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 | 713,037,992 | 14,135 | alexey-milovidov | ```
Alexey Milovidov, [20.10.20 19:28]
[In reply to Alexander Sapin]
Capabilities.
Alexey Milovidov, [20.10.20 19:28]
Если Docker не имеет capabilities, а бинарник имеет fs атрибут с capabilities, то ядро ОС отказывается его запускать.
Alexey Milovidov, [20.10.20 19:28]
cp сбрасывает эти атрибуты.
``` | 2020-10-20T17:59:50 | 2020-10-20T17:59:50 | {} | MEMBER |
ClickHouse | ClickHouse | 713,038,357 | 14,135 | alexey-milovidov | We need to incorporate a check that capabilities are working (we have this check in debian postinst script). | 2020-10-20T18:00:31 | 2020-10-20T18:01:39 | {} | MEMBER |
ClickHouse | ClickHouse | 713,102,693 | 16,191 | alesapin | > `double mapPiecewiseLinearToUnit`
>
> * make static?
It's already in the anonymous namespace. | 2020-10-20T19:56:08 | 2020-10-20T19:56:08 | {} | MEMBER |
ClickHouse | ClickHouse | 713,154,209 | 16,199 | alexey-milovidov | 1. The issue does not happen in the official build in CI.
2. We are using clang-11, not 10.
3. It fails compatibility check.
4. Maybe add a substitute from Musl? | 2020-10-20T21:36:40 | 2020-10-20T21:37:42 | {} | MEMBER |
ClickHouse | ClickHouse | 713,219,657 | 13,707 | alexey-milovidov | CC @excitoon @filimonov | 2020-10-21T00:38:37 | 2020-10-21T00:38:37 | {} | MEMBER |
ClickHouse | ClickHouse | 713,384,310 | 16,130 | leoones | > mysql -uptx -p -h**172.16.6.207** -P3306
> select * from mysql('**172.16.6.207**:3306',
> Can't connect to MySQL server on '**192.168.5.207'**
> (from 172.16.6.210:53228) (in query: select * from mysql( **192.168.5.207:** 3306',
>
> Clickhouse version **20.8.2.3**
> version **20.9.3.45**
already solved . Th... | 2020-10-21T07:57:35 | 2020-10-21T07:57:35 | {} | NONE |
ClickHouse | ClickHouse | 712,590,700 | 16,176 | naman2403 | LIMIT and OFFSET are used when you want to retrieve only a few records from your result of query.
LIMIT will retrieve only the number of records specified after the LIMIT keyword, unless the query itself returns fewer records than the number specified by LIMIT.
OFFSET is used to skip the number of records from t... | 2020-10-20T05:04:32 | 2020-10-20T05:04:32 | {} | NONE |
ClickHouse | ClickHouse | 712,674,498 | 16,183 | amosbird | We can provide two variant : `exceptStrict` and `replaceStrict`, which don't allow unmatched transformations. The error message should clearly show which column(s) are not used (matched). | 2020-10-20T08:08:25 | 2020-10-20T08:08:25 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 713,312,530 | 16,207 | YiuRULE | Also on a related subject, `X-ClickHouse-Exception-Code` is not returned if `send_progress_in_http_headers` is set to `True` | 2020-10-21T05:27:52 | 2020-10-21T05:27:52 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 712,528,038 | 16,148 | alexey-milovidov | > -z noexecstack, but seems it's a crutch
I think it's totally Ok.
> /usr/bin/gold
But we are using `lld` for all builds, maybe there is no such issue with `lld`? | 2020-10-20T01:16:00 | 2020-10-20T01:16:00 | {} | MEMBER |
ClickHouse | ClickHouse | 712,702,017 | 16,158 | alexey-milovidov | TSan: https://github.com/ClickHouse/ClickHouse/pull/15984 | 2020-10-20T08:57:43 | 2020-10-20T08:57:43 | {} | MEMBER |
ClickHouse | ClickHouse | 712,842,808 | 16,146 | alesapin | Flaky test in this branch. | 2020-10-20T13:17:58 | 2020-10-20T13:17:58 | {} | MEMBER |
ClickHouse | ClickHouse | 712,843,239 | 16,144 | alesapin | Ignore broken unbundled build and flaky test. | 2020-10-20T13:18:35 | 2020-10-20T13:18:35 | {} | MEMBER |
ClickHouse | ClickHouse | 712,848,074 | 16,142 | alesapin | `test_always_fetch_merged/test.py::test_replica_always_download` seems like "merge deadlock", but not related to changes. | 2020-10-20T13:25:21 | 2020-10-20T13:25:21 | {} | MEMBER |
ClickHouse | ClickHouse | 712,867,233 | 16,179 | den-crane | >This makes me very puzzled, Is there anything wrong with the configuration of Example_2 ?
For example you have a table T (sharded / replicated) and 4 servers (S1, S2, S3, S4)
T is sharded and replicated (S1 <-> S3) + (S2 <-> S4)
For real HighAvailability (Distributed table can switch to replica) configurati... | 2020-10-20T13:53:01 | 2020-10-20T13:53:35 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 712,911,470 | 16,178 | vitlibar | Documentation: [ALTER USER](https://clickhouse.tech/docs/en/sql-reference/statements/alter/user/), [DROP USER](https://clickhouse.tech/docs/en/sql-reference/statements/drop/#drop-user-statement) | 2020-10-20T14:53:17 | 2020-10-20T14:53:17 | {} | MEMBER |
ClickHouse | ClickHouse | 713,100,268 | 14,560 | alexey-milovidov | ```
SELECT i FROM generateRandom('i Array(Nullable(Enum8(\'hello\' = 1, \'world\' = 5)))', 1024, 1048575, 1048575) LIMIT 1048575
``` | 2020-10-20T19:51:31 | 2020-10-20T19:51:31 | {} | MEMBER |
ClickHouse | ClickHouse | 712,397,297 | 15,996 | den-crane | >INSERT INTO sfz3 SELECT *
>FROM sfz2
>Received exception from server (version 20.9.2):
>Code: 70. DB::Exception: Received from localhost:9000. DB::Exception: Conversion from Int128 to Decimal(38, 0) is not >supported: while converting source column code to destination column code.
>
>INSERT INTO sfz3 SELECT toDec... | 2020-10-19T19:37:27 | 2020-10-19T19:37:27 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 712,865,037 | 16,126 | KochetovNicolai | Works:
```
SELECT
key,
argMax(if(0, tuple(), tuple(* EXCEPT(key))), v1)
FROM kv
GROUP BY key
FORMAT TSVWithNames
Query id: 989214d2-deec-4cb3-98c3-158a2256f3b0
key argMax(tuple(v1, v2, v3, v4, v5), v1)
4 (10,20,10,20,30)
3 (70,20,10,20,30)
2 (11,20,10,20,30)
5 (10,20,10,20,30)
1 (20,20,10,20,3... | 2020-10-20T13:49:35 | 2020-10-20T13:52:24 | {} | MEMBER |
ClickHouse | ClickHouse | 713,081,913 | 16,176 | alexey-milovidov | @abyss7
> Why may you need offset for pagination? If it's a page number, then settings should be named like page_size, page_num.
It's better to keep the semantcs of the settings aligned with SQL.
> How do you get numbers 25, 1075 from 100, 1000?
From the description of the task, it should work as the follo... | 2020-10-20T19:16:11 | 2020-10-20T19:16:11 | {} | MEMBER |
ClickHouse | ClickHouse | 713,306,944 | 15,933 | filimonov | Backport to 20.3 needed | 2020-10-21T05:09:49 | 2020-10-21T05:09:49 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 712,389,103 | 15,073 | azat | >In bytedance, bigo, suning , they used rocksdb for caching the MergeTreeParts, it significantly reduces the loading time for large set tables. Because io sequence reads perform better than random reads in HDD.
@sundy-li does this implemented on top of clickhouse or inside the server code? | 2020-10-19T19:20:21 | 2020-10-19T19:20:21 | {} | MEMBER |
ClickHouse | ClickHouse | 712,579,670 | 6,772 | flyfoxCI | Is there any progress or roadmap on scaling? | 2020-10-20T04:24:09 | 2020-10-20T04:24:09 | {} | NONE |
ClickHouse | ClickHouse | 712,779,658 | 16,176 | abyss7 | > Suppose we have a query:
>
> SELECT * FROM table LIMIT 100 OFFSET 1000
> And the setting limit is set to 50 and offset is set to 75.
> Then we should execute this query as the following:
>
> SELECT * FROM table LIMIT 25 OFFSET 1075
>
1. Why may you need `offset` for pagination? If it's a page number, the... | 2020-10-20T11:19:37 | 2020-10-20T11:19:37 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 712,781,319 | 16,178 | abyss7 | Please, provide exact queries. | 2020-10-20T11:22:42 | 2020-10-20T11:22:42 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 712,851,665 | 16,044 | vzakaznikov | @vitlibar, TestFlows check is green. Could you please review it and merge it? | 2020-10-20T13:30:52 | 2020-10-20T13:30:52 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 713,081,845 | 16,191 | alesapin | Good ideas, but let's look at performance tests without any additional changes. | 2020-10-20T19:16:04 | 2020-10-20T19:18:06 | {} | MEMBER |
ClickHouse | ClickHouse | 713,100,108 | 16,191 | alesapin | Didn't help:
```
create 1 3.8490941524505615 OPTIMIZE TABLE mt_100_parts FINAL
create 0 3.725403070449829 OPTIMIZE TABLE mt_100_parts FINAL
create 0 154.7758858203888 OPTIMIZE TABLE mt_1000_parts FINAL
create 1 160.87323808670044 OPTIMIZE TABLE mt_1000_parts FINAL
... | 2020-10-20T19:51:13 | 2020-10-20T19:51:13 | {} | MEMBER |
ClickHouse | ClickHouse | 713,144,443 | 16,105 | alexey-milovidov | > Yandex synchronization check (only for Yandex employees) Pending — error, check will be restarted
Don't care. | 2020-10-20T21:16:22 | 2020-10-20T21:16:22 | {} | MEMBER |
ClickHouse | ClickHouse | 713,150,424 | 5,783 | alexey-milovidov | @avgolubev We have removed the notion of leader replicas since version 20.5 making this issue completely obsolete.
You should upgrade to the latest stable version (20.9). | 2020-10-20T21:28:39 | 2020-10-20T21:28:59 | {} | MEMBER |
ClickHouse | ClickHouse | 713,223,335 | 16,121 | alexey-milovidov | I like the idea, I have also tried to make something like that but then found it difficult. | 2020-10-21T00:52:32 | 2020-10-21T00:52:32 | {} | MEMBER |
ClickHouse | ClickHouse | 713,277,481 | 16,176 | hexiaoting | > The general intent is to add some simple building blocks into ClickHouse that will make creating some APIs easier.
I get the usage of setting's limit and offset, and I think it can be implemented in the parser phrase to modify limit and offset value when setting's limit and offset are not null.
But how to add si... | 2020-10-21T03:39:49 | 2020-10-21T03:39:49 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 713,353,211 | 16,192 | filimonov | Just a side note - it also seems that clickhouse-local can start merging parts of the attached tables on its own, so it may be potentially unsafe to use it to introspect the data of the offline server. | 2020-10-21T06:56:49 | 2020-10-21T06:56:49 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 712,716,751 | 16,152 | jiaoyang3 | > Do you mean producer (sending the same messages several times on error) or consumer (flushing to materialized view)?
>
> Can you explain the exact scenario you expect to have?
If the kafka consumer fails to send ACK, repeated consumption may occur,
So I want to know whether Kafka engine has done some processin... | 2020-10-20T09:21:59 | 2020-10-20T09:21:59 | {} | NONE |
ClickHouse | ClickHouse | 712,805,260 | 16,178 | Veryfirefly | my clickhouse-server version is `ClickHouse server version 20.9.3.45 (official build).` | 2020-10-20T12:11:57 | 2020-10-20T12:11:57 | {} | NONE |
ClickHouse | ClickHouse | 712,873,242 | 16,178 | den-crane | You can not use `alter table ` in this case. It's not Mysql.
You have to use `drop user clickhouse` & `alter clickhouse user drop IP 'address'`
| 2020-10-20T13:58:51 | 2020-10-20T13:58:51 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 712,946,405 | 16,191 | alesapin | Actually I cannot reproduce this error locally without a heavy background process like build. Perf top in bad case:
```
17.52% 17.52% BackgrProcPool libm-2.27.so [.] __log1p
10.33% 10.33% BackgrProcPool clickhouse_clang [.] DB::SimpleMergeSelector::select
6.13% 6.13% BackgrPr... | 2020-10-20T15:43:43 | 2020-10-20T15:43:43 | {} | MEMBER |
ClickHouse | ClickHouse | 713,040,135 | 14,135 | alexey-milovidov | I'll do it. | 2020-10-20T18:03:51 | 2020-10-20T18:03:51 | {} | MEMBER |
ClickHouse | ClickHouse | 712,438,159 | 12,736 | vitlibar | > @vitlibar, this PR seems to be ready for merge. Is there anything else that is holding it?
No, everything's fine now. | 2020-10-19T20:58:26 | 2020-10-19T20:58:26 | {} | MEMBER |
ClickHouse | ClickHouse | 712,565,533 | 16,157 | fernando1987 | It's bad judgment | 2020-10-20T03:29:51 | 2020-10-20T03:29:51 | {} | NONE |
ClickHouse | ClickHouse | 712,742,925 | 13,597 | tangxueming | > I have changed the CMakeLists.txt line 149, see below
>
> ```
> origin: find_program (OBJCOPY_PATH NAMES "llvm-objcopy" "llvm-objcopy-10" "llvm-objcopy-9" "llvm-objcopy-8" "objcopy")
>
> changed: find_program (OBJCOPY_PATH NAMES "llvm-objcopy" "llvm-objcopy-10" "llvm-objcopy-9" "llvm-objcopy-8" "objcopy" PATH... | 2020-10-20T10:06:28 | 2020-10-20T10:08:08 | {} | NONE |
ClickHouse | ClickHouse | 712,802,291 | 16,178 | Veryfirefly | > Please, provide exact queries.
I performed the update and delete operations, the code is as follows
```sql
alter table system.users delete where name='clickhouse';
```
```sql
alter table system.users update host_ip=array('::/0') where name='clickhouse'
``` | 2020-10-20T12:06:17 | 2020-10-20T12:06:17 | {} | NONE |
ClickHouse | ClickHouse | 712,539,036 | 15,073 | sundy-li | > > In bytedance, bigo, suning , they used rocksdb for caching the MergeTreeParts, it significantly reduces the loading time for large set tables. Because io sequence reads perform better than random reads in HDD.
>
> @sundy-li does this implemented on top of clickhouse or inside the server code?
Inside the codes... | 2020-10-20T01:57:11 | 2020-10-20T13:50:04 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 712,931,397 | 16,181 | 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-... | 2020-10-20T15:21:34 | 2020-10-20T15:21:34 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 713,129,377 | 16,105 | azat | @alexey-milovidov friendly ping | 2020-10-20T20:45:39 | 2020-10-20T20:45:39 | {} | MEMBER |
ClickHouse | ClickHouse | 713,204,874 | 13,707 | tooptoop4 | how to specify S3 access/secret key? I couldn't see in the docs | 2020-10-20T23:55:55 | 2020-10-20T23:55:55 | {} | NONE |
ClickHouse | ClickHouse | 712,553,592 | 15,883 | yanchaoguo | > @yanchaoguo The reason for this problem may be `binlog_checksum = NONE`, you can check it with `SHOW VARIABLES WHERE Variable_name ='binlog_checksum';`
mysql> SHOW VARIABLES WHERE Variable_name ='binlog_checksum'
-> ;
+-----------------+-------+
| Variable_name | Value |
+-----------------+-------+
| bi... | 2020-10-20T02:48:11 | 2020-10-20T02:48:11 | {} | NONE |
ClickHouse | ClickHouse | 712,754,851 | 15,977 | tavplubix | `DROP DATABASE test_database` hangs for unknown reason. Probably it's something similar to #13822. Let's try to capture stacktraces | 2020-10-20T10:28:47 | 2020-10-20T10:28:47 | {} | MEMBER |
ClickHouse | ClickHouse | 712,842,590 | 16,147 | alesapin | Ignore broken unbundled build. | 2020-10-20T13:17:39 | 2020-10-20T13:17:39 | {} | MEMBER |
ClickHouse | ClickHouse | 712,930,632 | 16,180 | alexey-milovidov | Changelog Ok. | 2020-10-20T15:20:32 | 2020-10-20T15:20:32 | {} | MEMBER |
ClickHouse | ClickHouse | 712,931,330 | 16,182 | alexey-milovidov | No significant changes according to the changelog. | 2020-10-20T15:21:27 | 2020-10-20T15:21:27 | {} | MEMBER |
ClickHouse | ClickHouse | 713,084,171 | 16,176 | alexey-milovidov | The general intent is to add some simple building blocks into ClickHouse that will make creating some APIs easier. | 2020-10-20T19:20:10 | 2020-10-20T19:20:10 | {} | MEMBER |
ClickHouse | ClickHouse | 713,160,480 | 16,198 | traceon | > Testflows broken here: #12736
Here it fails differently, if compared to other failures, that happened after merging #12736. There were no known instabilities or failures at the time of merging it, so, the general rule should apply: until proven otherwise, the first suspect is the code of the PR where it fails.
... | 2020-10-20T21:50:34 | 2020-10-20T21:50:34 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 713,380,193 | 16,191 | alesapin | Now I cannot reproduce anything even with concurrent clickhouse build. | 2020-10-21T07:49:48 | 2020-10-21T07:49:48 | {} | MEMBER |
ClickHouse | ClickHouse | 712,657,713 | 15,702 | bohutang | Hi,
We are working on #15951, this should be the next to be addressed. | 2020-10-20T07:37:16 | 2020-10-20T07:37:16 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 712,756,132 | 15,977 | TCeason | Ok | 2020-10-20T10:31:27 | 2020-10-20T10:31:27 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 712,811,026 | 13,956 | Akazz | We've discovered issues with test_multiple_disks from the "Integration tests" section. It will be addressed in a separate PR. | 2020-10-20T12:23:13 | 2020-10-20T12:23:13 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 712,873,528 | 16,126 | KochetovNicolai | It also doesn't work for subqueries like `select untuple(x) from (select (1, 2) as x)`. But I like this implementation better anyway ... | 2020-10-20T13:59:14 | 2020-10-20T13:59:14 | {} | MEMBER |
ClickHouse | ClickHouse | 712,931,367 | 16,181 | alexey-milovidov | No significant changes according to the changelog. | 2020-10-20T15:21:30 | 2020-10-20T15:21:30 | {} | MEMBER |
ClickHouse | ClickHouse | 713,035,778 | 16,191 | alexey-milovidov | `log1p` is needed for better precision of `log(1 + x)` expression for very small x.
But in our code we don't need it because x is integer.
Could you also try to replace log1p to log(1 + x) to see what will change? (and add a comment) | 2020-10-20T17:55:46 | 2020-10-20T17:55:46 | {} | MEMBER |
ClickHouse | ClickHouse | 713,236,230 | 16,191 | alexey-milovidov | Did not help. | 2020-10-21T01:36:57 | 2020-10-21T01:36:57 | {} | MEMBER |
ClickHouse | ClickHouse | 713,261,586 | 8,467 | huynhphuong10284 | Just give more information about duplicated records in clickHouse as below:
- We're consuming data from Kafka then saving to clickHouse (using Distributed table in cluster and ReplicatedMergeTree in each shard)
- Some times Kafka has duplicated data ( in producer if a writing process is not received responses from Ka... | 2020-10-21T02:56:37 | 2020-10-21T02:56:37 | {} | NONE |
ClickHouse | ClickHouse | 713,284,966 | 16,178 | Veryfirefly | > Documentation: [ALTER USER](https://clickhouse.tech/docs/en/sql-reference/statements/alter/user/), [DROP USER](https://clickhouse.tech/docs/en/sql-reference/statements/drop/#drop-user-statement)
Thanks, but the offical documentation are not very detailed.
I don’t know how to solve it. The guide of creation of use... | 2020-10-21T04:03:08 | 2020-10-21T04:03:08 | {} | NONE |
ClickHouse | ClickHouse | 713,356,790 | 16,191 | alesapin | Actually it helps, but not too much:
```
create 1 22.754340648651123 INSERT INTO mt_100_parts SELECT number, rand() % 10000, rand() FROM numbers_mt(100000000)
create 0 23.076173067092896 INSERT INTO mt_100_parts SELECT number, rand() % 10000, rand() FROM numbers_mt(100000000)
create 1 ... | 2020-10-21T07:04:22 | 2020-10-21T07:04:22 | {} | MEMBER |
ClickHouse | ClickHouse | 712,537,991 | 16,158 | alexey-milovidov | TODO: add an option to clamp settings instead of throw in HTTP interface. | 2020-10-20T01:53:31 | 2020-10-20T01:53:31 | {} | MEMBER |
ClickHouse | ClickHouse | 712,736,185 | 1,469 | SkyGra | when the window function will be supported in Clickhouse?
I hope to see the function early! | 2020-10-20T09:55:17 | 2020-10-20T09:55:17 | {
"+1": 17
} | NONE |
ClickHouse | ClickHouse | 713,107,400 | 16,191 | alesapin | On my local machine (without thin lto):
```
create 0 26.846017122268677 INSERT INTO mt_100_parts SELECT number, rand() % 10000, rand() FROM numbers_mt(100000000)
create 0 71.50743651390076 INSERT INTO mt_1000_parts SELECT number, rand() % 10000, rand() FROM numbers_mt(100000000)
create 0 ... | 2020-10-20T20:04:40 | 2020-10-20T20:04:40 | {} | MEMBER |
ClickHouse | ClickHouse | 713,143,979 | 16,105 | alexey-milovidov | I'm Ok making it on by default.
The only surprising case is filtering by rand... and even this case looks Ok. | 2020-10-20T21:15:29 | 2020-10-20T21:15:29 | {} | MEMBER |
ClickHouse | ClickHouse | 713,362,092 | 15,983 | alesapin | Stress:
```
2020.10.21 02:51:14.812613 [ 420272 ] {} <Error> Application: Caught exception while loading metadata: Code: 57, e.displayText() = DB::Exception: Mapping for table with UUID=00000738-1000-4000-8000-000000000001 al
ready exists: Cannot attach table `test_16`.`mv` from metadata file /var/lib/clickhouse/sto... | 2020-10-21T07:15:13 | 2020-10-21T07:15:13 | {} | MEMBER |
ClickHouse | ClickHouse | 712,417,216 | 7,462 | den-crane | 20.11.1.4897
```
┌─attr─┬─_id─┬─z─────┐
│ 1 │ 0 │ ['1'] │
└──────┴─────┴───────┘
┌─attr─┬─_id─┬─z─────┐
│ 1 │ 1 │ ['1'] │
└──────┴─────┴───────┘
``` | 2020-10-19T20:16:34 | 2020-10-19T20:16:34 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 712,516,802 | 16,131 | chang0391 | @den-crane
version：20.5.4
| 2020-10-20T00:35:25 | 2020-10-20T00:35:25 | {} | NONE |
ClickHouse | ClickHouse | 712,719,995 | 16,148 | socketpair | @alexey-milovidov bug not in the linker. Bug in hundreds of `.s` files with missing section note. So, every `.s` file without section for compatibility purposes treated as requiring executable stack. I'm unsure, who is analyzing presence/absence of section note. Possibly linker. Possibly `as`. So, anyway, even for `ll... | 2020-10-20T09:27:28 | 2020-10-20T09:27:41 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 712,863,681 | 16,152 | filimonov | So you saying about retrying commits. We already do that, but it not always possible (losing the connection / rebalance etc).
One of the possible solutions is to store offsets on clickhouse side, we considering that (but no timeline for that). | 2020-10-20T13:47:31 | 2020-10-20T13:47:31 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 713,241,854 | 15,996 | l1t1 | note:
when input`code` **Decimal128(38, 0)**, the error msg Decimal data type family must have **exactly two arguments: precision and scale.** is confusing, it should reply as yours
**code Decimal128(38, 0) -- is incorrect. You should use Decimal(38, 0) or Decimal128(0)** | 2020-10-21T01:56:54 | 2020-10-21T01:56:54 | {} | NONE |
ClickHouse | ClickHouse | 713,346,867 | 16,192 | filimonov | Thread sanitizer caught the data race in the context
Read access w/o lock:
https://github.com/ClickHouse/ClickHouse/blob/97671a17719eb9c101095e01cac7baa282908126/src/Interpreters/Context.h#L459-L461
Modification:
https://github.com/ClickHouse/ClickHouse/blob/72b13396561023bce3d45fe1a2dea8cec12a60dd/src/Interpre... | 2020-10-21T06:43:00 | 2020-10-21T09:43:59 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 712,679,717 | 15,883 | zhang2014 | @yanchaoguo 暂不支持binlog_checksum = None的MySQL主从消费，我将尝试修复它。 | 2020-10-20T08:17:38 | 2020-10-20T08:17:38 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 712,861,207 | 16,126 | KochetovNicolai | ```
SELECT
key,
untuple(argMax(tuple(* EXCEPT(key)), v1)),
untuple(argMin(tuple(* EXCEPT(key)), v1))
FROM kv
GROUP BY key
FORMAT TSVWithNames
Query id: b5f3276b-4109-41af-943e-fa2e11380854
Received exception from server (version 20.11.1):
Code: 179. DB::Exception: Received from localhost:900... | 2020-10-20T13:44:44 | 2020-10-20T13:44:44 | {} | MEMBER |
ClickHouse | ClickHouse | 712,931,391 | 16,182 | 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-... | 2020-10-20T15:21:33 | 2020-10-20T15:21:33 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 713,048,312 | 16,032 | Avogar | > Some queries like `select * from system.tables` or `select * from system.columns` still may fail for the same reason when MySQL is not available. Maybe we should return cached result from `DatabaseConnectionMySQL::getCreateTableQueryImpl(...)` and `DatabaseConnectionMySQL::getTablesIterator(...)`?
In queries like ... | 2020-10-20T18:19:33 | 2020-10-20T18:19:33 | {} | MEMBER |
ClickHouse | ClickHouse | 713,158,986 | 16,167 | alexey-milovidov | Pass without CI. | 2020-10-20T21:47:23 | 2020-10-20T21:47:23 | {} | MEMBER |
ClickHouse | ClickHouse | 713,194,623 | 14,560 | alexey-milovidov | Regression: https://github.com/ClickHouse/ClickHouse/pull/12182 | 2020-10-20T23:21:37 | 2020-10-20T23:21:37 | {} | MEMBER |
ClickHouse | ClickHouse | 713,280,667 | 16,178 | Veryfirefly | > You can not use `alter table ` in this case. It's not Mysql.
>
> You have to use `drop user clickhouse` & `alter clickhouse user drop IP 'address'`
There are two identical users, but the host_ip is different. How to delete one of them.
Excuse me, does alter have an example? The offical documentation and Google... | 2020-10-21T03:50:47 | 2020-10-21T03:50:47 | {} | NONE |
ClickHouse | ClickHouse | 712,472,689 | 15,988 | traceon | @den-crane Does this look like a good solution for the slow LDAP server operations, that you were pointing out?
```
verification_cooldown - a period of time, in seconds, after a successful bind attempt, during which a user will be assumed
to be successfully authenticated for all consecutiv... | 2020-10-19T22:17:06 | 2020-10-19T22:18:24 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 713,081,158 | 16,176 | alexey-milovidov | @naman2403 The task is about adding separate settings `limit` and `offset` that allow to adjust the query result in off-SQL way.
BTW, I also have similar thoughts about adding support for off-SQL prewhere expressions. To allow user to specify arbitrary SQL query but to restrict the available dataset. | 2020-10-20T19:14:45 | 2020-10-20T19:17:27 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 713,263,330 | 16,152 | jiaoyang3 | ok，It's great, and I'm thinking about trying to change the code myself | 2020-10-21T03:01:24 | 2020-10-21T03:01:24 | {} | NONE |
ClickHouse | ClickHouse | 712,497,810 | 16,171 | den-crane | STR:
```
20.11.1
CREATE TABLE errors_local(level LowCardinality(String))
ENGINE=ReplacingMergeTree ORDER BY level settings min_bytes_for_wide_part = '10000000';
insert into errors_local select toString(number) from numbers(10000);
SELECT toTypeName(level)
FROM errors_local FINAL
PREWHERE isNotNull(l... | 2020-10-19T23:32:33 | 2020-10-19T23:39:28 | {
"+1": 2
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 712,428,187 | 7,462 | den-crane | 20.11.1.4897
Now it needs `max_threads=1` to reproduce.
```
set max_threads=1;
SELECT attr, _id, arrayFilter(x -> (x IN (select '1')), attr_list) z
FROM TESTTABLE ARRAY JOIN z AS attr ORDER BY _id LIMIT 3 BY attr;
Cannot get value from Function: While executing MergeSortingTransform.
``` | 2020-10-19T20:38:18 | 2020-10-19T20:38:37 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 712,724,496 | 16,148 | socketpair | So, what should I do in order to move on the issue ? | 2020-10-20T09:34:58 | 2020-10-20T09:35:08 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 713,005,139 | 16,169 | alexey-milovidov | Leak in OpenSSL - known issue. | 2020-10-20T17:01:24 | 2020-10-20T17:01:24 | {} | MEMBER |
ClickHouse | ClickHouse | 713,100,944 | 15,685 | alexey-milovidov | We have reviewed the failures.
AST Fuzzer has found already known issue: https://github.com/ClickHouse/ClickHouse/issues/14560
Integration tests has showed that the `test_distributed_over_live_view` is flaky. @vzakaznikov will fix it.
Stress test (thread) showed an issue that is fixed in another PR.
Now everyth... | 2020-10-20T19:52:49 | 2020-10-20T19:52:49 | {} | MEMBER |
ClickHouse | ClickHouse | 713,157,773 | 16,148 | alexey-milovidov | This PR does not cross-compile for Darwin: https://clickhouse-builds.s3.yandex.net/16148/b6aa38a339d710324f611c1bc9f46dfcb5a1b945/clickhouse_special_build_check/build_log_800409216_1603124866.txt
Release build also failed: https://clickhouse-builds.s3.yandex.net/16148/b6aa38a339d710324f611c1bc9f46dfcb5a1b945/clickho... | 2020-10-20T21:44:46 | 2020-10-20T21:44:46 | {} | MEMBER |
ClickHouse | ClickHouse | 713,171,054 | 14,135 | alexey-milovidov | Wow, it's fixed. | 2020-10-20T22:14:50 | 2020-10-20T22:14:50 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 713,212,146 | 11,844 | alexey-milovidov | It's strange to have GCM but not to have CTR mode. | 2020-10-21T00:15:00 | 2020-10-21T00:15:00 | {} | MEMBER |
ClickHouse | ClickHouse | 712,387,982 | 15,073 | azat | >Now with the newest commit, this engine supports supportsIndexForIn
Great!
>The larger N is, the much more batch sequence read can merge tree engine achieve, the less IO call cost there will be, while in rocksdb engine, the keys are mostly random read. This engine may be useful in large datasets to filter some s... | 2020-10-19T19:18:03 | 2020-10-19T19:18:03 | {} | MEMBER |
ClickHouse | ClickHouse | 712,772,366 | 16,131 | abyss7 | You should provide at least your cluster configuration and the logs from all the shards with a part corresponding to your query execution - it would help to better analyze and understand timings.
Also, as mentioned earlier, the `global` join for local MergeTree is the same as non-`global` one. You may try the non-`g... | 2020-10-20T11:05:08 | 2020-10-20T11:05:08 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 713,155,750 | 16,198 | alexey-milovidov | Testflows broken here: #12736 | 2020-10-20T21:40:14 | 2020-10-20T21:40:14 | {} | MEMBER |
ClickHouse | ClickHouse | 713,198,497 | 14,560 | alexey-milovidov | \+ wrong logic and logical race condition. | 2020-10-20T23:33:49 | 2020-10-20T23:33:54 | {} | MEMBER |
ClickHouse | ClickHouse | 715,655,222 | 16,072 | baibaichen | @alexey-milovidov would you please revert this commit? My macos is catalina version 10.15 and AppClang is 11
> -- The C compiler identification is AppleClang 11.0.3.11030032
> -- The CXX compiler identification is AppleClang 11.0.3.11030032
This my cmake setting: `-DCMAKE_CXX_COMPILER=/usr/bin/clang++ -DCMAKE_... | 2020-10-24T01:54:25 | 2020-10-24T01:54:25 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 715,669,927 | 16,331 | den-crane | I was confused by
```
echo '{"value":5,"string_value":"6"}'|gzip -c| curl -sS --data-binary @- -H 'Content-Encoding: qqqq' 'http://localhost:8123/?query=insert+into+json_comressed_stream_sql+format+JSONEachRow'
Code: 48, e.displayText() = DB::Exception: Unknown compression method qqqq. Only 'auto', 'none', 'gzip'... | 2020-10-24T04:08:59 | 2020-10-24T04:08:59 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 715,939,256 | 12,632 | keliss | Cool, thank you | 2020-10-24T15:32:47 | 2020-10-24T15:32:47 | {} | NONE |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.