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 | 664,661,194 | 12,745 | azat | >Why is the enum to string conversion is the only one that needs this?
>E.g. cast(toNullable('1'), 'Nullable(UInt8)') or cast(toNullable('null'), 'Nullable(UInt8)') work w/o additional flags.
Because `String` is special, since it requires [`DataTypeEnum::getValue`](https://github.com/ClickHouse/ClickHouse/blob/4d76... | 2020-07-27T22:03:06 | 2020-07-27T22:03:06 | {} | MEMBER |
ClickHouse | ClickHouse | 664,938,004 | 12,756 | nikitamikhaylov | Please, add a testcase from the issue. | 2020-07-28T09:47:11 | 2020-07-28T09:47:11 | {} | MEMBER |
ClickHouse | ClickHouse | 664,618,649 | 12,911 | alexey-milovidov | Not backported due to significant changes in code. | 2020-07-27T20:23:11 | 2020-07-27T20:23:11 | {} | MEMBER |
ClickHouse | ClickHouse | 664,621,398 | 12,894 | alexey-milovidov | Better error messages is not for backport. | 2020-07-27T20:28:57 | 2020-07-27T20:28:57 | {} | MEMBER |
ClickHouse | ClickHouse | 664,973,918 | 12,670 | alexey-milovidov | > Docs check — Found 1 errors
Broken by @blinkov | 2020-07-28T10:58:19 | 2020-07-28T10:58:19 | {
"eyes": 1
} | MEMBER |
ClickHouse | ClickHouse | 665,114,984 | 13,026 | excitoon | Progress on #12188 | 2020-07-28T15:40:14 | 2020-07-28T15:40:14 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 664,527,079 | 12,934 | alexey-milovidov | Not relevant for 20.3. | 2020-07-27T17:17:57 | 2020-07-27T17:17:57 | {} | MEMBER |
ClickHouse | ClickHouse | 664,599,211 | 12,959 | alesapin | Just checked all possible scenarios. | 2020-07-27T19:42:06 | 2020-07-27T19:42:06 | {} | MEMBER |
ClickHouse | ClickHouse | 664,792,237 | 10,177 | PhantomPhreak | I found a root cause of the problem:
in my example
```
alter table <table> MODIFY TTL <column> + interval 6 month;
```
table `<table>` have been created using old (deprecated) syntax, when TTL setting supported when using new syntax only.
I've re-created that table using new syntax with TTL and it works.
Curr... | 2020-07-28T05:50:35 | 2020-07-28T05:51:12 | {} | NONE |
ClickHouse | ClickHouse | 664,957,719 | 12,327 | nikitamikhaylov | Timeout in stress test (memory) seems unrelated. Earlier in this PR it was Ok. | 2020-07-28T10:18:00 | 2020-07-28T10:18:00 | {} | MEMBER |
ClickHouse | ClickHouse | 665,081,369 | 13,003 | alexey-milovidov | Close because the bug is too rare. | 2020-07-28T14:41:27 | 2020-07-28T14:41:27 | {} | MEMBER |
ClickHouse | ClickHouse | 664,526,638 | 12,926 | alexey-milovidov | Don't pick because this bugfix is not significant enough. | 2020-07-27T17:17:04 | 2020-07-27T17:17:04 | {} | MEMBER |
ClickHouse | ClickHouse | 664,623,121 | 12,890 | alexey-milovidov | Not significant for backport. | 2020-07-27T20:32:30 | 2020-07-27T20:32:30 | {} | MEMBER |
ClickHouse | ClickHouse | 664,679,453 | 12,647 | azat | prometheus exporter exports only global metrics and hence there is no per-table metrics, besides this may be too specific.
Instead you can use custom http handlers (#7572) to export everything you want, i.e.:
```xml
<!-- cat /etc/clickhouse-server/config.d/http_handlers.xml -->
<yandex>
<http_handlers>
... | 2020-07-27T22:56:13 | 2020-07-28T08:08:58 | {
"+1": 2
} | MEMBER |
ClickHouse | ClickHouse | 664,905,761 | 12,975 | abyss7 | Can you please provide obfuscated CSV contents if possible? It will be much easier to reproduce and fix an issue.
I.e. you can just replace all the characters in strings with 'a' char, and all the digits in numbers with '1'. | 2020-07-28T09:12:44 | 2020-07-28T09:12:44 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 665,053,062 | 12,288 | filimonov | BTW: how does it work in SQL tests?
| 2020-07-28T13:51:56 | 2020-07-28T13:51:56 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 664,542,402 | 12,772 | alexey-milovidov | > Build failed without build log. It's inner CI problem. Maybe something wrong with docker or host. Just wait for build restart.
Sporadic build failure, reported. | 2020-07-27T17:46:09 | 2020-07-27T17:46:09 | {} | MEMBER |
ClickHouse | ClickHouse | 664,620,899 | 12,899 | alexey-milovidov | Not a significant fix. | 2020-07-27T20:27:57 | 2020-07-27T20:27:57 | {} | MEMBER |
ClickHouse | ClickHouse | 662,255,305 | 12,640 | xuyatian | I simplify the SQL1 and SQL2 as below:
### SQL1 and Result:
```sql
select count(e) from DatabaseName.DistributedTable where dateTime ='2020-07-13' and d ='100007252161' group by d
```
Result on 160 nodes:
### SQL2(removed group by) and Result:
```sql
select count(e) from DatabaseName.DistributedTable wher... | 2020-07-22T05:47:53 | 2020-07-28T02:53:56 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 665,035,586 | 12,708 | qoega | At least there was fail like this
```
../contrib/libcxx/include/vector:1557: _LIBCPP_ASSERT '__n < size()' failed. vector[] index out of bounds
2020.07.23 03:55:44.653795 [ 26433 ] {f515c801-27fb-4588-8889-67bcf8faec3d} <Debug> executeQuery: (from [::1]:38508) SELECT joinGet();
2020.07.23 03:55:44.655257 [ 20... | 2020-07-28T13:19:55 | 2020-07-28T13:19:55 | {} | MEMBER |
ClickHouse | ClickHouse | 665,081,302 | 13,000 | alexey-milovidov | Close because the bug is too rare. | 2020-07-28T14:41:20 | 2020-07-28T14:41:20 | {} | MEMBER |
ClickHouse | ClickHouse | 665,092,036 | 12,959 | alesapin | All affected tests passed. | 2020-07-28T14:59:45 | 2020-07-28T14:59:45 | {} | MEMBER |
ClickHouse | ClickHouse | 664,561,579 | 12,952 | alexey-milovidov | Looks like already known issue with OpenSSL initialization code. | 2020-07-27T18:23:43 | 2020-07-27T18:23:43 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 664,574,343 | 12,950 | alexey-milovidov | If the user is defined in `users.xml` (instead of being created by CREATE USER query) it's not possible to revoke grants.
You should
- add another user (with CREATE USER or inside users.xml) with grant option;
- then remove default user from users.xml;
- then create default user with CREATE USER;
Now you can a... | 2020-07-27T18:49:16 | 2020-07-27T18:51:25 | {} | MEMBER |
ClickHouse | ClickHouse | 664,603,697 | 12,959 | alexey-milovidov | By the way, what will happen if we
- perform backup;
- remove a column from the table;
- restore from backup? | 2020-07-27T19:51:18 | 2020-07-27T19:51:18 | {} | MEMBER |
ClickHouse | ClickHouse | 664,620,605 | 12,900 | alexey-milovidov | @myrrc Please help to resolve conflict. | 2020-07-27T20:27:23 | 2020-07-27T20:27:23 | {} | MEMBER |
ClickHouse | ClickHouse | 664,800,853 | 11,482 | bobrik | I've reorganized my data to read it in forward order rather than in reverse. Let's close this one and hope I was alone with my issue. | 2020-07-28T06:17:37 | 2020-07-28T06:17:37 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 664,805,365 | 12,973 | qoega | ```
2020.07.28 09:29:15.637743 [ 4890 ] {6acacf7f-9ef6-4aee-8e68-3c9a61bc046d} <Debug> executeQuery: (from [::1]:47570) INSERT INTO `table`(k) VALUES
2020.07.28 09:29:15.638060 [ 4890 ] {6acacf7f-9ef6-4aee-8e68-3c9a61bc046d} <Trace> ContextAccess (default): Access granted: INSERT(k) ON default.table
2020.07.28 09:29... | 2020-07-28T06:29:37 | 2020-07-28T06:29:37 | {} | MEMBER |
ClickHouse | ClickHouse | 664,838,045 | 12,960 | excitoon | غمگینی | 2020-07-28T07:49:27 | 2020-07-28T07:49:27 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 664,867,032 | 11,591 | filimonov | My initial plan was to fix that (i was even asking you about that :P ), but i used another approach to fix the problem, so the issue is not imporantant anymore.
There are still some inconsistencies here (different formats behave different way, roundtrips does not always work when passing data directly to parser), bu... | 2020-07-28T08:39:43 | 2020-07-28T08:39:43 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 664,923,937 | 12,238 | KochetovNicolai | Backported manually. | 2020-07-28T09:37:58 | 2020-07-28T09:37:58 | {} | MEMBER |
ClickHouse | ClickHouse | 664,924,556 | 11,891 | hieunguyen1809 | I got same issue on version 19.17.9.60 | 2020-07-28T09:38:31 | 2020-07-28T09:38:31 | {} | NONE |
ClickHouse | ClickHouse | 664,961,074 | 12,508 | qoega | 01307_multiple_leaders https://clickhouse-test-reports.s3.yandex.net/12953/4bd6261dc164c64172c641adafec271df4c12301/functional_stateless_tests_(debug).html#fail1
```
2020-07-27 21:56:48 01307_multiple_leaders: [ FAIL ] 5.02 sec. - having stderror:
2020-07-27 21:56:48 [... | 2020-07-28T10:25:50 | 2020-07-28T10:25:50 | {} | MEMBER |
ClickHouse | ClickHouse | 665,036,449 | 12,756 | sundy-li | > Please, add a testcase from the issue.
ok | 2020-07-28T13:21:33 | 2020-07-28T13:21:33 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 664,507,737 | 12,840 | KochetovNicolai | @azat tests are ok.
What do you think, do we have other reasons to keep this code?
Waiting for perftests now. | 2020-07-27T16:41:16 | 2020-07-27T16:41:42 | {} | MEMBER |
ClickHouse | ClickHouse | 664,524,356 | 12,918 | alexey-milovidov | Too many code was changed - don't pick. | 2020-07-27T17:12:41 | 2020-07-27T17:12:41 | {} | MEMBER |
ClickHouse | ClickHouse | 664,526,252 | 12,925 | alexey-milovidov | Don't pick because it's a feature, not a bugfix. | 2020-07-27T17:16:14 | 2020-07-27T17:16:14 | {} | MEMBER |
ClickHouse | ClickHouse | 664,621,592 | 12,893 | alexey-milovidov | Not significant (experimental feature). | 2020-07-27T20:29:22 | 2020-07-27T20:29:22 | {} | MEMBER |
ClickHouse | ClickHouse | 664,944,367 | 12,870 | 4ertus2 | Backported manually here https://github.com/ClickHouse/ClickHouse/commit/7aef15f9c5f5bb3db4898c79b7bcf04d00037e1b | 2020-07-28T09:53:16 | 2020-07-28T09:53:16 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 664,964,089 | 12,988 | Jiasm | > @Jiasm Thank you!
>
> CC @blinkov You should _never_ do machine translation for docs.
lol，I will check if the same problems exist on other pages and try to fix them. | 2020-07-28T10:33:19 | 2020-07-28T10:33:19 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 664,969,914 | 12,485 | alexey-milovidov | @Jokser Maybe we can get rid of `curl` library in our build? | 2020-07-28T10:48:03 | 2020-07-28T10:48:03 | {} | MEMBER |
ClickHouse | ClickHouse | 665,053,868 | 11,327 | alexey-milovidov | @KochetovNicolai It also reproduces with `now`, just wait for a minute. | 2020-07-28T13:53:30 | 2020-07-28T13:53:30 | {} | MEMBER |
ClickHouse | ClickHouse | 665,105,696 | 12,642 | vitlibar | Could you add to README.md information about how to run locally a specific test, not all of them. It's more likely that one need to run locally one specific test only. | 2020-07-28T15:23:24 | 2020-07-28T15:23:24 | {} | MEMBER |
ClickHouse | ClickHouse | 665,166,346 | 12,662 | ildus | Thanks! | 2020-07-28T17:15:45 | 2020-07-28T17:15:45 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 665,205,780 | 12,991 | alexey-milovidov | @filimonov
Something was changed in Kafka so the integration test failed. | 2020-07-28T18:31:52 | 2020-07-28T18:31:52 | {} | MEMBER |
ClickHouse | ClickHouse | 664,504,427 | 12,701 | qoega | It will be easier to check this if you provide scenario to reproduce this case | 2020-07-27T16:34:43 | 2020-07-27T16:34:43 | {} | MEMBER |
ClickHouse | ClickHouse | 664,618,579 | 12,912 | alexey-milovidov | Not backported because of low significance. | 2020-07-27T20:23:02 | 2020-07-27T20:23:02 | {} | MEMBER |
ClickHouse | ClickHouse | 664,625,462 | 12,760 | alexey-milovidov | > Docs check — Found 1 errors
Broken by @blinkov. | 2020-07-27T20:37:35 | 2020-07-27T20:37:35 | {} | MEMBER |
ClickHouse | ClickHouse | 664,712,385 | 12,768 | gubinjie | @abyss7
Yes, it is replicas.
sql:
alter table lndata.table1 on cluster mycluster
delete where sync_id in (select sync_id from ln.table1 where op_day >'2019-08-08') and id not in (select id from ln.table1 where op_day >'2019-08-08' order by updated_date desc limit 1 by sync_id)
Question 1. Will the leader replic... | 2020-07-28T00:50:13 | 2020-07-28T00:50:13 | {} | NONE |
ClickHouse | ClickHouse | 665,060,750 | 7,752 | stas-sl | As a workaround it is possible to use `WITH FILL STEP dateDiff('second', now()), now() + INTERVAL 1 HOUR)` to convert interval to seconds (not sure if there is a better way), but of course it would be more convenient to use time interval expressions directly. | 2020-07-28T14:05:42 | 2020-07-28T14:05:42 | {} | NONE |
ClickHouse | ClickHouse | 665,206,176 | 12,991 | alexey-milovidov | Most of Kafka tests are Ok, only single case has failed. Looks like something trivial... | 2020-07-28T18:32:34 | 2020-07-28T18:32:47 | {} | MEMBER |
ClickHouse | ClickHouse | 664,623,271 | 12,880 | alexey-milovidov | It's a feature, not a bugfix - no need to backport. | 2020-07-27T20:32:49 | 2020-07-27T20:33:04 | {} | MEMBER |
ClickHouse | ClickHouse | 664,669,283 | 12,540 | azat | @KochetovNicolai thanks! Closing this then (anyway it does not shows any real errors, need something completely different if any) | 2020-07-27T22:25:33 | 2020-07-27T22:25:33 | {} | MEMBER |
ClickHouse | ClickHouse | 662,207,127 | 12,640 | xuyatian | Hi @den-crane ,
Thanks for quick response. Let me give more real details. (shard 1-13/14-20 in my previous comments was just an example)
My cluster has 160 nodes (20shard X 8 replica)
### SQL1 and Result:
```sql
select uniq(e) ,
sum(l) ,
uniq(f) ,
sum(h) ,
d
from DatabaseName. DistributedTable
where dat... | 2020-07-22T02:36:13 | 2020-07-28T02:53:39 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 664,923,783 | 12,959 | alesapin | > The same question for MODIFY COLUMN (the answer is probably "it won't work"?)
No, it should work. We just apply conversions on fly in case of modify or ignore columns in case of drop. But we have very strange code https://github.com/ClickHouse/ClickHouse/pull/12959/commits/1bece3de0aa7e431fbf2ba86ce161b0d46430f00... | 2020-07-28T09:37:42 | 2020-07-28T09:37:42 | {} | MEMBER |
ClickHouse | ClickHouse | 665,091,876 | 12,975 | calebeaires | This csv take 1.6 GB size with many columns. I have tried to simulate all possible mistakes I could make, but i seems very clean and well formatted.
```csv
1,12,a,a,a,a,a,a,444,a,2016-04-01,2017-02-28,201604,201702,a,2017-03-01,201703,a,a,01185702,a,123.538.166-98,a,a,a,(34) 1234-1234,,2344-24,a,a,,,2002-11-26,,15,... | 2020-07-28T14:59:29 | 2020-07-28T16:32:46 | {} | NONE |
ClickHouse | ClickHouse | 664,512,689 | 12,739 | alesapin | integration OK. | 2020-07-27T16:50:50 | 2020-07-27T16:50:50 | {} | MEMBER |
ClickHouse | ClickHouse | 664,524,442 | 12,921 | alexey-milovidov | Too many code was changed - don't pick. | 2020-07-27T17:12:52 | 2020-07-27T17:12:52 | {} | MEMBER |
ClickHouse | ClickHouse | 664,527,566 | 12,937 | alexey-milovidov | The bug is not significant - too specific reproducible case. | 2020-07-27T17:18:46 | 2020-07-27T17:18:46 | {} | MEMBER |
ClickHouse | ClickHouse | 664,625,164 | 12,959 | alexey-milovidov | > Docs check — Found 1 errors
Broken by @blinkov. | 2020-07-27T20:37:01 | 2020-07-27T20:37:01 | {} | MEMBER |
ClickHouse | ClickHouse | 664,977,284 | 12,990 | alexey-milovidov | I'll fix it. | 2020-07-28T11:06:47 | 2020-07-28T11:06:47 | {} | MEMBER |
ClickHouse | ClickHouse | 665,026,019 | 2,156 | FrieNze | I have the same problem. version 20.5.2.7 (official build)
`DB::Exception: Block structure mismatch in function connect between Converting and Concat stream: different number of columns`
My query looks like:
```
SELECT
*
FROM
(
SELECT * FROM results_v1
UNION ALL
SELECT * ... | 2020-07-28T13:02:51 | 2020-07-28T13:03:30 | {} | NONE |
ClickHouse | ClickHouse | 665,178,480 | 12,642 | vzakaznikov | > Could you add to README.md information about how to run locally a specific test, not all of them. It's more likely that one need to run locally one specific test only.
Yes. I will add it. In general, this information is covered in https://testflows.com/handbook/#Filtering. | 2020-07-28T17:39:53 | 2020-07-28T17:39:53 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 664,571,786 | 3,207 | alexey-milovidov | In recent versions it's not possible:
```
Received exception from server (version 20.7.1):
Code: 524. DB::Exception: Received from localhost:9000. DB::Exception: Trying to ALTER DROP key a column which is a part of key expression.
``` | 2020-07-27T18:44:13 | 2020-07-27T18:44:13 | {} | MEMBER |
ClickHouse | ClickHouse | 664,620,388 | 12,903 | alexey-milovidov | Not a bug. | 2020-07-27T20:26:56 | 2020-07-27T20:26:56 | {} | MEMBER |
ClickHouse | ClickHouse | 664,995,885 | 12,898 | 4ertus2 | 20.3 has nothing to fix. There's no such optimisation there. | 2020-07-28T11:53:51 | 2020-07-28T11:53:51 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 665,018,880 | 12,773 | macEar | Sorry, only parts with 0 rows remained.
The corresponding issue already exists: https://github.com/ClickHouse/ClickHouse/issues/5491
Should we close this issue to continue on the existing one? | 2020-07-28T12:47:49 | 2020-07-28T12:47:49 | {} | NONE |
ClickHouse | ClickHouse | 665,020,316 | 12,638 | alesapin | Thread sanitizer failure is not related to changes. Restarted test. | 2020-07-28T12:51:01 | 2020-07-28T12:51:01 | {} | MEMBER |
ClickHouse | ClickHouse | 664,604,019 | 12,959 | alexey-milovidov | The same question for MODIFY COLUMN (the answer is probably "it won't work"?) | 2020-07-27T19:52:01 | 2020-07-27T19:52:01 | {} | MEMBER |
ClickHouse | ClickHouse | 664,619,151 | 12,904 | alexey-milovidov | Not significant. | 2020-07-27T20:24:20 | 2020-07-27T20:24:20 | {} | MEMBER |
ClickHouse | ClickHouse | 665,027,525 | 12,737 | CurtizJ | Also syntetic test from performance tests.
```
CREATE TABLE mt_comp_parts
ENGINE = MergeTree
ORDER BY (c1, c2)
SETTINGS min_rows_for_wide_part = 1000000000 AS
SELECT *
FROM generateRandom('c1 UInt32, c2 UInt64, s1 String, arr1 Array(UInt32), c3 UInt64, s2 String', 0, 30, 30)
LIMIT 50000000
```
```
CREATE T... | 2020-07-28T13:05:22 | 2020-07-28T13:05:22 | {
"+1": 2
} | MEMBER |
ClickHouse | ClickHouse | 664,526,056 | 12,924 | alexey-milovidov | Don't pick because too many code was changed. | 2020-07-27T17:15:51 | 2020-07-27T17:15:51 | {} | MEMBER |
ClickHouse | ClickHouse | 664,981,607 | 12,737 | alexey-milovidov | @kreuzerkrieg I don't have these numbers, maybe @CurtizJ has? | 2020-07-28T11:17:15 | 2020-07-28T11:17:15 | {} | MEMBER |
ClickHouse | ClickHouse | 665,048,196 | 12,288 | filimonov | > > Same in interactive mode - try to paste those into clickhouse-client -mn interactive session
>
> Does not reproduce.
Interesting. Check this out:
> It always worked this way.
... | 2020-07-28T13:42:54 | 2020-07-28T13:50:17 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 665,080,635 | 12,548 | Akazz | Merged in #12982 | 2020-07-28T14:40:06 | 2020-07-28T14:40:06 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 665,081,332 | 13,002 | alexey-milovidov | Close because the bug is too rare. | 2020-07-28T14:41:24 | 2020-07-28T14:41:24 | {} | MEMBER |
ClickHouse | ClickHouse | 665,217,172 | 12,751 | alexey-milovidov | Here is the test demonstrating the issue:
`SELECT materialize([4294967295]) AS arr, has(arr, -1)` | 2020-07-28T18:55:46 | 2020-07-28T18:55:46 | {} | MEMBER |
ClickHouse | ClickHouse | 664,509,027 | 12,840 | azat | >@azat tests are ok.
>What do you think, do we have other reasons to keep this code?
Nope, LGTM | 2020-07-27T16:43:54 | 2020-07-27T16:43:54 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 664,524,044 | 12,915 | alexey-milovidov | Not relevant for 20.3. | 2020-07-27T17:12:07 | 2020-07-27T17:12:07 | {} | MEMBER |
ClickHouse | ClickHouse | 664,852,881 | 12,857 | alesapin | ```
FAILED test_version_update_after_mutation/test.py::test_mutate_and_upgrade - ...
FAILED test_adaptive_granularity/test.py::test_version_update_two_nodes - Que...
```
Not related to changes, but require investigation... | 2020-07-28T08:19:49 | 2020-07-28T08:20:06 | {} | MEMBER |
ClickHouse | ClickHouse | 664,864,353 | 12,951 | alesapin | > Ok
>
> ... but I remember the use case that Kirill Shvakov @kshvakov told on the meetup:
> You can safely clear the last column of primary key. It will not affect the order of data. But after that, the data will be aggregated in SummingMergeTree/AggregatingMergeTree.
Actually it's a strange and dangerous use c... | 2020-07-28T08:37:53 | 2020-07-28T08:37:53 | {} | MEMBER |
ClickHouse | ClickHouse | 664,962,791 | 12,988 | alexey-milovidov | @Jiasm Thank you!
CC @blinkov You should *never* do machine translation for docs. | 2020-07-28T10:30:13 | 2020-07-28T10:30:34 | {} | MEMBER |
ClickHouse | ClickHouse | 664,978,067 | 12,737 | kreuzerkrieg | > Yes.
>
> Reading a column from big compact part will involve more disk seeks.
> It should be noticeably slower for HDD and just a little slower on SSD.
>
> I'm not 100% sure that 2x difference is Ok, though.
> I expect that #12492 will eliminate most of the difference, especially for SSD.
>
> But about 1.1... | 2020-07-28T11:08:33 | 2020-07-28T11:08:33 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 664,981,966 | 12,852 | 4ertus2 | It's hard to backport. | 2020-07-28T11:18:06 | 2020-07-28T11:18:06 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 665,053,397 | 11,327 | alexey-milovidov | Still relevant in 20.7. | 2020-07-28T13:52:36 | 2020-07-28T13:52:36 | {} | MEMBER |
ClickHouse | ClickHouse | 665,063,061 | 11,289 | alexey-milovidov | We don't need it. | 2020-07-28T14:09:51 | 2020-07-28T14:09:51 | {} | MEMBER |
ClickHouse | ClickHouse | 665,078,388 | 12,773 | alexey-milovidov | Yes, it's a duplicate. | 2020-07-28T14:36:09 | 2020-07-28T14:36:09 | {} | MEMBER |
ClickHouse | ClickHouse | 665,204,216 | 2,534 | alexey-milovidov | As #1851 is fixed, this is also fixed. | 2020-07-28T18:28:49 | 2020-07-28T18:28:49 | {} | MEMBER |
ClickHouse | ClickHouse | 664,565,994 | 12,951 | alexey-milovidov | Do you remeber when it was started to be forbidden? | 2020-07-27T18:32:09 | 2020-07-27T18:32:09 | {} | MEMBER |
ClickHouse | ClickHouse | 664,618,808 | 12,910 | alexey-milovidov | Not backported because there is no bug. | 2020-07-27T20:23:32 | 2020-07-27T20:23:32 | {} | MEMBER |
ClickHouse | ClickHouse | 664,843,196 | 10,006 | filimonov | Loosely related to https://github.com/ClickHouse/ClickHouse/issues/10729 | 2020-07-28T08:00:17 | 2020-07-28T08:00:17 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 664,983,708 | 12,485 | Jokser | @alexey-milovidov As I can see `curl` is used as a dependency in `mariadb-connector` and `sentry-native` | 2020-07-28T11:22:31 | 2020-07-28T11:22:31 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 664,993,038 | 12,998 | qoega | @blinkov
actually I can't understand how it works https://github.com/ClickHouse/ClickHouse/runs/912223591?check_suite_focus=true
and newer one fails. | 2020-07-28T11:46:27 | 2020-07-28T11:46:27 | {} | MEMBER |
ClickHouse | ClickHouse | 664,994,864 | 12,958 | qoega | ```
=================================================================
==30713==ERROR: AddressSanitizer: heap-buffer-overflow on address 0x60400222515c at pc 0x00000c4c0cf9 bp 0x7f86d7d26dc0 sp 0x7f86d7d26db8
READ of size 4 at 0x60400222515c thread T107 (QueryPipelineEx)
==30713==WARNING: invalid path to external sy... | 2020-07-28T11:51:16 | 2020-07-28T11:51:16 | {} | MEMBER |
ClickHouse | ClickHouse | 665,023,088 | 12,737 | CurtizJ | I have some.
E.g. query on table `hits` from test datasets: `select min(URL), min(UserID) from hits_wide group by CounterID % 1000`. `UserID` and `CounterID` have `UInt64` type, `URL` has `String` type.
```
SELECT count()
FROM hits_wide
┌─count()─┐
│ 8873898 │
└─────────┘
```
On VM with HDD:
```
clickho... | 2020-07-28T12:56:49 | 2020-07-28T12:56:49 | {} | MEMBER |
ClickHouse | ClickHouse | 664,497,956 | 12,589 | KochetovNicolai | `Performance — 7 faster, 1 slower, 249 unstable` - good enough.
Merging. | 2020-07-27T16:22:47 | 2020-07-27T16:22:47 | {} | MEMBER |
ClickHouse | ClickHouse | 664,618,349 | 12,913 | alexey-milovidov | Not backported due to significant changes in code. | 2020-07-27T20:22:33 | 2020-07-27T20:22:33 | {} | MEMBER |
ClickHouse | ClickHouse | 664,976,697 | 12,990 | alexey-milovidov | Minor defect in `rdkafka` library. | 2020-07-28T11:05:21 | 2020-07-28T11:05:21 | {} | MEMBER |
ClickHouse | ClickHouse | 671,356,046 | 13,564 | filimonov | Can I ask why do you need Nullable there? You expect that some messages will not have json object? | 2020-08-10T13:31:57 | 2020-08-10T13:31:57 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 671,372,635 | 13,564 | chu1070y | Hi :)
I'm using Debezium as kafka source connector in data lake project.
Here is whole process of data flow.
PostgreSQL or MariaDB (Source DB) -> kafka source connector(Debezium) -> kafka cluster -> kafka sink connector(Confluent) -> PostgreSQL or etc...(Target DB)
There is a tombstone message when i delete a row... | 2020-08-10T14:01:23 | 2020-08-10T14:04:07 | {
"+1": 2
} | NONE |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.