | ucasfl | > Related to #21170
Both of them will be fixed in #21246.
| 2021-03-01T13:05:54 | 2021-03-01T13:05:54 | {
"+1": 1
} | COLLABORATOR |
ClickHouse | ClickHouse | 787,502,068 | 20,149 | alexey-milovidov | Ok.
Btw,
> ClickHouse build check Pending — 0/16 builds finished
This check has hit GitHub "unicorn" page. | 2021-02-28T18:53:47 | 2021-02-28T18:53:47 | {
"laugh": 1
} | MEMBER |
ClickHouse | ClickHouse | 787,539,891 | 21,177 | alexey-milovidov | Just remove `-fno-pie` and `-Wl,-no-pie`. | 2021-02-28T22:55:19 | 2021-02-28T22:55:19 | {} | MEMBER |
ClickHouse | ClickHouse | 787,741,430 | 21,167 | KochetovNicolai | That's suspicious.
Backported commit: https://github.com/ClickHouse/ClickHouse/pull/19912/commits/09c42f1d4d46ffedb4b7e23304c3e860e5712b5f
Addr2line:
`./obj-x86_64-linux-gnu/../src/Client/Connection.cpp:753
ext::basic_scope_guard<DB::Connection::receivePacket(std::__1::function<void (Poco::Net::Socket&)>)::$_3>::i... | 2021-03-01T08:01:42 | 2021-03-01T08:01:42 | {} | MEMBER |
ClickHouse | ClickHouse | 788,031,229 | 21,334 | alesapin | race :( | 2021-03-01T15:22:32 | 2021-03-01T15:22:32 | {} | MEMBER |
ClickHouse | ClickHouse | 788,058,643 | 18,695 | filimonov | > It's non-trivial to manipulate partitions for insertion, however it's possible to modify existing partitions.
Yep, that idea sounds very similar to #13826.
So for example you first insert data into the minute partition.
When the minute ends minute partition gets attached/mutated to an hourly partition.
Wh... | 2021-03-01T15:54:53 | 2021-03-01T15:54:53 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 787,208,994 | 19,229 | alexey-milovidov | @chirag-applift I suggest to first optimize the query: do GROUP BY integer or Enum, not String, as suggested in https://github.com/ClickHouse/ClickHouse/issues/19229#issuecomment-762206584 | 2021-02-28T00:21:45 | 2021-02-28T00:21:45 | {} | MEMBER |
ClickHouse | ClickHouse | 787,224,953 | 21,079 | azat | >Should be addressed in #21264 (after nightly stress testing I cannot reproduce the failure anymore after applying patch from the referenced PR)
Rebased against upstream (that includes that PR) | 2021-02-28T03:05:42 | 2021-02-28T03:05:42 | {} | MEMBER |
ClickHouse | ClickHouse | 787,458,941 | 21,250 | vdimir | `Stress test (thread) — Sanitizer assert (in stderr.log)` -- resolved in master https://github.com/ClickHouse/ClickHouse/issues/19887 | 2021-02-28T14:13:01 | 2021-02-28T14:13:01 | {} | MEMBER |
ClickHouse | ClickHouse | 787,833,134 | 18,368 | filimonov | ```
CREATE TABLE test_table (Id Int32, Foo Int32) ENGINE = MergeTree ORDER BY Id;
CREATE TABLE test_table_merge AS test_table ENGINE = Merge(currentDatabase(), 'test_table');
INSERT INTO test_table (Id, Foo) VALUES (1, 1);
SELECT any(t2.Foo) FROM test_table_merge AS t1 LEFT JOIN test_table AS t2 ON t1.Id = t2.Id GR... | 2021-03-01T10:20:00 | 2021-03-01T10:28:12 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 787,542,537 | 8,295 | alexey-milovidov | No answer. | 2021-02-28T23:12:07 | 2021-02-28T23:12:07 | {} | MEMBER |
ClickHouse | ClickHouse | 787,543,541 | 5,751 | alexey-milovidov | Fixed here: #5728 | 2021-02-28T23:17:04 | 2021-02-28T23:17:04 | {} | MEMBER |
ClickHouse | ClickHouse | 787,837,205 | 18,368 | filimonov | @vdimir ^^ | 2021-03-01T10:26:52 | 2021-03-01T10:26:52 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 787,838,073 | 19,766 | ghost | @vdimir
I fixed source codes. and added comments.
It seems that the failed tests are irrelevant. | 2021-03-01T10:28:24 | 2021-03-01T10:28:24 | {} | NONE |
ClickHouse | ClickHouse | 787,840,918 | 21,333 | filimonov | One of the replicas which don't have that setting enabled.
So if you will enable it cluster-wide nobody will do merges.
Also enabling that setting may introduce excessive logging and zookeeper traffic increase. | 2021-03-01T10:33:07 | 2021-03-01T10:33:07 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 788,032,437 | 21,094 | akuzm | I thought only the insert was slow and changed it to use `numbers_mt`, but apparently both selects are also slow (tens of seconds), so let's merge like this. | 2021-03-01T15:24:09 | 2021-03-01T15:24:09 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 787,073,862 | 20,991 | alexey-milovidov | Ok.
> The same long test with json format works fine for zlib and zstd (checked with asan and msan). Should I add this test for them in this pr too?
Not necessarily, it's enough. | 2021-02-27T13:37:02 | 2021-02-27T13:37:02 | {} | MEMBER |
ClickHouse | ClickHouse | 787,076,856 | 9,502 | alexey-milovidov | Task description in russian:
> Разработать clickhouse-library-bridge - процесс для взаимодействия со сторонними библиотеками для внешних словарей.
> В ClickHouse существует понятие "внешние словари". У внешних словарей бывают разные источники. В качестве одного из источников есть ODBC. ODBC подгружает произвольны... | 2021-02-27T13:56:52 | 2021-02-27T13:56:52 | {} | MEMBER |
ClickHouse | ClickHouse | 787,082,807 | 21,278 | alexey-milovidov | Can be used to test geometry functions: #19257 @nikitamikhaylov | 2021-02-27T14:43:46 | 2021-02-27T14:43:46 | {} | MEMBER |
ClickHouse | ClickHouse | 787,083,170 | 21,278 | nikitamikhaylov | Nice 👍 | 2021-02-27T14:46:46 | 2021-02-27T14:46:46 | {} | MEMBER |
ClickHouse | ClickHouse | 787,203,380 | 20,334 | alexey-milovidov | > contrib/libdivide/libdivide.h 1250-1266
I checked the snippet. This code is using the same principle but is written for AVX-512. | 2021-02-27T23:26:20 | 2021-02-27T23:26:20 | {} | MEMBER |
ClickHouse | ClickHouse | 787,206,050 | 13,675 | alexey-milovidov | Still relevant in 21.3. | 2021-02-27T23:54:09 | 2021-02-27T23:54:09 | {} | MEMBER |
ClickHouse | ClickHouse | 787,205,454 | 10,128 | alexey-milovidov | > Add an introspection table similar to system.merges/mutations in order to monitor active deletes
and moves.
Done with the `merge_type` column in `system.merges`. | 2021-02-27T23:48:51 | 2021-02-27T23:48:51 | {} | MEMBER |
ClickHouse | ClickHouse | 787,428,901 | 20,149 | keen-wolf | > @keen-wolf Let's continue?
Sorry for the late reply! Just came back from the holiday.
| 2021-02-28T10:19:58 | 2021-02-28T10:19:58 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 787,774,311 | 21,167 | KochetovNicolai | ```
gdb ./21.2.4.6+debug/usr/bin/clickhouse
(gdb) disas 'DB::Connection::receivePacket(std::__1::function<void (Poco::Net::Socket&)>)::$_3::operator()() const'
Dump of assembler code for function DB::Connection::receivePacket(std::__1::function<void (Poco::Net::Socket&)>)::$_3::operator()() const:
0x000000001aa4... | 2021-03-01T08:51:24 | 2021-03-01T08:51:24 | {} | MEMBER |
ClickHouse | ClickHouse | 787,076,255 | 21,269 | alexey-milovidov | You have to explicitly cast to Decimal:
`SELECT round(CAST(1027 - 949, 'Decimal(10, 5)') / CAST(949, 'Decimal(10, 5)'), 4) * 100`
`8.22000`
| 2021-02-27T13:52:43 | 2021-02-27T13:52:54 | {} | MEMBER |
ClickHouse | ClickHouse | 787,460,545 | 21,308 | alexey-milovidov | Thank you. PS. You can squash commits in your branch or I can squash on merge. | 2021-02-28T14:23:29 | 2021-02-28T14:23:29 | {} | MEMBER |
ClickHouse | ClickHouse | 787,632,767 | 21,324 | den-crane | because you need to kill unfinished mutation mutation_5489.txt.
check `select * from system.mutaition` `kill mutation ...` https://clickhouse.tech/docs/en/sql-reference/statements/kill/#kill-mutation | 2021-03-01T04:45:57 | 2021-03-01T04:45:57 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 787,739,892 | 21,330 | kitaisreal | @taiyang-li currently functions conversions file is too complex and we do not want to add additional functionality before it will be easier to maintain.
From my perspective it would be better to rewrite this function as composition of accurateCastOrNull and if result is null use default value. What do you think ? | 2021-03-01T07:59:09 | 2021-03-01T07:59:09 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 787,049,560 | 19,673 | azat | >I'm in process of reviewing this PR.
@abyss7 thanks! By the way, I can split this PR into smaller parts, if this will help | 2021-02-27T10:18:12 | 2021-02-27T10:18:12 | {} | MEMBER |
ClickHouse | ClickHouse | 787,073,974 | 20,991 | alexey-milovidov | @kssenii Maybe add similar tests for zstd and zlib just in case, in separate PR? | 2021-02-27T13:37:54 | 2021-02-27T13:37:54 | {} | MEMBER |
ClickHouse | ClickHouse | 787,075,523 | 21,236 | alexey-milovidov | > Does it reproduce on recent release?
NO
So the solution is to install the recent release. | 2021-02-27T13:48:50 | 2021-02-27T13:48:50 | {} | MEMBER |
ClickHouse | ClickHouse | 787,202,607 | 20,334 | alexey-milovidov | > contrib/libdivide/libdivide.h 1250-1266
This is part of `libdivide`. We are using libdivide under the terms of its license. If you have questions, you can ask them to the author: https://ridiculousfish.com/
Also ClickHouse can be built and work without libdivide.
> website/js/sentry.js 704 - 718
This is par... | 2021-02-27T23:18:07 | 2021-02-27T23:18:07 | {} | MEMBER |
ClickHouse | ClickHouse | 787,417,742 | 21,308 | ucasfl | Should be `pr-not-for-changelog`. | 2021-02-28T08:43:04 | 2021-02-28T08:43:04 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 787,482,824 | 20,202 | itzikiusa | Long expected feature!
Waiting for PR to be approved, and the following steps, specially different order by
Thank you very much for your work! | 2021-02-28T16:54:53 | 2021-02-28T16:54:53 | {} | NONE |
ClickHouse | ClickHouse | 787,564,275 | 18,583 | alexey-milovidov | Maybe if only thing that I want is memcpy, it will be simpler to rewrite the code instead of trying to plug it as is? | 2021-03-01T00:47:06 | 2021-03-01T00:47:06 | {} | MEMBER |
ClickHouse | ClickHouse | 787,628,676 | 21,324 | peaksnail | eventDate type Date
I just want to delete the duplicated data | 2021-03-01T04:32:39 | 2021-03-01T04:32:39 | {} | NONE |
ClickHouse | ClickHouse | 787,635,882 | 21,324 | peaksnail | thanks, it works now | 2021-03-01T04:56:06 | 2021-03-01T04:56:06 | {} | NONE |
ClickHouse | ClickHouse | 787,752,058 | 20,946 | tavplubix | Docs check - broken in master
Functional stateless tests flaky check - 00600_replace_running_query - looks unrelated, the test was bad before current changes
Integration tests (thread) - test_zookeeper_config/test.py::test_secure_connection - flaky | 2021-03-01T08:19:07 | 2021-03-01T08:19:07 | {} | MEMBER |
ClickHouse | ClickHouse | 787,803,312 | 21,130 | alesapin | No related failures | 2021-03-01T09:30:54 | 2021-03-01T09:30:54 | {} | MEMBER |
ClickHouse | ClickHouse | 787,843,663 | 21,330 | kitaisreal | @taiyang-li actually I mean separate function that will call accurate cast or null. And if in result column we see nullable row then we replace It with default value. | 2021-03-01T10:37:29 | 2021-03-01T10:37:29 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 787,965,393 | 18,351 | vdimir | Continue in https://github.com/ClickHouse/ClickHouse/pull/21109 | 2021-03-01T13:53:31 | 2021-03-01T13:53:31 | {} | MEMBER |
ClickHouse | ClickHouse | 787,967,518 | 21,332 | DipalPrajapati | Yes thank you I found it. You can close it. I keep 10 minutes wait in between delete and create the same DB , or i can change the property database_atomic_delay_before_drop_table_sec to 0 right?
| 2021-03-01T13:56:25 | 2021-03-01T13:56:25 | {} | NONE |
ClickHouse | ClickHouse | 787,072,420 | 21,258 | alexey-milovidov | Binary hash is only for official builds, so I have to merge it before checking. | 2021-02-27T13:25:11 | 2021-02-27T13:25:11 | {} | MEMBER |
ClickHouse | ClickHouse | 787,076,522 | 21,167 | alexey-milovidov | @KochetovNicolai But the fix is already present in version 21.2.4. | 2021-02-27T13:55:01 | 2021-02-27T13:55:01 | {} | MEMBER |
ClickHouse | ClickHouse | 787,429,496 | 21,306 | azat | >Stress test (thread) — Killed by signal (in clickhouse-server.log)
https://clickhouse-test-reports.s3.yandex.net/21306/568a49dad59acbc036eb6dacecfac297d618e853/stress_test_(thread).html#fail1 - #21305 (`ZlibDeflatingWriteBuffer`) | 2021-02-28T10:24:40 | 2021-02-28T10:24:40 | {} | MEMBER |
ClickHouse | ClickHouse | 787,452,519 | 21,305 | alexey-milovidov | Not for changelog, because this bug was never released. | 2021-02-28T13:30:32 | 2021-02-28T13:30:32 | {} | MEMBER |
ClickHouse | ClickHouse | 787,468,858 | 21,308 | fuwhu | > Thank you. PS. You can squash commits in your branch or I can squash on merge.
thanks, done. | 2021-02-28T15:20:23 | 2021-02-28T15:20:23 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 787,553,957 | 18,583 | alexey-milovidov | @jart
It's unclear where to plug in initialization code like calls to `_init_kHalfCache3`, `_init_kCpuids` and `_init_memcpy` if we are using glibc (and not using startup code of cosmopolitan).
Looks like glibc does not automatically call functions from
```
[19] .init.300._init_m PROGBITS 00000000003... | 2021-03-01T00:15:43 | 2021-03-01T00:29:07 | {} | MEMBER |
ClickHouse | ClickHouse | 787,780,731 | 21,167 | KochetovNicolai | That's because this fix will be only in 21.2.5
| 2021-03-01T09:00:22 | 2021-03-01T09:00:22 | {} | MEMBER |
ClickHouse | ClickHouse | 787,795,554 | 21,203 | Akazz | Failures are irrelevant to this PR. Will merge it in. | 2021-03-01T09:20:12 | 2021-03-01T09:20:12 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 787,911,703 | 9,361 | filimonov | > By not having any order, you mean that it retains the order of the original data during insert, or data are inserted in random order?
No guarantees are provided here.
> @filimonov what does it mean?
It's just timestamp multiplied by 2^24 + rownumber in insert block, but using bit operations which are more e... | 2021-03-01T12:28:11 | 2021-03-01T12:28:11 | {
"+1": 3
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 787,928,932 | 20,876 | filimonov | Related to https://github.com/ClickHouse/ClickHouse/issues/21170 | 2021-03-01T12:57:06 | 2021-03-01T12:57:06 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 787,779,900 | 18,583 | alexey-milovidov | @jart I see that these sections are in the binary, but the functions are not called.
I tried to call them directly but the first attempt failed:
```
extern "C" void _init_kHalfCache3();
extern "C" void _init_kCpuids();
__attribute__((__constructor__)) void memcpy_init()
{
_init_kHalfCache3();
_init_... | 2021-03-01T08:59:23 | 2021-03-01T08:59:37 | {} | MEMBER |
ClickHouse | ClickHouse | 787,049,717 | 19,673 | abyss7 | > >I'm in process of reviewing this PR.
>
> @abyss7 thanks! By the way, I can split this PR into smaller parts, if this will help
Nope, everything is fine. Will finish today. | 2021-02-27T10:19:36 | 2021-02-27T10:19:36 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 787,191,361 | 20,182 | alexey-milovidov | #21292 | 2021-02-27T22:04:29 | 2021-02-27T22:04:29 | {} | MEMBER |
ClickHouse | ClickHouse | 787,440,797 | 21,149 | nikitamikhaylov | > Need to check with permutted and skipped fields.
> See the `input_format_skip_unknown_fields` and `input_format_with_names_use_header`.
I know, it would be necessary to convert this PR to draft. | 2021-02-28T11:58:35 | 2021-02-28T11:59:33 | {} | MEMBER |
ClickHouse | ClickHouse | 787,613,155 | 21,324 | den-crane | You need to upgrade to 20.8.13.15 | 2021-03-01T03:37:36 | 2021-03-01T03:37:36 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 787,615,048 | 21,324 | den-crane | > eventDate>='2021-2-27'
? typo?
Clickhouse mutations does not support such complicated conditions.
Try to use this way https://stackoverflow.com/a/55775479/11644308
| 2021-03-01T03:44:52 | 2021-03-01T03:44:52 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 787,072,453 | 21,258 | alexey-milovidov | Writing integration test is impossible for the same reason. | 2021-02-27T13:25:29 | 2021-02-27T13:25:29 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 787,541,600 | 19,958 | alexey-milovidov | @leosunli Something wrong with your toolchain.
I recommend installing the latest clang, llvm, lld on Ubuntu and compiling version 21.3 instead of 20.3. | 2021-02-28T23:06:21 | 2021-02-28T23:06:21 | {} | MEMBER |
ClickHouse | ClickHouse | 787,631,444 | 21,324 | peaksnail | now
```
alter table trade_log delete where tradelogId='bdd55bcf-3b48-4e86-9a8c-194e6553347e'
optimize table trade_log final
```
doesn't work!
```
select count() from trade_log final where tradelogId='bdd55bcf-3b48-4e86-9a8c-194e6553347e' and eventDate>='2021-02-27'
```
┌─count()─┐
│ 2
└─────┘... | 2021-03-01T04:41:35 | 2021-03-01T04:41:35 | {} | NONE |
ClickHouse | ClickHouse | 787,803,002 | 21,133 | alesapin | Failure not related to changes | 2021-03-01T09:30:29 | 2021-03-01T09:30:29 | {} | MEMBER |
ClickHouse | ClickHouse | 802,098,038 | 21,875 | iassasin | > но тогда надо больше cpu тратить на if else if, это деньги на лишние сервера и электричество
Согласен, платить больше никому не хочется. Еще раз спасибо за ответы. | 2021-03-18T16:41:26 | 2021-03-18T16:41:26 | {} | NONE |
ClickHouse | ClickHouse | 802,209,566 | 20,343 | djmarcin | I added some more data to https://github.com/ClickHouse/ClickHouse/issues/21851 but I don't think the workaround solves the issue in all cases, notably when the binary data section begins with a newline character (0x0a). There are potentially other problematic characters (0x03 seems to be one) as well. | 2021-03-18T18:59:24 | 2021-03-18T18:59:24 | {} | NONE |
ClickHouse | ClickHouse | 802,248,415 | 20,179 | Jokser | @vmihailenco Yes, each column is uploaded in parallel. But since 21.3 version you can limit the number of threads used in parallel copies and other async operations for S3 disk: `thread_pool_size` which defaults to 16. | 2021-03-18T20:00:24 | 2021-03-18T20:00:24 | {} | CONTRIBUTOR |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.