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 | 596,770,581 | 11 | waralex | Maybe this is a crazy idea, but that about the [Julia](https://julialang.org/) language as UDF language? It's fast, can broadcast function calls to arrays out of box and can be embedded into c/c++ code. | 2020-03-09T20:45:53 | 2020-03-09T20:45:53 | {
"+1": 11
} | NONE |
ClickHouse | ClickHouse | 597,488,571 | 9,410 | alexey-milovidov | ? | 2020-03-11T07:43:33 | 2020-03-11T07:43:33 | {} | MEMBER |
ClickHouse | ClickHouse | 597,080,639 | 9,583 | alesapin | @VadimPlh take a look. | 2020-03-10T13:18:48 | 2020-03-10T13:18:48 | {} | MEMBER |
ClickHouse | ClickHouse | 597,105,457 | 9,578 | den-crane | probably WSL has been fixed in the master two days ago. Try testing release https://github.com/ClickHouse/ClickHouse/releases which will be publish tomorrow or day after tomorrow | 2020-03-10T14:08:14 | 2020-03-10T14:08:41 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 597,230,730 | 7,019 | citrin | I see the same problem with 19.16.14.65.
Query hangs until connection is closed by a HTTP client (even when HTTP timeout is much higher than `max_execution_time` settings). | 2020-03-10T18:02:30 | 2020-03-10T18:02:30 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 597,270,455 | 9,591 | den-crane | https://clickhouse.tech/docs/en/query_language/table_functions/remote/
`insert into xxx select * from remote('addresses_expr', db, xxx, 'user', 'password')`
or
`insert into function remote('addresses_expr', db, xxx, 'user', 'password') select * from xxx`
or unix pipe
`clickhouse-client -h remotehost -u user ..... | 2020-03-10T19:27:20 | 2020-03-10T19:28:18 | {
"+1": 2
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 597,921,749 | 9,132 | CurtizJ | > Sometimes I also get another exception, but I couldn't reproduce it on these parts:
> DB::Exception: Invalid number of rows in Chunk column Int32: expected 13, got 65536
#9580 | 2020-03-11T23:00:35 | 2020-03-11T23:00:53 | {} | MEMBER |
ClickHouse | ClickHouse | 596,120,340 | 9,418 | alexey-milovidov | https://cwiki.apache.org/confluence/display/ZOOKEEPER/Client-Server+mutual+authentication | 2020-03-07T17:49:49 | 2020-03-07T17:49:49 | {} | MEMBER |
ClickHouse | ClickHouse | 596,140,774 | 9,556 | mbtolou | ```
select
T1.SERVER_ID ,
T1.DATA_ID ,
T1.TV ,
(T2.FA-T1.FA) USE_A,
(T2.FR-T1.FR) USE_R
from MDM.ENERGY_C_T T2 join MDM.ENERGY_C_T T1
on ( T2.TV = addHours(T1.TV, -1) and T2.SERVER_ID = T1.SERVER_ID
and T2.DATA_ID = T1.DATA_ID)
where T1.SERVER_ID =1
```
this is my final query for calculate... | 2020-03-07T21:46:56 | 2020-03-07T21:48:03 | {} | NONE |
ClickHouse | ClickHouse | 596,282,998 | 9,484 | alexey-milovidov | @filimonov Copy paste is Ok, let's finish this PR. | 2020-03-09T01:38:07 | 2020-03-09T01:38:07 | {} | MEMBER |
ClickHouse | ClickHouse | 596,574,744 | 9,566 | alexey-milovidov | clang-tidy is not installed in Docker image in CI.
Will install in next PR. | 2020-03-09T14:43:11 | 2020-03-09T14:43:11 | {} | MEMBER |
ClickHouse | ClickHouse | 596,922,030 | 9,505 | leoones |
> Did you consider `order by ... With fill`?
With fill and arrayfill functions are used to implement the！！！
Thank you for your reply ！！！ | 2020-03-10T06:13:55 | 2020-03-10T06:13:55 | {} | NONE |
ClickHouse | ClickHouse | 597,053,543 | 9,571 | huiwang | please note the flag is supported by the c compiler but not cpp. | 2020-03-10T12:12:13 | 2020-03-10T12:12:13 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 597,070,824 | 9,578 | filimonov | That is not officially supported but we know it should be possible. See:
https://github.com/ClickHouse/ClickHouse/issues/6480
ClickHouse is an open-source project, if you have some experience & goodwill - please
participate, you can become a supporter of ClickHouse on WSL. :) | 2020-03-10T12:56:40 | 2020-03-10T12:56:40 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 597,787,384 | 9,600 | alexey-milovidov | This PR will effectively invalidate old test results. This is fine. | 2020-03-11T18:11:30 | 2020-03-11T18:11:30 | {} | MEMBER |
ClickHouse | ClickHouse | 596,119,821 | 9,542 | alexey-milovidov | Squash before merge. | 2020-03-07T17:44:01 | 2020-03-07T17:44:01 | {} | MEMBER |
ClickHouse | ClickHouse | 596,126,619 | 9,418 | alexey-milovidov | > Builtin ACL Schemes
> ZooKeeeper has the following built in schemes:
> world has a single id, anyone, that represents anyone.
> auth is a special scheme which ignores any provided expression and instead uses the current user, credentials, and scheme. Any expression (whether user like with SASL authenti... | 2020-03-07T18:52:34 | 2020-03-07T18:53:05 | {} | MEMBER |
ClickHouse | ClickHouse | 596,146,070 | 9,556 | den-crane | >1697 rows in set. Elapsed: 4.089 sec. Processed 338.67 million rows
It works as expected because of CH JOIN implementation.
>table created after 2400 second but in CH this query not response after long time .
>other performance test on CH 50x to 400x faster than oracle .
>Why?
It works as expected because o... | 2020-03-07T22:59:49 | 2020-03-07T23:01:41 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 596,668,340 | 9,569 | dependabot-preview[bot] | OK, I won't notify you again about this release, but will get in touch when a new version is available. If you'd rather skip all updates until the next major or minor version, let me know by commenting `@dependabot ignore this major version` or `@dependabot ignore this minor version`.
If you change your mind, just re-... | 2020-03-09T17:26:28 | 2020-03-09T17:26:28 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 596,967,741 | 9,520 | alesapin | This error looks like a bug, but we need clickhouse version, your table definition, and data sample to investigate the error. | 2020-03-10T08:42:39 | 2020-03-10T08:42:39 | {} | MEMBER |
ClickHouse | ClickHouse | 597,171,185 | 9,585 | filimonov | Related #8429 | 2020-03-10T16:08:05 | 2020-03-10T16:08:05 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 597,204,623 | 4,132 | alexey-milovidov | @dmkoltsov It should be fixed in the latest version, as we moved from `readline` to `replxx`. | 2020-03-10T17:11:15 | 2020-03-10T17:11:15 | {} | MEMBER |
ClickHouse | ClickHouse | 597,271,848 | 9,593 | alexey-milovidov | Test missing? | 2020-03-10T19:30:30 | 2020-03-10T19:30:30 | {} | MEMBER |
ClickHouse | ClickHouse | 597,791,840 | 9,566 | alexey-milovidov | Also we can run build with `ninja -k 1000` option to show all the errors in log. | 2020-03-11T18:20:58 | 2020-03-11T18:20:58 | {} | MEMBER |
ClickHouse | ClickHouse | 596,285,678 | 9,410 | alexey-milovidov | @excitoon
```
Cannot quickly remove directory /var/lib/clickhouse/data/test/hits_indexed_by_time/delete_tmp_20140317_20140323_3_3_0 by removing files; fallback to recursive removal. Reason: Code: 458, e.displayText() = DB::ErrnoException: Cannot unlink file /var/lib/clickhouse/data/test/hits_indexed_by_time/delete_t... | 2020-03-09T01:40:01 | 2020-03-09T01:40:01 | {} | MEMBER |
ClickHouse | ClickHouse | 596,969,327 | 9,484 | filimonov | Yep. It look like image `yandex/clickhouse-stateless-test-with-coverage` was not rebuild | 2020-03-10T08:46:58 | 2020-03-10T08:46:58 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 597,163,556 | 4,132 | dmkoltsov | @alexey-milovidov , could this workaround above be transformed on your side to live patch for some modern branch of clickhouse (and upstreamed), e.g. 19.17, 19.18, 20.x.x ? | 2020-03-10T15:54:02 | 2020-03-10T16:12:21 | {} | NONE |
ClickHouse | ClickHouse | 597,193,451 | 9,132 | CurtizJ | > I can provide the parts if needed
Yes, it will be helpful. | 2020-03-10T16:50:27 | 2020-03-10T16:50:37 | {} | MEMBER |
ClickHouse | ClickHouse | 597,634,144 | 9,591 | Jack012a | Hello den-crane,
It works for me.
Thank you very much!
| 2020-03-11T13:28:32 | 2020-03-11T13:28:32 | {} | NONE |
ClickHouse | ClickHouse | 596,592,089 | 9,561 | filimonov | Software packaged for docker is executed as a single process (not as a systemd/initV service).
Once the process finishes it also stop docker container. So while you are able to run any command in the container - it's working :)
The proper way to check status of docker container is to use commands like
```
docke... | 2020-03-09T15:14:24 | 2020-03-09T15:15:14 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 596,791,979 | 9,571 | alexey-milovidov | @huiwang by the way, as MacOS is a second class platform, we are Ok with not having -Werror enabled on that platform. | 2020-03-09T21:37:16 | 2020-03-09T21:37:16 | {} | MEMBER |
ClickHouse | ClickHouse | 597,277,204 | 9,577 | alexey-milovidov | Integration test:
```
2020.03.10 21:20:17.458557 [ 1 ] {} <Warning> AWSClient: AWSErrorMarshaller: Encountered AWSError 'NoSuchBucket': The specified bucket does not exist
2020.03.10 21:20:17.458656 [ 1 ] {} <Error> AWSClient: AWSClient: HTTP response code: 404
Exception name: NoSuchBucket
Error message: The spe... | 2020-03-10T19:41:52 | 2020-03-10T19:41:52 | {} | MEMBER |
ClickHouse | ClickHouse | 597,356,904 | 9,566 | alexey-milovidov | ```
2020-03-10 20:23:01 /usr/bin/ccache /usr/bin/cmake -E __run_co_compile --tidy=/usr/bin/clang-tidy-8 --source=../dbms/programs/odbc-bridge/ColumnInfoHandler.cpp -- /usr/bin/clang++-8 -DAWS_SDK_VERSION_MAJOR=1 -DAWS_SDK_VERSION_MINOR=7 -DAWS_SDK_VERSION_PATCH=231 -DBOOST_SYSTEM_NO_DEPRECATED -DENABLE_CURL_CLIENT -D... | 2020-03-10T22:45:20 | 2020-03-10T22:45:20 | {} | MEMBER |
ClickHouse | ClickHouse | 597,383,362 | 9,588 | alexey-milovidov | > ClickHouse build check Pending — 13/16 builds finished
Today we have an issue with our internal infrastructure (Sandbox) that is used for CI and it is working slower. Everything is Ok. | 2020-03-11T00:23:07 | 2020-03-11T00:23:07 | {} | MEMBER |
ClickHouse | ClickHouse | 597,874,107 | 9,565 | akuzm | https://clickhouse-test-reports.s3.yandex.net/0/75754f309eaf1e2b4296390a54acc2538a3ea4d7/performance_comparison/report.html#4.82
joins oom now | 2020-03-11T20:55:26 | 2020-03-11T20:55:26 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 597,884,783 | 9,607 | qoega | Current cache implementation is unbounded? | 2020-03-11T21:22:21 | 2020-03-11T21:22:21 | {} | MEMBER |
ClickHouse | ClickHouse | 596,125,762 | 9,418 | alexey-milovidov | https://github.com/apache/zookeeper/blob/master/zookeeper-server/src/main/java/org/apache/zookeeper/server/auth/DigestAuthenticationProvider.java | 2020-03-07T18:43:39 | 2020-03-07T18:43:39 | {} | MEMBER |
ClickHouse | ClickHouse | 597,162,140 | 4,132 | dmkoltsov | @alexey-milovidov , hello.
Those efforts/changes, which helped me to successfully build Clickhouse on ppc64le platform, are hard to name patches - they mostly look like workaround to achieve the goal.
Nevertheless, here is the successfull sequence of steps to build ClickHouse on Ubuntu 18.04.4 LTS ppc64le:
```... | 2020-03-10T15:51:27 | 2020-03-10T17:11:14 | {
"+1": 1
} | NONE |
ClickHouse | ClickHouse | 597,555,999 | 9,587 | filimonov | It relates to `use_index_for_in_with_subqueries` (looks like disabling that fixes / improves the situation) | 2020-03-11T10:29:49 | 2020-03-11T10:29:49 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 597,921,076 | 9,607 | CurtizJ | Also cache resets on every part. | 2020-03-11T22:58:22 | 2020-03-11T22:58:22 | {} | MEMBER |
ClickHouse | ClickHouse | 597,136,393 | 9,132 | SaltTan | @CurtizJ I can provide the parts if needed | 2020-03-10T15:04:58 | 2020-03-10T15:04:58 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 597,409,256 | 9,593 | amosbird | > (that use system.numbers or system.numbers_mt but don't use the number column)
Can also be replaced by using `rowNumberInAllBlocks`. | 2020-03-11T02:17:51 | 2020-03-11T02:18:39 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 597,642,217 | 9,489 | alesapin | > > CREATE TABLE test (a DEFAULT b, b DEFAULT a) ENGINE = Memory
>
> I can execute this query and it will prevent the server to start!
https://github.com/ClickHouse/ClickHouse/pull/9603 Fix is here. | 2020-03-11T13:41:58 | 2020-03-11T13:41:58 | {} | MEMBER |
ClickHouse | ClickHouse | 596,126,841 | 9,418 | alexey-milovidov | Existing code was perfectly valid. No fix needed. | 2020-03-07T18:55:08 | 2020-03-07T18:55:08 | {} | MEMBER |
ClickHouse | ClickHouse | 596,977,854 | 9,566 | alesapin | Wow! Looks great. I have added clang-tidy manually https://github.com/ClickHouse/ClickHouse/pull/9575 | 2020-03-10T09:08:45 | 2020-03-10T09:08:45 | {} | MEMBER |
ClickHouse | ClickHouse | 597,049,327 | 9,571 | huiwang | in the meantime, i continue the investigation. according the source code here, the program used to check flags seems to be platform independence.
https://github.com/Kitware/CMake/blob/master/Modules/CheckCXXCompilerFlag.cmake#L49
` check_cxx_source_compiles("int main() { return 0; }" ${_var}`
however, as explai... | 2020-03-10T12:01:23 | 2020-03-10T12:01:23 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 597,080,977 | 9,583 | alesapin | It's almost impossible to create a test :( | 2020-03-10T13:19:30 | 2020-03-10T13:19:30 | {} | MEMBER |
ClickHouse | ClickHouse | 597,134,438 | 9,132 | SaltTan | The segfault is fixed in 20.1.6.30, but the exception is still there:
```
2020.03.10 14:46:27.997915 [ 59 ] {} <Error> HTTPHandler: Code: 9, e.displayText() = DB::Exception: Size of filter doesn't match size of column., Stack trace (when copying this message, always include the lines below):
0. 0x100ac1bc Poco::Ex... | 2020-03-10T15:01:24 | 2020-03-10T15:01:24 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 597,781,450 | 9,566 | alexey-milovidov | `clang-tidy` will be enabled not by default, because it slows down build too much.
Instead, it will be activated inside one of CI builds (most likely for clang debug variant). | 2020-03-11T17:58:48 | 2020-03-11T17:58:48 | {} | MEMBER |
ClickHouse | ClickHouse | 596,113,734 | 9,552 | alexey-milovidov | > Functional stateless tests (thread)
No diagnostics. Let's proceed. | 2020-03-07T17:39:44 | 2020-03-07T17:39:44 | {} | MEMBER |
ClickHouse | ClickHouse | 596,126,410 | 9,418 | alexey-milovidov | https://stackoverflow.com/questions/52292389/how-to-access-a-protected-znode-from-zookeeper-using-zkcli | 2020-03-07T18:50:13 | 2020-03-07T18:50:13 | {} | MEMBER |
ClickHouse | ClickHouse | 596,395,650 | 6,462 | DaiDaDao | > I also encountered the same problem. I modified the ttl of the original table to shorten it, and then used OPTIMIZE FINAL, but the expired data was not deleted
So do I , how about make friend In weixin or QQ, then we can talk about this problem | 2020-03-09T08:33:41 | 2020-03-09T08:33:41 | {} | NONE |
ClickHouse | ClickHouse | 596,708,179 | 9,484 | filimonov | Check now (not sure if the docker image will be rebuilt automatically). | 2020-03-09T18:32:02 | 2020-03-09T18:32:02 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 597,048,052 | 9,571 | huiwang | @alexey-milovidov i understand MacOS is a second class platform and ignore some warnings is not big deals. But i still see the value of enabling this flag.
If it's really too hard to fix the issue, at least we can make it consistent between CI and local for Mac developers.
For example, in CMakeList.txt here, we... | 2020-03-10T11:57:55 | 2020-03-10T12:14:21 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 597,443,755 | 9,598 | amosbird | The example is a bit misleading, as I interpret it as supplying two ranges : `[1, 2]` and `[2, 3]` at first sight. | 2020-03-11T04:54:14 | 2020-03-11T04:54:14 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 596,248,104 | 9,534 | alexey-milovidov | Normalize performance tests: in next PR. | 2020-03-08T20:22:29 | 2020-03-08T20:22:29 | {} | MEMBER |
ClickHouse | ClickHouse | 596,979,312 | 9,573 | alesapin | You use very old version `1.1.54388`. I recommend you to update. | 2020-03-10T09:12:31 | 2020-03-10T09:12:31 | {} | MEMBER |
ClickHouse | ClickHouse | 597,079,714 | 9,583 | alesapin | Fix for an extremely rare situation appeared in https://clickhouse-test-reports.s3.yandex.net/0/f28172406650e786826f86b3c4cd30aa6f55bb3a/functional_stateless_tests_(debug).html. Now we shouldn't duplicate queue entries when cloning replica. | 2020-03-10T13:16:51 | 2020-03-10T13:16:51 | {} | MEMBER |
ClickHouse | ClickHouse | 597,485,121 | 9,234 | alexey-milovidov | Changed, thank you! | 2020-03-11T07:33:32 | 2020-03-11T07:33:32 | {} | MEMBER |
ClickHouse | ClickHouse | 597,786,917 | 9,600 | alexey-milovidov | > Functional stateless tests (debug) — fail: 20, passed: 1, skipped: 2
Was in master, fixed. | 2020-03-11T18:10:27 | 2020-03-11T18:10:27 | {} | MEMBER |
ClickHouse | ClickHouse | 597,829,906 | 9,581 | excitoon | Somehow I forgot to save that file in my IDE o_O. | 2020-03-11T19:39:56 | 2020-03-11T19:39:56 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 597,873,907 | 9,132 | SaltTan | @CurtizJ Here are the parts
[i9132.tar.gz](https://github.com/ClickHouse/ClickHouse/files/4320828/i9132.tar.gz)
```
ATTACH TABLE test.i9132
(
`access_time` DateTime,
`access_day` Date DEFAULT toDate (access_time),
`time_key` Int32,
`seller_key` Int32,
`cmp_key` Int32,
`rcount` ... | 2020-03-11T20:54:57 | 2020-03-11T20:54:57 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 596,124,284 | 9,556 | mbtolou | after change query and set partial_merge_join = 1 :
This query not end normaly
```
SELECT
TEMP_ENRG.FA,
T2.FA
FROM
MDM.ENERGY_C_T T2,
MDM.ENERGY_C_T TEMP_ENRG
where
TEMP_ENRG.SERVER_ID = 1
AND TEMP_ENRG.DATA_ID = '711688325'
AND T2.TV = TEMP_ENRG.TV
AND T2.SERVER_ID = TEMP_ENRG.SERVER_ID
AND... | 2020-03-07T18:28:24 | 2020-03-07T18:29:23 | {} | NONE |
ClickHouse | ClickHouse | 596,126,363 | 9,418 | alexey-milovidov | http://web.archive.org/web/20191212214652/https://zookeeper.apache.org/doc/current/zookeeperProgrammers.html | 2020-03-07T18:49:44 | 2020-03-07T18:49:44 | {} | MEMBER |
ClickHouse | ClickHouse | 596,235,179 | 9,563 | alexey-milovidov | Stateful tests Ok. | 2020-03-08T18:22:48 | 2020-03-08T18:22:48 | {} | MEMBER |
ClickHouse | ClickHouse | 596,628,553 | 9,544 | shayrybak | Hi @filimonov thank you for your quick reply.
All of kafka_max_block_size, stream_poll_timeout_ms, stream_flush_interval_ms are left default.
We use the JSONEachRow format.
each message is one row.
Here is the table scheme
`CREATE TABLE default.kafka (`args` String, `browser_hash` String, `captcha_vld_st... | 2020-03-09T16:18:22 | 2020-03-09T16:18:22 | {} | NONE |
ClickHouse | ClickHouse | 597,213,513 | 9,558 | alexey-milovidov | It has awful performance, revert. | 2020-03-10T17:28:11 | 2020-03-10T17:28:11 | {} | MEMBER |
ClickHouse | ClickHouse | 597,510,673 | 9,489 | filimonov | Maybe we should make a note in changelog - just in case if somebody was using circular aliases. | 2020-03-11T08:43:35 | 2020-03-11T08:43:35 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 597,680,060 | 8,573 | dstruck | Feedback: latest stable version 20.1.6.30 runs fine with intra-cluster TLS connections (secure=1) and client -> clickhouse server TLS connections. | 2020-03-11T14:52:09 | 2020-03-11T14:52:09 | {
"+1": 1,
"hooray": 1
} | NONE |
ClickHouse | ClickHouse | 596,205,738 | 9,407 | ghost | @CurtizJ
Thank you for your code-review.
I fixed that line to set can_be_false always to true.
And I do forced commit the codes for simplification of commit logs. | 2020-03-08T13:27:05 | 2020-03-08T13:28:32 | {} | NONE |
ClickHouse | ClickHouse | 596,646,826 | 9,544 | filimonov | > We will try upgrading to the 20.1.6 version and see if that helps as well
It will not help with performance, but it will Improve stability.
For performance - main direction of adjustments is `kafka_max_block_size`. Try values in range 512k - 2M.
Rather leave `stream_poll_timeout_ms` and `stream_flush_in... | 2020-03-09T16:50:01 | 2020-03-09T16:50:01 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 596,920,811 | 9,505 | filimonov | Did you consider `order by ... With fill`? | 2020-03-10T06:08:27 | 2020-03-10T06:08:27 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 597,196,103 | 9,566 | alexey-milovidov | `clang-tidy` is activated successfully: https://proxy.sandbox.yandex-team.ru/1392983742/build.out.log | 2020-03-10T16:55:14 | 2020-03-10T16:55:14 | {} | MEMBER |
ClickHouse | ClickHouse | 597,451,484 | 9,598 | hczhcz | This one might be better.
```sql
SELECT arrayReduceInRanges(
'sum',
[1, 2, 3, 4],
[5, 3, 4, 4],
[1000000, 200000, 30000, 4000, 500, 60, 7]
) AS res
```
```text
┌─res─────────────────────────┐
│ [1234500,234000,34560,4567] │
└─────────────────────────────┘
``` | 2020-03-11T05:30:37 | 2020-03-11T05:30:37 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 597,496,029 | 9,598 | qoega | Will it be more clear if you use array of tuples?
```sql
SELECT arrayReduceInRanges(
'sum',
[(1, 5), (2, 3), (3, 4), (4, 4)],
[1000000, 200000, 30000, 4000, 500, 60, 7]
) AS res
```
If you have a case, when index and length are in separate columns you can just use arrayZip
```sql
SELECT
[1,... | 2020-03-11T08:05:09 | 2020-03-11T08:05:09 | {} | MEMBER |
ClickHouse | ClickHouse | 597,629,174 | 9,574 | qoega | There are failures in tests related to this PR changes. Can you analyze the reason of failures? | 2020-03-11T13:20:45 | 2020-03-11T13:20:45 | {} | MEMBER |
ClickHouse | ClickHouse | 597,799,660 | 9,574 | alexey-milovidov | ```
2020-03-10 13:10:34 00718_format_datetime: [ FAIL ] 0.18 sec. - result differs with reference:
2020-03-10 13:10:34 --- /usr/share/clickhouse-test/queries/0_stateless/00718_format_datetime.reference 2020-03-10 11:29:18.000000000 +0300
2020-03-10 13:10:34 +++ /tmp/c... | 2020-03-11T18:36:58 | 2020-03-11T18:36:58 | {} | MEMBER |
ClickHouse | ClickHouse | 596,217,017 | 9,563 | alesapin | Image already updated | 2020-03-08T15:22:38 | 2020-03-08T15:22:38 | {} | MEMBER |
ClickHouse | ClickHouse | 596,791,510 | 9,571 | alexey-milovidov | @abyss7 Maybe checks are done by trying to compile and run the binary on the host system that is build for the target system (it will obviously fail). | 2020-03-09T21:35:59 | 2020-03-09T21:35:59 | {} | MEMBER |
ClickHouse | ClickHouse | 597,263,701 | 9,595 | alexey-milovidov | Здравствуйте. Жаль, что лог для потока 86 приведён не полностью и в нём не виден запрос.
Скорее всего, запрос делался к табличной функции odbc.
У нас была проблема с владением объектом Context при использовании табличных функций, и ваш репорт напоминает об этой проблеме. Её недавно исправлял @tavplubix, он подскаже... | 2020-03-10T19:12:57 | 2020-03-10T19:13:48 | {} | MEMBER |
ClickHouse | ClickHouse | 597,267,102 | 9,596 | den-crane | I remember that I saw something similar but I cannot find the existing issue. | 2020-03-10T19:20:01 | 2020-03-10T19:20:23 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 597,376,160 | 9,596 | alexey-milovidov | `caseWithExpression` is implemented with function `transform`
See caseWithExpression.cpp, executeImpl.
The function `transform` is using "default implementation for nulls", that means if one of the arguments is NULL then the result is also NULL.
See transform.cpp, and
`virtual bool useDefaultImplementationForNull... | 2020-03-10T23:55:34 | 2020-03-10T23:58:16 | {} | MEMBER |
ClickHouse | ClickHouse | 597,891,150 | 9,607 | alexey-milovidov | If it's per query, you should not worry. Only logarithmic amount of marks are touched, it should never be a big number. | 2020-03-11T21:38:47 | 2020-03-11T21:38:47 | {} | MEMBER |
ClickHouse | ClickHouse | 596,145,807 | 9,556 | den-crane | READ CAREFULLY:
**1. Join does not use indexes.**
**2. It's your responsibility to write SQL in proper way.**
Try
```
SELECT TEMP_ENRG.FA, T2.FA
FROM MDM.ENERGY_C_T T2 join
(SELECT SERVER_ID , DATA_ID , TV, addHours(TV, -1) PRE_TV, FA, FR
from MDM.ENERGY_C_T T1
where T1.SERVER_ID = 1 AND T1... | 2020-03-07T22:56:03 | 2020-03-07T22:57:25 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 596,179,275 | 8,994 | amosbird | > But what is the sense of doing this if the user can just SELECT everything from system.numbers?
Ah, good point.
@qoega I intend to use clickhouse to generate test data files that can be used in other applications. I have made StorageTinyLog to not compress data and the result files can be directly mmaped into ... | 2020-03-08T08:24:51 | 2020-03-08T08:24:51 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 597,068,359 | 9,579 | filimonov | `0\r\n` - это часть chunked-encoding
https://clickhouse.tech/docs/ru/interfaces/http/#buferizatsiia-otveta
см. `wait_end_of_query` | 2020-03-10T12:50:44 | 2020-03-10T12:52:17 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 597,069,999 | 9,568 | excitoon | Probably after you did ALTER TTL there were old parts which did not contain new information about TTL time. There is new functionality for `ALTER TABLE nnn MATERIALIZE TTL` introduced in #8775 in 20.3. Can you try it out? | 2020-03-10T12:54:44 | 2020-03-10T12:54:44 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 597,264,770 | 9,596 | alexey-milovidov | Known issue: lack of proper support for Nullable in function `transform`. | 2020-03-10T19:15:00 | 2020-03-10T19:15:00 | {} | MEMBER |
ClickHouse | ClickHouse | 597,358,881 | 9,593 | alexey-milovidov | Let's add a test that it actually returns zeros and merge.
Then make a next PR to switch most of performance tests (that use system.numbers or system.numbers_mt but don't use the number column) to the new tables. | 2020-03-10T22:52:02 | 2020-03-10T22:52:12 | {} | MEMBER |
ClickHouse | ClickHouse | 597,612,362 | 9,595 | tavplubix | Похоже, это другая проблема, которую я раньше не заметил, она не исправлена. Вот [здесь](https://github.com/ClickHouse/ClickHouse/blob/e13bb06bbbbb7b060645e84faac76dd08ab01ac2/dbms/src/Interpreters/InterpreterSelectQuery.cpp#L296) табличная функция выполняется для временного контекста, который сразу уничтожается. Пробл... | 2020-03-11T12:47:42 | 2020-03-11T12:47:42 | {} | MEMBER |
ClickHouse | ClickHouse | 597,765,787 | 9,579 | filimonov | Интересная идея, но это потребовало делать хаки в http сервере, это поломает http keepalive, и кажется не все клиенты на это верно отреагируют.
Трейлеры с этой точки зрения кажутся лучше /правильнее.
Как вариант - вот это могло бы решить проблему:
#6294 | 2020-03-11T17:28:43 | 2020-03-11T17:30:49 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 596,125,808 | 9,418 | alexey-milovidov | It was not implemented in original ZooKeeper C library. | 2020-03-07T18:44:02 | 2020-03-07T18:44:02 | {} | MEMBER |
ClickHouse | ClickHouse | 596,139,927 | 9,556 | mbtolou | > Try to swap table order in FROM!!!!
> `MDM.ENERGY_C_T T2, MDM.ENERGY_C_T TEMP_ENRG`
>
> CH does not have proper cost-based query optimizer yet. It's your responsibility to write SQL in proper way.
>
> Try ANSI conventional way :
>
> ```
> SELECT TEMP_ENRG.FA, T2.FA
> FROM MDM.ENERGY_C_T T2 join
> (S... | 2020-03-07T21:36:40 | 2020-03-07T21:39:38 | {} | NONE |
ClickHouse | ClickHouse | 596,152,601 | 9,560 | alexey-milovidov | Is the cluster size and data volume up to date? | 2020-03-08T00:38:03 | 2020-03-08T00:38:03 | {} | MEMBER |
ClickHouse | ClickHouse | 596,393,144 | 9,568 | DaiDaDao | Oh, I found the issue [6462](https://github.com/ClickHouse/ClickHouse/issues/6462),maybe it can answer my question, | 2020-03-09T08:27:51 | 2020-03-09T08:27:51 | {} | NONE |
ClickHouse | ClickHouse | 596,644,034 | 9,544 | shayrybak | Also tried setting the kafka_max_block_size parameter to 512k as well at the `stream_poll_timeout_ms =100, stream_flush_interval_ms = 800` settings and we were able to pull at about 60k rows per second. which is a big improvement it appears. so this might have been the issue. I'll update if we get more data. | 2020-03-09T16:44:54 | 2020-03-09T16:44:54 | {} | NONE |
ClickHouse | ClickHouse | 597,480,248 | 9,234 | dmyers | FYI the docs still say to use the `stable` branch:
https://clickhouse.tech/docs/en/development/build_osx/ | 2020-03-11T07:18:00 | 2020-03-11T07:18:00 | {} | NONE |
ClickHouse | ClickHouse | 597,561,327 | 9,579 | MakarovYaroslav | `wait_end_of_query ` не подходит, если по HTTP-интерфейсу хотим потоково быстро получать большие объемы данных.
Про `0\r\n` я имел в виду, что можно таким образом сигнализировать об ошибочном завершении запроса, искусственно делать ответ незавершенным.
Из RFC 7230:
> A message body that uses the chunked transfer c... | 2020-03-11T10:42:07 | 2020-03-11T10:42:07 | {} | NONE |
ClickHouse | ClickHouse | 597,672,308 | 2,829 | kekekekule | maybe we'll take it
@evillique @namikhnenko | 2020-03-11T14:38:25 | 2020-03-11T14:38:25 | {
"+1": 2
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 569,565,842 | 8,461 | Inasayang | I have resolved the question.
https://stackoverflow.com/questions/59518102/clickhouse-create-jdbc-engine-table/59523918#59523918
| 2019-12-30T02:30:58 | 2019-12-30T02:30:58 | {} | NONE |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.