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 | 515,833,701 | 6,148 | DaisyFlower | Hello @blinkov
I've seen the error details
`ERROR:mkdocs.structure.pages:File not found: data_types/domains/ipv4.md
ERROR:mkdocs.commands.build:Error reading page 'data_types/domains/ipv4.md': [Errno 36] File name too long: '/place/sandbox-data/tasks/8/7/474097778/ClickHouse/docs/zh/data_types/domains/ipv4.md'`
I ... | 2019-07-29T03:44:24 | 2019-07-29T03:44:24 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 516,931,553 | 6,227 | CurtizJ | You can set `CODEC(NONE)` for column and data will be stored without compression. Is it actually what do you want? | 2019-07-31T16:52:48 | 2019-07-31T16:52:48 | {} | MEMBER |
ClickHouse | ClickHouse | 516,972,923 | 5,972 | BayoNet | @blinkov Do you have any comments? | 2019-07-31T18:45:20 | 2019-07-31T18:45:20 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 515,778,507 | 6,195 | alexey-milovidov | @vitlibar That's trivial bug in code. | 2019-07-28T17:01:25 | 2019-07-28T17:01:25 | {} | MEMBER |
ClickHouse | ClickHouse | 515,919,107 | 6,200 | alexey-milovidov | Yes, it makes sense.
Another option: specify large enough `merge_with_ttl_timeout` (e.g. `1000000`) in `/yandex/merge_tree` in config.xml (it's a place for global MergeTree settings; you can find all the settings in `MergeTreeSettings.h` file; if you have not specified a setting in `SETTINGS` clause of CREATE TABLE,... | 2019-07-29T09:31:54 | 2019-07-29T09:31:54 | {} | MEMBER |
ClickHouse | ClickHouse | 516,281,744 | 6,214 | aopetrov86 | Without `abs`:
```sql
SELECT
sum(ASD) AS asd,
intDiv(toInt64(asd), abs(toInt64(asd))) AS int_div_with_abs,
intDiv(toInt64(asd), toInt64(asd)) AS int_div_without_abs
FROM
(
SELECT ASD
FROM
(
SELECT [-1000, -1000] AS asds
)
ARRAY JOIN asds AS ASD
)
```
```
┌... | 2019-07-30T06:26:18 | 2019-07-30T06:26:18 | {} | NONE |
ClickHouse | ClickHouse | 516,432,494 | 6,191 | CurtizJ | Performance tests have been failed because of CI problems. | 2019-07-30T14:01:49 | 2019-07-30T14:01:49 | {} | MEMBER |
ClickHouse | ClickHouse | 516,432,202 | 6,139 | alexey-milovidov | For non-optimized build:
- first thrown exception: 1.5 sec;
- subsequent exceptions (with different stack traces): 0.75 sec;
For optimized build:
- first thrown exception: 100 ms;
- subsequent exceptions (with different stack traces): 10..20 ms;
| 2019-07-30T14:01:10 | 2019-07-30T14:04:32 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 516,634,163 | 6,231 | alexey-milovidov | Only the b627501 commit will be backported. | 2019-07-30T23:30:17 | 2019-07-30T23:30:17 | {} | MEMBER |
ClickHouse | ClickHouse | 516,673,000 | 5,547 | alexey-milovidov | Fixed in master. | 2019-07-31T02:54:22 | 2019-07-31T02:54:22 | {} | MEMBER |
ClickHouse | ClickHouse | 516,924,318 | 6,216 | filimonov | ClickHouse currently doesn't support subsecond time precision (but there are some plans for that https://github.com/yandex/ClickHouse/issues/4860 )
For achieving similar results in numerical case you can try to use `range(N)` function.
```
SELECT arrayJoin(arrayMap(x -> (toFloat32(now()) + (0.2 * x)), range(10)))... | 2019-07-31T16:31:51 | 2019-07-31T16:32:20 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 516,971,424 | 5,992 | BayoNet | @blinkov Do you have any comments? | 2019-07-31T18:41:32 | 2019-07-31T18:41:32 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 515,712,968 | 6,187 | alexey-milovidov | FYI. Does not reproduce in master build with clang 9:
```
milovidov-nix :) CREATE TEMPORARY TABLE t (a UInt8)
CREATE TEMPORARY TABLE t
(
`a` UInt8
)
Ok.
0 rows in set. Elapsed: 0.007 sec.
milovidov-nix :) select * from t join t using a;
SELECT *
FROM t
INNER JOIN t USING (a)
Ok.
```
| 2019-07-27T20:54:44 | 2019-07-27T20:54:44 | {} | MEMBER |
ClickHouse | ClickHouse | 515,715,610 | 6,124 | alexey-milovidov | But it's by default. | 2019-07-27T21:43:56 | 2019-07-27T21:43:56 | {} | MEMBER |
ClickHouse | ClickHouse | 515,915,648 | 2,381 | millin | I had the same error and I spent a lot of time searching for its causes.
Finally I discovered that this happens when getting empty message from Kafka :man_facepalming: | 2019-07-29T09:22:14 | 2019-07-29T09:22:14 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 516,043,556 | 6,203 | alexey-milovidov | It works as expected. | 2019-07-29T15:36:26 | 2019-07-29T15:36:26 | {} | MEMBER |
ClickHouse | ClickHouse | 516,407,717 | 6,167 | KochetovNicolai | I can't reproduce your issue on master. Test doesn't fail (except it can return different result as limit is not deterministic without order by).
This fix seems to be correct but I couldn't create test case which fails now. | 2019-07-30T12:58:56 | 2019-07-30T12:58:56 | {} | MEMBER |
ClickHouse | ClickHouse | 516,911,130 | 6,215 | filimonov | Alternaitve is smth like that:
```
select ..., arrayReduce( 'sumMap', [payMode.code],[payMode.fee] ), ...
```
Will work properly on real table, but please be careful - i've found that there is some bug in processing constant parameters for that type of function call combination (see #6242). | 2019-07-31T15:56:27 | 2019-07-31T15:56:27 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 515,713,059 | 6,187 | alexey-milovidov | Does not reproduce with gcc 8.3.0 on master. | 2019-07-27T20:56:20 | 2019-07-27T20:56:20 | {} | MEMBER |
ClickHouse | ClickHouse | 515,713,179 | 6,187 | alexey-milovidov | Del: -- @4ertus2 Backport needed? -- | 2019-07-27T20:58:27 | 2019-07-27T21:02:56 | {} | MEMBER |
ClickHouse | ClickHouse | 515,769,037 | 6,124 | alexey-milovidov | Backported changes to query profiler to 19.12, because it was in unusable state there. | 2019-07-28T14:50:12 | 2019-07-28T14:50:12 | {} | MEMBER |
ClickHouse | ClickHouse | 516,482,073 | 6,139 | kreuzerkrieg | you are da man!
but I loved these numbers before the edit :)
| 2019-07-30T15:56:57 | 2019-07-30T15:59:32 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 516,658,006 | 6,139 | alexey-milovidov | Available in master.
PS. If you install from packages, don't forget to install `clickhouse-common-dbg` package that ships debug info. | 2019-07-31T01:38:06 | 2019-07-31T01:38:06 | {} | MEMBER |
ClickHouse | ClickHouse | 516,949,737 | 5,042 | MahmoudGoda0 | Thanks, I have 2 questions please
1 - Is this is a stable version ? if no, When it will be stable version ?
2- I installed this version already "http://repo.yandex.ru/clickhouse/deb/stable/ main/", If i install the testing release, so i will have 2 releases on the same time? How i can run the second one ?
Sorry f... | 2019-07-31T17:42:07 | 2019-07-31T17:42:07 | {} | NONE |
ClickHouse | ClickHouse | 517,014,389 | 6,248 | den-crane | It looks like a bug.
What CH version do you use?
Can you show DDL (show create table myTable)? | 2019-07-31T20:46:17 | 2019-07-31T20:46:43 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 515,770,911 | 6,190 | sundy-li | LGTM. | 2019-07-28T15:14:35 | 2019-07-28T15:14:35 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 515,955,951 | 6,052 | millin | Solved | 2019-07-29T11:37:06 | 2019-07-29T11:37:06 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 516,009,881 | 6,180 | akuzm | Looks like it is already merged to 19.12. | 2019-07-29T14:15:24 | 2019-07-29T14:15:24 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 516,190,561 | 6,139 | alexey-milovidov | I'm trying to add lines printing in my branch https://github.com/yandex/ClickHouse/pull/6201
It will look like this:
```
2019.07.30 01:43:23.281095 [ 57 ] {4d608dcb-4a8a-4e53-b61b-7e5c288d67b1} <Error> executeQuery: Code: 395, e.displayText() = DB::Exception: Value passed to 'throwIf' function is non zero (versi... | 2019-07-29T22:46:18 | 2019-07-29T22:46:18 | {} | MEMBER |
ClickHouse | ClickHouse | 516,604,319 | 3,143 | alexey-milovidov | ```
│ MemoryTrackingInBackgroundProcessingPool │ -5.87 GiB │
│ MemoryTrackingInBackgroundSchedulePool │ -53.49 GiB │
```
It's technically Ok, because BackgroundPool frees memory that was allocated outside.
Example:
- SELECT from MergeTree is loading index for data parts;
- "cle... | 2019-07-30T21:33:00 | 2019-07-30T21:35:04 | {} | MEMBER |
ClickHouse | ClickHouse | 516,611,234 | 6,225 | alexey-milovidov | Affects all versions. | 2019-07-30T21:56:06 | 2019-07-30T21:56:06 | {} | MEMBER |
ClickHouse | ClickHouse | 516,645,495 | 6,214 | alexey-milovidov | This is how C++ works:
```
$ g++ -xc++ -include cstdint -include stdlib.h - <<< "int main(int, char ** argv) { return int8_t(atoi(argv[1])) / uint64_t(atoi(argv[2])); }"
$ ./a.out -1 3; echo $?
85
```
Will fix. | 2019-07-31T00:26:55 | 2019-07-31T00:26:55 | {} | MEMBER |
ClickHouse | ClickHouse | 516,977,454 | 6,165 | BayoNet | @blinkov Do you have any comments? | 2019-07-31T18:57:45 | 2019-07-31T18:57:45 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 515,745,222 | 6,186 | alexey-milovidov | > ../dbms/src/Common/new_delete.cpp:41:20: error: always_inline function might not be inlinable [-Werror=attributes]
ALWAYS_INLINE void untrackMemory(void * ptr [[maybe_unused]], std::size_t size [[maybe_unused]] = 0) noexcept | 2019-07-28T08:53:24 | 2019-07-28T08:53:24 | {} | MEMBER |
ClickHouse | ClickHouse | 515,788,425 | 6,199 | alexey-milovidov | > Is it actually a issue to have large partition size?
No, it's not an issue. You will have higher number of data parts inside partition, because ClickHouse limits the size of a single data part to 150 GB by default.
The system is designed in a way that you can even have single partition (a table without `PARTITI... | 2019-07-28T19:28:20 | 2019-07-28T19:28:20 | {
"heart": 1
} | MEMBER |
ClickHouse | ClickHouse | 516,207,862 | 6,201 | alexey-milovidov | Now query profiling looks like this: https://gist.github.com/alexey-milovidov/92758583dd41c24c360fdb8d6a4da194 | 2019-07-30T00:08:28 | 2019-07-30T00:08:28 | {} | MEMBER |
ClickHouse | ClickHouse | 516,370,051 | 6,154 | dimarub2000 | But we have `CREATE TABLE [IF NOT EXISTS] [db.]table_name ENGINE = engine AS SELECT ...` | 2019-07-30T10:55:54 | 2019-07-30T10:55:54 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 516,795,818 | 6,200 | CurtizJ | I suggest to introduce query `SYSTEM STOP TTL MERGES`. | 2019-07-31T10:39:50 | 2019-07-31T10:39:50 | {} | MEMBER |
ClickHouse | ClickHouse | 516,836,540 | 5,846 | simpl1g | @Rostish we are waiting when https://github.com/yandex/ClickHouse/pull/6075 will be released to give it a try | 2019-07-31T12:55:50 | 2019-07-31T12:55:50 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 516,022,656 | 6,209 | dimarub2000 | Single column ValuesList queries doesn't work yet. Trying to implement.
UPD: Fixed in `parseAndInsertValues`. But not sure if it looks OK. | 2019-07-29T14:46:05 | 2019-07-29T16:38:32 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 516,199,475 | 6,201 | alexey-milovidov | TODO: open correct file when debugging information is placed separate (in `/usr/lib/debug`). | 2019-07-29T23:25:53 | 2019-07-29T23:25:53 | {} | MEMBER |
ClickHouse | ClickHouse | 516,281,571 | 6,191 | svladykin | I click on Details for the failed Performance test and it time outs. | 2019-07-30T06:25:36 | 2019-07-30T06:25:36 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 516,639,068 | 6,228 | alexey-milovidov | There is no upper bound on sleep time. I think it should be no more than one second.
Otherwise it's too easy to hang a query with `SET max_execution_speed = 1`.
| 2019-07-30T23:54:05 | 2019-07-30T23:54:05 | {} | MEMBER |
ClickHouse | ClickHouse | 516,682,622 | 6,124 | alexey-milovidov | Cross-reference: https://github.com/libunwind/libunwind/issues/16 | 2019-07-31T03:46:06 | 2019-07-31T03:46:06 | {} | MEMBER |
ClickHouse | ClickHouse | 516,907,993 | 6,209 | dimarub2000 | Latest commits also fix problem #6217, fix is implemented in `parseColumnsListFromString` function which is used in `ITableFunctionFileLike.cpp` and `TableFunctionValues.cpp`. | 2019-07-31T15:48:55 | 2019-07-31T15:48:55 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 516,951,255 | 5,042 | alexey-milovidov | > 1 - Is this is a stable version ?
Packages in testing repository have passed all CI tests but are not considered stable before they appear in the stable repository.
> if no, When it will be stable version ?
We have stable release every two weeks, most of the time.
> 2- I installed this version already "ht... | 2019-07-31T17:46:25 | 2019-07-31T17:46:25 | {} | MEMBER |
ClickHouse | ClickHouse | 516,885,011 | 5,042 | alexey-milovidov | It's already available in testing release:
https://repo.yandex.ru/clickhouse/deb/testing/main/ | 2019-07-31T14:54:39 | 2019-07-31T14:54:39 | {} | MEMBER |
ClickHouse | ClickHouse | 516,933,909 | 6,238 | den-crane | There is no query cache in CH.
Do you mean Linux cache? It normal Unix-like OS use all free memory for the disk cache. | 2019-07-31T16:59:11 | 2019-07-31T16:59:11 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 515,713,405 | 6,187 | alexey-milovidov | Does not reproduce in 19.9.2.4 (official build). | 2019-07-27T21:02:25 | 2019-07-27T21:02:25 | {} | MEMBER |
ClickHouse | ClickHouse | 515,770,065 | 6,187 | zlobober | Let me collect more information on how to reproduce this... | 2019-07-28T15:03:35 | 2019-07-28T15:03:35 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 515,990,528 | 5,880 | alexey-milovidov | I don't remember. | 2019-07-29T13:26:47 | 2019-07-29T13:26:47 | {
"laugh": 1
} | MEMBER |
ClickHouse | ClickHouse | 516,043,398 | 6,203 | alexey-milovidov | To put backslash in regexp, you have to escape it in a string literal with another backslash. Example:
```
select replaceRegexpOne('smc.exec.ds.alpha.testcase.testcase4g.epsfallback.epsfallback_5g2eps_test.go','(.*)\\..+\\.go','\\1')
``` | 2019-07-29T15:36:02 | 2019-07-29T15:36:02 | {} | MEMBER |
ClickHouse | ClickHouse | 516,288,697 | 6,063 | rangez | thank you very much @zhang2014 | 2019-07-30T06:50:30 | 2019-07-30T06:50:30 | {} | NONE |
ClickHouse | ClickHouse | 516,511,927 | 6,214 | aopetrov86 | More strange:
```sql
SELECT
ID,
sum(ASD) AS asd,
intDiv(abs(asd), asd) AS int_div_abs_first,
intDiv(asd, abs(asd)) AS int_div_abs_second
FROM
(
SELECT
ID,
ASD
FROM
(
SELECT
[-1000, -1000, 1000, 1000] AS asds,
[1, 1, 2,... | 2019-07-30T17:17:26 | 2019-07-30T17:17:26 | {} | NONE |
ClickHouse | ClickHouse | 516,567,383 | 3,143 | Nikolo | ```
SELECT
metric,
formatReadableSize(value)
FROM system.metrics
WHERE metric LIKE 'Memory%'
┌─metric───────────────────────────────────┬─formatReadableSize(value)─┐
│ MemoryTracking │ 18.56 KiB │
│ MemoryTrackingInBackgroundProcessingPool │ -5.87 GiB ... | 2019-07-30T19:46:56 | 2019-07-30T19:46:56 | {} | NONE |
ClickHouse | ClickHouse | 516,603,149 | 6,016 | alexey-milovidov | `Compatibility check` is still broken. | 2019-07-30T21:29:20 | 2019-07-30T21:29:20 | {} | MEMBER |
ClickHouse | ClickHouse | 516,633,043 | 6,124 | alexey-milovidov | @laplab
PS. If you are interested, it was continued in https://github.com/yandex/ClickHouse/pull/6201 | 2019-07-30T23:25:29 | 2019-07-30T23:25:29 | {} | MEMBER |
ClickHouse | ClickHouse | 516,732,137 | 6,127 | Enmk | > geohashCoverAreaWithBoxes() seems confusing to me in 'WithBoxes' part. It assumes there are multiple boxes and we do something with them. But there is single 'box argument'. Bounding Box is a commonly used term. Probably we could name it geohashArrayFromBoundingBox() if you want a long descriptive name?
As agreed ... | 2019-07-31T07:28:05 | 2019-07-31T07:28:05 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 516,961,744 | 6,236 | alesapin | Segfault in function cast. Doesn't connect with changes. | 2019-07-31T18:15:29 | 2019-07-31T18:15:29 | {} | MEMBER |
ClickHouse | ClickHouse | 515,713,519 | 6,187 | alexey-milovidov | Does the issue relate to using gcc 8.2 instead of gcc 8.3? | 2019-07-27T21:04:40 | 2019-07-27T21:04:40 | {} | MEMBER |
ClickHouse | ClickHouse | 515,803,987 | 6,200 | alexey-milovidov | Probably SYSTEM STOP MERGES will help (but it will also disable all merges). | 2019-07-28T23:26:08 | 2019-07-28T23:26:08 | {} | MEMBER |
ClickHouse | ClickHouse | 515,913,690 | 6,200 | simPod | Yup, seems to do the thing but also disables mutations which breaks the tests in other way.
Does it makes sense to have option in CH to disable TTL or should I find some workaround? | 2019-07-29T09:16:28 | 2019-07-29T09:16:28 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 516,587,211 | 6,228 | alexey-milovidov | Though it works perfectly for system.numbers.
But I have complex Merge table of MergeTree tables. | 2019-07-30T20:40:53 | 2019-07-30T20:40:53 | {} | MEMBER |
ClickHouse | ClickHouse | 516,638,868 | 6,228 | alexey-milovidov | It doesn't work if SAMPLE is specified. | 2019-07-30T23:52:56 | 2019-07-30T23:52:56 | {} | MEMBER |
ClickHouse | ClickHouse | 516,679,928 | 6,133 | alexey-milovidov | If you store IPv4 address in UInt32, `intHash64` will suffice,
otherwise use `cityHash64`. | 2019-07-31T03:31:52 | 2019-07-31T03:31:52 | {} | MEMBER |
ClickHouse | ClickHouse | 516,724,096 | 5,846 | Rostish | Good day collegues!
What news about the problem?
Got this bug on 19.11.3.11, nothing is help, only table and engine reacreation. | 2019-07-31T06:59:27 | 2019-07-31T06:59:27 | {} | NONE |
ClickHouse | ClickHouse | 516,988,163 | 6,224 | nikvas0 | This is an error in the MergeTreeIndexConditionSet.
```
ThinkPad-E570 :) insert into a values ('Jon','alive'),('Ramsey','rip');
INSERT INTO a VALUES
... | 2019-07-31T19:29:22 | 2019-07-31T20:16:08 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 516,631,497 | 6,231 | alexey-milovidov | No tests for this PR. | 2019-07-30T23:18:30 | 2019-07-30T23:18:30 | {} | MEMBER |
ClickHouse | ClickHouse | 516,807,936 | 5,042 | MahmoudGoda0 | Hi Guys,
Can I know when this fix will be available ?
This fix really urgent for my company to use CH as main DB engine. | 2019-07-31T11:22:21 | 2019-07-31T11:22:21 | {} | NONE |
ClickHouse | ClickHouse | 516,885,431 | 6,200 | alexey-milovidov | @CurtizJ Ok. | 2019-07-31T14:55:34 | 2019-07-31T14:55:34 | {} | MEMBER |
ClickHouse | ClickHouse | 517,005,713 | 6,236 | alesapin | Performance failures also not connected with changes. | 2019-07-31T20:22:24 | 2019-07-31T20:22:24 | {} | MEMBER |
ClickHouse | ClickHouse | 517,006,925 | 6,249 | 4ertus2 | #3145 | 2019-07-31T20:25:19 | 2019-07-31T20:25:19 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 515,715,449 | 6,124 | alexey-milovidov | > Note that the address that libunwind wants to read (and it think that it's the value of `rbp` register) corresponds to the value of function argument a few stack frames above:
>
This probably requires `-fasynchronous-unwind-tables` option. | 2019-07-27T21:40:34 | 2019-07-27T21:40:34 | {} | MEMBER |
ClickHouse | ClickHouse | 515,845,884 | 6,148 | zhang2014 | @DaisyFlower 谢谢你的贡献， 或许你可以在翻译一下[domain](https://clickhouse.yandex/docs/zh/data_types/domains/overview/)么？我们还没有它的翻译。这两个页面一起提交可能会更好一点 ：） | 2019-07-29T05:05:08 | 2019-07-29T05:05:08 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 516,240,395 | 5,670 | shangshujie365 | @alesapin , In old code, 'drop partition' is send to leader replica, can you give us some example why the old should be changed? Because our ch is a little old(19.5), and we want to add this patch to our ch release in case some bugs occur. | 2019-07-30T03:00:49 | 2019-07-30T03:00:49 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 516,631,346 | 6,231 | alexey-milovidov | The error exists from 19.3.3. Before this version there was also the case for unclean server shutdown but there was no issue that the server can't shutdown due to remaining queries. | 2019-07-30T23:17:47 | 2019-07-30T23:17:54 | {} | MEMBER |
ClickHouse | ClickHouse | 516,633,708 | 6,231 | alexey-milovidov | Queries like `SELECT count() FROM system.numbers` are correctly killed on shutdown.
I've also tested with the query `SELECT (SELECT count() FROM system.numbers) AS x` - I've used the fact that evaluation of scalar subqueries does not send progress packets. | 2019-07-30T23:28:26 | 2019-07-30T23:28:26 | {} | MEMBER |
ClickHouse | ClickHouse | 516,872,789 | 6,226 | akuzm | Test failures look unrelated to the patch. | 2019-07-31T14:27:10 | 2019-07-31T14:27:10 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 516,964,845 | 6,191 | CurtizJ | Yes, tests failures are not related to your changes.
It's hard to check performance with TTL using CI tests. I think it's reasonable to write unit tests for TTLMergeSelector, because TTL tests use `OPTIMIZE ... FINAL` and selector is stayed almost unused. But it's not exactly part of this task, because test is nee... | 2019-07-31T18:24:03 | 2019-07-31T18:24:03 | {} | MEMBER |
ClickHouse | ClickHouse | 515,713,154 | 6,187 | alexey-milovidov | Does not reproduce on 19.11.3 stable. | 2019-07-27T20:58:02 | 2019-07-27T20:58:02 | {} | MEMBER |
ClickHouse | ClickHouse | 515,792,668 | 6,199 | simPod | Thanks a lot for clarification! | 2019-07-28T20:28:24 | 2019-07-28T20:28:24 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 516,053,931 | 6,139 | kreuzerkrieg | You are walking thin ice. Start with `folly` `symbolizer` and you will find yourself rewriting CH threading model to use `folly` `fibers` :) | 2019-07-29T16:02:52 | 2019-07-29T16:02:52 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 516,258,000 | 6,139 | kreuzerkrieg | Looks like just what the doctor ordered!
How about time it takes? Something one can live with that?
| 2019-07-30T04:38:46 | 2019-07-30T04:38:46 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 516,679,471 | 1,344 | alexey-milovidov | Some optimizations still remain: exponential growth of read window from some small value. | 2019-07-31T03:29:37 | 2019-07-31T03:29:37 | {} | MEMBER |
ClickHouse | ClickHouse | 516,699,805 | 6,139 | kreuzerkrieg | Thanks! | 2019-07-31T05:19:42 | 2019-07-31T05:19:42 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 516,842,607 | 6,135 | vitlibar | Can you try the following:
```
cat Test.txt | curl 'http://localhost:8123/?query=INSERT%20INTO%20default.proto%20FORMAT%20Protobuf%20SETTINGS%20format_schema%3D%27test.proto%3ATest%27' --data-binary @-
```
Does it work? | 2019-07-31T13:12:16 | 2019-07-31T13:14:42 | {} | MEMBER |
ClickHouse | ClickHouse | 516,932,363 | 6,191 | svladykin | I reverted it because there were other tests failures. I was thinking that this fix caused them.
Maybe they happened because of the CI problems as well?
I do not see the previous CI run results, I'll give another try to push the fix.
As for testing, do you have any ideas how to test this stuff better? The new fla... | 2019-07-31T16:55:04 | 2019-07-31T16:55:04 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 515,970,817 | 5,880 | avasiliev | Does server send FIN packet to the client, or connection is closed forcefully? If former, is latter possible? | 2019-07-29T12:29:40 | 2019-07-29T12:29:40 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 516,042,418 | 6,139 | alexey-milovidov | There is quite a good code with appropriate license:
https://github.com/facebook/folly/blob/master/folly/experimental/symbolizer/Dwarf.h
https://github.com/facebook/folly/blob/master/folly/experimental/symbolizer/Dwarf.cpp
Maybe we can reuse it. | 2019-07-29T15:33:27 | 2019-07-29T15:33:38 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 516,679,339 | 1,344 | alexey-milovidov | Fixed in master. | 2019-07-31T03:28:57 | 2019-07-31T03:28:57 | {} | MEMBER |
ClickHouse | ClickHouse | 516,842,226 | 6,213 | filimonov | In ClickHouse materialized view behaves more like BEFORE INSERT TRIGGER, each time processing new block arrived with insert.
So that is quite natural limitation as inserts to 2 different table will come asynchronously and you usually expect to see in JOINs whole table not only newly arrived blocks. | 2019-07-31T13:11:11 | 2019-07-31T13:11:11 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 516,932,051 | 6,227 | den-crane | I want to remove CODEC from the column and use default compression from config.xml
It just feature request. I don't really need this right now. | 2019-07-31T16:54:09 | 2019-07-31T16:55:04 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 516,314,816 | 3,950 | kaijianding | echo this.
I found the same error when I set insert_quorum=2, these 2 exceptions are raised in version: 19.7.3.1:
Another quorum insert has been already started
Quorum for previous write has not been satisfied yet | 2019-07-30T08:14:01 | 2019-07-30T08:14:01 | {} | NONE |
ClickHouse | ClickHouse | 516,609,711 | 6,219 | alexey-milovidov | Perfect.
FYI. You can also test for rows read with `max_rows_to_read` setting.
It can be done using only `.sql` tests. | 2019-07-30T21:50:45 | 2019-07-30T21:50:45 | {} | MEMBER |
ClickHouse | ClickHouse | 516,614,025 | 6,217 | alexey-milovidov | It was implemented in deficient way from the beginning. Need to apply real parser to parse table declaration, similar to CREATE query. | 2019-07-30T22:05:27 | 2019-07-30T22:05:27 | {} | MEMBER |
ClickHouse | ClickHouse | 515,664,041 | 6,166 | kreuzerkrieg | Ok, I will PR during the upcoming week.
| 2019-07-27T08:03:42 | 2019-07-27T08:03:42 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 515,913,632 | 6,202 | blinkov | cc @zhang2014 @sundy-li | 2019-07-29T09:16:18 | 2019-07-29T09:16:18 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 516,283,835 | 6,214 | den-crane | ```
SELECT intDivOrZero( CAST(-1000, 'Int64') , CAST(1000, 'UInt64') )
--
18446744073709550
SELECT intDivOrZero( CAST(-1000, 'Int64') , CAST(1000, 'Int64') )
--
-1
``` | 2019-07-30T06:33:55 | 2019-07-30T06:34:06 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 517,019,335 | 6,248 | den-crane | And can you check a result of
select id, toFloat64(id) = 439385272.0 as compare_indicator
from myTable
where identity(id) = '439385272' | 2019-07-31T21:00:53 | 2019-07-31T21:00:53 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 500,900,567 | 5,585 | den-crane | @dgrr are you trolling ? REALLY ?
your table HAS to have **37** columns
```
CREATE TABLE `x28` (
`field1` Float64,
`field2` Float64,
`field3` Float64,
`field4` String,
`field5` String,
`field6` Int64,
`field7` String,
`field8` String,
`field9` String,
`field10` String,
`field11` String,
`field12` Int... | 2019-06-11T15:42:33 | 2019-06-11T15:42:33 | {} | CONTRIBUTOR |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.