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 | 601,862,674 | 9,717 | alexey-milovidov | #6044 | 2020-03-20T18:58:32 | 2020-03-20T18:58:32 | {} | MEMBER |
ClickHouse | ClickHouse | 601,862,843 | 9,765 | alexey-milovidov | #6044 | 2020-03-20T18:58:58 | 2020-03-20T18:58:58 | {} | MEMBER |
ClickHouse | ClickHouse | 601,941,296 | 9,721 | alexey-milovidov | FYI other regressions of replxx:
https://github.com/ClickHouse/ClickHouse/issues/9475
https://github.com/ClickHouse/ClickHouse/issues/9474
| 2020-03-20T22:43:28 | 2020-03-20T22:43:28 | {} | MEMBER |
ClickHouse | ClickHouse | 602,004,766 | 9,673 | azat | >Functional stateless tests (address)
Looks flacky:
- 01006_simpod_empty_part_single_column_write
- 01019_alter_materialized_view_consistent | 2020-03-21T06:50:24 | 2020-03-21T06:50:24 | {} | MEMBER |
ClickHouse | ClickHouse | 602,042,369 | 9,794 | alexey-milovidov | Perfect! But it's only added for farsi (fa) documentation, not english (en), chinese (zh)... | 2020-03-21T13:16:38 | 2020-03-21T13:16:38 | {} | MEMBER |
ClickHouse | ClickHouse | 602,043,952 | 9,475 | alexey-milovidov | Fixed in one of PRs, thank you! | 2020-03-21T13:32:12 | 2020-03-21T13:32:12 | {} | MEMBER |
ClickHouse | ClickHouse | 602,092,339 | 9,784 | zlobober | > Ok, but why shared_ptr instead of unique_ptr?
That was done similarly to QueryStatusInfo::profile_counters. I do not see any strong reason to expose both of them as shared_ptr insted of unique_ptr though.
> Also missing comment that the method is used in 3rd party code (otherwise the method will be removed beca... | 2020-03-21T19:33:00 | 2020-03-21T19:33:00 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 602,291,996 | 9,808 | alexey-milovidov | If we're asking remote servers to process the query to "Complete" stage, we cannot merge their results... | 2020-03-22T23:15:11 | 2020-03-22T23:15:11 | {} | MEMBER |
ClickHouse | ClickHouse | 601,584,366 | 5,485 | leoones | > [neighbor](https://clickhouse.tech/docs/en/query_language/functions/other_functions/#neighbor) function introduced in #5925 is avaliable since 19.14 address that issue.
>
> But it works currently inside each block separately:
>
> ```
> SELECT
> number,
> neighbor(toNullable(number), 1)
> FROM numb... | 2020-03-20T08:27:49 | 2020-03-20T08:28:50 | {} | NONE |
ClickHouse | ClickHouse | 602,061,347 | 9,796 | filimonov | Is there a way to run performance test locally now? | 2020-03-21T15:35:22 | 2020-03-21T15:35:22 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 602,188,505 | 8,846 | azat | @alexey-milovidov So, I came up with a cleaner approach, and removed `DRAFT` from the subject, please take a look! (but keep in mind the conflict with #9804 ) | 2020-03-22T12:00:55 | 2020-03-22T12:00:55 | {} | MEMBER |
ClickHouse | ClickHouse | 602,292,444 | 5,562 | knoguchi | Now MV supports subqueries. | 2020-03-22T23:16:17 | 2020-03-22T23:16:17 | {
"+1": 2
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 601,855,786 | 9,762 | alexey-milovidov | @alesapin Yes - that's exactly what this test is supposed to check :)
| 2020-03-20T18:42:44 | 2020-03-20T18:42:44 | {} | MEMBER |
ClickHouse | ClickHouse | 601,868,684 | 9,739 | 4ertus2 | For now V2 works but leads to regression in #6767
OK without predicate pushdown and not OK if pushdown enabled.
I think it would be OK to merge with disabled setting after tests pass. | 2020-03-20T19:13:22 | 2020-03-20T19:22:56 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 601,922,374 | 9,786 | filimonov | /cc @Enmk | 2020-03-20T21:41:31 | 2020-03-20T21:41:31 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 602,022,523 | 9,792 | filimonov | @blinkov What is going on? He optimize the same images few times per day.
https://github.com/ClickHouse/ClickHouse/pulls?q=is%3Apr+author%3Aapp%2Fimgbot+%22%2Fdocs%2Ffa%2Fimages%2Flogo.svg%22 | 2020-03-21T10:07:11 | 2020-03-21T10:07:42 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 602,023,114 | 9,792 | blinkov | > @blinkov What is going on? He optimize the same images few times per day.
I probably have some issue with my local software that adds a useless newline or something and bot tries to get rid of it. I'll take a closer look on how it happens later. | 2020-03-21T10:14:47 | 2020-03-21T10:14:47 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 602,356,389 | 1,417 | knoguchi | It works in 20.3.4.10.
```
-- Expected
localhost :) SELECT count() AS cnt, groupArray(d) AS ds FROM (SELECT d FROM test.table GROUP BY d);
SELECT
count() AS cnt,
groupArray(d) AS ds
FROM
(
SELECT d
FROM test.table
GROUP BY d
)
┌─cnt─┬─ds────┐
│ 2 │ [0,1] │
└─────┴───────┘
-... | 2020-03-23T03:00:50 | 2020-03-23T03:00:50 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 602,516,910 | 9,814 | alexey-milovidov | > forgot ifdef under Darwin?
Yes, thank you! | 2020-03-23T10:45:14 | 2020-03-23T10:45:14 | {} | MEMBER |
ClickHouse | ClickHouse | 602,686,187 | 9,821 | alexey-milovidov | Does the issue present if you update to 20.1.8 or 20.3.4? | 2020-03-23T15:47:54 | 2020-03-23T15:47:54 | {} | MEMBER |
ClickHouse | ClickHouse | 601,901,898 | 5,514 | den-crane | is not relevant anymore.
Predicate pushdown works against views with FINAL
```
create table bill (no String, fee Int64 , date Date) engine = ReplacingMergeTree(date,(no,date),8192)
insert into bill select toString(number), number, '2020-01-01' from numbers(1000000);
insert into bill select toString(number... | 2020-03-20T20:42:03 | 2020-03-20T20:43:58 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 601,975,847 | 9,475 | amosbird | I cannot reproduce this one | 2020-03-21T01:47:58 | 2020-03-21T01:47:58 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 602,058,578 | 9,794 | simPod | @blinkov should be resolved | 2020-03-21T15:13:10 | 2020-03-21T15:13:10 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 602,062,193 | 9,790 | alexey-milovidov | Me - not. | 2020-03-21T15:42:40 | 2020-03-21T15:42:40 | {} | MEMBER |
ClickHouse | ClickHouse | 602,325,405 | 9,799 | alexey-milovidov | #6044 | 2020-03-23T02:13:20 | 2020-03-23T02:13:20 | {} | MEMBER |
ClickHouse | ClickHouse | 602,419,967 | 9,813 | qoega | https://clickhouse-test-reports.s3.yandex.net/9813/622bca235848bf1a7950d6eeb10241204aa069eb/stress_test_(thread)/stderr.log
```
ThreadSanitizer: data race
#6 DB::TablesBlockSource::generate() /build/obj-x86_64-linux-gnu/../dbms/src/Storages/System/StorageSystemTables.cpp:258:41 (clickhouse+0xf6a561d)
and
... | 2020-03-23T07:00:30 | 2020-03-23T07:00:30 | {} | MEMBER |
ClickHouse | ClickHouse | 602,455,114 | 9,801 | den-crane | It does not matter.
It does not break anything. And it happened before. For example 'Ontime
example' parts had modern create syntax but documnetation had old syntax.
On Mon, Mar 23, 2020, 04:36 filimonov <notifications@github.com> wrote:
> But distribution in form of parts it will still have old format?
>
> —
> You a... | 2020-03-23T08:34:19 | 2020-03-23T08:34:19 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 601,889,099 | 5,122 | den-crane | 20.4.1.2722 still relevant | 2020-03-20T20:08:05 | 2020-03-20T20:08:05 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 601,941,051 | 9,721 | alexey-milovidov | Ok. | 2020-03-20T22:42:38 | 2020-03-20T22:42:38 | {} | MEMBER |
ClickHouse | ClickHouse | 601,994,157 | 9,474 | amosbird | It's not easy to have complete readline history editing persistency feature in replxx. However simply prevent `Accidentially up and down` can be achieved.
https://github.com/ClickHouse-Extras/replxx/pull/4
| 2020-03-21T04:46:34 | 2020-03-21T04:46:34 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 602,042,840 | 9,795 | simPod | I forgot there are more versions. Done.
~BTW `ja` was already done this way :)~ `ja` links to english, got it | 2020-03-21T13:21:08 | 2020-03-21T13:24:24 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 602,044,508 | 9,475 | alexey-milovidov | I'm using kUbuntu 19.10 (eoan) with Konsole as a terminal emulator and Kate as a text editor.
I'm copying a multiline text from Kate with selection and Ctrl+C.
```
SELECT *
FROM system.replicas
WHERE table = 'elog'
```
Then I'm pasting it into `clickhouse-cleint` with Ctrl+Shift+V or with middle mouse button o... | 2020-03-21T13:37:31 | 2020-03-21T13:38:26 | {} | MEMBER |
ClickHouse | ClickHouse | 602,149,519 | 9,568 | sunny19930321 | I have the same problem，Is there a good solution? @DaiDaDao,how about make friend In weixin or QQ, then we can talk about this problem | 2020-03-22T05:16:09 | 2020-03-22T05:34:57 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 602,237,731 | 8,846 | azat | >but keep in mind the conflict with #9804
It does not conflicts with that PR anymore | 2020-03-22T16:46:44 | 2020-03-22T16:46:44 | {} | MEMBER |
ClickHouse | ClickHouse | 602,554,934 | 9,821 | alexey-milovidov | Could you please also share the textual stack trace from log?
It is located right after the messages
```
BaseDaemon: ########################################
BaseDaemon: (version 20.1.3.7 (official build)) (from thread 56) (query_id: 85e3b2c2-4c96-49b7-bcd5-ae7d60ef4f2a) Received signal Segmentation fault (11).
``... | 2020-03-23T12:14:39 | 2020-03-23T12:14:39 | {} | MEMBER |
ClickHouse | ClickHouse | 602,685,331 | 9,822 | alexey-milovidov | Currently there is no way to flush it manually. | 2020-03-23T15:46:36 | 2020-03-23T15:46:36 | {} | MEMBER |
ClickHouse | ClickHouse | 601,692,006 | 9,776 | tetafro | First I drop distributed table and materialized view trigger who writes in this table (if there is any), and only then I drop the table itself.
Maybe dropping distributed tables and views is async operation and I should check somewhere that it is completed? | 2020-03-20T13:10:07 | 2020-03-20T13:10:49 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 601,891,296 | 9,766 | den-crane | https://stackoverflow.com/questions/60768819/using-arraytuplelowcardinalitystring-int32-in-clickhouse | 2020-03-20T20:14:28 | 2020-03-20T20:14:28 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 602,062,452 | 9,796 | alexey-milovidov | Yes, it's described in:
`/docker/test/performance-comparison/performance_comparison.md`
I will be happy if you will try. | 2020-03-21T15:44:27 | 2020-03-21T15:44:27 | {} | MEMBER |
ClickHouse | ClickHouse | 602,292,657 | 9,809 | alexey-milovidov | ```
2020-03-23 01:35:16 00993_system_parts_race_condition_drop_zookeeper: [ FAIL ] 31.46 sec. - having stderror:
2020-03-23 01:35:16 [1a3c49b01305] 2020.03.23 01:35:15.505487 [ 1017 ] {6cb05d0f-f756-45fa-aba0-243bee9733b2} <Error> default.alter_table_0: Cannot quickly remove directory /var/lib/c... | 2020-03-22T23:17:46 | 2020-03-22T23:17:46 | {} | MEMBER |
ClickHouse | ClickHouse | 601,634,304 | 9,673 | azat | >Functional stateless tests (thread) — fail: 20, passed: 1522, skipped: 8
00965_live_view_watch_heartbeat (does not looks relevant)
P.S. I can rebase to force CI | 2020-03-20T10:41:07 | 2020-03-20T10:41:07 | {} | MEMBER |
ClickHouse | ClickHouse | 601,679,362 | 9,778 | qoega | Actually we don't know your specific case.
The best way to answer your question is to try your estimated rps with close to real data in tables. And we don't know how big your cluster is and whether you have very hot keys.
What you should be conserned of is whether you want to make updates in this table or not. It is n... | 2020-03-20T12:39:36 | 2020-03-20T12:39:36 | {} | MEMBER |
ClickHouse | ClickHouse | 601,685,735 | 9,776 | filimonov | Is it possible that the table you dropped was used by some distibuted tables? | 2020-03-20T12:55:04 | 2020-03-20T12:55:04 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 601,790,313 | 9,715 | filimonov | Will you be able to predict (and explain to the user the behavior) that query?
```
WITH revenue + 3 as revenue
SELECT sum(revenue) as revenue,
max(revenue) as max_revenue -- use revenue column
FROM ....
WHERE revenue > 1 -- column ?
GROUP BY xyz
HAVING revenue > 3 and max_revenue > 4 and mi... | 2020-03-20T16:25:04 | 2020-03-20T16:27:17 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 601,856,986 | 9,768 | alexey-milovidov | @YiuRULE Thank you! It's production grade feature, just not documented yet. | 2020-03-20T18:45:21 | 2020-03-20T18:45:21 | {} | MEMBER |
ClickHouse | ClickHouse | 601,862,749 | 9,575 | alexey-milovidov | #6044 | 2020-03-20T18:58:43 | 2020-03-20T18:58:43 | {} | MEMBER |
ClickHouse | ClickHouse | 602,071,646 | 9,796 | akuzm | > Is there a way to run performance test locally now?
There is a simple python script that will run a single test on the servers you specify and output raw times. For analysis, it's really unwieldy now -- it's a sequence of clickhouse-local queries ran by a bash script. There -is- a way to run a full comparison scri... | 2020-03-21T16:51:59 | 2020-03-21T16:51:59 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 602,239,637 | 9,046 | azat | Nope, affected (though not with the first try of https://github.com/ClickHouse/ClickHouse/issues/9046#issuecomment-586011358) | 2020-03-22T17:00:59 | 2020-03-22T17:00:59 | {} | MEMBER |
ClickHouse | ClickHouse | 602,290,205 | 3,450 | den-crane | this issue is about sub-queries in MV, fixed in 19.14.3.3 | 2020-03-22T22:59:29 | 2020-03-22T22:59:29 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 602,292,791 | 9,809 | alexey-milovidov | Builds are Ok but I'm not sure if Docker images are rebuilt in CI. | 2020-03-22T23:18:54 | 2020-03-22T23:18:54 | {} | MEMBER |
ClickHouse | ClickHouse | 601,616,986 | 8,640 | azat | Any progress on this one? | 2020-03-20T09:57:15 | 2020-03-20T09:57:15 | {} | MEMBER |
ClickHouse | ClickHouse | 601,992,696 | 9,789 | alexey-milovidov | #6044 | 2020-03-21T04:25:49 | 2020-03-21T04:25:49 | {} | MEMBER |
ClickHouse | ClickHouse | 602,060,891 | 9,793 | filimonov | Looks like numeric value of Date (numer of days) compared with numeric value of DateTime (number of seconds).
Workaround is trivial. | 2020-03-21T15:31:50 | 2020-03-21T15:31:50 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 602,238,989 | 9,739 | alexey-milovidov | @zhang2014 We are struggle to find why we have a regression in #6767 | 2020-03-22T16:56:15 | 2020-03-22T16:56:15 | {} | MEMBER |
ClickHouse | ClickHouse | 602,644,794 | 9,424 | 4ertus2 | There's workaround in master, not released yet: `set multiple_joins_rewriter_version = 2`
```
SET multiple_joins_rewriter_version = 2;
SELECT c1, c2, c3
FROM (SELECT dummy AS c1 FROM system.one) AS t1
INNER JOIN (SELECT dummy AS c2 FROM system.one) AS t2 ON t1.c1 = t2.c2
INNER JOIN (SELECT dummy AS c3 FROM sy... | 2020-03-23T14:49:12 | 2020-03-23T14:49:12 | {
"heart": 2
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 601,558,840 | 2,515 | filimonov | Related #975 #6528 #8326 | 2020-03-20T06:50:47 | 2020-03-20T07:28:02 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 602,042,262 | 9,795 | alexey-milovidov | Sorry for confusion, I mean documentation languages like `en`, `zh`, `fa`... | 2020-03-21T13:15:32 | 2020-03-21T13:15:32 | {} | MEMBER |
ClickHouse | ClickHouse | 602,044,750 | 9,475 | alexey-milovidov | Fixed in https://github.com/ClickHouse-Extras/replxx/pull/5 | 2020-03-21T13:39:35 | 2020-03-21T13:39:35 | {} | MEMBER |
ClickHouse | ClickHouse | 602,238,340 | 9,046 | azat | FWIW upstream (6550aa5e6cf26deb7935f29c3a75df3cf194937c) does not affected | 2020-03-22T16:51:10 | 2020-03-22T16:51:10 | {} | MEMBER |
ClickHouse | ClickHouse | 601,601,031 | 9,762 | alesapin | ```2020-03-20 01:50:31 01201_read_single_thread_in_order: [ FAIL ] 1.09 sec. - result differs with reference:
2020-03-20 01:50:31 --- /usr/share/clickhouse-test/queries/0_stateless/01201_read_single_thread_in_order.reference 2020-03-20 00:09:15.000000000 +0300
2020-03-20 01:50:31 ... | 2020-03-20T09:15:42 | 2020-03-20T18:40:38 | {} | MEMBER |
ClickHouse | ClickHouse | 601,862,714 | 9,625 | alexey-milovidov | #6044 | 2020-03-20T18:58:38 | 2020-03-20T18:58:38 | {} | MEMBER |
ClickHouse | ClickHouse | 602,620,513 | 9,822 | alexey-milovidov | Yes, the schemas are cached.
We can introduce special SYSTEM query to flush this cache. | 2020-03-23T14:12:53 | 2020-03-23T14:12:53 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 601,628,907 | 9,767 | KochetovNicolai | Segfault from asan stress test:
```
2020.03.20 10:54:54.450913 [ 1043762 ] {} <Fatal> BaseDaemon: ########################################
2020.03.20 10:54:54.451010 [ 1043762 ] {} <Fatal> BaseDaemon: (version 20.4.1.2721) (from thread 604) (query_id: 2f20d51c-cc78-4dd3-bb63-9e0d8aa1472f) Received signal Segmentatio... | 2020-03-20T10:28:11 | 2020-03-20T10:28:11 | {} | MEMBER |
ClickHouse | ClickHouse | 601,862,813 | 9,719 | alexey-milovidov | #6044 | 2020-03-20T18:58:53 | 2020-03-20T18:58:53 | {} | MEMBER |
ClickHouse | ClickHouse | 601,597,168 | 9,595 | AlexAkulov | Ну и получается, что эта проблема в версии 20.1.6.30 сохранилась. | 2020-03-20T09:05:23 | 2020-03-20T09:05:23 | {} | NONE |
ClickHouse | ClickHouse | 601,659,963 | 9,775 | alesapin | @hczhcz Take a look, please. | 2020-03-20T11:48:21 | 2020-03-20T11:48:21 | {} | MEMBER |
ClickHouse | ClickHouse | 601,825,671 | 9,776 | den-crane | Drop MT table causes `mark cache dropping` for all tables. Probably the cache invalidation causes queries timeouts and a Distributed table switches to other replica. | 2020-03-20T17:34:24 | 2020-03-20T17:35:37 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 601,862,775 | 9,718 | alexey-milovidov | #6044 | 2020-03-20T18:58:48 | 2020-03-20T18:58:48 | {} | MEMBER |
ClickHouse | ClickHouse | 602,058,329 | 9,794 | blinkov | Lgtm, but can't merge from phone due to conflicts | 2020-03-21T15:11:09 | 2020-03-21T15:11:09 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 602,236,224 | 8,331 | Vxider | @azat Thank you for your review, I have removed all hard code database from tests. Your comments might disappear by a force-push. | 2020-03-22T16:35:50 | 2020-03-22T16:35:50 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 602,291,496 | 2,878 | knoguchi | Tested and worked in 20.3.4.10. It's possibly working since 19.14.3.3
```
localhost :) select Period, CounterId, uniqMerge(Uniq) from temp_view group by Period, CounterId;
SELECT
Period,
CounterId,
uniqMerge(Uniq)
FROM temp_view
GROUP BY
Period,
CounterId
┌──────────────Period─┬─Co... | 2020-03-22T23:11:35 | 2020-03-22T23:11:35 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 602,291,789 | 9,808 | alexey-milovidov | But what will happen if `optimize_skip_unused_shards` is set but the number of shards after filtering is still more than one? | 2020-03-22T23:13:50 | 2020-03-22T23:13:50 | {} | MEMBER |
ClickHouse | ClickHouse | 602,292,634 | 9,809 | alexey-milovidov | The main motivation is to have more fresh version of `clang-tidy`. | 2020-03-22T23:17:37 | 2020-03-22T23:17:37 | {} | MEMBER |
ClickHouse | ClickHouse | 602,560,079 | 9,821 | muyuqianjia | > Could you please also share the textual stack trace from log?
> It is located right after the messages
>
> ```
> BaseDaemon: ########################################
> BaseDaemon: (version 20.1.3.7 (official build)) (from thread 56) (query_id: 85e3b2c2-4c96-49b7-bcd5-ae7d60ef4f2a) Received signal Segmentation f... | 2020-03-23T12:26:50 | 2020-03-23T12:26:50 | {} | NONE |
ClickHouse | ClickHouse | 601,684,204 | 9,778 | filimonov | Clickhouse is not good for a single row lookup. Clickhouse is not key-value storage, it's an analytical database that fills the best in conditions when a wide range of rows need to be processed.
> When NOT to use ClickHouse
> ✕ Transactional workloads (OLTP)
> ✕ Key-value requests with a high rate
> ✕... | 2020-03-20T12:51:24 | 2020-03-20T12:52:30 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 601,902,986 | 7,609 | den-crane | No info. Closing | 2020-03-20T20:44:58 | 2020-03-20T20:44:58 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 601,939,924 | 9,784 | alexey-milovidov | Also missing comment that the method is used in 3rd party code (otherwise the method will be removed because it's unused). | 2020-03-20T22:38:05 | 2020-03-20T22:38:05 | {} | MEMBER |
ClickHouse | ClickHouse | 602,040,521 | 9,795 | simPod | @alexey-milovidov ok, I'll look into it. Do you want to do it in 1 MR? I'll have to explore each platform to provide best solution per each. I'm familiar with few of them. | 2020-03-21T12:58:52 | 2020-03-21T12:58:52 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 602,085,963 | 9,793 | den-crane | Duplicate https://github.com/ClickHouse/ClickHouse/issues/6287
You compare Date VS DateTime
```
SELECT
toUInt32(now()),
toUInt16(today())
┌─toUInt32(now())─┬─toUInt16(today())─┐
│ 1584816796 │ 18342 │
└─────────────────┴───────────────────┘
```
OR `where toDate(TV) BETWEEN toD... | 2020-03-21T18:46:36 | 2020-03-21T19:02:31 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 602,415,425 | 9,814 | qoega | forgot ifdef under Darwin?
```
dbms/CMakeFiles/clickhouse_common_io.dir/src/Common/ThreadFuzzer.cpp.o.d -o dbms/CMakeFiles/clickhouse_common_io.dir/src/Common/ThreadFuzzer.cpp.o -c ../dbms/src/Common/ThreadFuzzer.cpp
../dbms/src/Common/ThreadFuzzer.cpp:117:5: error: use of undeclared identifier 'pthread_mutex_lock_b... | 2020-03-23T06:45:16 | 2020-03-23T06:45:16 | {} | MEMBER |
ClickHouse | ClickHouse | 602,552,833 | 2,078 | dinhthengoc2021998 | Same problem, i cant use sumMerge or anything else when met that error, so how can i deal with that. Could i change the setting just like i did with max_memory_usage? I want to use Array Calculation because it's very fast. Thank you so much. Hope u reply soon!
| 2020-03-23T12:09:16 | 2020-03-23T12:09:16 | {} | NONE |
ClickHouse | ClickHouse | 602,635,355 | 9,822 | tgdkere2 | Thanks for the answer.
In the meantime, is there any way I can flush the cache manually ?
| 2020-03-23T14:37:39 | 2020-03-23T14:37:39 | {} | NONE |
ClickHouse | ClickHouse | 601,807,765 | 9,715 | 4ertus2 | > If prefer_on_column_clashes=column, would the query like below work:
>
> ```
> SELECT sum(revenue) as revenue,
> max(revenue) as max_revenue -- use revenue column
> FROM ....
> ORDER by revenue -- use alias
> ```
To support such queries we need redefinition in scope: use `revenue` as col... | 2020-03-20T16:59:38 | 2020-03-20T17:10:58 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 601,899,032 | 3,942 | den-crane | enable_optimize_predicate_expression enabled by default | 2020-03-20T20:34:42 | 2020-03-20T20:34:42 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 602,061,710 | 9,790 | filimonov | BTW: did anybody test custom prompts with replxx? | 2020-03-21T15:38:27 | 2020-03-21T15:38:27 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 602,290,815 | 9,799 | alexey-milovidov | Squashing is Ok. | 2020-03-22T23:05:20 | 2020-03-22T23:05:20 | {} | MEMBER |
ClickHouse | ClickHouse | 602,620,589 | 9,822 | alexey-milovidov | @oandrew | 2020-03-23T14:12:59 | 2020-03-23T14:12:59 | {} | MEMBER |
ClickHouse | ClickHouse | 601,945,343 | 9,786 | UnamedRus | VALUES works without dot.
```
CREATE TABLE dt
(
timestamp DateTime64(9, 'Europe/Moscow'),
event_id UInt8
) ENGINE = TinyLog
INSERT INTO dt Values (1584009186942540518, 1), ('2019-01-01 00:00:00', 2)
┌─────────────────────timestamp─┬─event_id─┐
│ 2020-03-12 13:33:06.942540518 │ 1 │
│ 2019-01-01... | 2020-03-20T22:59:35 | 2020-03-20T23:11:04 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 602,040,207 | 9,795 | alexey-milovidov | Ok. Could you please also update other languages? (now the diff only contains `fa`). | 2020-03-21T12:55:56 | 2020-03-21T12:55:56 | {} | MEMBER |
ClickHouse | ClickHouse | 602,287,842 | 3,450 | knoguchi | I tested this in ClickHouse version 20.3.4.10, official debian package. It worked fine.
Create the table, mv, and insert new data, then run the following query.
```
localhost :) select date_time, site, sumMerge(lan_byte) lan_byte, countMerge(cnt) as cnt from test.mv group by date_time, site;
SELECT
date... | 2020-03-22T22:40:30 | 2020-03-22T22:42:24 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 602,323,823 | 6,273 | knoguchi | Tested in 20.3.4.10, it works. | 2020-03-23T02:05:16 | 2020-03-23T02:05:16 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 602,327,979 | 448 | knoguchi | It worked in 20.3.4.10. Btw table b needs data before `insert into a` to trigger the MV.
| 2020-03-23T02:25:28 | 2020-03-23T02:25:28 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 601,683,988 | 9,767 | alesapin | https://github.com/ClickHouse/ClickHouse/issues/9775 | 2020-03-20T12:50:56 | 2020-03-20T12:50:56 | {} | MEMBER |
ClickHouse | ClickHouse | 602,008,281 | 9,606 | alesapin | test failure is ok, just need to fix the test.
| 2020-03-21T07:31:44 | 2020-03-21T07:31:44 | {} | MEMBER |
ClickHouse | ClickHouse | 602,082,632 | 9,793 | mbtolou | When remove
`
PARTITION BY toYYYYMM(TV)
`
in
```
CREATE TABLE TESTDB.TAB_DATE100 (
`SERVER_ID` Int32,
`TV` DateTime
) ENGINE = MergeTree ORDER BY (SERVER_ID) SETTINGS index_granularity = 8192;
```
All query work correctly.
also when change **toDate('2020-01-01')** to **toDateTime('2020-01-01 ... | 2020-03-21T18:19:18 | 2020-03-21T18:31:10 | {} | NONE |
ClickHouse | ClickHouse | 602,242,628 | 9,806 | den-crane | >INDEX_TYPE.mrk' doesn't exist,
Recently an index adaptive granularity was introduced. Parts with adaptive granules does not have .mrk files but do have .mrk2 instead. So old versions CH does not able to attach such parts because they does not support index adaptive granularity and .mrk2 marks files. | 2020-03-22T17:22:22 | 2020-03-22T17:22:44 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 602,292,469 | 9,808 | azat | >If we're asking remote servers to process the query to "Complete" stage, we cannot merge their results...
Hm, didn't though about it, so maybe you have ideas except for filtering unused shards in the `getQueryProcessingStage`? | 2020-03-22T23:16:29 | 2020-03-22T23:16:29 | {} | MEMBER |
ClickHouse | ClickHouse | 602,432,024 | 9,801 | filimonov | But distribution in form of parts it will still have old format? | 2020-03-23T07:35:58 | 2020-03-23T07:35:58 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 602,463,334 | 9,817 | blinkov | @qoega "es" docs probably need this patch as well? | 2020-03-23T08:53:05 | 2020-03-23T08:53:05 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 611,788,495 | 1,604 | alexey-milovidov | As `MsgPack` format was merged to master, this issue is fixed. | 2020-04-09T22:41:41 | 2020-04-09T22:41:41 | {} | MEMBER |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.