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 | 418,103,962 | 2,825 | filimonov | Related #208 and #2999 | 2018-09-03T12:41:13 | 2018-09-03T12:43:16 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 418,286,200 | 3,014 | filimonov | What platform and golang version do you use? For me it rather looks like something is wrong with golang installation or system inself because on the top of stacktrace there are 2 go core modules (garbage collector and bufio Writer). Normally core modules don't die in a such weird way. | 2018-09-04T08:33:32 | 2018-09-04T08:33:32 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 418,295,684 | 3,014 | filimonov | Yep, i was afraid of something like that :D Windows host, multiple containers in docker, pretty old go version... And I bet containers communicate via windows-host (virtualbox / Hyper-V) provided network...
It's hard to imagine what exactly went bad there, but I can say that it doesn't look like [yandex/ClickHouse... | 2018-09-04T09:07:05 | 2018-09-04T09:07:05 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 418,858,618 | 2,958 | kmatt | Built stable with this PR, and while it skips undefined columns from JSON messages in a Kafka consumer table, adding a materialized view from the consumer to a shard table starts the following to be logged. Uncertain if this is from the PR, or another issue:
```
2018.09.05 15:46:10.769379 [ 21 ] <Error> void DB::St... | 2018-09-05T19:48:42 | 2018-09-05T19:49:08 | {} | NONE |
ClickHouse | ClickHouse | 419,030,180 | 3,039 | TianyiLu | After testing, I find it difficult to use FINAL or CollapsingMergeTree.
1. FINAL performance is really poor due to my query is very heavy. Basically, my query will hit 200 million rows and FINAL performance is even worse than query from the result of sub-query.
2. In fact, CollapsingMergeTree seems to be an increme... | 2018-09-06T09:35:56 | 2018-09-06T09:35:56 | {} | NONE |
ClickHouse | ClickHouse | 419,356,498 | 1,514 | cesarmarinhorj | @theseusyang is possible doing some simple experiments with graph with clickhouse? It helps in situations where there are too many edges in same node? Thanks. | 2018-09-07T07:59:35 | 2018-09-07T07:59:35 | {} | NONE |
ClickHouse | ClickHouse | 418,274,445 | 3,014 | lightee2 | @filimonov @kshvakov
the problem is so strange, i run the golang program on the server which is running clickhouse container successfully., | 2018-09-04T07:51:53 | 2018-09-04T07:51:53 | {} | NONE |
ClickHouse | ClickHouse | 418,379,065 | 3,031 | alexey-milovidov | 00411_long_accurate_number_comparison: FAIL - Timeout! | 2018-09-04T14:03:59 | 2018-09-04T14:03:59 | {} | MEMBER |
ClickHouse | ClickHouse | 418,389,832 | 3,033 | proller | tcp_ssl_port was renamed to tcp_port_secure
| 2018-09-04T14:33:32 | 2018-09-04T14:33:32 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 418,638,739 | 3,039 | TianyiLu | For example, I have three keys {key1, key2, key3} and a value {value} and a version {version}
I may need to write query like this:
select sum(T), key2 from {select key2, argMax(value, version) as T from SomeTable where XXX condition group by key1, key2, key3} group by key2.
The problem is that only when you u... | 2018-09-05T08:07:29 | 2018-09-05T08:07:29 | {} | NONE |
ClickHouse | ClickHouse | 418,723,422 | 3,039 | TianyiLu | Hi,
Your suggestion seems good!
Could you please give me an example on how to use CollapsingMergeTree on my case? Basically, I just want to pick every unique key-value pair that is the top version(or latest version) and aggregate all of them by some index, and avoid selecting from subselect result as well. | 2018-09-05T13:06:43 | 2018-09-05T13:06:43 | {} | NONE |
ClickHouse | ClickHouse | 418,749,731 | 2,990 | alesapin | Failed test 00634_performance_introspection_and_logging_shard. It's currently broken. | 2018-09-05T14:23:38 | 2018-09-05T14:23:38 | {} | MEMBER |
ClickHouse | ClickHouse | 418,793,077 | 3,039 | filimonov | Please, check that message: https://groups.google.com/d/msg/clickhouse/VixyOUD-K68/Km8EpkCyAQAJ | 2018-09-05T16:23:22 | 2018-09-05T16:23:22 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 418,813,719 | 3,041 | VadimPlh | Hi) I try to reproduce it. There is not in version 18.10.3 | 2018-09-05T17:28:22 | 2018-09-05T17:28:22 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 419,005,339 | 867 | BayoNet | Ready. | 2018-09-06T08:11:56 | 2018-09-06T08:11:56 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 419,451,119 | 1,514 | theseusyang | yes, It is feasible that clickhouse is begarded as a backend storage engine for distributed graph database. | 2018-09-07T14:08:51 | 2018-09-07T14:08:51 | {} | NONE |
ClickHouse | ClickHouse | 417,966,252 | 2,754 | alexey-milovidov | Just for reference:
https://lemire.me/blog/2012/05/31/data-alignment-for-speed-myth-or-reality/ | 2018-09-02T23:08:30 | 2018-09-02T23:08:30 | {} | MEMBER |
ClickHouse | ClickHouse | 418,658,283 | 3,040 | dsfishchen | I use a older jdbc driver release,it works!!
Thank you very much!!
<dependency>
<groupId>ru.yandex.clickhouse</groupId>
<artifactId>clickhouse-jdbc</artifactId>
<version>0.1.13</version>
</dependency> | 2018-09-05T09:12:06 | 2018-09-05T09:12:06 | {} | NONE |
ClickHouse | ClickHouse | 418,881,238 | 3,051 | blinkov | For some reason your patch is larger than it's supposed to be + I've already done this in another branch that will be merged soon. | 2018-09-05T21:05:30 | 2018-09-05T21:05:30 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 419,019,492 | 867 | blinkov | @francisdb it's kind of scattered across many parts of documentation on other topics, but they are easy to find by searching "Null". Also the issue is closed a bit early as that patch in English is not live yet, but will be hopefully today or maybe tomorrow. | 2018-09-06T09:00:20 | 2018-09-06T09:00:20 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 419,062,360 | 167 | blinkov | FYI: this feature is currently on code review at #3035 | 2018-09-06T11:38:21 | 2018-09-06T11:38:21 | {
"heart": 3,
"hooray": 5
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 419,146,400 | 3,054 | igor-sh8 | Hi, @alexey-milovidov.
Thanks for the info.
Maybe you could also suggest something about adding a column to the aggregated columns like:
`CREATE TABLE Test ( Date Date, Test1 Int32, Test2 Int32, Test4 Int32, Test3 Int32) ENGINE = SummingMergeTree(Date, (Date, Test1), 8192, (Test2, Test3,Test4))`
Is it possibl... | 2018-09-06T15:52:55 | 2018-09-06T15:52:55 | {} | NONE |
ClickHouse | ClickHouse | 418,265,247 | 3,014 | kshvakov | @lightee2 Hi, are you connected to the container from the local host or is it remote server?
What version of the ClickHouse server and Go client do you use?
PS: Check the network connection from the client to server using clickhouse-client. | 2018-09-04T07:14:55 | 2018-09-04T07:14:55 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 418,266,944 | 3,014 | lightee2 | @kshvakov thank you for your reply.
my golang program from local connected to another remote server which is running the clickhouse container .
i use telnet to test the remote server port 9000, the result is ok. | 2018-09-04T07:22:08 | 2018-09-04T07:22:08 | {} | NONE |
ClickHouse | ClickHouse | 418,297,765 | 3,014 | filimonov | > i use telnet to test the remote server port 9000, the result is ok.
golang also connects properly and start communication properly, but dies after few packets. So telnet connection is bad test. You should try to connect using clickhouse-client from one container to another. | 2018-09-04T09:14:00 | 2018-09-04T09:14:00 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 418,329,850 | 2,918 | blinkov | There're a lot of issues, but I'll fix 'em separately before publishing. | 2018-09-04T11:18:51 | 2018-09-04T11:18:51 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 418,550,847 | 2,441 | alexey-milovidov | @alesapin will take this task. | 2018-09-04T23:33:08 | 2018-09-04T23:33:08 | {} | MEMBER |
ClickHouse | ClickHouse | 418,733,943 | 2,441 | alexey-milovidov | We can pause this task till the end of September in favor of your changes. | 2018-09-05T13:38:54 | 2018-09-05T13:38:54 | {} | MEMBER |
ClickHouse | ClickHouse | 418,957,377 | 2,441 | zhang2014 | OK, thanks. I will finish it as soon as possible. | 2018-09-06T04:01:50 | 2018-09-06T04:01:50 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 419,187,345 | 3,059 | alexey-milovidov | It should work.
```
milovidov-Pro-P30 :) SELECT avg(x) FROM (SELECT arrayJoin([1, NULL]) AS x)
SELECT avg(x)
FROM
(
SELECT arrayJoin([1, NULL]) AS x
)
┌─avg(x)─┐
│ 1 │
└────────┘
1 rows in set. Elapsed: 0.047 sec.
milovidov-Pro-P30 :) SELECT avgArray(x) FROM (SELECT [1, NULL] AS x)
... | 2018-09-06T18:01:16 | 2018-09-06T18:01:16 | {
"+1": 2
} | MEMBER |
ClickHouse | ClickHouse | 418,101,106 | 3,014 | filimonov | Todo section of [readme](https://github.com/kshvakov/clickhouse#todo) of go driver by @kshvakov says that compression is not supported yet. Try remove `&compress=true` from connection string.
| 2018-09-03T12:28:38 | 2018-09-03T14:22:53 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 418,314,244 | 2,996 | blinkov | @fizerkhan do you have any further questions? | 2018-09-04T10:13:12 | 2018-09-04T10:13:12 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 418,958,318 | 3,040 | dsfishchen | Ok，thank you.
becouse of need to connect to clickhouse by DBeaver, so i choose an older jdbc driver. | 2018-09-06T04:08:52 | 2018-09-06T04:08:52 | {} | NONE |
ClickHouse | ClickHouse | 419,132,098 | 3,054 | alexey-milovidov | Good news are that recently we have added this feature to our plans.
| 2018-09-06T15:13:38 | 2018-09-06T15:13:38 | {} | MEMBER |
ClickHouse | ClickHouse | 418,321,085 | 2,996 | fizerkhan | Done https://stackoverflow.com/questions/52164354/safe-way-to-filters-including-account-id-or-user-id-in-sql-query
| 2018-09-04T10:40:51 | 2018-09-04T10:40:51 | {
"+1": 1
} | NONE |
ClickHouse | ClickHouse | 418,329,135 | 3,027 | fizerkhan | +1
| 2018-09-04T11:15:40 | 2018-09-04T11:15:40 | {} | NONE |
ClickHouse | ClickHouse | 418,352,301 | 3,027 | igor-sh8 | Ok.
I understand.
Thanks. | 2018-09-04T12:42:14 | 2018-09-04T12:42:14 | {} | NONE |
ClickHouse | ClickHouse | 418,415,031 | 3,033 | Farfaday | Many thanks !
(but i could not find the string tcp_port_secure in the documentation at https://clickhouse.yandex/docs/en/ ) | 2018-09-04T15:37:36 | 2018-09-04T15:37:36 | {} | NONE |
ClickHouse | ClickHouse | 418,620,561 | 3,040 | blinkov | You just need to consider updating ClickHouse server. The version you named is pretty old, not even from this year.
Or at least use older jdbc driver release, it won't rely on relatively recent features. | 2018-09-05T06:58:23 | 2018-09-05T06:58:23 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 418,800,436 | 3,029 | alexey-milovidov | This change breaks the test
`00634_performance_introspection_and_logging_shard`
metrics became inconsistent.
| 2018-09-05T16:46:43 | 2018-09-05T16:46:43 | {} | MEMBER |
ClickHouse | ClickHouse | 419,209,985 | 2,958 | kmatt | @alexey-milovidov Thanks. I rebuilt from a clean v18.12.3-testing without applying the PR and get the same exception. | 2018-09-06T19:15:40 | 2018-09-06T19:15:40 | {} | NONE |
ClickHouse | ClickHouse | 417,982,211 | 2,801 | amosbird | @VadimPE Hi, I didn't see failures in test 678 and there isn't a related commit in recent commit log. What problem is it? | 2018-09-03T02:16:29 | 2018-09-03T02:16:29 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 418,135,028 | 2,801 | VadimPlh | @amosbird
In this case
`CREATE TABLE test.remote_test(uid String, its UInt32, action_code String, day Date) ENGINE = MergeTree(day, (uid, its), 8192);`
`INSERT INTO test.remote_test SELECT toString(number) AS uid, number % 3 AS its, toString(number % 3) AS action_code, '2000-01-01' FROM system.numbers LIMIT 1000... | 2018-09-03T14:42:38 | 2018-09-03T14:43:31 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 418,669,830 | 3,040 | serebrserg | Actually you can use recent versions of clickhouse-jdbc as well with additional settings
either through ClickHouseProperties
```java
properties.setUseServerTimeZone(false);
```
or through Properties
```java
properties.setProperty("use_server_time_zone", "false")
```
and passing it to
ru.yandex.clickhouse.Cl... | 2018-09-05T09:46:46 | 2018-09-05T09:46:46 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 419,304,154 | 3,039 | TianyiLu | Yep... Your example query performs as same as my previous query. Both of them query from a sub-query result. When the query is heavy, you must load the sub-query result into memory, which costs a lot.
So, is there any other way to avoid query from a sub-query in my situation? | 2018-09-07T02:46:52 | 2018-09-07T02:46:52 | {} | NONE |
ClickHouse | ClickHouse | 418,146,108 | 3,026 | alexey-milovidov | Incorrect. | 2018-09-03T15:30:42 | 2018-09-03T15:30:42 | {} | MEMBER |
ClickHouse | ClickHouse | 418,350,155 | 3,027 | VadimPlh | @igor-sh8
1) If you do not stop CH, you will not be sure about the end of operations in CH. So it is bad solution.
2) Unfortunately, not | 2018-09-04T12:34:29 | 2018-09-04T12:34:29 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 418,605,968 | 3,040 | blinkov | @dsfishchen Which ClickHouse version do you use? | 2018-09-05T05:41:55 | 2018-09-05T05:41:55 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 418,731,662 | 3,044 | blinkov | Usually untrusted users are just marked as readonly: https://clickhouse.yandex/docs/en/operations/settings/query_complexity/#readonly | 2018-09-05T13:32:14 | 2018-09-05T13:32:14 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 418,896,196 | 2,995 | filimonov | #3052 | 2018-09-05T22:02:44 | 2018-09-05T22:02:44 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 419,390,954 | 867 | blinkov | English version is now also available on website. | 2018-09-07T10:06:05 | 2018-09-07T10:06:05 | {
"+1": 2
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 419,392,480 | 2,995 | blinkov | @filimonov looks way better now, thanks.
`sitemap_static.xml` has not been added to the list of what's published to website, but it'll get fixed in next release. | 2018-09-07T10:12:29 | 2018-09-07T10:12:29 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 418,047,312 | 3,023 | VadimPlh | #2758 | 2018-09-03T08:53:28 | 2018-09-03T08:53:28 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 418,101,941 | 2,999 | filimonov | Related #208 and #2825 | 2018-09-03T12:32:33 | 2018-09-03T12:43:46 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 419,128,199 | 3,060 | alexey-milovidov | 1. Leaked submodule.
2. Typo.
3. Test. | 2018-09-06T15:03:33 | 2018-09-06T15:03:33 | {} | MEMBER |
ClickHouse | ClickHouse | 419,283,348 | 3,066 | champtar | issue present on 18.12.6 | 2018-09-07T00:31:45 | 2018-09-07T00:31:45 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 418,106,197 | 3,001 | filimonov | There is no such a way currently.
There is similar issue about user agent: https://github.com/yandex/ClickHouse/issues/157#issuecomment-258265684
If you don't need high performance from that dictionary you can create http / or executable wrapper which will get list of key from CH, do matches / regexp agains exte... | 2018-09-03T12:51:05 | 2018-09-03T12:51:05 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 418,315,582 | 2,801 | VadimPlh | @amosbird #3032 | 2018-09-04T10:18:24 | 2018-09-04T10:18:24 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 418,319,487 | 2,996 | fizerkhan | Thanks @filimonov @blinkov. I will post it in the stackoverflow. | 2018-09-04T10:33:52 | 2018-09-04T10:33:52 | {} | NONE |
ClickHouse | ClickHouse | 418,416,439 | 3,033 | proller | But it described in default config https://github.com/yandex/ClickHouse/blob/master/dbms/programs/server/config.xml#L18
| 2018-09-04T15:41:24 | 2018-09-04T15:41:24 | {
"heart": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 418,608,014 | 3,040 | dsfishchen | I contrast this clickhouse to another normal clickhouse,the following functions was not found.
how can i repair it.
│ UUIDNumToString │ 0 │
│ UUIDStringToNum │ 0 │
│ arrayReverseSort │ 0 │
│ arraySort │ 0... | 2018-09-05T05:54:47 | 2018-09-05T05:54:47 | {} | NONE |
ClickHouse | ClickHouse | 418,636,094 | 3,039 | amosbird | That doesn't make sense to me as `argMax()` would collapse different values in the same key group. Can you elaborate with an example? | 2018-09-05T07:57:56 | 2018-09-05T07:57:56 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 419,023,705 | 2,990 | alesapin | Accidentally commited from wrong account :( | 2018-09-06T09:14:24 | 2018-09-06T09:14:24 | {} | MEMBER |
ClickHouse | ClickHouse | 418,229,609 | 3,001 | Slind14 | With the cache this might even work in this case. | 2018-09-04T03:19:31 | 2018-09-04T03:19:31 | {} | NONE |
ClickHouse | ClickHouse | 418,347,164 | 3,027 | igor-sh8 | Hi.
Thanks, @VadimPE.
As I understand, for copying data from the database directory I ideally need to stop clickhouse server ?
There is no other way? | 2018-09-04T12:23:17 | 2018-09-04T12:23:17 | {} | NONE |
ClickHouse | ClickHouse | 418,379,301 | 3,031 | alexey-milovidov | Do you want to implement the same solution for clickhouse-clang and clickhouse-lld? | 2018-09-04T14:04:39 | 2018-09-04T14:04:39 | {} | MEMBER |
ClickHouse | ClickHouse | 418,801,889 | 3,029 | alexey-milovidov | I reverted it before investigation. | 2018-09-05T16:51:07 | 2018-09-05T16:51:07 | {} | MEMBER |
ClickHouse | ClickHouse | 419,016,191 | 867 | francisdb | Thanks @BayoNet , would you mind adding a link to the specific pages? | 2018-09-06T08:49:40 | 2018-09-06T08:49:40 | {} | NONE |
ClickHouse | ClickHouse | 419,130,098 | 3,054 | alexey-milovidov | #147
#2210 | 2018-09-06T15:08:32 | 2018-09-06T15:08:32 | {} | MEMBER |
ClickHouse | ClickHouse | 419,449,364 | 3,016 | alexey-milovidov | There is no such metric.
Explanation of related metrics:
ProfileEvents.Query
ProfileEvents.SelectQuery
ProfileEvents.InsertQuery
- number of queries started in time interval (for example, one minute, if you export per-minute metrics to Graphite)
Metrics.Query
- number of queries currently executing and the... | 2018-09-07T14:03:01 | 2018-09-07T14:03:42 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 418,240,291 | 3,001 | filimonov | It will work quite good if most of the used keys (>99.9%) will fit into memory. Otherwise performance will not be good. | 2018-09-04T04:49:51 | 2018-09-04T04:49:51 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 418,289,743 | 3,014 | lightee2 | @filimonov thank you for your reply :)
OS :windows 10 64bit
GO:go version go1.10.3 windows/386
docker + ku8s
Client:
Version: 1.12.6
API version: 1.24
Go version: go1.7.4
Git commit: 78d1802
Built: Tue Mar 14 09:47:15 2017
OS/Arch: linux/amd64
Server:
Version: 1.12.6
A... | 2018-09-04T08:46:31 | 2018-09-04T08:46:31 | {} | NONE |
ClickHouse | ClickHouse | 418,924,702 | 2,822 | alexey-milovidov | Thank you!
Merged in #3048
@alesapin have added support for UUID in all remaining data sources: MySQL, MongoDB, etc.
We have a test suite for external dictionaries. It covers MySQL, MongoDB, ClickHouse, File, HTTP.
Though it doesn't cover ODBC... but the existing tests should be enough. | 2018-09-06T00:33:36 | 2018-09-06T00:33:36 | {} | MEMBER |
ClickHouse | ClickHouse | 419,053,769 | 867 | filimonov | https://clickhouse.yandex/docs/ru/data_types/nullable/
vs
https://clickhouse.yandex/docs/en/data_types/nullable/ ? | 2018-09-06T11:01:13 | 2018-09-06T11:01:13 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 419,467,994 | 3,061 | 4ertus2 | Reproduced with MergeTree on master
`USE test;
CREATE TABLE IF NOT EXISTS uuid
(
created_at DateTime,
id0 String,
id1 FixedString(36)
)
ENGINE = MergeTree
PARTITION BY toDate(created_at)
ORDER BY (created_at);
INSERT INTO uuid VALUES ('2018-01-01 01:02:03', '00000000-0000-03f8-9cb8-cb1b82fb39... | 2018-09-07T15:02:38 | 2018-09-07T15:02:38 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 419,491,116 | 3,061 | sjoerdmulder | Great :-) but one suggestion for the test is that the FixedString should be
16 chars (128bit) since it can store the binary representation.
Op vr 7 sep. 2018 17:02 schreef Artem Zuikov <notifications@github.com>:
> Reproduced with MergeTree on master
>
> `USE test;
> CREATE TABLE IF NOT EXISTS uuid
> (
> created_at D... | 2018-09-07T16:18:27 | 2018-09-07T16:18:27 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 418,046,730 | 3,023 | VadimPlh | Hi
It stores metadata about inserted chunks.
znodes inside log are cleaned after all replicas have processed these log entries.
If you have replicas that are abandoned forever, log will not be cleaned up. You have to remove abandoned replicas inside replicas znode.
Now you can delete some old entries inside log m... | 2018-09-03T08:51:17 | 2018-09-03T08:51:17 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 418,148,304 | 3,024 | 4ertus2 | Broken test: 00693_max_block_size_system_tables_columns
SELECT avg(blockSize()) <= 10000 FROM system.columns LIMIT 10 SETTINGS max_block_size = 10;
0
SELECT avg(blockSize()) FROM system.columns LIMIT 10 SETTINGS max_block_size = 10;
nan | 2018-09-03T15:41:15 | 2018-09-03T15:41:15 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 418,208,686 | 3,014 | lightee2 | @filimonov, thank you for your reply so soon,the clickserver i am using is the docker "https://hub.docker.com/r/yandex/clickhouse-server/" which point here "https://clickhouse.yandex/docs/en/getting_started/",the go program error output :+1:
Exception 0xc0000005 0x0 0x10 0x10
PC=0x10
runtime: unknown pc 0x10
st... | 2018-09-04T00:18:42 | 2018-09-04T00:18:42 | {} | NONE |
ClickHouse | ClickHouse | 418,576,523 | 2,441 | zhang2014 | Maybe this can be left to me? I have opened another pull request #2586 : ) | 2018-09-05T02:12:13 | 2018-09-05T02:12:13 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 419,053,963 | 867 | filimonov | > Also the issue is closed a bit early as that patch in English is not live yet, but will be hopefully today or maybe tomorrow.
aha, ok. | 2018-09-06T11:01:57 | 2018-09-06T11:01:57 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 418,036,085 | 3,005 | blinkov | Alternatively you could set up your favorite http load balancer (nginx, haproxy, etc.) and use it instead of going to ClickHouse directly. Also there's https://github.com/Vertamedia/chproxy | 2018-09-03T08:10:34 | 2018-09-03T08:10:34 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 418,118,405 | 2,996 | filimonov | That is general question, not Clickhouse related (answer would be the same for any DB) - please use StackOverflow for finding answers for such a problems.
Short answer: using user input directly is very unsafe, and subquery itself does not give you any guarantees and protects you from nothing, because user can type ... | 2018-09-03T13:39:05 | 2018-09-04T08:53:08 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 418,305,400 | 3,014 | lightee2 | @filimonov @kshvakov
thank you for you all reply,it is my turn to handle my own env problem. | 2018-09-04T09:39:12 | 2018-09-04T09:39:12 | {} | NONE |
ClickHouse | ClickHouse | 418,319,695 | 2,996 | blinkov | @fizerkhan when you will, leave here a link, please. | 2018-09-04T10:34:43 | 2018-09-04T10:34:43 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 418,456,613 | 3,033 | Farfaday | Could it be possible to use tcp_port_secure only and not tcp_port anymore ?
When I tried that, I got
`<Error> raw_events_masterdata.Channel_Adgroup_UserId (ReplicatedMergeTreeRestartingThread): void DB::ReplicatedMergeTreeRestartingThread::run(): Poco::Exception. Code: 1000, e.code() = 0, e.displayText() = Not fo... | 2018-09-04T17:42:20 | 2018-09-04T17:42:41 | {} | NONE |
ClickHouse | ClickHouse | 418,734,634 | 3,044 | shacuros | I think it makes sense to add a level for allowing inserts but not DDL statements like alter or drop or the like... | 2018-09-05T13:40:50 | 2018-09-05T13:40:50 | {} | NONE |
ClickHouse | ClickHouse | 418,929,065 | 2,990 | alexey-milovidov | It doesn't compile in Travis. | 2018-09-06T01:01:19 | 2018-09-06T01:01:19 | {} | MEMBER |
ClickHouse | ClickHouse | 418,946,875 | 2,958 | alexey-milovidov | It's not related to this PR.
And the branch `stable` doesn't contain any features related to `ThreadStatus`, `CurrentThread`, etc. It looks like you have built master or testing version. | 2018-09-06T02:50:41 | 2018-09-06T02:50:41 | {} | MEMBER |
ClickHouse | ClickHouse | 419,006,547 | 2,798 | BayoNet | Documentation is ready. | 2018-09-06T08:16:22 | 2018-09-06T08:16:22 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 418,743,686 | 3,044 | champtar | @shacuros +1, my "ingest" user doesn't need to do any DDL, and not even SELECT actually, just INSERT | 2018-09-05T14:06:01 | 2018-09-05T14:06:01 | {
"+1": 2
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 419,179,477 | 3,059 | VadimPlh | https://clickhouse.yandex/docs/en/query_language/agg_functions/reference/#avgx
NULL is not number. | 2018-09-06T17:37:09 | 2018-09-06T17:37:09 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 419,285,507 | 3,064 | blinkov | It has already been done in #3063 | 2018-09-07T00:46:33 | 2018-09-07T00:46:33 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 418,133,284 | 3,014 | filimonov | I've check that:
```go
package main
import (
"database/sql"
"fmt"
"log"
"github.com/kshvakov/clickhouse"
)
func main() {
connect, err := sql.Open("clickhouse", "tcp://remote_server:9000?username=&compress=true&debug=true")
if err != nil {
l... | 2018-09-03T14:35:30 | 2018-09-03T14:35:47 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 418,142,737 | 2,801 | amosbird | Well that's expected. The correct output should be
```
1 - 6667
2 - 3333
```
Old version doesn't work when the some filter includes others, and this is what my patch meant to fix. | 2018-09-03T15:15:27 | 2018-09-03T15:15:27 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 418,338,979 | 3,027 | VadimPlh | igor-sh8
Hi!
https://clickhouse.yandex/docs/en/query_language/alter/#backups-and-replication
You can copy data from the `/var/lib/clickhouse/data/database/table`, but it isn't a good way. | 2018-09-04T11:53:01 | 2018-09-04T11:53:01 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 418,607,071 | 3,040 | dsfishchen | ClickHouse server version 1.1.54023. | 2018-09-05T05:48:52 | 2018-09-05T05:48:52 | {} | NONE |
ClickHouse | ClickHouse | 418,668,746 | 3,039 | filimonov | 1) You can use FINAL keyword to get finally merged state of the table. It has some performance penalty, but may be it will be bit faster (or slower) than subselect, try just ```select sum(T), key2 from SomeTable FINAL where XXX condition group by key2```
2) That is one of cases where CollapsingMergeTree is better than... | 2018-09-05T09:42:58 | 2018-09-05T10:11:15 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 419,115,135 | 3,039 | filimonov | > So, do I must run this "old value fetching" query before I can update the value by CollapsingMergeTree?
yes, if you like to go that direction.
If you don't like both approaches you can also scale up you cluster to improve performance.
In general case it's not possible to execute mutiple "aggregation functio... | 2018-09-06T14:30:54 | 2018-09-06T15:03:34 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 419,431,136 | 3,016 | lfundaro | Hey guys, I didn't see/understand in this documentation, if one can know what's the time a select query takes. From my clickhouse logs I see how much it takes, but I was wondering if i can get the same in metrics sent to graphite. Thanks ! | 2018-09-07T12:58:03 | 2018-09-07T12:58:03 | {} | NONE |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.