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 | 460,196,002 | 4,194 | alesapin | @Rakovskiy28 Напишите версию сервера пожалуйста. | 2019-02-04T10:21:53 | 2019-02-04T10:21:53 | {} | MEMBER |
ClickHouse | ClickHouse | 460,264,595 | 4,251 | alexey-milovidov | Merged manually. | 2019-02-04T14:16:54 | 2019-02-04T14:16:54 | {} | MEMBER |
ClickHouse | ClickHouse | 460,279,909 | 3,574 | alexey-milovidov | Implemented in #4230 | 2019-02-04T15:00:15 | 2019-02-04T15:00:15 | {} | MEMBER |
ClickHouse | ClickHouse | 460,302,135 | 4,258 | alexey-milovidov | `pageviewId = 'eda647528b9084461a5b2025cbb0e991\0\0\0\0'` will work as expected. | 2019-02-04T15:58:23 | 2019-02-04T15:58:30 | {} | MEMBER |
ClickHouse | ClickHouse | 460,764,538 | 4,253 | alexey-milovidov | It is possible if you use independent MergeTree tables and Distributed table just looks at them as replicas. For data replication, you can INSERT into Distributed table (cluster should has `internal_replication` set to false) or INSERT data to both replicas directly. Data consistency between replicas will not be mainta... | 2019-02-05T19:09:58 | 2019-02-05T19:09:58 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 460,833,131 | 4,283 | alexey-milovidov | > Is there any recommendation if we want to have the queries waiting instead of timeout.
I don't recomment to have a queue of waiting queries at all.
If the system cannot keep up with load, a queue will not help much.
> What is the default value for `queue_max_wait_ms`
Five seconds.
```
:) SELECT * FROM sys... | 2019-02-05T22:41:25 | 2019-02-05T22:41:25 | {} | MEMBER |
ClickHouse | ClickHouse | 460,967,260 | 4,046 | filimonov | > It can be related to insertion to `Distributed` tables.
I was also thinking about that, but i can't reproduce that in simple synthetic test:
```bash
# setup tables
cat <<END_HEREDOC | clickhouse-client -n
CREATE DATABASE IF NOT EXISTS tests;
DROP TABLE IF EXISTS tests.issue4046;
DROP TABLE IF EXISTS test... | 2019-02-06T10:07:06 | 2019-02-06T10:27:29 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 461,179,914 | 4,265 | mcarbonneaux | https://www.tecmint.com/convert-from-rpm-to-deb-and-deb-to-rpm-package-using-alien/
This link explain the problem. | 2019-02-06T20:43:22 | 2019-02-06T20:43:22 | {
"+1": 1
} | NONE |
ClickHouse | ClickHouse | 461,513,637 | 3,779 | filimonov | > not sure my problem is the same
But it looks like you have a normal client error, it says that number of columns in source and destination tables not match.
List the exact columns to insert and to select. | 2019-02-07T17:07:13 | 2019-02-07T17:08:36 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 460,048,935 | 4,147 | alexey-milovidov | First part: #4245 | 2019-02-03T12:50:52 | 2019-02-03T12:50:52 | {} | MEMBER |
ClickHouse | ClickHouse | 460,295,652 | 4,259 | alesapin | Now it's possible to write
`VERSION_MAJOR=19 VERSION_MINOR=1 VERSION_PATCH=7 NO_PUSH=1 bash -c './release --version env'` | 2019-02-04T15:41:31 | 2019-02-04T15:41:31 | {} | MEMBER |
ClickHouse | ClickHouse | 460,458,222 | 4,268 | rekby | Yes, it's help and debian not accept install without --allow-unauthenticated | 2019-02-04T23:42:53 | 2019-02-04T23:42:53 | {} | NONE |
ClickHouse | ClickHouse | 460,584,015 | 4,270 | alesapin | Wait for test | 2019-02-05T10:18:59 | 2019-02-05T10:18:59 | {} | MEMBER |
ClickHouse | ClickHouse | 460,764,061 | 4,263 | greenwolf-nsk | >addr2line -iCe /usr/bin/clickhouse-server 0x32c7a4b
```
/usr/include/c++/7/bits/shared_ptr_base.h:151
/usr/include/c++/7/bits/shared_ptr_base.h:684
/usr/include/c++/7/bits/shared_ptr_base.h:1123
/usr/include/c++/7/bits/shared_ptr.h:93
/tmp/buildd/clickhouse-18.16.1/build/../dbms/src/Core/NamesAndTypes.h:16
/usr... | 2019-02-05T19:08:39 | 2019-02-05T19:08:39 | {} | NONE |
ClickHouse | ClickHouse | 460,976,473 | 4,265 | filimonov | > they don't install for now... without correcting the packaging process they are useless...
Yes, that is the reason why they are not listed / supported officially.
> but why not integrate the rpm packaging from altinity to your master and generated rpm in https://repo.yandex.ru/ with it ?
Currently, it's a l... | 2019-02-06T10:37:51 | 2019-02-06T10:39:29 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 461,502,725 | 3,779 | mkarmona | the same version is having problems with compressed cache as it is currently disabled | 2019-02-07T16:38:43 | 2019-02-07T16:38:43 | {} | NONE |
ClickHouse | ClickHouse | 460,251,598 | 4,213 | nvartolomei | @alexey-milovidov the test is in, let me know if it works.
re readonly guard, I didn't want to alter the behaviour for pending inserts, alters or any other similar queries. Although this better be handled in other places (it might be actually), I wanted to be extra cautious. (also, initially it was implemented witho... | 2019-02-04T13:35:02 | 2019-02-04T13:35:02 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 460,254,389 | 4,250 | den-crane | I think MV solves **test JOIN test** over inserted buffer not over real table.
And this a bad idea because CH's join places a right table to the memory, so eventually it will stop working with out of memory.
Try another approach
https://gist.github.com/den-crane/49ce2ae3a688651b9c2dd85ee592cb15
https://gist.gith... | 2019-02-04T13:44:28 | 2019-02-04T13:44:28 | {
"+1": 3
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 460,302,651 | 4,258 | alexey-milovidov | Because your `pageviewId` have just 32 significant bytes, but you store them in `FixedString(36)`. | 2019-02-04T15:59:36 | 2019-02-04T15:59:36 | {} | MEMBER |
ClickHouse | ClickHouse | 460,604,094 | 3,145 | 4ertus2 | > Actually the problem that * with using should not return columns from right table at all.
Current behaviour * with using is return all columns from left and right table except duplicates. It's almost like MySQL workd with using(). So this result (master) is correct: all the columns except duplicates.
```
create ta... | 2019-02-05T11:28:37 | 2019-02-05T16:49:46 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 461,100,286 | 4,292 | dusze | Thanks! | 2019-02-06T16:58:14 | 2019-02-06T16:58:14 | {} | NONE |
ClickHouse | ClickHouse | 461,119,845 | 4,282 | alexey-milovidov | Is the issue anyhow related to the Kafka table?
(Does it reproduce if you first remove the Kafka table and materialized view and only then issue an ALTER command?)
Could you please install `clickhouse-common-dbg` package, reproduce the issue and run
`sudo gdb -batch -ex 't apply all bt' -p $(pidof clickhouse-serve... | 2019-02-06T17:51:31 | 2019-02-06T17:51:57 | {} | MEMBER |
ClickHouse | ClickHouse | 461,812,759 | 4,312 | den-crane | the same with ClickHouse server version 19.1.6
| 2019-02-08T14:04:25 | 2019-02-08T14:04:25 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 460,250,870 | 4,253 | den-crane | Replicated*MergeTree -- No.
You can use smaller partitions (partition by daily). In this case [where EventDate = ] will scan only one partition. | 2019-02-04T13:32:35 | 2019-02-04T13:32:35 | {
"+1": 2
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 460,748,848 | 4,283 | alexey-milovidov | There is `queue_max_wait_ms` setting.
It is user level setting, that can be set up per user profile in `users.xml` configuration file.
When the max wait time is exceeded, ClickHouse will throw an exception nevertheless. | 2019-02-05T18:26:22 | 2019-02-05T18:26:22 | {} | MEMBER |
ClickHouse | ClickHouse | 461,801,892 | 4,301 | alesapin | Unfortunately I had to modify ASTInsertQuery outside of parser https://github.com/yandex/ClickHouse/pull/4301/files#diff-387f58c9305eb5dcc6cdb110eaf1382fR172. Is it OK, or it should be avoided? | 2019-02-08T13:26:22 | 2019-02-08T13:26:22 | {} | MEMBER |
ClickHouse | ClickHouse | 460,565,893 | 4,265 | filimonov | I don't recommend using rpms from https://repo.yandex.ru/ for now (as i understand it's just repacked deb packages with some automatic converting tool).
For now it's better to use that repo https://packagecloud.io/Altinity/clickhouse
Or build rpm by your own with scripts from https://github.com/Altinity/clickhou... | 2019-02-05T09:19:47 | 2019-02-05T09:20:06 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 460,620,618 | 3,852 | alexey-milovidov | @bgranvea When you SELECT from AggregatingMergeTree (in contrast to ReplacingMergeTree), you usually do additional aggregation with `-Merge` aggregate functions.
As the alternative, you can try to select partially aggregated values without final aggregation by using `finalizeAggregation` function. This is ordinary f... | 2019-02-05T12:29:53 | 2019-02-05T12:30:12 | {} | MEMBER |
ClickHouse | ClickHouse | 460,645,023 | 4,215 | abyss7 | The `stateless.00411_long_accurate_number_comparison` has timed out on Jenkins, while passed fine on Sandbox. | 2019-02-05T13:53:10 | 2019-02-05T13:53:10 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 461,628,092 | 4,308 | daledude | btw, in my case a few queues would take about 15 minutes to "fill" and be committed. So, thats 15 minutes the DROP and a shutdown (hard kill happens before this time, however) would "freeze." | 2019-02-07T23:01:26 | 2019-02-07T23:01:26 | {} | NONE |
ClickHouse | ClickHouse | 461,911,502 | 4,317 | alexey-milovidov | Proposed test:
```
cat 00839_client_exit.sh
#!/usr/bin/env bash
set -e
CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
. $CURDIR/../shell_config.sh
echo '
spawn '$CLICKHOUSE_CLIENT'
expect -ex ":)"
send "exit\r"
expect eof
' | /usr/bin/env expect -
```
But I cannot catch segfault from `exp... | 2019-02-08T19:07:52 | 2019-02-08T19:07:52 | {} | MEMBER |
ClickHouse | ClickHouse | 460,076,002 | 4,213 | alexey-milovidov | Let's write a test with cancellation of infinite query like `SELECT count() FROM system.numbers`. | 2019-02-03T18:31:55 | 2019-02-03T18:31:55 | {} | MEMBER |
ClickHouse | ClickHouse | 460,536,219 | 4,200 | DarkWanderer | Added input format support + test for output format.
TODO: test for input format | 2019-02-05T07:14:53 | 2019-02-05T07:14:53 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 461,425,450 | 4,198 | edonin | So, you are totally right :
This pull request adds support for Nullable types in `mysql` table function (type inference from mysql tables).
A new setting has been added called 'external_tables_use_nulls' (enable by default) for disabling this null management (currently used only in mysql function).
TODO: integrat... | 2019-02-07T13:48:21 | 2019-02-07T13:48:21 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 461,830,392 | 2,829 | fessmage | Same problem with same database - GeoLite2-City-Locations. We definitely need this feature for dictionaries. | 2019-02-08T15:00:45 | 2019-02-08T15:00:45 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 461,869,632 | 3,299 | Crusader4Christ | I can`t use LowCardinality(UInt8) in dictionaries?
Because I get the "Unknown type LowCardinality(UInt8)" | 2019-02-08T16:54:03 | 2019-02-08T16:54:03 | {} | NONE |
ClickHouse | ClickHouse | 460,098,557 | 4,246 | alexey-milovidov | > Нужно запустить сервер - он у меня запускается, но с какими-то подозрительными сообщениями: http://bit.ly/2UEHrCk
Судя по выводу, всё Ок.
Include not found - сообщение про подстановки в конфигурационном файле config.xml, которые могут загружаться из других файлов, но других файлов нет.
> Нужно скопироват... | 2019-02-03T23:18:31 | 2019-02-03T23:20:05 | {} | MEMBER |
ClickHouse | ClickHouse | 460,563,166 | 4,200 | filimonov | BTW: if there some chance to wrap AggregateFunction states somehow in RowBinary? For example - add LEB128-encoded length prefix before the value, just to be able to skip that column if it comes in the output.
If it will be added just for new format ( RowBinaryWithNamesAndTypes ) then no backward compatibility issues... | 2019-02-05T09:10:14 | 2019-02-05T09:11:12 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 460,952,493 | 3,779 | bgranvea | Strangely, in my case there is a crash instead of an exception if I execute the statement over HTTP:
2019.02.06 10:14:28.135442 [ 2 ] {} <Error> BaseDaemon: ########################################
2019.02.06 10:14:28.135572 [ 2 ] {} <Error> BaseDaemon: (from thread 53) Received signal Segmentation fault (11).
201... | 2019-02-06T09:18:22 | 2019-02-06T09:18:22 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 461,321,541 | 4,265 | filimonov | For repacking also there are
https://github.com/jordansissel/fpm
https://github.com/goreleaser/nfpm
| 2019-02-07T08:02:12 | 2019-02-07T08:02:19 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 461,815,222 | 3,985 | Crusader4Christ | Same problem with MySQL dictionaries:
```
<dictionary>
<name>tt</name>
<source>
<mysql>
...
<table>tt</table>
</mysql>
</source>
...
<structure>
<attribute>
<name>plan</name>
<type>Enum8(...... | 2019-02-08T14:13:03 | 2019-02-08T14:13:03 | {
"confused": 1
} | NONE |
ClickHouse | ClickHouse | 460,239,492 | 4,177 | alexey-milovidov | Fixed in master. | 2019-02-04T12:54:52 | 2019-02-04T12:54:52 | {} | MEMBER |
ClickHouse | ClickHouse | 460,301,855 | 4,258 | alexey-milovidov | Values in FixedString data are padded by 0-bytes. | 2019-02-04T15:57:41 | 2019-02-04T15:57:41 | {} | MEMBER |
ClickHouse | ClickHouse | 460,947,453 | 3,779 | bgranvea | I get the same error with another test case. I've just tested with latest stable version 19.1.6, it still fails but with another message:
Code: 271. DB::Exception: Data compressed with different methods, given method byte 69, previous method byte 82: while receiving packet from localhost:9000, 127.0.0.1
| 2019-02-06T09:00:53 | 2019-02-06T09:00:53 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 460,988,767 | 4,286 | nikvas0 | https://github.com/yandex/ClickHouse/pull/4143#discussion_r253052648
MergeTreeReaderStream is moved MergeTreeDataReader::Stream, so I decided not to touch this code in order not to make more bugs. | 2019-02-06T11:21:21 | 2019-02-06T11:21:21 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 461,515,014 | 4,305 | proller | ```
../dbms/src/IO/AIO.cpp:115:6: warning: variable 'r' may be uninitialized when used here [-Wconditional-uninitialized]
if (r < 0) {
^
../dbms/src/IO/AIO.cpp:93:10: note: initialize the variable 'r' to silence this warning
int r;
^
= 0
../dbms/src/IO/AIOContextPool... | 2019-02-07T17:10:58 | 2019-02-07T17:10:58 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 461,520,339 | 3,779 | mkarmona | @filimonov there are the same columns there; the only diff is I am grouping some of them as nested structure. I am probably missing something here?
```
DESCRIBE TABLE ot.v2d_by_chrpos
... | 2019-02-07T17:25:02 | 2019-02-07T17:26:28 | {} | NONE |
ClickHouse | ClickHouse | 461,813,176 | 4,312 | alesapin | :( | 2019-02-08T14:05:56 | 2019-02-08T14:05:56 | {} | MEMBER |
ClickHouse | ClickHouse | 460,180,221 | 4,244 | k-lopatin | https://github.com/yandex/ClickHouse/pull/4251/files | 2019-02-04T09:27:06 | 2019-02-04T09:27:06 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 460,360,190 | 4,253 | nvartolomei | You can create a materialized view with a different sorting/primary key if that is required. | 2019-02-04T18:34:31 | 2019-02-04T18:34:31 | {
"+1": 2
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 460,678,660 | 4,267 | artbeglaryan | Thanks!! Looks I confused with some docs and late night config. | 2019-02-05T15:24:46 | 2019-02-05T15:24:46 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 461,127,962 | 4,277 | alexey-milovidov | Confirmed.
It can be easily reproduced by the following command:
```
for i in {1..1000}; do clickhouse-client --max_memory_usage 100000000 <<< "SELECT uniqExactState(number) FROM system.numbers GROUP BY number % 1000"; done
``` | 2019-02-06T18:13:55 | 2019-02-06T18:13:55 | {
"+1": 2
} | MEMBER |
ClickHouse | ClickHouse | 461,178,606 | 4,265 | mcarbonneaux | I found it beceause is listed in documentation... Before the altinity repo....
https://clickhouse.yandex/docs/en/getting_started/
> Packages from https://repo.yandex.ru/clickhouse/rpm/stable/x86_64/ are generated from official deb packages by Yandex and have byte-identical binaries.
But you have standalone b... | 2019-02-06T20:39:21 | 2019-02-06T20:41:48 | {} | NONE |
ClickHouse | ClickHouse | 460,303,188 | 4,258 | alexey-milovidov | This is documented in https://clickhouse.yandex/docs/en/data_types/fixedstring/ | 2019-02-04T16:01:01 | 2019-02-04T16:01:01 | {} | MEMBER |
ClickHouse | ClickHouse | 460,577,682 | 4,255 | alexey-milovidov | `test.py::test_reconnect` - flappy test. | 2019-02-05T09:58:11 | 2019-02-05T09:58:11 | {} | MEMBER |
ClickHouse | ClickHouse | 460,766,167 | 4,283 | dsr301 | Thanks for the quick reply.
Is there any recommendation if we want to have the queries waiting instead of timeout. Will ch_proxy help?
What is the default value for `queue_max_wait_ms`, I don't see it mentioned in users.xml | 2019-02-05T19:14:38 | 2019-02-05T19:14:38 | {} | NONE |
ClickHouse | ClickHouse | 461,495,621 | 4,305 | proller | ```
../dbms/src/IO/AIO.cpp: In function 'int io_getevents(int, long int, long int, kevent*, timespec*)':
../dbms/src/IO/AIO.cpp:123:32: warning: parameter 'min_nr' set but not used [-Wunused-but-set-parameter]
int io_getevents(int ctx, long min_nr, long max_nr, struct kevent * events, struct timespec * timeout)
... | 2019-02-07T16:22:12 | 2019-02-07T16:22:12 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 462,020,274 | 4,207 | andyyzh | clang-7 build mode: the files I submitted has passed the compiled. An error occurred when linker libprotobuf.a. I don't know the reason. The error was reported as follows:
> /opt/rh/devtoolset-7/root/usr/lib/gcc/x86_64-redhat-linux/7/../../../../include/c++/7/iostream:74: undefined reference to `std::ios_base::Init::~... | 2019-02-09T07:00:58 | 2019-02-09T07:00:58 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 460,051,902 | 2,684 | alexey-milovidov | @CompadreP
The index you have in ClickHouse is wrong.
The index should be: `ORDER BY (kkt_sn, created_at_datetime)`
But you have `ORDER BY (created_at_datetime, kkt_sn)`
PS. Can you please share your script for data generation?
I will try to reproduce these results.
@filimonov
> And 'key-value' usage doe... | 2019-02-03T13:30:23 | 2019-02-03T23:23:51 | {} | MEMBER |
ClickHouse | ClickHouse | 460,576,367 | 4,255 | alexey-milovidov | This PR closes #2870 and #2565. | 2019-02-05T09:54:03 | 2019-02-05T09:54:50 | {} | MEMBER |
ClickHouse | ClickHouse | 460,733,595 | 4,282 | kmatt | Example table structure for shards and Kafka consumers. Distributed engine is not used in this case:
```sql
CREATE TABLE logs_shard
(
shard_key UInt64,
id String,
dateTime DateTime,
...
)
ENGINE = MergeTree
PARTITION BY (toStartOfHour(dateTime), id)
ORDER BY (cid, dateTime)
SETTINGS index_... | 2019-02-05T17:43:31 | 2019-02-05T17:43:31 | {} | NONE |
ClickHouse | ClickHouse | 461,188,403 | 4,265 | mcarbonneaux | You can use rpmrebuild for filtering files of genereted rpm. | 2019-02-06T21:09:59 | 2019-02-06T21:09:59 | {} | NONE |
ClickHouse | ClickHouse | 460,246,256 | 4,252 | alexey-milovidov | Можно сделать вторым, необязательным параметром.
`topK(10, 3)(x)` | 2019-02-04T13:16:39 | 2019-02-04T13:16:55 | {} | MEMBER |
ClickHouse | ClickHouse | 460,612,844 | 971 | 4ertus2 | Reproduced on master with PREWHERE only. WHERE section knows about qualified name.
No matter was there ALTER before or not. | 2019-02-05T12:00:51 | 2019-02-05T12:03:16 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 460,703,025 | 4,263 | vitlibar | Can you please install `clickhouse-common-static-dbg` and `binutils` and then show what output the following two lines produce:
```
addr2line -iCe /usr/bin/clickhouse-server 0x32c7a4b
addr2line -iCe /usr/bin/clickhouse-server 0x6f9d50a
```
| 2019-02-05T16:24:28 | 2019-02-05T16:24:28 | {} | MEMBER |
ClickHouse | ClickHouse | 461,501,346 | 3,779 | mkarmona | not sure my problem is the same here but having a
```
INSERT INTO ot.v2d_by_chrpos SELECT *
FROM ot.v2d_log
Received exception from server (version 19.1.6):
Code: 20. DB::Exception: Received from localhost:9000, ::1. DB::Exception: Number of columns doesn't match.
0 r... | 2019-02-07T16:35:13 | 2019-02-07T16:37:49 | {} | NONE |
ClickHouse | ClickHouse | 460,138,564 | 2,061 | alniks | Looks like it is fixed in 19.1.6 | 2019-02-04T05:53:13 | 2019-02-04T05:53:13 | {} | NONE |
ClickHouse | ClickHouse | 460,200,126 | 4,047 | alesapin | @andywow Unfortunately I can't get any useful information from your log :( Seems like you get some other problem. Can you describe your shard configuration? Do you update all shard servers to 19.1.6? Can you provide logs from other server? Maybe we can discuss this issue in details in telegram? | 2019-02-04T10:36:16 | 2019-02-04T10:42:18 | {} | MEMBER |
ClickHouse | ClickHouse | 460,627,463 | 4,200 | alexey-milovidov | > BTW: if there some chance to wrap AggregateFunction states somehow in RowBinary? For example - add LEB128-encoded length prefix before the value, just to be able to skip that column if it comes in the output.
> If it will be added just for new format ( RowBinaryWithNamesAndTypes ) then no backward compatibility is... | 2019-02-05T12:54:36 | 2019-02-05T12:54:36 | {} | MEMBER |
ClickHouse | ClickHouse | 460,632,818 | 4,198 | alexey-milovidov | @edonin Could you please add a setting `external_tables_use_nulls`? | 2019-02-05T13:13:17 | 2019-02-05T13:13:17 | {} | MEMBER |
ClickHouse | ClickHouse | 460,764,981 | 4,244 | alexey-milovidov | Thank you! | 2019-02-05T19:11:19 | 2019-02-05T19:11:19 | {} | MEMBER |
ClickHouse | ClickHouse | 460,765,221 | 4,263 | greenwolf-nsk | Also, there are several different addresses appearing in segfaults:
> grep "Received signal Segmentation fault" -A 3 clickhouse-server.err.log
```
2019.02.05 17:12:01.536792 [ 2 ] {} <Error> BaseDaemon: (from thread 115263) Received signal Segmentation fault (11).
2019.02.05 17:12:01.536815 [ 2 ] {} <Error> BaseDa... | 2019-02-05T19:11:58 | 2019-02-05T19:11:58 | {
"+1": 1
} | NONE |
ClickHouse | ClickHouse | 460,848,361 | 3,145 | den-crane | some examples https://den-crane.github.io/CHvsPG.html | 2019-02-05T23:39:16 | 2019-02-05T23:39:16 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 461,414,204 | 3,920 | alesapin | @amosbird What is 'hashfile' you referenced in your test? | 2019-02-07T13:16:04 | 2019-02-07T13:16:04 | {} | MEMBER |
ClickHouse | ClickHouse | 461,626,720 | 4,308 | daledude | Setting kafka_max_block_size per kafka table does keep the data flowing through the materialized view.
The problem of waiting for the queue to fill and be written also prevents clickhouse from being shut down. Ultimately the init script does a hard kill of clickhouse because it's waiting for the queue to fill. | 2019-02-07T22:56:08 | 2019-02-07T22:56:08 | {} | NONE |
ClickHouse | ClickHouse | 460,261,420 | 4,194 | den-crane | ЕМНИП, вы храните текстовые логи? Какая средняя длина строки? | 2019-02-04T14:07:01 | 2019-02-04T14:07:01 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 460,577,480 | 2,870 | alexey-milovidov | Fixed in master. | 2019-02-05T09:57:32 | 2019-02-05T09:57:32 | {} | MEMBER |
ClickHouse | ClickHouse | 460,627,720 | 4,200 | alexey-milovidov | Input format reminds me the weird issue #3469 that should be fixed first. | 2019-02-05T12:55:38 | 2019-02-05T12:55:38 | {} | MEMBER |
ClickHouse | ClickHouse | 460,631,363 | 4,198 | alexey-milovidov | Clarification: it was already supported for MySQL table engine if you specify Nullable types explicitly.
See https://github.com/yandex/ClickHouse/pull/3362#issuecomment-429184331
This pull request adds support for Nullable types in `mysql` table function (type inference from mysql tables).
I'd like to put this fea... | 2019-02-05T13:08:04 | 2019-02-05T13:08:04 | {} | MEMBER |
ClickHouse | ClickHouse | 460,712,159 | 4,265 | mcarbonneaux | > I don't recommend using rpms from https://repo.yandex.ru/ for now (as i understand it's just repacked deb packages with some automatic converting tool).
they don't install for now... without correcting the packaging process they are useless...
i've successfully installed the altinity version.
but why not int... | 2019-02-05T16:47:45 | 2019-02-05T17:30:25 | {} | NONE |
ClickHouse | ClickHouse | 461,749,996 | 4,143 | nikvas0 | Fixed in https://github.com/yandex/ClickHouse/pull/4286 | 2019-02-08T09:57:35 | 2019-02-08T09:57:35 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 461,808,476 | 3,982 | alexey-milovidov | Broken in https://github.com/yandex/ClickHouse/pull/3609 | 2019-02-08T13:49:13 | 2019-02-08T13:49:13 | {} | MEMBER |
ClickHouse | ClickHouse | 461,854,288 | 3,918 | halayli | I would also consider `t1ha` hash family. `t1ha_aes` performs particularly well on aes supported cpus.
https://github.com/leo-yuriev/t1ha | 2019-02-08T16:09:47 | 2019-02-08T16:10:01 | {} | NONE |
ClickHouse | ClickHouse | 460,089,468 | 4,246 | k3box | Здравствуйте, я вроде как написал код по задаче "Произвольные константные выражения в LIMIT", но не могу его протестировать :/
1. Нужно запустить сервер - он у меня запускается, но с какими-то подозрительными сообщениями: http://bit.ly/2UEHrCk
2. Нужно скопировать клиент в /usr/bin, но на маке, кажется, нельзя сделат... | 2019-02-03T21:19:04 | 2019-02-03T21:19:04 | {} | NONE |
ClickHouse | ClickHouse | 460,456,737 | 4,268 | vitlibar | Try to install the package dirmngr:
```
sudo apt-get install dirmngr
```
then run
```
sudo apt-key adv --keyserver keyserver.ubuntu.com --recv E0C56BD4
```
| 2019-02-04T23:35:56 | 2019-02-04T23:35:56 | {} | MEMBER |
ClickHouse | ClickHouse | 460,606,506 | 4,271 | 4ertus2 | Possibly related errors #4222 #971 | 2019-02-05T11:38:16 | 2019-02-05T12:21:31 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 460,799,586 | 4,271 | 4ertus2 | fixed in master | 2019-02-05T20:54:26 | 2019-02-05T20:54:26 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 461,053,329 | 4,292 | filimonov | > What am I doing wrong?
Remove ```incl="zookeeper-servers" optional="true"``` part from zookeeper tag. So just:
```xml
<zookeeper>
<node index="1">
<host>ent5</host>
<port>2181</port>
</node>
<node index="2">
<host>ent6</host>
<port>2181</po... | 2019-02-06T14:58:30 | 2019-02-06T15:10:42 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 461,504,774 | 4,305 | proller | ```
../dbms/src/IO/WriteBufferAIO.cpp:202:32: warning: use of old-style cast [-Wold-style-cast]
bytes_written = aio_return((struct aiocb *)event.udata);
^ ~~~~~~~~~~~
1 warning generated.
[1093/1771] Building CXX object dbms/CMakeFiles/clickhouse_common_io.dir/src/I... | 2019-02-07T16:44:09 | 2019-02-07T16:44:09 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 461,644,144 | 3,779 | mkarmona | @filimonov thank you very much! I didn't realise about the materialized columns aren't selected with the `select *` | 2019-02-08T00:12:30 | 2019-02-08T00:12:30 | {} | NONE |
ClickHouse | ClickHouse | 461,962,926 | 4,305 | alexey-milovidov | Missing changelog.
Added. | 2019-02-08T22:05:21 | 2019-02-08T22:06:37 | {} | MEMBER |
ClickHouse | ClickHouse | 462,042,023 | 2,805 | dsetrakyan | I like the ability to change the query used for non-materialized views:
* ALTER VIEW view_name() AS new_query
Has this been implemented yet? | 2019-02-09T12:50:30 | 2019-02-09T12:50:30 | {} | NONE |
ClickHouse | ClickHouse | 460,252,627 | 4,194 | Rakovskiy28 | > @Rakovskiy28 Напишите версию сервера пожалуйста.
ClickHouse server version 18.16.1 revision 54412.
Ещё заметил такой момент. Работал скрипт, данные писались в БД, на этот момент юзалось 70гб RAM, после чего остановил скрипт и часов 5-7 КХ вообще не юзал, но htop показывает, что КХ так и использует 70гб RAM. Т.е... | 2019-02-04T13:38:27 | 2019-02-04T13:38:27 | {} | NONE |
ClickHouse | ClickHouse | 460,471,864 | 4,267 | den-crane | remote_servers -- it's for Engine=Distributed it's not related to ReplicatedMergeTree
replication works over http < interserver_http_port > (9009)
and hostname of replica < interserver_http_host > to get the data (and it is placed to zk)
```
<!-- Port for communication between replicas. Used for data exch... | 2019-02-05T00:46:58 | 2019-02-05T00:47:33 | {
"hooray": 2
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 460,731,536 | 4,258 | xmariachi | Perfect, thanks! Apologies. | 2019-02-05T17:37:59 | 2019-02-05T17:39:41 | {} | NONE |
ClickHouse | ClickHouse | 460,798,339 | 4,216 | 4ertus2 | #4218 | 2019-02-05T20:50:47 | 2019-02-05T20:50:47 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 461,569,103 | 4,226 | aahmed-se | pulsar cpp has been ported to the buckaroo package manager
https://github.com/njlr/apache-pulsar-demo | 2019-02-07T19:43:44 | 2019-02-07T19:43:44 | {} | NONE |
ClickHouse | ClickHouse | 461,624,132 | 3,779 | filimonov | > @filimonov there are the same columns there; the only diff is I am grouping some of them as nested structure. I am probably missing something here?
>
According to [docs](https://clickhouse.yandex/docs/en/query_language/create/#default-values): Materialized column is not substituted when using an asterisk in a SE... | 2019-02-07T22:45:48 | 2019-02-07T22:48:26 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 461,723,545 | 3,779 | bgranvea | just to go back to the initial issue, here is my test case if it can help:
```
CREATE TABLE table1 (A String, B String, ts DateTime) ENGINE = MergeTree PARTITION BY toStartOfDay(ts) ORDER BY (ts, A, B);
CREATE TABLE table2 (B String, ts DateTime) ENGINE = MergeTree PARTITION BY toStartOfDay(ts) ORDER BY (ts, B);
... | 2019-02-08T08:12:47 | 2019-02-08T08:12:47 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 461,826,396 | 4,312 | den-crane | set optimize_move_to_prewhere=0
```
create table foo_c1(d DateTime) engine = MergeTree order by tuple();
insert into foo_c1 values
('2019-02-06 01:01:01'),('2019-02-07 01:01:01'),('2019-02-08 01:01:01'),('2088-05-09 01:01:01'),
('2093-05-29 01:01:01'),('2100-06-06 01:01:01'),
('2100-10-14 01:01:01'),('2100-... | 2019-02-08T14:48:40 | 2019-02-08T14:48:40 | {} | CONTRIBUTOR |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.