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
389,913,126
2,260
ludv1x
> How can I rerun the test? Just rebase the branch to master to update the last commit hash.
2018-05-17T15:44:22
2018-05-17T15:44:32
{}
CONTRIBUTOR
ClickHouse
ClickHouse
391,677,259
2,401
VerusK
@ludv1x Вставка происходит с `FORMAT TabSeparated` и вставка происходит не руками из консоли, а через clickhouse-client и на входе файлы.
2018-05-24T11:14:06
2018-05-24T11:14:06
{}
NONE
ClickHouse
ClickHouse
391,678,548
2,407
silviucpp
Note: seems at some point there is a timeout but in case you have a pool off connections 20 in my case the server is started in few minutes even if no query is sent on that server
2018-05-24T11:19:23
2018-05-24T11:19:23
{}
CONTRIBUTOR
ClickHouse
ClickHouse
388,704,933
2,336
zzl-wy
we have solved it by update new version.
2018-05-14T05:55:54
2018-05-14T05:55:54
{}
NONE
ClickHouse
ClickHouse
390,383,354
1,053
theseusyang
@prog8 We have implemented the integration ClickHouse with HDFS, next week We will push codes into github.
2018-05-19T06:24:13
2018-05-19T06:24:13
{}
NONE
ClickHouse
ClickHouse
392,116,231
2,403
dsr301
The port was in use by another process. I had to change port in /etc/clickhouse-server/config.xml and pass it as a parameter when launching client as below: clickhouse-client --port 9011
2018-05-25T16:46:47
2018-05-25T16:46:47
{ "+1": 1, "-1": 1 }
NONE
ClickHouse
ClickHouse
392,576,755
2,431
alexey-milovidov
Please merge with master first.
2018-05-28T17:42:17
2018-05-28T17:42:17
{}
MEMBER
ClickHouse
ClickHouse
393,344,572
2,443
meshpaul
That makes sense. Thank you.
2018-05-30T22:49:54
2018-05-30T22:49:54
{}
NONE
ClickHouse
ClickHouse
393,455,998
2,218
meshpaul
Wow great effort! One thing I would like to mention there are some Java based tools that are easily used all they need is ClickHouse JDBC driver. Ex Netbeans, Eclipse or DBVisualizer (not sure if it's open source) but I use it . Anyhow wanted to mention some here but for basic SQL queries and desenet grid to display...
2018-05-31T08:32:07
2018-05-31T08:32:07
{}
NONE
ClickHouse
ClickHouse
388,992,313
167
shockabsorber
@alexey-milovidov awesome product! there is any chance for update / delete in the near future ?
2018-05-14T23:15:44
2018-05-14T23:15:44
{ "+1": 2 }
NONE
ClickHouse
ClickHouse
389,072,845
2,156
jpiper
This is also happening for me on 1.1.54381
2018-05-15T07:36:42
2018-05-15T07:36:42
{}
NONE
ClickHouse
ClickHouse
389,113,021
2,351
kingbacks
<attribute> <name>tax_name</name> <type>String</type> <expression>assumeNotNull(tax_name)</expression> <null_value></null_value> <injective>true</injective> </attribute>
2018-05-15T10:01:19
2018-05-15T10:01:19
{}
NONE
ClickHouse
ClickHouse
389,871,261
2,381
filimonov
Can you find and provide that problematic row?
2018-05-17T13:45:09
2018-05-17T13:45:09
{}
CONTRIBUTOR
ClickHouse
ClickHouse
391,692,437
2,407
ludv1x
How ClickHouse handles SIGSTOP and SIGCONT signals: - If you make ```sudo kill -s 19 `pidof clickhouse-server` ``` than ClickHouse freezes and does not close connections. - If you unfreeze ClickHouse via ```sudo kill -s 18 `pidof clickhouse-server` ``` it will continue working as expected. I am not sure that Cli...
2018-05-24T12:16:00
2018-05-24T12:36:28
{}
CONTRIBUTOR
ClickHouse
ClickHouse
391,950,281
2,414
meshpaul
Also what are you using as metric collector to store in data in DB?
2018-05-25T05:58:47
2018-05-25T05:58:47
{}
NONE
ClickHouse
ClickHouse
392,168,622
2,354
ztlpn
Superseded by #2426
2018-05-25T20:18:03
2018-05-25T20:18:03
{}
CONTRIBUTOR
ClickHouse
ClickHouse
392,572,617
2,422
den-crane
The right PK order allows to reduce data size by column compression. PK it's not an index it just a sort order of data in columns. so you have data field0 | field1 | field3 -- | -- | -- | 1 | 1 | longstring1 2 | 2 | anotherstring1 3 | 3 | thirdstring1 4 | 1 | longstring2 5 | 2 | anotherstring2 6 | 3 | thi...
2018-05-28T17:13:01
2018-05-28T17:19:55
{ "+1": 1, "heart": 1 }
CONTRIBUTOR
ClickHouse
ClickHouse
393,294,519
2,417
kmatt
I'm seeing something similar. The startup time before a node comes online is a problem. Any relation to #1758 ?
2018-05-30T19:49:26
2018-05-30T19:49:26
{}
NONE
ClickHouse
ClickHouse
389,836,627
2,385
filimonov
Known (and very nasty) issue. Currently clickhouse calculate both branches, and only afterwards choose one. Workaround usually possible, in your case you can try: ``` SELECT CASE WHEN empty(time_logged) = 0 THEN toDateTime(substring(concat(logged,'0000-00-00 00:00:00') ,1,19)) ELSE created END as created_at from...
2018-05-17T11:37:03
2018-05-17T11:42:29
{}
CONTRIBUTOR
ClickHouse
ClickHouse
391,995,596
2,413
kshvakov
Из документации > Числа типа UInt64 и Int64, для совместимости с JavaScript, по умолчанию выводятся в двойных кавычках, чтобы они выводились без кавычек можно установить конфигурационный параметр output_format_json_quote_64bit_integers равным 0. https://clickhouse.yandex/docs/ru/single/#json
2018-05-25T09:24:08
2018-05-25T09:24:08
{ "+1": 1 }
CONTRIBUTOR
ClickHouse
ClickHouse
392,185,913
2,394
alexey-milovidov
Ok, but: - `pointInPolygon` is not documented; by the way, it is the most well developed and useful geo function; - now `File` table engine should also work with files in `user_files_path` directory, as `file` table function, see https://github.com/yandex/ClickHouse/pull/2164 But I'm not sure, we don't have a test fo...
2018-05-25T20:59:52
2018-05-25T20:59:52
{}
MEMBER
ClickHouse
ClickHouse
388,705,019
2,338
zzl-wy
solve it by update new version.
2018-05-14T05:56:28
2018-05-14T05:56:28
{}
NONE
ClickHouse
ClickHouse
389,320,941
2,372
jonatasfreitasv
Solution: ``` select session_id, groupArray(details) from ( select session_id, details as details from rt_analytics.user_responses ) group by session_id ```
2018-05-15T21:34:32
2018-05-15T21:34:44
{}
CONTRIBUTOR
ClickHouse
ClickHouse
389,494,329
2,371
filimonov
Actually concrete SQL depends on what you want to do in case if same eventid will appear twice. If it can't happen you can do just ``` SELECT key, anyIf(payload, event = 'idEventoInicial') as idEventoInicial, anyIf(payload, event = 'idEnd2') as idEnd2 FROM table GROUP BY key ```
2018-05-16T12:03:37
2018-05-16T12:03:37
{ "+1": 1 }
CONTRIBUTOR
ClickHouse
ClickHouse
389,660,331
2,074
rpfennin
optimize final does not appear to delete all pairs of rows that it should. in my tests, it leaves about half of the rows that should get purged.
2018-05-16T20:47:49
2018-05-16T20:47:49
{}
NONE
ClickHouse
ClickHouse
389,841,968
2,385
alniks
Closing as a duplicate and there is a workaround
2018-05-17T11:59:38
2018-05-17T11:59:38
{}
NONE
ClickHouse
ClickHouse
389,853,384
2,375
rpfennin
I'm having an issue getting the setting to be respected. i have tried both <output_format_json_quote_64bit_integers>false</output_format_json_quote_64bit_integers> and <output_format_json_quote_64bit_integers>0</output_format_json_quote_64bit_integers> in the config.xml but i'm not having any luck. i do ...
2018-05-17T12:44:39
2018-05-17T12:45:07
{}
NONE
ClickHouse
ClickHouse
391,957,486
2,414
meshpaul
Also what ClickHouse user profile are you using? ex https://github.com/yandex/ClickHouse/edit/master/docs/en/operations/settings/settings_profiles.md. In addition can you share the details on how you came up with such numbers? and mention things like hardware/vm and whole app stack for each use case?
2018-05-25T06:42:13
2018-05-25T07:14:33
{}
NONE
ClickHouse
ClickHouse
392,028,840
2,420
tru1
Vote for this issue
2018-05-25T11:48:09
2018-05-25T11:48:09
{}
NONE
ClickHouse
ClickHouse
393,430,733
2,218
jonatasfreitasv
Version 1.0.0-beta is open!!!! https://github.com/HouseOps/HouseOps
2018-05-31T07:06:48
2018-05-31T07:06:48
{ "+1": 1 }
CONTRIBUTOR
ClickHouse
ClickHouse
388,616,920
2,352
alexey-milovidov
Good feature. Thank you!
2018-05-13T10:33:21
2018-05-13T10:33:21
{}
MEMBER
ClickHouse
ClickHouse
389,353,924
1,535
alexey-milovidov
For history: #1538
2018-05-16T00:11:55
2018-05-16T00:11:55
{}
MEMBER
ClickHouse
ClickHouse
389,736,724
2,379
rpfennin
Following up on this. The following does not do what I would expect (remove all rows): create table testcp (p date, v Int32, sign Int8) Engine = CollapsingMergeTree(p, v, 8192, sign) insert into testcp values ('2018-05-16', 33, 1); insert into testcp values ('2018-05-16', 33, -1); insert into testcp values ('2...
2018-05-17T03:54:44
2018-05-17T14:53:31
{}
NONE
ClickHouse
ClickHouse
390,414,530
2,388
alexey-milovidov
:+1:
2018-05-19T16:03:09
2018-05-19T16:03:09
{}
MEMBER
ClickHouse
ClickHouse
390,767,422
2,396
alexey-milovidov
Nullable columns are not permitted in primary key. The issue is that MODIFY PRIMARY KEY lacks check for it. You can rollback changes manually, by editing `.sql` file. Tell me if you will need help with rolling back changes. PS. MODIFY PRIMARY KEY is unofficial feature: it is intentionally undocumented and shoul...
2018-05-21T20:09:24
2018-05-21T20:09:24
{}
MEMBER
ClickHouse
ClickHouse
391,988,506
2,414
agaurav
@meshpaul Both dashboard and metric collector are our own implementation at [archsaber](https://archsaber.io). Metric collector is in golang & dashboard in React+[dygraphs](https://github.com/danvk/dygraphs)+[crossfilter](http://crossfilter.github.io/crossfilter/)+d3 The given number are for [flowlog](https://docs.a...
2018-05-25T08:57:24
2018-05-25T08:57:24
{}
NONE
ClickHouse
ClickHouse
392,491,998
867
opavader
@alexey-milovidov any updates on this ?
2018-05-28T10:52:29
2018-05-28T10:52:29
{}
NONE
ClickHouse
ClickHouse
392,654,263
867
BayoNet
I'll try to add the documentation today.
2018-05-29T05:08:20
2018-05-29T05:08:20
{}
CONTRIBUTOR
ClickHouse
ClickHouse
393,466,938
2,218
sundy-li
@meshpaul By the way, we are working on Java Native jdbc library, see [here](https://github.com/houseflys/ClickHouse-Native-JDBC)
2018-05-31T09:10:29
2018-05-31T09:10:29
{}
CONTRIBUTOR
ClickHouse
ClickHouse
389,249,125
1,522
YindSoft
so do I
2018-05-15T17:30:47
2018-05-15T17:30:47
{}
NONE
ClickHouse
ClickHouse
389,460,595
1,201
AlexMAS
Hello again! :) I would like to add one more example which illustrates what I want to see in the nearest future (I hope). [TimeScale](https://www.timescale.com/) database offers [`time_bucket()`](https://docs.timescale.com/v0.9/api#time_bucket) function which works as described above. Using this function you can agg...
2018-05-16T09:45:26
2018-05-16T09:45:26
{}
CONTRIBUTOR
ClickHouse
ClickHouse
389,500,568
1,201
sundy-li
Maybe the function like `roundTime( time, '5m' )` , I think it's much better than `toStartOfInterval(time, 5, 'minutes')`
2018-05-16T12:27:59
2018-05-16T12:27:59
{}
CONTRIBUTOR
ClickHouse
ClickHouse
390,221,469
2,390
filimonov
> CH:1.1.54236 Эта версия была выпущена год назад, поддержка Null в ней была на очень ранних стадиях. С тех пор было ОЧЕНЬ много улучшений и поправок, повышающих стабильность ClickHouse вообще и поддержки Null в частности. Пожалуйста, обновитесь. Скрипты для сборки свежего ClickHouse для Centos есть тут: https://...
2018-05-18T14:16:48
2018-05-18T14:40:25
{}
CONTRIBUTOR
ClickHouse
ClickHouse
392,583,809
2,432
proller
Can you run script with command print: ```bash -x ./release``` and show output? And you can try create pbuilder image first: ```sudo DIST=bionic pbuilder create --configfile debian/.pbuilderrc```
2018-05-28T18:33:29
2018-05-28T18:33:29
{}
CONTRIBUTOR
ClickHouse
ClickHouse
389,653,644
2,368
bobrik
#1821 is loosely related.
2018-05-16T20:24:55
2018-05-16T20:24:55
{}
CONTRIBUTOR
ClickHouse
ClickHouse
391,714,450
2,407
silviucpp
Hello @ludv1x , On My mac at CTRL+C I see: 2018.05.24 13:26:11.498059 [ 8 ] <Information> Application: Received termination signal (Interrupt: 2) And server stop accepting new connections but it won't close after all existing connections are closed by the client. Also I see that on a client that's using htt...
2018-05-24T13:28:10
2018-05-24T13:28:10
{}
CONTRIBUTOR
ClickHouse
ClickHouse
392,147,081
2,400
alexey-milovidov
Thank you! The fix is totally correct.
2018-05-25T18:44:36
2018-05-25T18:44:36
{}
MEMBER
ClickHouse
ClickHouse
392,601,467
2,432
AlessandroSechi
Create the pbuilder image first, after doing `sudo apt-get install ubuntu-archive-keyring ` fixed the problem. Thanks
2018-05-28T20:50:19
2018-05-28T20:50:19
{}
NONE
ClickHouse
ClickHouse
393,435,733
1,805
marklit
This looks like #2298. There's some work ongoing to fix this issue.
2018-05-31T07:26:28
2018-05-31T07:26:28
{}
NONE
ClickHouse
ClickHouse
389,489,513
1,201
filimonov
It's very easy to workaround. Use that as a template ``` toDateTime(intDiv(toUnixTimestamp(timestamp), 300) * 300) ``` replace 300 with your time interval (in seconds). But that question is quite popular, may be the shortcut function should be added.
2018-05-16T11:43:29
2018-05-16T11:43:29
{ "+1": 1 }
CONTRIBUTOR
ClickHouse
ClickHouse
389,501,449
2,367
filimonov
Clickhouse doesn't support deep structures in JSON like `{"x": {"a":"1", "b":"2"}}`. If your input data doesn't contain such a deep structures, and ClickHouse still complain that something's wrong, please provide some sample of your JSON.
2018-05-16T12:31:17
2018-05-16T12:31:32
{}
CONTRIBUTOR
ClickHouse
ClickHouse
389,504,943
2,357
filimonov
That's because standart defined only one type of escaping quotes - by doubling it. Check p. 2.7 of https://tools.ietf.org/html/rfc4180 > Any notes on how to make it work ? Try to reformat that according to Clickhouse rules. It's very easy to do for example with python (that simple script read from stdin and writ...
2018-05-16T12:44:08
2018-05-16T12:52:28
{ "heart": 1 }
CONTRIBUTOR
ClickHouse
ClickHouse
390,221,337
2,390
proller
Это давно исправлено. Используйте последний релиз. тест: ``` echo 'CREATE TABLE IF NOT EXISTS `test_stat` (`event_date` Date, `time` Int32, `user_id` UInt32, `counter_id` Nullable (UInt32), `ip` Int64, `click` UInt8, `is_unique` UInt8, `click_id` String) ENGINE = MergeTree(event_date, (user_id), 8192);' | ${CLICK...
2018-05-18T14:16:19
2018-05-18T14:16:19
{ "+1": 1 }
CONTRIBUTOR
ClickHouse
ClickHouse
390,373,668
2,337
kingbacks
init the linux system and It's OK , possible is the linux OS makes this error !
2018-05-19T02:58:25
2018-05-19T02:58:25
{}
NONE
ClickHouse
ClickHouse
391,676,639
2,401
ludv1x
Два раза экранировать не надо: ``` vludv-ub :) create table test.string (key String) ENGINE = Memory CREATE TABLE test.string ( key String ) ENGINE = Memory Ok. 0 rows in set. Elapsed: 0.012 sec. vludv-ub :) insert into test.string FORMAT JSONEachRow {"key":"data-md-component=\"navigation\"><di...
2018-05-24T11:11:41
2018-05-24T11:11:41
{}
CONTRIBUTOR
ClickHouse
ClickHouse
391,775,192
2,401
ludv1x
При вставке в формате `TabSeparated` пары символов `\<symbol>` заменяются на `<symbol>`. Соответственно, при вставке строки ```{"key":"data-md-component=\"navigation\"><div class=\"md-sidebar__scrollwrap\">"}``` в формате TSV реально вставится ```{"key":"data-md-component="navigation"><div class="md-sidebar__scro...
2018-05-24T16:20:54
2018-05-24T16:21:42
{}
CONTRIBUTOR
ClickHouse
ClickHouse
392,575,370
2,422
simPod
Thank you very much!
2018-05-28T17:32:32
2018-05-28T17:32:32
{}
CONTRIBUTOR
ClickHouse
ClickHouse
392,876,081
2,437
kshvakov
Hi, you can find them in https://github.com/yandex/ClickHouse/tree/master/dbms/src/Parsers
2018-05-29T17:58:46
2018-05-29T17:58:46
{}
CONTRIBUTOR
ClickHouse
ClickHouse
393,093,965
2,443
meshpaul
Hi We actually use these as part of our Ansible management to see if we need to bounce ClickHouse. So what will be new behavior to tell us Config changes have been processed?
2018-05-30T09:27:57
2018-05-30T09:27:57
{}
NONE
ClickHouse
ClickHouse
388,938,966
2,362
kmatt
This may be a due to a problem with my table design? Adding a column of type Date and partitioning on that populates parts with the min_date and max_date columns with the expected values.
2018-05-14T19:44:59
2018-05-14T19:44:59
{}
NONE
ClickHouse
ClickHouse
389,377,225
2,366
alexey-milovidov
To specifically pay attention on the implementation method: it will do heavyweight data rewrite in partitions.
2018-05-16T02:43:44
2018-05-16T02:43:44
{}
MEMBER
ClickHouse
ClickHouse
389,544,164
2,369
zhkvia
I'm currently working on Protobuf support. The first testable implementation should be available at the end of this week. Stay tuned (:
2018-05-16T14:42:10
2018-05-16T14:42:10
{}
CONTRIBUTOR
ClickHouse
ClickHouse
389,820,901
2,375
vpanfilov
You can see the explanation of this behaviour here: https://github.com/yandex/ClickHouse/issues/114 There is an option in config to disable quotes for Int64: https://clickhouse.yandex/docs/en/operations/settings/settings/#output_format_json_quote_64bit_integers
2018-05-17T10:27:55
2018-05-17T10:27:55
{}
NONE
ClickHouse
ClickHouse
389,891,722
2,386
the-tinderbox
@alexey-milovidov said that it's in a sense logical and confirmed that `PREWHERE` runs before `FINAL`. So, be careful with these two statements!
2018-05-17T14:43:50
2018-05-17T14:43:50
{}
NONE
ClickHouse
ClickHouse
390,955,931
2,378
simPod
That clears it up for me. Thanks, much appreciated!
2018-05-22T11:24:19
2018-05-22T11:24:19
{}
CONTRIBUTOR
ClickHouse
ClickHouse
392,760,533
2,373
fangyc
The issue cannot be reproduced with very simple MergeTree and Distributed tables. The case where I found this problem is the OnTime test data on ClickHouse documentation. I changed it to cluster version (on 2 machines) with the following statements: CREATE TABLE test.`ontime` on cluster test_shard ( `Year` UI...
2018-05-29T12:37:28
2018-05-30T03:51:31
{}
NONE
ClickHouse
ClickHouse
389,070,378
2,352
Jingyi244
cheer 4 this.
2018-05-15T07:27:05
2018-05-15T07:27:05
{}
NONE
ClickHouse
ClickHouse
389,499,166
2,370
filimonov
Problem is caused by the fact that that your decoder expects to see utf8 in binary stream. Try to do it like that: ``` clickhouse-client --query "SELECT IPv6StringToNum('2001:4860:4860:0:0:0:0:8888') FORMAT JSONEachRow" | python -c "import json,sys,pprint; pprint.pprint( json.loads(sys.stdin.read(),encoding='lat...
2018-05-16T12:22:53
2018-05-16T12:24:03
{}
CONTRIBUTOR
ClickHouse
ClickHouse
389,499,909
2,369
filimonov
Please check ClickHouse roadmap https://clickhouse.yandex/docs/en/roadmap/#q2-2018 ;) And you can always help with that by contributing. :)
2018-05-16T12:25:47
2018-05-16T12:26:31
{}
CONTRIBUTOR
ClickHouse
ClickHouse
389,506,564
1,201
AlexMAS
@sundy-li I don't mind. I just used [the existing 'naming convention'](https://clickhouse.yandex/docs/en/functions/date_time_functions/#tostartofmonth).
2018-05-16T12:49:33
2018-05-16T12:49:33
{}
CONTRIBUTOR
ClickHouse
ClickHouse
389,607,968
2,367
filimonov
Clickhouse use quick and dirty JSON parser, which does not how to read complex deep structures. So it can't skip that field as it does not know where that nested structure ends. Sorry. :/ So you should preprocess your json with some external tools, of you can contribute to Clickhouse and improve JSON parser.
2018-05-16T17:50:56
2018-05-16T17:50:56
{}
CONTRIBUTOR
ClickHouse
ClickHouse
392,112,785
2,376
proller
Can you show your cmake output? lz4 should be used from contrib: -- Using lz4: /Users/.../ClickHouse/contrib/lz4/lib : lz4 cmake will use libmariadb or libmysql lib: -- Using mysqlclient=1: /usr/local/include : ; staticlib=/usr/local/lib/libmariadbclient.a
2018-05-25T16:33:35
2018-05-25T16:33:35
{}
CONTRIBUTOR
ClickHouse
ClickHouse
393,056,607
2,171
linuxgood1230
nobody nobody no. close
2018-05-30T07:20:43
2018-05-30T07:20:43
{}
NONE
ClickHouse
ClickHouse
393,467,612
2,218
perez1987
Congratulations! I'm going to try it and test it.
2018-05-31T09:12:56
2018-05-31T09:12:56
{ "+1": 1 }
NONE
ClickHouse
ClickHouse
388,616,731
2,294
alexey-milovidov
Could you please also add something into `dbms/tests/integration`?
2018-05-13T10:30:10
2018-05-13T10:30:10
{}
MEMBER
ClickHouse
ClickHouse
389,507,629
2,357
AbdealiLoKo
Hm, got it. Most common tools allow both (pandas, excel, mysql, etc.) - would be beneficial for a bunch of users to avoid doing an additional step. My data was very large and if I tried doing that I would have gotten a Harddisk full error so I wan unable to do any good formatting - got sed to remove those for now thou...
2018-05-16T12:53:23
2018-05-16T12:53:23
{}
NONE
ClickHouse
ClickHouse
389,621,867
2,370
leoluk
~This works with Python 2, but not with Python 3:~ ``` clickhouse-client --query "SELECT IPv6StringToNum('2001:4860:4860:0:0:0:0:8888') FORMAT JSONEachRow" | python3 -c "import json,sys,pprint; pprint.pprint( json.loads(sys.stdin.read(),encoding='latin1') )" Traceback (most recent call last): File "<string>", l...
2018-05-16T18:35:25
2018-05-16T18:37:28
{}
CONTRIBUTOR
ClickHouse
ClickHouse
392,896,016
2,437
jonatasfreitasv
Not have in a simple text list ? hehe Thx a lot!
2018-05-29T18:52:05
2018-05-29T18:52:05
{}
CONTRIBUTOR
ClickHouse
ClickHouse
393,174,880
2,443
proller
Preprocessed configs will be written to /var/lib/clickhouse/config_preprocessed
2018-05-30T14:05:14
2018-05-30T14:05:14
{}
CONTRIBUTOR
ClickHouse
ClickHouse
388,617,513
2,349
alexey-milovidov
Maybe just use `Poco::URI::addQueryParameter`?
2018-05-13T10:43:13
2018-05-13T10:43:13
{}
MEMBER
ClickHouse
ClickHouse
388,981,397
2,316
harlinb
I am having the same issue. After upgrading from 1.1.54543 to 1.1.54578 the metric is logging ~30 "zookeeper exceptions" per second, with no sign of anything going wrong in the logs or behaviour of the application.
2018-05-14T22:20:09
2018-05-14T22:20:09
{}
NONE
ClickHouse
ClickHouse
389,503,580
2,362
filimonov
That fields are filled only if some of the column used to calculate partition expression has `Date` type. In your case it's `DateTime`. If you will add date column it will work. ```sql CREATE TABLE testttt ( dttm Datetime, key UInt64, data UInt64, date Date DEFAULT toDate(dttm) ) ENGINE =...
2018-05-16T12:39:23
2018-05-16T12:40:10
{}
CONTRIBUTOR
ClickHouse
ClickHouse
389,998,152
2,381
yctn
i tryed but rows are coming in at a very high rate so thats hard. and i was unable to find the row causing it. i even wrote all the json to a file while inserting in clickhouse. and when clickhouse fails search the row and insert manualy but then it didnt fail.... so i was unable to find the row causing it. but it come...
2018-05-17T20:21:52
2018-05-17T20:21:52
{}
NONE
ClickHouse
ClickHouse
390,142,955
2,375
vpanfilov
@rpfennin, here is a simple example without touching config: ```sql Connected to ClickHouse server version 1.1.54380. :) select toInt64(12345) as testInt64 format JSONEachRow SELECT toInt64(12345) AS testInt64 FORMAT JSONEachRow {"testInt64":"12345"} 1 rows in set. Elapsed: 0.083 sec. :) set output_...
2018-05-18T08:59:13
2018-05-18T08:59:13
{ "+1": 2 }
NONE
ClickHouse
ClickHouse
390,232,297
2,378
filimonov
PARTITION BY - declare how to split data in table to a parts which can be manipulated separately. Partitions should be nor too small nor too big. Usually it's months / week / days - that allow you to drop old partitions, or do other manipulation with partition level. Clickhouse allows you to drop / optimize / freeze / ...
2018-05-18T14:52:06
2018-05-18T14:53:15
{ "+1": 26 }
CONTRIBUTOR
ClickHouse
ClickHouse
391,732,887
2,407
silviucpp
I found why the connection is cut of after a response : keep_alive_timeout setting, but is not working as it suppose to: For example set : keep_alive_timeout = 3 and create a connection. This one is not close even after minutes.The send a request and after you get the response connection is closed after that 3 se...
2018-05-24T14:21:09
2018-05-24T14:21:09
{}
CONTRIBUTOR
ClickHouse
ClickHouse
388,616,968
2,294
sundy-li
@alexey-milovidov thanks for your detail comment, I will do those improvement.
2018-05-13T10:34:03
2018-05-13T10:34:03
{}
CONTRIBUTOR
ClickHouse
ClickHouse
389,650,433
2,171
yctn
Did you manage to resolve this? it seems input_format_allow_errors_ratio input_format_allow_errors_num doesnt affect kafka
2018-05-16T20:13:06
2018-05-16T20:13:06
{}
NONE
ClickHouse
ClickHouse
391,334,495
2,400
dqminh
The bug seems to only trigger for table created implicitly by materialized view which didnt specify the column type. For example we see the bug happen when querying timeslot field. ``` CREATE MATERIALIZED VIEW IF NOT EXISTS r0.affected_mv ENGINE = ReplicatedAggregatingMergeTree('zookeeper', '{replica}', date, (date,...
2018-05-23T12:50:10
2018-05-23T12:50:10
{}
CONTRIBUTOR
ClickHouse
ClickHouse
391,817,026
2,401
antonio-antuan
Считается ли такое поведение корректным? Нет ли в планах это изменить? Спрашиваю чтобы в будущем, после гипотетического обновления версии сервера, не пришлось данные переливать.
2018-05-24T18:37:02
2018-05-24T18:37:02
{}
NONE
ClickHouse
ClickHouse
391,949,709
2,414
meshpaul
What are you using for dashboard grafana?
2018-05-25T05:54:55
2018-05-25T05:54:55
{}
NONE
ClickHouse
ClickHouse
391,996,860
2,413
negasus
Спасибо. Буду внимательнее изучать документацию)
2018-05-25T09:28:50
2018-05-25T09:28:50
{}
NONE
ClickHouse
ClickHouse
389,366,655
2,366
zhang2014
Why not `delete table` ?
2018-05-16T01:35:22
2018-05-16T01:35:22
{}
CONTRIBUTOR
ClickHouse
ClickHouse
389,597,089
2,367
maxulan
I don't use nested fields in target table however nested field exists in my input JSON. I use --input_format_skip_unknown_fields=1 specifically to ignore that nested field. However it doesn't work. See exception above. If I don't use that field then why Clickhouse cares if it's nested or not?
2018-05-16T17:15:50
2018-05-16T17:15:50
{}
CONTRIBUTOR
ClickHouse
ClickHouse
389,837,519
2,385
filimonov
See also #1562 #1768
2018-05-17T11:41:00
2018-05-17T11:41:00
{}
CONTRIBUTOR
ClickHouse
ClickHouse
390,228,126
2,381
filimonov
I bet something wrong with quotes or backslashes in your data. There is only one case when that error can happen https://github.com/yandex/ClickHouse/blob/adfc743e060dca26d6fa0bcb7fd93320de0655c5/dbms/src/DataStreams/JSONEachRowRowInputStream.cpp#L104 And it can means that clickhouse thinks that it succesfully...
2018-05-18T14:38:22
2018-05-18T14:56:04
{}
CONTRIBUTOR
ClickHouse
ClickHouse
390,550,155
2,393
zhang2014
#2226 seems like a more complete solution.
2018-05-21T04:28:55
2018-05-21T04:28:55
{}
CONTRIBUTOR
ClickHouse
ClickHouse
390,764,338
2,381
yctn
Nothing goes before time. Time is the first thing i start a line with. which means the previous line was not closed proberly. but there are a few issues going on. when i juse a kafka consume and pipe it into jq i do not get any json errors. but still clickhouse-kafka gives me an error. would it be possible to print the...
2018-05-21T19:57:44
2018-05-21T19:57:44
{ "confused": 2 }
NONE
ClickHouse
ClickHouse
392,446,420
167
AlekSi
See https://github.com/yandex/ClickHouse/pull/2366 for initial DELETE implementation.
2018-05-28T07:41:04
2018-05-28T07:41:04
{}
NONE
ClickHouse
ClickHouse
393,108,160
2,414
agaurav
@alexey-milovidov any thoughts on this ? Some insights on what might be the throttling the small read queries.
2018-05-30T10:15:04
2018-05-30T10:15:04
{}
NONE