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 | 707,568,885 | 15,853 | yanchaoguo | I know
do like this
grant CREATE TEMPORARY TABLE, MYSQL ON *.* TO bigdata WITH GRANT OPTION ;
| 2020-10-13T08:06:25 | 2020-10-13T08:06:25 | {} | NONE |
ClickHouse | ClickHouse | 707,639,435 | 15,806 | hexiaoting | @CurtizJ
Hi, I build these codes in my environment successfully with gcc10 or clang11.
How can I reproduce the build error in "ClickHouse build check" in my environment? | 2020-10-13T10:12:16 | 2020-10-13T10:12:33 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,735,848 | 15,872 | socketpair | https://fedoraproject.org/wiki/Packaging_tricks#Executable_stack
`In typical use, gcc generates code that disables execution on the stack.`
```sh
$ objdump -h /usr/bin/clickhouse | fgrep gcc
16 .gcc_except_table 0040b594 000000000e8cae7c 000000000e8cae7c 0e4cae7c 2**2
```
Seems binary made by GCC
```
... | 2020-10-13T13:24:41 | 2020-10-13T13:35:19 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,787,218 | 15,930 | kitaisreal | > By the way, does it already work with shell redirection, e.g. `clickhouse-client < queries.sql`?
> Looks like this is the only way in MySQL, and Posgres supports both but they are "subtly different"
>
> https://www.postgresql.org/docs/current/app-psql.html (see the --file option)
> https://dev.mysql.com/doc/refm... | 2020-10-13T14:42:28 | 2020-10-13T14:42:28 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,941,485 | 15,874 | nikitamikhaylov | Code for test generation [here](https://gist.github.com/nikitamikhaylov/88ee523f01bfa026793a59ac6224e8cb) | 2020-10-13T18:52:58 | 2020-10-13T18:52:58 | {} | MEMBER |
ClickHouse | ClickHouse | 708,094,640 | 15,884 | den-crane | https://clickhouse.tech/docs/en/sql-reference/functions/#strong-typing
Strong Typing
In contrast **to standard SQL, ClickHouse has strong typing**. In other words, it doesn’t make implicit conversions between types. Each function works for a specific set of types. This means that sometimes you need to use type c... | 2020-10-14T01:14:07 | 2020-10-14T01:14:07 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 708,113,672 | 15,265 | bohutang | Better to return an error instead fatal here:
```
2020.10.14 09:40:05.923962 [ 21489 ] {fa38ae2b-c3e8-448d-a97e-c0989354f8fc} <Error> Application: mysqlxx::ConnectionFailed
2020.10.14 09:40:06.876883 [ 21489 ] {fa38ae2b-c3e8-448d-a97e-c0989354f8fc} <Error> DatabaseMaterializeMySQL: Cannot load MySQL nested database ... | 2020-10-14T02:21:35 | 2020-10-14T02:23:01 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,494,403 | 15,793 | aadant | I guess it only makes sense for Strings. | 2020-10-13T05:18:43 | 2020-10-13T05:18:43 | {} | NONE |
ClickHouse | ClickHouse | 707,708,977 | 15,898 | den-crane | duplicate: https://github.com/ClickHouse/ClickHouse/issues/10471
as a workaround, specify a database in `from db.table` | 2020-10-13T12:36:06 | 2020-10-13T13:14:58 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,755,074 | 15,855 | CurtizJ | In general it's ok, but in your implementation this one passes
`SELECT xxx FROM xxx ORDER BY xxx OFFSET $num row/rows fetch first/next $num2;`
| 2020-10-13T13:54:34 | 2020-10-13T13:54:34 | {} | MEMBER |
ClickHouse | ClickHouse | 707,821,114 | 15,872 | den-crane | it seems the issue is in rpm .spec file | 2020-10-13T15:28:55 | 2020-10-13T15:29:14 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,885,191 | 15,199 | filimonov | `test_storage_kafka/test.py::test_kafka_produce_key_timestamp (✕2)` failed again (btw this is the one which not relates to timing in any way). I will check that later #15936 | 2020-10-13T17:09:21 | 2020-10-13T17:09:21 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,419,121 | 15,835 | den-crane | >This despite the fact that with larger index granularity the bloom filter cannot be as precise in dropping granules.
Analyzing skip indexes is a super-slow process and they are heavy and need to read them from disk.
When an index granula covers many rows it speeds up the index analyzes but decreases number of drop... | 2020-10-13T00:50:46 | 2020-10-13T00:50:46 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,443,044 | 15,855 | hexiaoting | @CurtizJ There's a fast test "01272_offset_without_limit" failed with my code .
Do all below SQL queries should support ? If so, I should fix this.
```
SELECT xxx FROM xxx ORDER BY xxx OFFSET $num;
SELECT xxx FROM xxx ORDER BY xxx OFFSET $num row/rows;
SELECT xxx FROM xxx ORDER BY xxx OFFSET $num row/rows fetch ... | 2020-10-13T02:20:22 | 2020-10-13T02:21:37 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,707,928 | 15,872 | alexey-milovidov | @filimonov We should not have executable stack. It's build issue. Do we have this issue in recent versions in official build?
@socketpair `capability: warning` and this is not an issue at all. | 2020-10-13T12:34:08 | 2020-10-13T12:34:08 | {} | MEMBER |
ClickHouse | ClickHouse | 707,780,973 | 15,930 | filimonov | You need to create at least some basic tests for that.
Check that as an example:
https://github.com/ClickHouse/ClickHouse/blob/c48da8d544fe271a0c2b16c7e96d583365442f1c/tests/queries/0_stateless/00366_multi_statements.sh
https://github.com/ClickHouse/ClickHouse/blob/c48da8d544fe271a0c2b16c7e96d583365442f1c/tests/q... | 2020-10-13T14:34:05 | 2020-10-13T14:35:11 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,829,152 | 15,806 | ildus | Have you considered using separate columns for each key? In that case it would be easy to implement combinators and such stuff, since then it would be no need to collect values by keys first. | 2020-10-13T15:40:35 | 2020-10-13T15:40:35 | {
"+1": 3
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 708,030,769 | 15,944 | alexey-milovidov | Not a bugfix. | 2020-10-13T21:55:32 | 2020-10-13T21:55:32 | {} | MEMBER |
ClickHouse | ClickHouse | 708,035,003 | 15,607 | alexey-milovidov | We should allow to specify CLICKHOUSE source without host and port:
```
SOURCE(CLICKHOUSE(TABLE 'dict_source' DB 'issue15607'))
``` | 2020-10-13T22:05:32 | 2020-10-13T22:05:32 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 707,492,463 | 15,872 | filimonov | https://linux.die.net/man/8/execstack
| 2020-10-13T05:12:36 | 2020-10-13T05:12:36 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,494,916 | 13,597 | filimonov | Guys, can you make a PR with some cmake check / warning, and a line in doc about this? | 2020-10-13T05:20:20 | 2020-10-13T05:20:20 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,500,082 | 15,847 | DBA-Ray | I executed these following statements:
**set allow_experimental_database_materialize_mysql=1;
CREATE DATABASE xxx ENGINE = MaterializeMySQL('127.0.0.1:3377','xxx','xxx','xxx');**
Then clickhouse started to full synchronization by MaterializeMySQL from mysql.
But after a time,the log was:
**2020.10.12 11:10:2... | 2020-10-13T05:35:21 | 2020-10-13T05:35:21 | {} | NONE |
ClickHouse | ClickHouse | 707,705,620 | 15,610 | alexey-milovidov | @aubweb9 But it's totally different issue, let's move it to another task.
And please provide more details - about SSL certificate on MySQL server... | 2020-10-13T12:29:36 | 2020-10-13T12:29:36 | {} | MEMBER |
ClickHouse | ClickHouse | 707,761,053 | 15,928 | socketpair | So, I think, CH should use modern means if they are available. And some crutches/kludges for the old systems. | 2020-10-13T14:03:29 | 2020-10-13T14:03:29 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,797,409 | 15,903 | borisovcode | Thank you! | 2020-10-13T14:56:03 | 2020-10-13T14:56:03 | {} | NONE |
ClickHouse | ClickHouse | 708,030,820 | 15,942 | alexey-milovidov | Not a bugfix. | 2020-10-13T21:55:38 | 2020-10-13T21:55:38 | {} | MEMBER |
ClickHouse | ClickHouse | 707,434,029 | 15,882 | den-crane | Check this https://github.com/ClickHouse/ClickHouse/issues/11933#issuecomment-654286261
export SERVER_JVMFLAGS="-Djute.maxbuffer=50000000" bin/zkServer.sh restart | 2020-10-13T01:48:53 | 2020-10-13T01:48:53 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,538,801 | 15,883 | yanchaoguo | > Can you provide some information on how to reproduce this problem?
this is source table schema in mysql
CREATE TABLE `dashboard_datasource` (
`datasource_id` bigint(20) NOT NULL AUTO_INCREMENT,
`user_id` varchar(50) NOT NULL,
`source_name` varchar(100) NOT NULL,
`source_type` varchar(100) NOT NULL,... | 2020-10-13T07:07:43 | 2020-10-13T07:07:43 | {} | NONE |
ClickHouse | ClickHouse | 707,720,177 | 15,882 | alexey-milovidov | You should not get this error in normal ClickHouse usage scenario.
It may indicate an error with your tables or queries. | 2020-10-13T12:57:37 | 2020-10-13T12:57:37 | {} | MEMBER |
ClickHouse | ClickHouse | 707,828,071 | 15,932 | den-crane | 20.11
```
for i in `seq 1 5000`; do curl -s -o /dev/null "http://localhost:8123/?query=select+*+from+system.query_log+limit+100+format+JSON" ; echo -n 'ps rss:' `ps -C clickhouse-serv -o rss=`' | '; echo "select metric,':', formatReadableSize(value) from system.metrics where metric like 'MemoryTracking'"|clickhous... | 2020-10-13T15:39:02 | 2020-10-13T15:59:43 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 708,001,093 | 15,938 | UnamedRus | > Define a new setting do_not_merge_across_partitions_select_final. If it's true we will merge parts only with the same partition, moreover, if there is only one part in partition and it's level > 0 we won't merge it.
There is a possibility that partition was attached from ./detached dir or from another table and a... | 2020-10-13T20:49:47 | 2020-10-13T20:49:47 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 708,024,121 | 15,896 | alexey-milovidov | I strongly recommend using clang-11 instead of gcc-9. | 2020-10-13T21:39:03 | 2020-10-13T21:39:03 | {} | MEMBER |
ClickHouse | ClickHouse | 708,038,653 | 15,420 | alexey-milovidov | Does it reproduce on different servers? | 2020-10-13T22:14:58 | 2020-10-13T22:14:58 | {} | MEMBER |
ClickHouse | ClickHouse | 708,042,647 | 15,879 | azat | >00966_live_view_watch_events_http looks flacky
Here is the fix #15945
P.S. AFAICS the problem is in http client so this patch should not be related | 2020-10-13T22:25:32 | 2020-10-13T22:25:32 | {} | MEMBER |
ClickHouse | ClickHouse | 707,697,806 | 15,199 | alexey-milovidov | Let's @tavplubix take a look.
PS. It's Ok to merge. | 2020-10-13T12:14:14 | 2020-10-13T12:14:37 | {} | MEMBER |
ClickHouse | ClickHouse | 707,718,663 | 15,872 | filimonov | > Do we have this issue in recent versions in official build?
I don't think so:
```
# 20.9.3.45 (official build)
execstack -q /usr/bin/clickhouse
- /usr/bin/clickhouse
# 20.7.2.30 (official build)
execstack -q /usr/bin/clickhouse
- /usr/bin/clickhouse
```
@socketpair what kind of build do you use? | 2020-10-13T12:54:51 | 2020-10-13T12:54:51 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,720,492 | 15,065 | Realist007 | > User defined data types in ClickHouse
Hello, can I do this task? @alexey-milovidov | 2020-10-13T12:58:13 | 2020-10-13T12:58:13 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,757,218 | 15,929 | den-crane | I am working on it. I have it half-ready. Try to finish ASAP. | 2020-10-13T13:57:46 | 2020-10-13T13:57:46 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,764,007 | 15,830 | akuzm | I lost it :fearful:
You might be able to build it using the Android toolchain as described here: https://github.com/ClickHouse/ClickHouse/blob/95c6efbdc1d607306c5bee79b18a27885efaa218/website/blog/en/2020/pixel-benchmark.md#using-the-android-toolchain
It might be helpful to add `-DENABLE_LIBRARIES=0` cmake flag, ... | 2020-10-13T14:08:01 | 2020-10-13T14:08:01 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,978,134 | 15,935 | alexey-milovidov | Maybe something is wrong with the "Magic Links" markdown plugin in Python. | 2020-10-13T20:06:15 | 2020-10-13T20:06:15 | {} | MEMBER |
ClickHouse | ClickHouse | 707,494,349 | 15,841 | filimonov | Duplicate of https://github.com/ClickHouse/ClickHouse/issues/13597#issuecomment-680464522
| 2020-10-13T05:18:32 | 2020-10-13T05:18:32 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,530,241 | 15,852 | alesapin | Reported https://github.com/ClickHouse/ClickHouse/issues/15889 | 2020-10-13T06:49:21 | 2020-10-13T06:49:21 | {} | MEMBER |
ClickHouse | ClickHouse | 707,531,638 | 15,236 | alesapin | > But now fast test failed.
No idea why. No changes in fast test. | 2020-10-13T06:52:21 | 2020-10-13T06:52:21 | {} | MEMBER |
ClickHouse | ClickHouse | 707,696,045 | 15,896 | filimonov | Most probably you want to disable some (or most of) external libraries used by clickhouse, to get 'vanilla' / mimimalistic clickhouse build.
Try for example
```
cmake .. \
-DENABLE_LIBRARIES=0 -DUSE_UNWIND=1 -DENABLE_REPLXX=1 -DENABLE_JEMALLOC=1 -DENABLE_TESTS=0 -DENABLE_UTILS=0
``` | 2020-10-13T12:10:49 | 2020-10-13T12:11:40 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,804,689 | 15,792 | den-crane | this also provides the wrong result
```sql
select count() from xp_d prewhere B > -1 where toYYYYMM(A) global in (select toYYYYMM(min(A)) from xp_d);
┌─count()─┐
│ 0 │
└─────────┘
``` | 2020-10-13T15:06:10 | 2020-10-13T15:06:26 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,704,138 | 15,806 | CurtizJ | You can try to merge with fresh master.
`/build/src/Functions/FunctionsConversion.h:2201:19: error: no member named 'getByPosition' in 'std::__1::vector<DB::ColumnWithTypeAndName, std::__1::allocator<DB::ColumnWithTypeAndName>>'`
Seems that error appears because recently `Block` from `Block.h` was replaced to `Col... | 2020-10-13T12:26:50 | 2020-10-13T12:26:50 | {} | MEMBER |
ClickHouse | ClickHouse | 707,711,350 | 15,872 | alexey-milovidov | ```
-Wa,--noexecstack
-Wl,'-z noexecstack'
```
should help. | 2020-10-13T12:40:47 | 2020-10-13T12:40:55 | {} | MEMBER |
ClickHouse | ClickHouse | 707,720,251 | 15,903 | den-crane | Your clickhouse-client is outdated and does not support DateTime64.
Install clickhouse-client with version 20.9.2
In case of
`clickhouse-client --query "INSERT INTO spnav.log_viewing FORMAT TabSeparatedWithNames" <log_viewing-0.tsv`
clickhouse-client reads a stream and does a stream parsing and sends parsed dat... | 2020-10-13T12:57:45 | 2020-10-13T13:08:25 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,730,254 | 15,855 | CurtizJ | This query
`SELECT xxx FROM xxx ORDER BY xxx OFFSET $num`
should work as incomplete part of `SELECT xxx FROM xxx ORDER BY xxx LIMIT $num1 OFFSET $num;` query.
These queries shouldn't work according to standard, because it says that `row/rows` and `only/with ties` are required keywords.
`SELECT xxx FROM xxx ORDER... | 2020-10-13T13:15:38 | 2020-10-13T13:53:44 | {} | MEMBER |
ClickHouse | ClickHouse | 708,052,169 | 15,874 | nikitamikhaylov | > BTW, you can write test in python directly.
Maybe it will be better, but we will have to lock the precision. For example `our_answer - scipy_answer < 1e-10`. Is it Ok? | 2020-10-13T22:52:57 | 2020-10-13T22:52:57 | {} | MEMBER |
ClickHouse | ClickHouse | 708,085,652 | 15,884 | chang0391 | @den-crane
Since they are both int type, why need to convert them? I know that other databases not need that like MySQL. | 2020-10-14T00:41:46 | 2020-10-14T00:41:46 | {} | NONE |
ClickHouse | ClickHouse | 706,751,707 | 15,065 | gonzalezjo | > ClickHouse has support for LZ4 and ZSTD as generic compression methods. The choice of these particular methods is justified: these methods are pareto-optimal for compression level and speed across well known libraries. Nevertheless, there exist less well known compression libraries that can be somewhat better in cert... | 2020-10-11T18:57:34 | 2020-10-13T00:53:55 | {
"+1": 3,
"heart": 2
} | NONE |
ClickHouse | ClickHouse | 707,493,289 | 15,847 | zhang2014 | Can you provide some information on how to reproduce this problem? | 2020-10-13T05:15:14 | 2020-10-13T05:15:14 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,560,819 | 12,455 | azat | @alexey-milovidov what is the status of this PR? Are there some concerns about it? | 2020-10-13T07:52:04 | 2020-10-13T07:52:04 | {} | MEMBER |
ClickHouse | ClickHouse | 707,677,217 | 15,890 | alesapin | Tests OK | 2020-10-13T11:29:52 | 2020-10-13T11:29:52 | {} | MEMBER |
ClickHouse | ClickHouse | 707,733,168 | 15,928 | socketpair | And also, I don't see any reasons not to include libcap as a git submodule. | 2020-10-13T13:20:33 | 2020-10-13T13:20:33 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 708,030,789 | 15,943 | alexey-milovidov | Not a bugfix. | 2020-10-13T21:55:35 | 2020-10-13T21:55:35 | {} | MEMBER |
ClickHouse | ClickHouse | 707,643,652 | 15,236 | alesapin | Ok Again | 2020-10-13T10:20:28 | 2020-10-13T10:20:28 | {} | MEMBER |
ClickHouse | ClickHouse | 707,625,130 | 15,610 | aubweb9 | Hi @KochetovNicolai,
I upgraded to 20.9.3.45
It's much better but now clickouse is not crashing anymore but I am getting those errors 15% of the time during a load test (600 queries) :
2020.10.13 11:37:56.521677 [ 13555 ] {071b9ae9-d9e8-43b6-bcf6-6c061b1b535a} <Error> Application: mysqlxx::ConnectionFailed
... | 2020-10-13T09:44:37 | 2020-10-13T12:14:08 | {
"+1": 1
} | NONE |
ClickHouse | ClickHouse | 707,725,754 | 15,872 | socketpair | We use standard Fedora 32 build system (RPM). spec file: https://pastebin.com/tNHsYssE
| 2020-10-13T13:07:46 | 2020-10-13T13:07:46 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,752,329 | 15,872 | alexey-milovidov | As official build doesn't have this issue, it indicates that you have errors in your own spec file (custom build),
and we cannot do anything with it. | 2020-10-13T13:50:06 | 2020-10-13T13:50:06 | {} | MEMBER |
ClickHouse | ClickHouse | 707,787,813 | 15,930 | kitaisreal | > You need to create at least some basic tests for that.
>
> Check that as an example:
> https://github.com/ClickHouse/ClickHouse/blob/c48da8d544fe271a0c2b16c7e96d583365442f1c/tests/queries/0_stateless/00366_multi_statements.sh
> https://github.com/ClickHouse/ClickHouse/blob/c48da8d544fe271a0c2b16c7e96d583365442f1... | 2020-10-13T14:43:13 | 2020-10-13T15:16:42 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,861,767 | 15,897 | nikitamikhaylov | You can install `lld` and create symlink, for example `sudo ln -s /usr/bin/lld-10 /usr/bin/ld.lld`.
Or simply try not to build tests if you don't need them (try ninja `clickhouse-bundle`). | 2020-10-13T16:30:25 | 2020-10-13T16:30:25 | {} | MEMBER |
ClickHouse | ClickHouse | 707,982,363 | 11,217 | alexey-milovidov | It depends on query optimization - whether the conditions in WHERE will be pushed into joined tables.
Looks like it works in simple case and doesn't work with NOT.
How to check it:
1. Install ClickHouse version 20.9.
2. Use `EXPLAIN SYNTAX SELECT ...`
3. Check the difference between these two queries. | 2020-10-13T20:14:40 | 2020-10-13T20:14:40 | {} | MEMBER |
ClickHouse | ClickHouse | 708,023,531 | 15,897 | alexey-milovidov | gcc 9 is not supported.
The only supported build configurations listed here: https://clickhouse-builds.s3.yandex.net/15937/db4db42b65bed60512379e9ebc4dc009b7fc3666/clickhouse_build_check/report.html
The minimal version of gcc is 10. | 2020-10-13T21:37:44 | 2020-10-13T21:37:44 | {} | MEMBER |
ClickHouse | ClickHouse | 708,030,852 | 15,941 | alexey-milovidov | Not a bugfix. | 2020-10-13T21:55:41 | 2020-10-13T21:55:41 | {} | MEMBER |
ClickHouse | ClickHouse | 707,444,681 | 15,846 | den-crane | https://clickhouse.tech/docs/en/sql-reference/statements/alter/partition/#alter_freeze-partition
You can use `alter table T.. freeze`.
FREEZE creates a snapshot in /var/lib/clickhouse/shadow
You can copy this snapshot by rsync or any other tool to another server or into backup. | 2020-10-13T02:26:28 | 2020-10-13T02:26:28 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,042,179 | 14,094 | filimonov | ~~Regression in 20.4 (?)~~
@livace can you check that? | 2020-10-12T10:44:36 | 2020-10-13T06:41:39 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,611,728 | 10,417 | rupertevill | Any update on this? | 2020-10-13T09:19:56 | 2020-10-13T09:19:56 | {
"+1": 2
} | NONE |
ClickHouse | ClickHouse | 707,613,967 | 15,872 | socketpair | Also, I see ```capability: warning: `clickhouse-serv' uses 32-bit capabilities (legacy support in use)```. Is it a separate issue ? | 2020-10-13T09:24:02 | 2020-10-13T09:24:30 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,713,650 | 15,891 | den-crane | JOINs and primary keys are not related. CH does not use indexes or keys for joins. There is no restrictions which columns can be used.
```sql
1)
SELECT DISTINCT name
FROM test_all AS a
GLOBAL INNER JOIN credit_ga.test_all_2 AS b ON a.name = b.name
ORDER BY name ASC
2)
SELECT DISTINCT name
FROM test_all AS ... | 2020-10-13T12:45:22 | 2020-10-13T12:46:46 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,780,232 | 15,930 | akuzm | By the way, does it already work with shell redirection, e.g. `clickhouse-client < queries.sql`?
Looks like this is the only way in MySQL, and Posgres supports both but they are "subtly different"
https://www.postgresql.org/docs/current/app-psql.html (see the --file option)
https://dev.mysql.com/doc/refman/8.0/en/... | 2020-10-13T14:33:04 | 2020-10-13T14:33:04 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,821,321 | 15,462 | BayoNet | Комментарии обработаем в другой задаче. | 2020-10-13T15:29:14 | 2020-10-13T15:29:14 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,844,124 | 15,880 | BayoNet | Also, you didn't mention that ClickHouse supports the ILIKE operator, which works the same as the function. | 2020-10-13T16:03:06 | 2020-10-13T16:03:06 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 708,055,733 | 15,874 | alexey-milovidov | Yes, it's ok to compare up to 10 digits... | 2020-10-13T23:03:35 | 2020-10-13T23:03:35 | {} | MEMBER |
ClickHouse | ClickHouse | 708,084,158 | 15,883 | yanchaoguo | 即便是表里面没有数据 clickhouse也报下面的错误 @zhang2014
Code: 33. DB::Exception: Received from localhost:9000. DB::Exception: Cannot read all data. Bytes read: 1. Bytes expected: 4. | 2020-10-14T00:36:14 | 2020-10-14T00:36:14 | {} | NONE |
ClickHouse | ClickHouse | 707,435,085 | 15,884 | den-crane | Expected behavior , you should use type casting -- toInt64
**select * from aa a join bb b on toInt64(a.id)=b.id**
```
┌─id─┬─name─┬─b.id─┬─b.name─┐
│ 1 │ 1 │ 1 │ 1 │
└────┴──────┴──────┴────────┘
``` | 2020-10-13T01:52:38 | 2020-10-13T01:52:59 | {
"+1": 1,
"heart": 2
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,493,334 | 15,885 | zhang2014 | Can you provide some information on how to reproduce this problem? | 2020-10-13T05:15:24 | 2020-10-13T05:15:24 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,694,207 | 9,071 | chelsEg | I also encountered this problem, the version is 20.6.7.4 | 2020-10-13T12:06:59 | 2020-10-13T12:06:59 | {} | NONE |
ClickHouse | ClickHouse | 707,696,731 | 15,514 | alesapin | Actually implementation is much simpler. | 2020-10-13T12:12:14 | 2020-10-13T12:12:14 | {} | MEMBER |
ClickHouse | ClickHouse | 707,707,288 | 15,888 | den-crane | ```sql
1)
select *
from test_all a
join (select distinct name from credit_ga.test_all_2 where id<5) b on a.name=b.name
order by a.name
2)
select *
from (select name from test_all) a
join (select distinct name from credit_ga.test_all_2 where id<5) b on a.name=b.name
order by a.name
```
>I th... | 2020-10-13T12:32:54 | 2020-10-13T12:34:04 | {
"+1": 4
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,779,140 | 15,928 | alexey-milovidov | I *almost* agree but not quite. It will increase the number of branches in code and it will be difficult to test. | 2020-10-13T14:31:20 | 2020-10-13T14:31:20 | {} | MEMBER |
ClickHouse | ClickHouse | 707,797,131 | 15,872 | socketpair | No, I have just repeated steps in a clear Fedora 32 machine (in YandexCloud :) ) exactly as written here: https://clickhouse.tech/docs/ru/development/build/
```
$ execstack -q clickhouse
X clickhouse
... | 2020-10-13T14:55:40 | 2020-10-13T15:03:40 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,840,185 | 15,932 | den-crane | clickhouse-client instead of curl
```
for i in `seq 1 5000`; do clickhouse-client --query 'select * from system.query_log limit 100 format TSV' >/dev/null ; echo -n 'ps rss:' `ps -C clickhouse-serv -o rss=`' | '; echo "select metric,':', formatReadableSize(value) from system.metrics where metric like 'MemoryTrack... | 2020-10-13T15:57:24 | 2020-10-13T15:57:24 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,985,522 | 15,935 | alexey-milovidov | https://github.com/facelessuser/pymdown-extensions/blob/master/pymdownx/magiclink.py | 2020-10-13T20:21:02 | 2020-10-13T20:21:02 | {} | MEMBER |
ClickHouse | ClickHouse | 707,493,460 | 15,883 | zhang2014 | Can you provide some information on how to reproduce this problem? | 2020-10-13T05:15:47 | 2020-10-13T05:15:47 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,520,262 | 15,853 | yanchaoguo | > When I execute the following sentence, I get an error
>
> GRANT CREATE TEMPORARY TABLE on app.* to bigdata
>
> Exception on client:
> Code: 509. DB::Exception: CREATE TEMPORARY TABLE cannot be granted on the database level
@den-crane
| 2020-10-13T06:28:32 | 2020-10-13T06:28:32 | {} | NONE |
ClickHouse | ClickHouse | 707,521,091 | 15,848 | qinglok | CH version is 20.8.2.3-2.
The disk is normal. Because I can see that there are still files in the data directory of CH.
However, after the CH service is restarted, no data can be found by using SELECT.
But now I can’t reproduce the same situation.
| 2020-10-13T06:30:15 | 2020-10-13T06:30:15 | {} | NONE |
ClickHouse | ClickHouse | 707,703,930 | 15,845 | alexey-milovidov | Now everything looks alright :) | 2020-10-13T12:26:27 | 2020-10-13T12:26:27 | {} | MEMBER |
ClickHouse | ClickHouse | 708,110,812 | 15,951 | liangli1213 | @BohuTANG @zhang2014 | 2020-10-14T02:12:24 | 2020-10-14T02:12:24 | {} | NONE |
ClickHouse | ClickHouse | 707,663,918 | 15,845 | kitaisreal | @alexey-milovidov added tests. | 2020-10-13T11:01:15 | 2020-10-13T11:01:15 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,753,307 | 15,928 | alexey-milovidov | > I don't see any reasons not to include libcap as a git submodule.
It has GPL license.
> capget, capset - set/get capabilities of thread(s)
It depends on new `glibc` version.
> Even Centos 6 uses 2.4.32
We support Centos 5. | 2020-10-13T13:51:39 | 2020-10-13T13:51:39 | {} | MEMBER |
ClickHouse | ClickHouse | 707,977,167 | 15,937 | alexey-milovidov | We can apply this command repeatedly:
`find . -name '*.md' -type f | xargs sed -i -r 's/\\([_#])/\1/g'`
until the root cause of the bug will be fixed. | 2020-10-13T20:04:48 | 2020-10-13T20:05:12 | {} | MEMBER |
ClickHouse | ClickHouse | 707,678,322 | 15,199 | filimonov | Stress test failure is unrelated.
It finished, but restart at the end failed.
Seems releated somehow to Database=Atomic (it was slowly starting but 120 sec was not enough, lot of messages like `Trying load partially dropped table test_fdtq0h.t2 (a8206a26-17bc-4838-a680-4233a74d1d6a) from /var/lib/clickhouse/meta... | 2020-10-13T11:32:19 | 2020-10-13T11:32:47 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,719,447 | 15,887 | 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-... | 2020-10-13T12:56:20 | 2020-10-13T12:56:20 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 707,756,067 | 15,855 | CurtizJ | AST Fuzzer error: I suppose it's the same as #14974. | 2020-10-13T13:56:07 | 2020-10-13T13:56:07 | {} | MEMBER |
ClickHouse | ClickHouse | 708,011,342 | 15,938 | alexey-milovidov | Maybe better to have simple rule: if one part in partition - do not postprocess it. | 2020-10-13T21:11:08 | 2020-10-13T21:11:16 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 733,795,392 | 17,403 | audricbizouarn | It happens randomly. We don't understand how to reproduce it. | 2020-11-25T15:59:07 | 2020-11-25T15:59:07 | {} | NONE |
ClickHouse | ClickHouse | 733,833,893 | 17,387 | DmitriZhukov | i have
use_server_time_zone | true
use_time_zone |
use_server_time_zone_for_dates | false
Почему у меня все сдвигается на час
in my SQL client I do
insert into temp (datetime) values('2020-11-24 15:22:00')
but see what it stores
select * from temp order by datetime
results
2020-11-24 14:22:00
| 2020-11-25T17:06:09 | 2020-11-25T17:06:09 | {} | NONE |
ClickHouse | ClickHouse | 733,909,889 | 15,385 | nikitamikhaylov | I think now it will work, because we wrap column names with backticks. | 2020-11-25T19:34:54 | 2020-11-25T19:34:54 | {} | MEMBER |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.