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 | 721,608,453 | 16,575 | alexey-milovidov | > Testflows check — failed: 369, passed: 1838, other: 22
@vzakaznikov Could you please look what's wrong with Testflows? | 2020-11-04T09:10:28 | 2020-11-04T09:10:28 | {} | MEMBER |
ClickHouse | ClickHouse | 721,675,263 | 16,664 | alexey-milovidov | Also you probably want to switch the commit sha of the submodule. | 2020-11-04T11:23:15 | 2020-11-04T11:23:15 | {} | MEMBER |
ClickHouse | ClickHouse | 721,041,793 | 15,938 | CurtizJ | `00840_long_concurrent_select_and_drop_deadlock` is broken in master. | 2020-11-03T10:49:39 | 2020-11-03T10:49:39 | {} | MEMBER |
ClickHouse | ClickHouse | 721,046,815 | 16,484 | filimonov | > They are almost unusable especially with a new sorting behavior.
1) that 'new sorting behavior' which @den-crane referenced was introduced in 19.14 (see #6054 #6629) - it happens when sorting key and ORDER BY in query matches. Can be disabled with `optimize_read_in_order=0` if needed.
2) the way how clickhouse... | 2020-11-03T11:00:07 | 2020-11-03T11:00:07 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 721,136,986 | 16,636 | alexey-milovidov | Let's also add a test (`tests/queries/0_stateless`) that proves that these functions exist and give correct result up to 3 digits (`round`). | 2020-11-03T14:07:06 | 2020-11-03T14:07:06 | {} | MEMBER |
ClickHouse | ClickHouse | 721,408,008 | 16,636 | alexey-milovidov | 2020-11-03 20:29:23 ld.lld-10: error: undefined symbol: DB::registerFunction**AS**inh(DB::FunctionFactory&)
2020-11-03 20:29:23 >>> referenced by registerFunctionsMath.cpp
2020-11-03 20:29:23 >>> registerFunctionsMath.cpp.o:(DB::registerFunctionsMath(DB::FunctionFactory&)) in archive src/Functions/libcl... | 2020-11-03T22:30:40 | 2020-11-03T22:30:40 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 721,466,273 | 16,653 | den-crane | duplicate https://github.com/ClickHouse/ClickHouse/issues/13958
upgrade to 20.5.5.74
```sql
version 20.5.5
select b, c from tmp.logs order by b, c
┌─b────┬─c─┐
│ aaaa │ 1 │
│ aaaa │ 2 │
│ aaaa │ 2 │
│ aaaa │ 1 │
│ bbbb │ 1 │
│ bbbb │ 1 │
│ bbbb │ 2 │
│ bbbb │ 2 │
└──────┴───┘
``` | 2020-11-04T01:46:59 | 2020-11-04T01:46:59 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 721,600,866 | 16,613 | alexey-milovidov | One test has to be updated. | 2020-11-04T08:56:34 | 2020-11-04T08:56:34 | {} | MEMBER |
ClickHouse | ClickHouse | 721,647,800 | 16,123 | alexey-milovidov | It leads to 6x performance drop even if the argument is not nullable:
```
milovidov-desktop :) SELECT sum(number) FROM system.numbers
SELECT sum(number)
FROM system.numbers
Query id: 2f21a1b4-c469-4a0b-a8ed-e86efe2edd43
Cancelling query.
Ok.
Query was cancelled.
0 rows in set. Elapsed: 2.906 sec. Pro... | 2020-11-04T10:26:53 | 2020-11-04T10:26:53 | {} | MEMBER |
ClickHouse | ClickHouse | 721,667,036 | 16,661 | alexey-milovidov | The following test has failed.
The reference:
```
SELECT
count(),
countOrNull(),
sum(x),
sumOrNull(x)
FROM
(
SELECT number AS x
FROM numbers(10)
WHERE number > 10
)
Query id: 9205a751-3d15-4d1b-8408-5bac06059256
┌─count()─┬─countOrNull()─┬─sum(x)─┬─sumOrNull(x)─┐
│ ... | 2020-11-04T11:05:52 | 2020-11-04T11:06:26 | {} | MEMBER |
ClickHouse | ClickHouse | 721,684,282 | 16,123 | alexey-milovidov | Look at the function properties (from AggregateFunctionFactory) and don't rewrite functions that should not return NULL. | 2020-11-04T11:44:09 | 2020-11-04T11:44:24 | {} | MEMBER |
ClickHouse | ClickHouse | 719,917,075 | 13,862 | UnamedRus | It looks like it was fixed in master and 20.10 release (not fixed in 20.9), so that query is using all available threads.
```
SELECT
t.key_1,
count()
FROM
(
SELECT
intDiv(number, 5000000) AS key_1,
toUInt64(rand() % 75000000) AS key_2
FROM system.numbers_mt
LIMIT 1500000... | 2020-10-31T10:50:31 | 2020-11-03T13:23:09 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 721,319,926 | 1,802 | azat | >We are going to simplify ALIAS processing that will make indices to work.
@alexey-milovidov is this in someones todo list already? | 2020-11-03T19:05:52 | 2020-11-03T19:05:52 | {} | MEMBER |
ClickHouse | ClickHouse | 721,414,269 | 1,802 | alexey-milovidov | No, it's not scheduled. | 2020-11-03T22:48:00 | 2020-11-03T22:48:00 | {} | MEMBER |
ClickHouse | ClickHouse | 721,480,404 | 16,654 | den-crane | You don't need neighbor.
```
SELECT arrayJoin(arrayMap(i -> (toDate('2019-01-01') + toIntervalMonth(i)), range(12))) AS m
┌──────────m─┐
│ 2019-01-01 │
│ 2019-02-01 │
│ 2019-03-01 │
│ 2019-04-01 │
│ 2019-05-01 │
│ 2019-06-01 │
│ 2019-07-01 │
│ 2019-08-01 │
│ 2019-09-01 │
│ 2019-10-01 │
│ 2019-11-01 ... | 2020-11-04T02:43:38 | 2020-11-04T02:43:38 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 721,551,443 | 16,639 | ucasfl | Hi, I want to do this. @alexey-milovidov | 2020-11-04T07:02:50 | 2020-11-04T07:02:50 | {
"rocket": 1
} | COLLABORATOR |
ClickHouse | ClickHouse | 721,626,403 | 16,575 | amosbird | > @amosbird I've edited the changelog entry, is it correct?
Might need to denote that we can only refer to previous ones, similar to how c/c++ find declarations. | 2020-11-04T09:44:45 | 2020-11-04T09:44:45 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 721,678,488 | 7,088 | alexey-milovidov | @amosbird Function `transform` allows to specify simple inline transformation without creating a dictionary.
But currently it does not support all the ClickHouse types. | 2020-11-04T11:30:21 | 2020-11-04T11:30:21 | {} | MEMBER |
ClickHouse | ClickHouse | 721,062,501 | 16,632 | alexey-milovidov | `hypot` can also be added just in case. | 2020-11-03T11:33:39 | 2020-11-03T11:33:39 | {} | MEMBER |
ClickHouse | ClickHouse | 721,169,627 | 16,626 | den-crane | >but when I set the parameter to auto, it didn't use hash method when there was enough memory
It because CH uses a prediction to switch to PARTIAL before memory overusage by HASH. | 2020-11-03T14:54:19 | 2020-11-03T14:54:19 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 721,333,679 | 16,529 | azat | >Functional stateless tests (thread) — fail: 1, passed: 2388, skipped: 14
`00840_long_concurrent_select_and_drop_deadlock` looks flacky | 2020-11-03T19:34:55 | 2020-11-03T19:34:55 | {} | MEMBER |
ClickHouse | ClickHouse | 721,483,515 | 16,654 | den-crane | ```
create table events (a Date, some String) Engine=Memory;
insert into events values('2020-01-01' , 'x'), ('2020-06-01' , 'x'), ('2020-06-15' , 'x'), ('2020-08-15' , 'x');
SELECT toStartOfMonth(a) AS m, sum(c)
FROM ( SELECT a, count() AS c
FROM events
GROUP BY a
ORDE... | 2020-11-04T02:56:13 | 2020-11-04T03:03:19 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 721,499,034 | 16,539 | den-crane | What CH version did you use? | 2020-11-04T04:00:46 | 2020-11-04T04:00:46 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 721,587,747 | 16,635 | alexey-milovidov | @vzakaznikov Please help to debug TestFlows check. | 2020-11-04T08:30:10 | 2020-11-04T08:30:10 | {} | MEMBER |
ClickHouse | ClickHouse | 721,622,980 | 16,486 | alexey-milovidov | Squashing is Ok. | 2020-11-04T09:38:03 | 2020-11-04T09:38:03 | {} | MEMBER |
ClickHouse | ClickHouse | 721,634,136 | 16,636 | hombit | Sorry for the typo. I fixed it and regenerated `../../src/Functions/ya.make` | 2020-11-04T09:59:51 | 2020-11-04T09:59:51 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 721,036,931 | 16,602 | alexey-milovidov | > 01493_alter_remove_properties_zookeeper
| 2020-11-03T10:39:32 | 2020-11-03T10:39:32 | {} | MEMBER |
ClickHouse | ClickHouse | 721,040,689 | 16,597 | alexey-milovidov | It has failed. | 2020-11-03T10:47:30 | 2020-11-03T10:47:30 | {} | MEMBER |
ClickHouse | ClickHouse | 721,065,100 | 16,578 | alexey-milovidov | We usually write our own CMakeLists for libraries. Because original CMakeLists are intended to build a library itself, not to build it for another project. | 2020-11-03T11:39:22 | 2020-11-03T11:39:22 | {} | MEMBER |
ClickHouse | ClickHouse | 721,084,086 | 16,633 | filimonov | ```sql
SELECT
toStartOfMonth(time1) month,
count()
FROM t1
WHERE time1>= toDateTime('2019-01-01 00:00:00') and time1< toDateTime('2020-01-01 00:00:00')
GROUP BY month;
``` | 2020-11-03T12:22:06 | 2020-11-03T12:22:51 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 721,481,699 | 15,998 | den-crane | @winter7
```
<?xml version="1.0" ?>
<yandex>
<users>
<default>
<access_management>1</access_management>
</default>
</users>
</yandex>
```
https://clickhouse.tech/docs/en/operations/access-rights/#enabling-access-control
**By default, SQL-driven access control and account mana... | 2020-11-04T02:49:06 | 2020-11-04T02:49:16 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 721,561,116 | 16,020 | hexiaoting | > @hexiaoting Did you test with MSan?
sorry, I haven't test this query with MSan yesterday.
I try it with MSan today, it failed. | 2020-11-04T07:28:20 | 2020-11-04T07:28:20 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 721,603,314 | 16,613 | alexey-milovidov | arrayJoin of Array(Nothing) can be allowed (and was allowed) as soon as no value of type Nothing appears in the result.
It was already working in some scenarios.
The issue is when joining with empty array appears during the calculation of WITH TOTALS.
But in this case there is nothing special about Array(Nothing... | 2020-11-04T09:01:12 | 2020-11-04T09:01:12 | {} | MEMBER |
ClickHouse | ClickHouse | 721,610,620 | 16,575 | alexey-milovidov | @den-crane Coult you please commit all the tests with a separate PR? | 2020-11-04T09:14:24 | 2020-11-04T09:14:24 | {} | MEMBER |
ClickHouse | ClickHouse | 721,669,424 | 16,123 | alexey-milovidov | It's tricky that the function `count` and `count(DISTINCT ...)` and some other functions have to continue to return zero value when `aggregate_functions_null_for_empty` is enabled (if we aim for SQL compatibility), in contrast to countOrNull, uniqExactOrNull, etc... | 2020-11-04T11:11:06 | 2020-11-04T11:11:32 | {} | MEMBER |
ClickHouse | ClickHouse | 721,677,876 | 16,575 | alexey-milovidov | Ok. | 2020-11-04T11:29:05 | 2020-11-04T11:29:05 | {} | MEMBER |
ClickHouse | ClickHouse | 721,170,020 | 16,622 | alexey-milovidov | What is the point to have LZ4 frame format if it will be wrapped into our own frame format and the final file will be incompatible with LZ4? | 2020-11-03T14:54:33 | 2020-11-03T14:54:33 | {} | MEMBER |
ClickHouse | ClickHouse | 721,341,108 | 16,612 | alesapin | Tests ok (failures are known) | 2020-11-03T19:50:35 | 2020-11-03T19:50:35 | {} | MEMBER |
ClickHouse | ClickHouse | 721,468,826 | 16,605 | bohutang | The problem does exist, but it's not the MySQLHandler/MySQL problem.
When we execute `INSERT INTO default.t1(n) SELECT * FROM numbers(1000) where number % 3`, the packet send to MySQL client is not through the MySQLOutputFormat.cpp:
https://github.com/ClickHouse/ClickHouse/blob/64bd63ca4999297e041fe853c6e4b6a2fa5bd5f... | 2020-11-04T01:57:54 | 2020-11-04T01:57:54 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 721,039,699 | 16,604 | alexey-milovidov | test_reload_zookeeper:
| 2020-11-03T10:45:25 | 2020-11-03T10:45:25 | {} | MEMBER |
ClickHouse | ClickHouse | 721,060,764 | 16,632 | alexey-milovidov | Can be added very easily, see `src/Functions/pow.cpp` | 2020-11-03T11:29:58 | 2020-11-03T11:29:58 | {} | MEMBER |
ClickHouse | ClickHouse | 721,098,844 | 16,610 | zhang2014 | CC: @KochetovNicolai | 2020-11-03T12:55:01 | 2020-11-03T12:55:01 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 721,121,975 | 12,085 | KochetovNicolai | Yes, it was backported to 20.3.13+
Commit: https://github.com/ClickHouse/ClickHouse/commit/f806a7e9f3fceda78f41275455f781fbb4c8df27 | 2020-11-03T13:41:36 | 2020-11-03T13:41:36 | {} | MEMBER |
ClickHouse | ClickHouse | 721,423,734 | 16,559 | alexey-milovidov | I expected that the changelog for version 20.10 will be created :( | 2020-11-03T23:15:20 | 2020-11-03T23:15:20 | {} | MEMBER |
ClickHouse | ClickHouse | 721,424,101 | 16,559 | alexey-milovidov | @abyss7 How is it possible that the bugfix release contains any "Improvements"?
Should check carefully, maybe changelog is incorrect. | 2020-11-03T23:16:28 | 2020-11-03T23:16:28 | {} | MEMBER |
ClickHouse | ClickHouse | 721,604,117 | 16,613 | alexey-milovidov | Also look at the https://github.com/ClickHouse/ClickHouse/pull/14500
It appears that it's just @KochetovNicolai has not completed the fix :) | 2020-11-04T09:02:41 | 2020-11-04T09:02:41 | {} | MEMBER |
ClickHouse | ClickHouse | 721,665,910 | 16,123 | alexey-milovidov | Please cherry-pick this commit: c7618ea
I reordered the settings and removed some of the obsolete settings. | 2020-11-04T11:03:36 | 2020-11-04T11:03:36 | {} | MEMBER |
ClickHouse | ClickHouse | 721,677,521 | 16,636 | alexey-milovidov | ACos, ATan also? | 2020-11-04T11:28:16 | 2020-11-04T11:28:16 | {} | MEMBER |
ClickHouse | ClickHouse | 721,679,824 | 16,539 | alexey-milovidov | @lambgong I have edited the instruction accordingly: https://github.com/ClickHouse/ClickHouse/commit/7986dbdfc7a6b57e9924171b46ca9899beb2ad8c | 2020-11-04T11:33:37 | 2020-11-04T11:33:37 | {} | MEMBER |
ClickHouse | ClickHouse | 721,121,324 | 15,951 | liangli1213 | @TCeason , thanks very much, let me have a try, then back to you. | 2020-11-03T13:40:27 | 2020-11-03T13:40:27 | {} | NONE |
ClickHouse | ClickHouse | 721,147,222 | 16,629 | den-crane | https://github.com/ClickHouse/ClickHouse/pull/13252#issuecomment-667734024 | 2020-11-03T14:25:39 | 2020-11-03T14:25:39 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 721,415,831 | 16,578 | fibersel | > `2020-11-04 01:46:00 -- Using CMake version 3.13.4`
Is seems like after updating base image to version 20.04 cmake is not being updated to 3.16 | 2020-11-03T22:52:24 | 2020-11-03T22:52:24 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 721,484,788 | 16,622 | hczhcz | @alexey-milovidov
LZ4F is slightly faster in decompression on our servers. I am wondering if it actually works.
| 2020-11-04T03:01:12 | 2020-11-04T03:01:12 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 721,569,734 | 16,654 | vegastar002 | yes, this way is wonderful, but it can't count every month ? | 2020-11-04T07:49:49 | 2020-11-04T07:50:21 | {} | NONE |
ClickHouse | ClickHouse | 721,580,312 | 16,639 | alexey-milovidov | This is wonderful, let's do it! | 2020-11-04T08:14:27 | 2020-11-04T08:14:35 | {} | MEMBER |
ClickHouse | ClickHouse | 721,591,497 | 16,622 | alexey-milovidov | > LZ4F is slightly faster in decompression on our servers. I am wondering if it actually works.
It should not be related to the framing format. | 2020-11-04T08:37:44 | 2020-11-04T08:37:44 | {} | MEMBER |
ClickHouse | ClickHouse | 721,063,346 | 5,326 | alexey-milovidov | But it's already fixed in version 20.10.
Just enable `insert_quorum_parallel`. | 2020-11-03T11:35:33 | 2020-11-03T11:35:33 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 721,139,760 | 16,629 | xjewer | there's an umbrella ticket covering TTL issues https://github.com/ClickHouse/ClickHouse/issues/10128
tl;dr 19.16 version is quite old for it, consider an upgrade. | 2020-11-03T14:12:03 | 2020-11-03T14:13:48 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 721,415,874 | 16,020 | alexey-milovidov | @hexiaoting Did you test with MSan? | 2020-11-03T22:52:32 | 2020-11-03T22:52:32 | {} | MEMBER |
ClickHouse | ClickHouse | 721,495,106 | 13,252 | kekekedeng | @CurtizJ How is the progress? Seems waiting for review? | 2020-11-04T03:43:37 | 2020-11-04T03:43:37 | {} | NONE |
ClickHouse | ClickHouse | 721,566,008 | 16,648 | alexey-milovidov | This change will impose backward incompatibility when different ClickHouse versions run on the same cluster (uniqExact values will double). We should either postpone it until #12552 or highlight the changes very carefully. | 2020-11-04T07:40:40 | 2020-11-04T07:40:40 | {} | MEMBER |
ClickHouse | ClickHouse | 721,681,416 | 16,647 | alexey-milovidov | This PR does not shed any light on the byte order :( | 2020-11-04T11:37:16 | 2020-11-04T11:37:16 | {} | MEMBER |
ClickHouse | ClickHouse | 721,063,600 | 15,998 | winter7 | Could you tell me how to create a user with grant option? | 2020-11-03T11:36:06 | 2020-11-03T11:36:06 | {} | NONE |
ClickHouse | ClickHouse | 721,204,184 | 16,265 | liangli1213 | So far, it is hard to reproduce this issue, but I use some workaround to by pass it, it seems only decimal type data has such problem. | 2020-11-03T15:36:31 | 2020-11-03T15:36:31 | {} | NONE |
ClickHouse | ClickHouse | 721,408,071 | 16,636 | alexey-milovidov | https://clickhouse-test-reports.s3.yandex.net/16636/7d363d26954035f8361bf5cffffa019a922d7d5f/style_check/output.txt | 2020-11-03T22:30:53 | 2020-11-03T22:30:53 | {} | MEMBER |
ClickHouse | ClickHouse | 721,418,829 | 8,772 | alexey-milovidov | I have the following numbers on my machine on version 20.11:
```
select d, count() from a group by d;
0.091 sec.
select os, count() from a group by os;
0.353 sec.
select d,os, count() from a group by d,os;
0.703 sec.
```
It means that the issue is already fixed. | 2020-11-03T23:00:56 | 2020-11-03T23:01:10 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 721,611,071 | 16,575 | alexey-milovidov | @amosbird I've edited the changelog entry, is it correct? | 2020-11-04T09:15:15 | 2020-11-04T09:15:15 | {} | MEMBER |
ClickHouse | ClickHouse | 721,669,905 | 16,123 | alexey-milovidov | It is controlled by `AggregateFunctionProperties`. | 2020-11-04T11:12:10 | 2020-11-04T11:12:10 | {} | MEMBER |
ClickHouse | ClickHouse | 721,142,777 | 15,998 | den-crane | > Could you tell me how to create a user with grant option?
with grant option
https://clickhouse.tech/docs/en/sql-reference/statements/grant/#grant-privigele-syntax
`grant all on *.* to bird with grant option` | 2020-11-03T14:17:32 | 2020-11-03T14:17:41 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 721,409,636 | 16,637 | azat | >If it's performance improvement, why "bug fix"?
@alexey-milovidov It is not a performance improvement, but lack of description for this PR lead to this though I guess.
It was before but I had to remove it, because the initial "fix" was wrong.
And I just pushed updated patch, and it should contains the correct fix... | 2020-11-03T22:35:05 | 2020-11-03T22:35:05 | {} | MEMBER |
ClickHouse | ClickHouse | 721,605,365 | 16,613 | alexey-milovidov | @KochetovNicolai You forgot to test the case https://github.com/ClickHouse/ClickHouse/pull/14500#pullrequestreview-483010299
Totals should simply disappear when joining with empty arrays. | 2020-11-04T09:04:50 | 2020-11-04T09:04:50 | {} | MEMBER |
ClickHouse | ClickHouse | 721,606,949 | 16,586 | alexey-milovidov | Also the reference for the `01458_named_tuple_millin` test should be updated. | 2020-11-04T09:07:43 | 2020-11-04T09:07:43 | {} | MEMBER |
ClickHouse | ClickHouse | 721,037,590 | 16,614 | alexey-milovidov | > 00840_long_concurrent_select_and_drop_deadlock
- flaky.
> Yandex synchronization check (only for Yandex employees)
Arcadia tests does not support remote connections. | 2020-11-03T10:40:48 | 2020-11-03T10:40:48 | {} | MEMBER |
ClickHouse | ClickHouse | 721,042,353 | 16,593 | alexey-milovidov | We decided that this implementation of statistical tests will be mostly unusable.
We should implement them as an aggregate function with two arguments:
Float64 or any other numeric type - value of the element of the sample.
UInt8 - a number to determine between two samples (0 - first sample, non-zero - second samp... | 2020-11-03T10:50:47 | 2020-11-03T10:51:13 | {} | MEMBER |
ClickHouse | ClickHouse | 721,118,015 | 16,610 | KochetovNicolai | > I'm not sure if I should fix this. because the second argument type is not UInt8
Agree. It is expected. | 2020-11-03T13:33:49 | 2020-11-03T13:33:49 | {} | MEMBER |
ClickHouse | ClickHouse | 721,167,743 | 16,625 | dependabot-preview[bot] | OK, I won't notify you again about this release, but will get in touch when a new version is available. If you'd rather skip all updates until the next major or minor version, let me know by commenting `@dependabot ignore this major version` or `@dependabot ignore this minor version`.
If you change your mind, just re-... | 2020-11-03T14:53:19 | 2020-11-03T14:53:19 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 721,251,512 | 4,679 | johnprakashgithub | Seeing a similar issue in my environment
↘ Progress: 7.52 million rows, 1.53 GB (39.74 thousand rows/s., 8.11 MB/s.) ████████████████████▋ 10%
Exc... | 2020-11-03T16:54:47 | 2020-11-03T17:04:29 | {} | NONE |
ClickHouse | ClickHouse | 721,471,441 | 16,633 | vegastar002 | thank u , it's useful.
but I want use neighbor() to get it.
eg:
SELECT
d2 AS start,
neighbor(d2, -1) AS end
FROM
(
WITH toDate('2020-04-01') AS d1
SELECT toStartOfMonth(d1 - (number * 32)) AS d2
FROM numbers(16)
)
┌──────start─┬────────end─┐
│ 2020-04-01 │ 1970-01-01 │
│ 2020-02-01... | 2020-11-04T02:08:07 | 2020-11-04T02:08:07 | {} | NONE |
ClickHouse | ClickHouse | 721,677,417 | 16,123 | ucasfl | > It's tricky that the function `count` and `count(DISTINCT ...)` and some other functions have to continue to return zero value when `aggregate_functions_null_for_empty` is enabled (if we aim for SQL compatibility), in contrast to countOrNull, uniqExactOrNull, etc...
So, what is the best way to resolve this? | 2020-11-04T11:28:01 | 2020-11-04T11:28:01 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 721,202,686 | 15,951 | liangli1213 | The issue has been resolved. | 2020-11-03T15:34:10 | 2020-11-03T15:34:10 | {} | NONE |
ClickHouse | ClickHouse | 721,227,563 | 16,424 | filimonov | > To be honest, I don't find this feature appealing.
Yes, it's not very common to have a bottleneck for merges on the CPU side: it runs on single-core, and normally it IO bounded.
It was requested by the following use case:
1) performant disks (NVMe)
2) hundreds of replicated tables (~1000), 3 replicas
3) ins... | 2020-11-03T16:14:35 | 2020-11-03T16:17:52 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 721,479,159 | 15,998 | winter7 | > grant all on \*.* to bird with grant option
I don't have any user with enough privileges to execute this query. I wonder how to create the first user with grant option.
```
Code: 497. DB::Exception: Received from localhost:9000. DB::Exception: default: Not enough privileges. To execute this query it's necessary ... | 2020-11-04T02:38:25 | 2020-11-04T02:40:03 | {} | NONE |
ClickHouse | ClickHouse | 721,586,874 | 16,540 | ghost | @l1t1
My plan is
```SQL
EXPLAIN
SELECT
action1,
uniqMerge(uv) AS uv,
uniqMerge(session) AS session
FROM test_merge
GROUP BY action1
ORDER BY uv DESC
LIMIT 10
FORMAT PrettySpace
explain
Expression (Projection) ... | 2020-11-04T08:28:18 | 2020-11-04T08:40:33 | {} | NONE |
ClickHouse | ClickHouse | 721,623,768 | 7,088 | amosbird | > for function `transform`
Hmm, what's the purpose of this one? I feel that dictGet / joinGet is equivalent. | 2020-11-04T09:39:35 | 2020-11-04T09:39:35 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 721,419,760 | 16,624 | alexey-milovidov | Why table should live in ram? If it's a tiny table, maybe `StripeLog` will suffice? | 2020-11-03T23:03:30 | 2020-11-03T23:03:38 | {} | MEMBER |
ClickHouse | ClickHouse | 721,466,903 | 8,772 | sundy-li | > It means that the issue is already fixed.
After verification in version 20.11, got an impressive improvement. | 2020-11-04T01:49:36 | 2020-11-04T01:49:36 | {
"hooray": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 721,502,441 | 16,578 | azat | >Is seems like after updating base image to version 20.04 cmake is not being updated to 3.16
Yeah, this is because fasttest image is not updated on per-PR bases (as most of the images) | 2020-11-04T04:16:19 | 2020-11-04T04:16:37 | {} | MEMBER |
ClickHouse | ClickHouse | 721,038,709 | 16,604 | alexey-milovidov | > test_multiple_disks
- bad test, @excitoon
> test_reload_zookeeper
- unknown, @alesapin
> 00840_long_concurrent_select_and_drop_deadlock
- flaky.
> 01493_alter_remove_properties_zookeeper
.
Compact data parts has been enabled by default since version 20.10. | 2020-11-03T18:43:59 | 2020-11-03T18:43:59 | {} | MEMBER |
ClickHouse | ClickHouse | 721,348,684 | 1,463 | alexey-milovidov | How to reproduce:
```
SELECT quantileMerge(arrayJoin(arrayMap(x -> state, range(10000)))) FROM (SELECT quantileState(rand()) AS state FROM numbers(10000))
``` | 2020-11-03T20:06:54 | 2020-11-03T20:06:54 | {} | MEMBER |
ClickHouse | ClickHouse | 721,471,193 | 15,994 | spihiker | Update to version(20.8.5.45) ,It's ok. | 2020-11-04T02:07:08 | 2020-11-04T02:07:08 | {} | NONE |
ClickHouse | ClickHouse | 721,598,801 | 16,540 | ghost | @l1t1
After inserting data of another partition into `test_merge`, ClickHouse uses multi-core for the query. | 2020-11-04T08:52:27 | 2020-11-04T08:53:02 | {} | NONE |
ClickHouse | ClickHouse | 721,611,828 | 16,570 | alexey-milovidov | Need to add a test (`tests/queries/0_stateless`) that will guarantee that `farmFingerprint64` won't change in future. | 2020-11-04T09:16:44 | 2020-11-04T09:16:44 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 721,624,539 | 10,894 | Weilor | The problem is same with mine. Is there any progress? | 2020-11-04T09:41:09 | 2020-11-04T09:41:09 | {} | NONE |
ClickHouse | ClickHouse | 721,711,123 | 16,666 | alexey-milovidov | `WITH [1, 2, 3] AS a, ['a', 'b', 'c'] AS b SELECT arrayJoin(arrayZip(a, b))` | 2020-11-04T12:46:01 | 2020-11-04T12:46:01 | {
"laugh": 1
} | MEMBER |
ClickHouse | ClickHouse | 721,804,470 | 16,176 | alexey-milovidov | @hexiaoting The point of these settings is that they should not affect parser - they should not interfere SQL language in any way. | 2020-11-04T15:36:21 | 2020-11-04T15:36:21 | {} | MEMBER |
ClickHouse | ClickHouse | 721,804,819 | 16,176 | alexey-milovidov | > But how to add simple building blocks into ClickHouse？
To implement pagination in a safe way when query is given by the user. | 2020-11-04T15:37:01 | 2020-11-04T15:37:01 | {} | MEMBER |
ClickHouse | ClickHouse | 721,811,840 | 9,537 | alexey-milovidov | No more deadlocks possible with table-level locks.
Also we have removed locks from DROP PARTITION. | 2020-11-04T15:49:06 | 2020-11-04T15:49:06 | {} | MEMBER |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.