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 | 431,289,580 | 3,423 | blinkov | @sundy-li @zhang2014 basically I just did what has been discussed in other PRs, but maybe you'll have any further suggestions/comments | 2018-10-19T08:39:56 | 2018-10-19T08:39:56 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 431,442,265 | 3,413 | artfabrique | > it's too complicated to debug via public comments without access to real ClickHouse cluster
Yeah, I know :( I thought maybe there are some fundamental mistakes it the query and maybe somebody will have an idea of some radically different approach. For example, pre-generate something when we collect data etc. | 2018-10-19T17:44:00 | 2018-10-19T17:44:09 | {} | NONE |
ClickHouse | ClickHouse | 431,616,024 | 3,411 | denisov-vlad | We are having the same issue.
ch 18.14.9 / ubuntu 18.04. | 2018-10-20T20:34:50 | 2018-10-20T20:34:50 | {} | NONE |
ClickHouse | ClickHouse | 431,813,254 | 3,432 | ztlpn | duplicate of #3433 | 2018-10-22T11:55:29 | 2018-10-22T11:55:29 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 432,033,449 | 3,410 | den-crane | t.me/clickhouse_ru/71285
ch0 :) select eventDate, count() from Events where eventDate = today()-1 group by eventDate **settings compile_expressions = 0;**
┌──eventDate─┬──count()─┐
│ 2018-10-22 │ 70056042 │
└────────────┴──────────┘
1 rows in set. Elapsed: 0.069 sec. Processed 70.06 million rows, ... | 2018-10-23T00:00:01 | 2018-10-23T00:01:43 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 432,043,784 | 3,436 | amosbird | also, `Field f = 1;` won't work as this pr is only used for copy assignments, not for constructions. Do you also want it also be implicitly constructable? | 2018-10-23T00:53:39 | 2018-10-23T00:54:51 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 432,045,330 | 3,436 | amosbird | and yes, there are tons of construction call sites suitable for simplification https://la.wentropy.com/1_lp | 2018-10-23T01:01:24 | 2018-10-23T01:01:24 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 430,867,739 | 3,408 | zhang2014 | @Bolinov makes it easier to read after commits. Because most of the content is now unreadable. | 2018-10-18T03:59:38 | 2018-10-18T04:01:23 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 427,866,336 | 3,310 | healiseu | The reason I am interested in `Clickhouse Join Engine`, `Dictionary Engine` and other related Clickhouse features is that I have developed and demonstrated a technique where you can extrapolate a hypergraph from any set of tables that share common columns. I called the technique [Associative-Semiotic-Hypergraph][1] and... | 2018-10-08T14:54:04 | 2018-10-18T06:39:09 | {} | NONE |
ClickHouse | ClickHouse | 431,047,675 | 2,156 | VaBezruchko | I reproduce same issue on 18.14.9 version with select query .
Error:
Received exception from server (version 18.14.9):
Code: 171. DB::Exception: Received from 127.0.0.1:9000. DB::Exception: Block structure mismatch in UNION stream: different names of columns:
But removing the limit section make it works. | 2018-10-18T15:12:08 | 2018-10-18T15:12:08 | {} | NONE |
ClickHouse | ClickHouse | 431,347,171 | 3,417 | lamberken | @sundy-li , @zhang2014, @blinkov thanks for review and suggestion. :+1: | 2018-10-19T12:31:27 | 2018-10-19T12:31:27 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 431,382,928 | 3,413 | artfabrique | @blinkov above is the query that we are using. | 2018-10-19T14:30:43 | 2018-10-19T14:30:43 | {} | NONE |
ClickHouse | ClickHouse | 431,448,146 | 3,410 | den-crane | select count(*) from dw.testx where not a and t < (select tkey from dt);
┌─count()─┐
│ 1 │
└─────────┘
1 rows in set. Elapsed: 0.002 sec.
**set compile_expressions = 0;**
select count(*) from dw.testx where not a and t < (select tkey from dt);
0 rows in set. Elapsed: 0.002 sec. | 2018-10-19T18:01:57 | 2018-10-19T18:01:57 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 431,451,135 | 3,410 | den-crane | In our system we don't drop/create dt.
It's a replacingmergetree table and it's data is different for every query.
And more complicated
_time_key < (select time_key from dim_time where time_stamp = (select max(last_record_time) from data_status_current))_
dim_time is constant, data_status_current has different max... | 2018-10-19T18:12:10 | 2018-10-19T18:12:10 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 432,103,912 | 3,435 | StratoCasterFen | календарь - просто список дат попавших в требуемый период. | 2018-10-23T05:59:06 | 2018-10-23T06:26:12 | {} | NONE |
ClickHouse | ClickHouse | 432,155,991 | 3,139 | alexey-milovidov | Do you know something similar in Matlab, R, Python, ...? | 2018-10-23T08:48:40 | 2018-10-23T08:48:40 | {} | MEMBER |
ClickHouse | ClickHouse | 431,118,571 | 3,418 | blinkov | ```
SET date_time_input_format = 'best_effort'
Ok.
```
```
DESCRIBE TABLE dt
FORMAT TSV
x DateTime
```
```
:) insert into dt(x) values ('2018-10-18T23:01:26+05:00');
INSERT INTO dt (x) VALUES
Ok.
```
```
SELECT *
FROM dt
FORMAT TSV
2018-10-18 21:01:26 | 2018-10-18T18:42:24 | 2018-10-18T18:42:24 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 431,119,083 | 3,417 | blinkov | @lamber-ken <strike>now PR shows that it has way too many unrelated changes</strike> UPD: some bug, now looks ok, @sundy-li @zhang2014 can you take a look please? | 2018-10-18T18:44:01 | 2018-10-18T18:46:06 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 431,572,641 | 3,420 | YuryBolkonsky | Same error. Please fix.
It is a problem to use enable_optimize_predicate_expression with DataGrip or connection through python not native clickhouse client | 2018-10-20T11:36:12 | 2018-10-20T11:36:12 | {} | NONE |
ClickHouse | ClickHouse | 431,937,290 | 3,436 | alexey-milovidov | The only non-obvious part is that when you write something like
Field f = 1;
Int64 value will be created,
but it can be transparently inserted into ColumnUInt64. | 2018-10-22T18:57:03 | 2018-10-22T18:57:03 | {} | MEMBER |
ClickHouse | ClickHouse | 431,971,733 | 3,304 | alexey-milovidov | If data loss happens (for example, data was lost on all replicas; or if data was lost on a single replica just after insertion due to hardware failure), ClickHouse will write it to log:
```
LOG_ERROR(log, "Part " << part_name << " is lost forever.");
ProfileEvents::increment(ProfileEvents::ReplicatedDataLoss... | 2018-10-22T20:27:29 | 2018-10-22T20:29:25 | {} | MEMBER |
ClickHouse | ClickHouse | 432,134,316 | 3,436 | alexey-milovidov | > If you approve this refactor process, I'll finish the left parts, which contains hundreds of changes.
Yes, I approve. | 2018-10-23T07:48:13 | 2018-10-23T07:48:13 | {} | MEMBER |
ClickHouse | ClickHouse | 431,024,554 | 3,413 | panyushkins | > @artfabrique can you show this 3-minutes-long query?
Hello!
SELECT
count() as trials,
uniq(uid) as users,
n_step,
step
FROM (
WITH
groupArray(_interface) as event_names,
groupArray(toInt32(date)) as event_dates,
arrayJoin(arrayFilter((... | 2018-10-18T14:12:08 | 2018-10-18T14:13:49 | {} | NONE |
ClickHouse | ClickHouse | 431,247,051 | 3,417 | blinkov | > @blinkov I have a question, should we keep the `[Original Artical](https://clickhouse.yandex/docs/en/data_types/boolean/) <!--hide-->` links in zh docs?
@sundy-li It's supposed to be a link to where the rendered version of current article is published. So you can translate the words in `[...]` and replace `/en/` w... | 2018-10-19T05:14:24 | 2018-10-19T05:14:24 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 431,332,966 | 3,346 | blinkov | @sundy-li @zhang2014 this PR touches few articles that are already have Chinese versions | 2018-10-19T11:30:12 | 2018-10-19T11:30:12 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 431,540,605 | 3,427 | sdlcwangsong | thank you | 2018-10-20T02:11:03 | 2018-10-20T02:11:03 | {} | NONE |
ClickHouse | ClickHouse | 431,933,265 | 3,439 | alexey-milovidov | Ok, but I don't understand how it is related to jemalloc
(it is related to glibc and libressl) | 2018-10-22T18:47:05 | 2018-10-22T18:47:05 | {} | MEMBER |
ClickHouse | ClickHouse | 430,972,116 | 3,405 | alesapin | failed test is stateless.00411_long_accurate_number_comparison | 2018-10-18T11:23:39 | 2018-10-18T11:23:39 | {} | MEMBER |
ClickHouse | ClickHouse | 431,341,104 | 3,427 | ztlpn | Hi, this message means that the update succeeded, but updated data parts for different replicas were not byte-identical. So one replica downloaded the updated part from another. There can be several reasons for this: different compression settings for different replicas; non-deterministic UPDATE expression or MATERIALI... | 2018-10-19T12:06:48 | 2018-10-19T12:06:55 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 431,481,203 | 3,410 | alesapin | Can you provide `select * from system.asynchronous_metrics` ? | 2018-10-19T19:56:10 | 2018-10-19T19:56:10 | {} | MEMBER |
ClickHouse | ClickHouse | 431,578,857 | 3,420 | YuryBolkonsky | Fixed. Add <enable_optimize_predicate_expression>0</enable_optimize_predicate_expression> to users.xml.
But i think this setting should be 0 by default | 2018-10-20T13:03:39 | 2018-10-20T13:03:39 | {} | NONE |
ClickHouse | ClickHouse | 431,816,783 | 3,435 | ztlpn | Давайте пример таблицы, данных в ней, и что хотите получить в итоге. Попробуем придумать запрос. | 2018-10-22T12:12:44 | 2018-10-22T12:12:44 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 431,883,006 | 3,304 | igor-sh8 | Hi.
Maybe our problem looks like this bug fix description in 18.14.9 version:
`Если после слияния кусков данных, у результирующего куска чексумма отличается от результата того же слияния на другой реплике, то результат слияния удаляется, и вместо этого кусок скачивается с другой реплики (это правильное поведение). Н... | 2018-10-22T16:14:25 | 2018-10-22T16:14:25 | {} | NONE |
ClickHouse | ClickHouse | 431,282,305 | 3,408 | blinkov | There was some weird issue on merge, I'll revert and fix it separately. | 2018-10-19T08:13:01 | 2018-10-19T08:13:15 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 431,585,862 | 3,420 | zhang2014 | Now it's closed in the master, Sorry to open it too optimistically.
https://github.com/yandex/ClickHouse/blob/master/CHANGELOG_RU.md#%D0%9E%D0%B1%D1%80%D0%B0%D1%82%D0%BD%D0%BE-%D0%BD%D0%B5%D1%81%D0%BE%D0%B2%D0%BC%D0%B5%D1%81%D1%82%D0%B8%D0%BC%D1%8B%D0%B5-%D0%B8%D0%B7%D0%BC%D0%B5%D0%BD%D0%B5%D0%BD%D0%B8%D1%8F-1 | 2018-10-20T14:19:17 | 2018-10-20T14:19:17 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 431,949,810 | 3,441 | alexey-milovidov | Let's rename `keep_alive_timeout` to `tcp_keep_alive_timeout` everywhere to avoid potential confusion with HTTP Keep-Alive. | 2018-10-22T19:30:12 | 2018-10-22T19:30:12 | {} | MEMBER |
ClickHouse | ClickHouse | 431,968,207 | 3,304 | alexey-milovidov | No, it could not happen.
Replacing of data parts to the merged or fetched part is completely atomic.
If merge or fetch of data part has failed, the set of data parts will remain in previous state. | 2018-10-22T20:18:14 | 2018-10-22T20:18:14 | {} | MEMBER |
ClickHouse | ClickHouse | 432,155,738 | 3,139 | alexey-milovidov | We can merge it if we will find appropriate name.
`boundingBoxRatio`, `boundingRatio`, `boundaryRatio`, `minmaxSlope`... | 2018-10-23T08:48:02 | 2018-10-23T08:48:02 | {} | MEMBER |
ClickHouse | ClickHouse | 430,706,598 | 1,448 | SaltTan | That was the only time I saw corrupted data in system.columns.
But I have never tried to reproduce this, and I don't manipulate partitions very often.
We are on 1.1.54385 at the moment, 18.12 is in testing.
| 2018-10-17T16:56:10 | 2018-10-17T16:56:10 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 430,857,786 | 3,410 | den-crane | Workaround: move all other conditions to prewhere and leave the condition with subquery in where section.
select count(*) from testx
prewhere not a
where t < (select tkey from dt); | 2018-10-18T02:53:34 | 2018-10-18T02:54:15 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 431,476,515 | 3,410 | alesapin | Den, I can't reproduce the problem from your example:
My steps:
```
:) CREATE TABLE dt(tkey Int32) ENGINE = MergeTree order by tuple();
Ok.
0 rows in set. Elapsed: 0.002 sec.
:) insert into dt values (300000);
Ok.
1 rows in set. Elapsed: 0.002 sec.
:) CREATE TABLE testx(t Int32, a UInt8) ENGINE = Me... | 2018-10-19T19:40:50 | 2018-10-19T19:40:50 | {} | MEMBER |
ClickHouse | ClickHouse | 431,498,803 | 3,410 | alesapin | https://github.com/yandex/ClickHouse/pull/3430/files hotfix | 2018-10-19T21:05:47 | 2018-10-19T21:05:47 | {} | MEMBER |
ClickHouse | ClickHouse | 431,825,304 | 275 | ztlpn | It is [the limitation of the Poco configuration reader](https://github.com/ClickHouse-Extras/poco/blob/d7a4383c4d85b51938b62ed5812bc0935245edb3/Util/include/Poco/Util/XMLConfiguration.h#L76). It needs some character to act as the delimiter of hierarchy levels and it is thus forbidden in the configuration keys. The easi... | 2018-10-22T12:47:28 | 2018-10-22T12:47:28 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 431,936,913 | 3,436 | alexey-milovidov | I've added the missing comment. | 2018-10-22T18:56:06 | 2018-10-22T18:56:06 | {} | MEMBER |
ClickHouse | ClickHouse | 432,146,086 | 3,400 | alexey-milovidov | Thank you! | 2018-10-23T08:21:54 | 2018-10-23T08:21:54 | {} | MEMBER |
ClickHouse | ClickHouse | 430,837,106 | 3,141 | tungdam | I have quite the same symptom with 18.12.17
Here's my insert query ( simply using clickhouse-client cli from a remote host )
`clickhouse-client --config-file my_config.xml --port 9500 -d my_db -h my_host --compression True --max_memory_usage 21474836480 -q 'INSERT INTO distributed_events FORMAT Native'`
The er... | 2018-10-18T00:45:44 | 2018-10-18T01:01:40 | {} | NONE |
ClickHouse | ClickHouse | 431,349,936 | 3,346 | zhang2014 | Ok. I'll do this task : ) | 2018-10-19T12:42:38 | 2018-10-19T12:43:41 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 431,483,141 | 3,410 | den-crane | yes, I think optimize_move_to_prewhere 1
transform it to
select count(*) from testx prewhere t <(subquery) where not a | 2018-10-19T20:03:53 | 2018-10-19T20:03:53 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 430,559,303 | 1,432 | RoyBellingan | Hi, you are right date is a materialized column
date Date MATERIALIZED toDate(timestamp)
Maybe that is the reason...
@George3d6 | 2018-10-17T09:34:08 | 2018-10-17T09:34:08 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 431,320,647 | 3,425 | proller | duplicate https://github.com/yandex/ClickHouse/issues/3415 | 2018-10-19T10:36:16 | 2018-10-19T10:36:16 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 431,480,371 | 3,410 | den-crane | select name, value from system.settings where changed;
┌─name──────────────────────────────────────┬─value────────────┐
│ connect_timeout_with_failover_ms │ 1000 │
│ use_uncompressed_cache │ 1 │
**│ optimize_move_to_prewhere │ 0** ... | 2018-10-19T19:53:01 | 2018-10-19T19:53:01 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 431,482,274 | 3,410 | alesapin | yes, reproduced with:
```
optimize_move_to_prewhere 0
enable_optimize_predicate_expression 0
``` | 2018-10-19T20:00:18 | 2018-10-19T20:00:18 | {} | MEMBER |
ClickHouse | ClickHouse | 431,848,073 | 3,437 | Tasselmi | thank you a lot. I am a newer in big data. SO much thanks for your warm heart help.
Github 's server is not stable~~ I offten get a 404 page today. | 2018-10-22T14:15:35 | 2018-10-22T14:15:35 | {} | NONE |
ClickHouse | ClickHouse | 431,888,926 | 1,848 | digitalist | Для меня - нет. Я пока на линуксы вернулся. Вряд ли будет время проверить на Фре в ближайшем будущем | 2018-10-22T16:32:40 | 2018-10-22T16:32:40 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 431,937,691 | 3,436 | alexey-milovidov | Shouldn't you prove this approach by removing explicit casts across the code? | 2018-10-22T18:58:09 | 2018-10-22T18:58:09 | {} | MEMBER |
ClickHouse | ClickHouse | 431,119,901 | 3,418 | chipitsine | я пробовал в clickhouse-local, завтра еще попробую, отпишусь | 2018-10-18T18:46:39 | 2018-10-18T18:46:39 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 431,481,493 | 3,410 | den-crane | ```
SELECT *
FROM system.asynchronous_metrics
┌─metric──────────────────────────────────┬──────value─┐
│ jemalloc.background_thread.run_interval │ 0 │
│ jemalloc.background_thread.num_runs │ 0 │
│ jemalloc.background_thread.num_threads │ 0 │
│ jemalloc.retained ... | 2018-10-19T19:57:21 | 2018-10-19T19:57:21 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 431,483,945 | 3,410 | alesapin | Yes. This problem seems to be very special case. Сondition `t < (subquery)` executes only when subquery returns single element. | 2018-10-19T20:07:12 | 2018-10-19T20:07:28 | {} | MEMBER |
ClickHouse | ClickHouse | 431,966,793 | 3,434 | daledude | Thank you @vavrusa! My msg loss is unrelated then. | 2018-10-22T20:14:22 | 2018-10-22T20:14:22 | {} | NONE |
ClickHouse | ClickHouse | 430,497,926 | 1,788 | alexey-milovidov | Now it works perfectly. | 2018-10-17T05:49:00 | 2018-10-17T05:51:14 | {} | MEMBER |
ClickHouse | ClickHouse | 430,513,124 | 3,399 | zhang2014 | Done. | 2018-10-17T07:02:13 | 2018-10-17T07:02:13 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 427,839,653 | 3,310 | healiseu | @filimonov I take your points, I have just started getting familiar with clickhouse DBMS but I like a lot their architectural design and user features, that's why I am here after all trying to learn more. From the user point of view I think **what I am missing is some better connection of clickhouse tools, i.e. buildin... | 2018-10-08T13:46:45 | 2018-10-17T21:27:27 | {} | NONE |
ClickHouse | ClickHouse | 428,178,463 | 3,310 | healiseu | Thanks, I will let you know... it's all about indexing and playing with references | 2018-10-09T12:48:33 | 2018-10-18T06:42:01 | {} | NONE |
ClickHouse | ClickHouse | 431,020,162 | 3,417 | blinkov | Looks like you haven't merged master to your branch for a while and your PR does not show diff from current translation...
cc @sundy-li @zhang2014 | 2018-10-18T14:00:58 | 2018-10-18T14:00:58 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 431,036,433 | 3,417 | lamberken | I'm trying to solve the conflicts. | 2018-10-18T14:43:24 | 2018-10-18T14:43:24 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 431,442,034 | 3,413 | panyushkins | > @artfabrique it looks roughly as I expected, but frankly speaking it's too complicated to debug via public comments without access to real ClickHouse cluster. Does it look CPU bound? If it does, what do you see in profiler? What kind of hardware do you use for this test?
I will answer at Monday, when I got this in... | 2018-10-19T17:43:08 | 2018-10-19T17:43:08 | {} | NONE |
ClickHouse | ClickHouse | 431,481,239 | 3,410 | den-crane | 
| 2018-10-19T19:56:16 | 2018-10-19T19:56:16 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 431,948,463 | 3,434 | vavrusa | This just means that Kafka consumer reached EOF for given partition. Background consumer uses this as a signal for closing current batch, instead of waiting for timeout. It's useful for tracing/debugging (which is why it's emitted at trace level). | 2018-10-22T19:26:39 | 2018-10-22T19:26:39 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 430,712,442 | 2,427 | arthurdk | Hello, are you still planning on working on that feature? It would be really awesome :smiley: | 2018-10-17T17:13:57 | 2018-10-17T17:13:57 | {
"+1": 1
} | NONE |
ClickHouse | ClickHouse | 430,910,802 | 3,162 | mdonkers | We are experiencing a similar issue, using the same ClickHouse version (18.12.17).
We are running a multi-tenant setup with separate databases, all having the same tables. The issue occurs on different databases, but thus far only with the below specific query (according to our logs).
We have checked local / dist... | 2018-10-18T07:52:22 | 2018-10-18T07:57:23 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 430,913,120 | 2,567 | mdonkers | Closing issue, was due to the way the `select` statement was constructed.
Thanks for the assistance. | 2018-10-18T08:00:10 | 2018-10-18T08:00:10 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 430,937,681 | 3,413 | sundy-li | This is a more complex funnel demand than windowFunnel , but I am looking forward to see your current query in solving this problem. | 2018-10-18T09:20:06 | 2018-10-18T09:20:06 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 431,212,319 | 3,417 | sundy-li | @blinkov I have a question, should we keep the `[Original Artical](https://clickhouse.yandex/docs/en/data_types/boolean/) <!--hide-->` links in zh docs? | 2018-10-19T01:07:15 | 2018-10-19T01:07:15 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 431,333,153 | 3,426 | blinkov | @BayoNet merge master into this branch, please | 2018-10-19T11:31:06 | 2018-10-19T11:31:06 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 431,825,527 | 3,433 | ztlpn | Yes, looks like a transient network error. | 2018-10-22T12:48:24 | 2018-10-22T12:48:24 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 431,916,704 | 3,435 | den-crane | >Можно ли получить такие данные запросом в ClickHouse?
в общем случае нельзя, КХ не умеет window functions.
```
Чисто так по приколу:
create table transactions(acc String, d DateTime, amount Float64) Engine = MergeTree partition by toYYYYMM(d) order by (acc,d);
insert into transactions values('a1', now(), ... | 2018-10-22T18:00:22 | 2018-10-22T18:01:55 | {
"+1": 2
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 430,950,688 | 1,861 | Farfaday | Seems that we are also affected with "OPTIMIZE TABLE". We are running version 18.12.14. | 2018-10-18T10:02:01 | 2018-10-18T10:02:01 | {} | NONE |
ClickHouse | ClickHouse | 430,956,205 | 3,413 | blinkov | @artfabrique can you show this 3-minutes-long query? | 2018-10-18T10:21:28 | 2018-10-18T10:21:28 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 431,020,540 | 3,408 | blinkov | @Bolinov could you please fix the comments above? | 2018-10-18T14:01:51 | 2018-10-18T14:01:51 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 431,830,461 | 3,437 | Tasselmi | > help me , plz . I am struggling~~~ | 2018-10-22T13:08:20 | 2018-10-22T13:08:20 | {} | NONE |
ClickHouse | ClickHouse | 430,663,632 | 3,393 | proller | Why you change CMAKE_INSTALL_PREFIX ? It brokes config loading. | 2018-10-17T14:59:28 | 2018-10-17T14:59:28 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 431,279,824 | 3,408 | blinkov | I'll fix punctuation/markup after merge. | 2018-10-19T08:03:16 | 2018-10-19T08:03:16 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 431,410,710 | 3,413 | blinkov | @artfabrique it looks roughly as I expected, but frankly speaking it's too complicated to debug via public comments without access to real ClickHouse cluster. Does it look CPU bound? If it does, what do you see in profiler? What kind of hardware do you use for this test? | 2018-10-19T15:54:21 | 2018-10-19T15:54:21 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 431,415,574 | 3,424 | den-crane | https://clickhouse.yandex/docs/en/data_types/nested_data_structures/nested/ | 2018-10-19T16:09:58 | 2018-10-19T16:09:58 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 431,438,221 | 3,184 | den-crane | notes for me:
worked in 18.6.0
fixed in 18.14? | 2018-10-19T17:29:31 | 2018-10-19T17:29:31 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 432,099,503 | 3,435 | StratoCasterFen | Если взять таблицу из ответа выше
┌acc┬──────dt─┬─sum─┐
│ a1 │ 2018-10-18 │ -2 │
│ a1 │ 2018-10-22 │ 10 │
│ a2 │ 2018-10-20 │ 1 │
│ a2 │ 2018-10-22 │ 1 │
└──┴────────┴─────┘
то за период с 17,10,18 по 22,10,18 должны получиться сл. данные:
┬──────dt─┬sum─┐
│ 2018-10-17 │ ... | 2018-10-23T05:37:21 | 2018-10-23T05:43:23 | {} | NONE |
ClickHouse | ClickHouse | 432,144,461 | 3,436 | alexey-milovidov | > Do you also want it also be implicitly constructable?
Why not. (Or I don't understand the motivation of these changes.) | 2018-10-23T08:17:31 | 2018-10-23T08:17:31 | {} | MEMBER |
ClickHouse | ClickHouse | 430,751,490 | 2,427 | blinkov | @luc1ph3r any news? | 2018-10-17T19:07:44 | 2018-10-17T19:07:44 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 431,480,934 | 3,410 | den-crane | it's tricky and works with engine = memory and sometimes caches other value
<img width="720" alt="screen shot 2018-10-19 at 4 50 28 pm" src="https://user-images.githubusercontent.com/19737682/47240740-babd9100-d3bf-11e8-880a-64adb5e4655f.png">
| 2018-10-19T19:55:08 | 2018-10-19T19:55:08 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 431,844,729 | 3,437 | ztlpn | The error means that you have configured ClickHouse to use the ZooKeeper cluster on servers 77.88.55.80 and 111.0.1.2 but it cannot connect to them. If you are just experimenting with ClickHouse on a single node, you can just comment out the `<zookeeper>` section in `config.xml`. | 2018-10-22T14:02:52 | 2018-10-22T14:02:52 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 431,851,575 | 3,309 | alexey-gl | Thanks a lot! | 2018-10-22T14:27:53 | 2018-10-22T14:27:53 | {} | NONE |
ClickHouse | ClickHouse | 431,863,042 | 3,141 | tungdam | Upgrading to 18.14.9 helped ( but then met [this](https://github.com/yandex/ClickHouse/issues/3411) issue ). JFYI | 2018-10-22T15:08:19 | 2018-10-22T15:09:32 | {} | NONE |
ClickHouse | ClickHouse | 432,043,329 | 3,436 | amosbird | > Shouldn't you prove this approach by removing explicit casts across the code?
I planed to do that, but there seems to be no simple explicit cast assignment. Here are all copy assignments reported by clang
```
DataTypeAggregateFunction.cpp:69: field = String();
DataTypeArray.cpp:50: field = Arr... | 2018-10-23T00:51:29 | 2018-10-23T00:51:29 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 432,074,091 | 3,436 | amosbird | > The only non-obvious part is that when you write something like
> Field f = 1;
> Int64 value will be created,
> but it can be transparently inserted into ColumnUInt64.
In fact, this makes more sense to me as it's consistent to c/c++ itself. If the conversion must happen, it will happen. And currently there is n... | 2018-10-23T03:16:32 | 2018-10-23T03:27:30 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 430,677,074 | 3,408 | blinkov | cc @zhang2014 @sundy-li | 2018-10-17T15:33:21 | 2018-10-17T15:33:21 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 427,728,988 | 3,310 | healiseu | @iFoton thanks, I got confused with naming, Start* and *Star,
```sql
SELECT
StartStationRAM.name,
count(*) AS trips
FROM BicycleTripStar
ANY INNER JOIN StartStationRAM ON BicycleTripStar.start_station_id = StartStationRAM.id
WHERE member_id = 1
GROUP BY StartStationRAM.name
HAVING trips < 10
ORDER... | 2018-10-08T06:03:59 | 2018-10-18T06:23:49 | {} | NONE |
ClickHouse | ClickHouse | 427,964,485 | 3,310 | healiseu | @filimonov it is a star schema only that it can be extended with as many tables as you like. This is achieved by creating a model, i.e. schema of tables and fields, I call them entities and attributes to keep consistency with the ER diagram. The trick here is to give numerical addresses to everything including the dist... | 2018-10-08T20:14:21 | 2018-10-18T07:03:39 | {} | NONE |
ClickHouse | ClickHouse | 431,279,585 | 3,417 | blinkov | I'll apply comments above after merge. | 2018-10-19T08:02:22 | 2018-10-19T08:02:22 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 430,660,159 | 3,310 | healiseu | @filimonov this is an update of a test I am currently running on a 10 x 1,226,766 flat table with reference to our previous discussion
```txt
TripFlat
┌─pk─┬─duration─┬──────────start_date─┬────────────end_date─┬─from_station_id─┬─from_station_name──────────────┬─to_station_id─┬─to_station_name─────────────┬─bike_... | 2018-10-17T14:51:28 | 2018-10-20T13:03:34 | {} | NONE |
ClickHouse | ClickHouse | 431,980,163 | 3,252 | alexey-milovidov | Non obvious. | 2018-10-22T20:50:06 | 2018-10-22T20:50:06 | {} | MEMBER |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.