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 | 497,550,046 | 5,479 | yingfeng | It's pretty easy to reproduce:
````
CREATE TABLE trips ( \
id UInt64, \
vendor_id String, \
pickup_datetime DateTime \
) engine=MergeTree() \
PARTITION BY toYYYYMM(pickup_datetime) \
ORDER BY (pickup_datetime, id);
````
After inserting several records, then perform the query:
````
SELECT count(*... | 2019-05-31T02:15:48 | 2019-05-31T02:15:48 | {} | NONE |
ClickHouse | ClickHouse | 497,565,928 | 5,478 | alexey-milovidov | If you have MySQL protocol released in any previous version, you must backport this change, because otherwise it is totally unusable. | 2019-05-31T03:54:25 | 2019-05-31T03:54:25 | {} | MEMBER |
ClickHouse | ClickHouse | 497,710,361 | 4,248 | abyss7 | Look's like that's it. The new v19.8 version with set "intermediate_commit=0" on Kafka table drastically lowers the number of commits and the initial issue can't be reproduced. Will wait for confirmation when someone tries this version. | 2019-05-31T13:33:53 | 2019-05-31T13:33:53 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 497,974,971 | 5,473 | lenstr | Wow, that was fast, thanks! | 2019-06-01T20:23:26 | 2019-06-01T20:23:26 | {} | NONE |
ClickHouse | ClickHouse | 498,066,851 | 5,514 | den-crane | > order by date desc
Pushdown disabled for "order by" due https://github.com/yandex/ClickHouse/issues/3885 | 2019-06-02T21:27:59 | 2019-06-02T21:27:59 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 498,092,475 | 5,514 | peaksnail | I find that when I create view with sql "select * from bill FINAL", the predicate expression pushdown disabled,But it works fine with sql "select no, argMax(fee, date) as fee, max(date) as date from bill group by no" | 2019-06-03T02:26:07 | 2019-06-03T02:26:07 | {} | NONE |
ClickHouse | ClickHouse | 495,760,536 | 3,174 | simPod | `select * from system.mutations where is_done = 0;` | 2019-05-24T19:18:53 | 2019-06-03T13:59:14 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 498,296,887 | 4,308 | abyss7 | Closing for now. | 2019-06-03T15:07:49 | 2019-06-03T15:07:49 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 498,298,517 | 4,356 | abyss7 | Closing for now. | 2019-06-03T15:11:30 | 2019-06-03T15:11:30 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 498,656,924 | 146 | perez1987 | @blinkov Is it possible to rename fields between source-destination tables between tables using clickhouse-copier? | 2019-06-04T12:48:17 | 2019-06-04T12:48:17 | {} | NONE |
ClickHouse | ClickHouse | 498,850,982 | 3,271 | den-crane | @iameugenejo and what is an engine of logs_local_20181212 ? | 2019-06-04T21:29:49 | 2019-06-04T21:29:49 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 499,089,185 | 5,512 | lzy305 | Good idea, Thanks | 2019-06-05T13:45:22 | 2019-06-05T13:45:22 | {} | NONE |
ClickHouse | ClickHouse | 498,020,291 | 5,511 | proller | You should fetch submodules before running cmake: `git submodule update --init --recursive` | 2019-06-02T10:55:09 | 2019-06-02T10:55:09 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 498,582,109 | 5,527 | filimonov | About implementation - if it will be done on log level only - then query_log (for example) or error messages will still contain the unsafe strings. Probably for error messages, it's acceptable (?), but for query_log it sounds like it should be done too.
So it should be either filter added in two places, either spec... | 2019-06-04T08:49:24 | 2019-06-04T08:53:00 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 498,673,788 | 5,417 | amosbird | Thanks for reviewing this! The formating and style adjustments are to settle down the changes introduced from my previous FixedHashMap PR. I agree those should be separate commits, however this feature is a long over due and has been refactored multiple times. The original commits aren't reviewable at all thus I've squ... | 2019-06-04T13:34:02 | 2019-06-04T13:40:20 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 498,715,109 | 5,530 | den-crane | > Thank you, @den-crane it works like a charm!
> It was not clear from the documentation.
> Is HTTP used only for nodes communication? Data replication is going by TCP?
It's used by Replicated* Engines only.
A replica announces itself (writes into some ZK znode) and uses this interserver_http_host
Other replicas... | 2019-06-04T15:13:34 | 2019-06-04T15:14:54 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 497,542,355 | 5,472 | mcarbonneaux | It's work based on date column if y enderstand!
Where not more performante if it can be base on partition creation date ? | 2019-05-31T01:29:32 | 2019-05-31T01:41:04 | {} | NONE |
ClickHouse | ClickHouse | 498,009,585 | 5,511 | liaorui | Can anyone help? thanks. | 2019-06-02T08:13:04 | 2019-06-02T08:13:04 | {} | NONE |
ClickHouse | ClickHouse | 498,025,142 | 5,511 | proller | This bug already fixed in master : https://github.com/yandex/ClickHouse/commit/b9385f9620d26808279b330e7bea8b03f48adbc6
without this patch you can try build only clickhouse binaries without tests:
`ninja clickhouse-bundle` | 2019-06-02T12:05:04 | 2019-06-02T12:05:57 | {
"+1": 2
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 498,202,880 | 5,384 | manish53 | > @manish53 there are still some conflicts, are you going to fix them?
I didn't modified those files which are listed as conflicts, so can't resolved it. | 2019-06-03T10:29:41 | 2019-06-03T10:29:41 | {} | NONE |
ClickHouse | ClickHouse | 498,290,253 | 5,513 | proller | Yes, please make pull request - it will be automatically tested. | 2019-06-03T14:52:07 | 2019-06-03T14:52:07 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 498,376,241 | 4,631 | hasnat | > Can you tell if the initial number of consumers (`kafka_num_consumers`) exceeds at any point of time the total number of partitions for all consumed topics?
@abyss7
No before it was setup as 3 consumers for a 3 partitions topic
We have another 8 consumers table on 8 partitions topic which never gave us any ... | 2019-06-03T18:39:47 | 2019-06-03T18:39:58 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 498,726,831 | 5,514 | den-crane | My idea was that you don't need view because just need to add FINAL. So view is unnecessary. | 2019-06-04T15:41:00 | 2019-06-04T15:41:00 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 498,739,603 | 4,062 | den-crane | @simPod check that you don't have parts with a size bigger than max_bytes_to_merge_at_max_space_in_pool
```
SELECT
table,
partition,
name,
round(((bytes_on_disk / 1024) / 1024) / 1024) AS size
FROM system.parts
WHERE bytes_on_disk >
(
SELECT toInt64(value)
FROM system.merge_tree_set... | 2019-06-04T16:12:38 | 2019-06-04T16:13:27 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 497,566,852 | 5,477 | alexey-milovidov | dataparts_500 - bad test @proller | 2019-05-31T04:01:08 | 2019-05-31T04:01:08 | {} | MEMBER |
ClickHouse | ClickHouse | 497,573,413 | 5,294 | alexey-milovidov | > No radix sort in sortBlock
Why? | 2019-05-31T04:44:35 | 2019-05-31T04:44:35 | {} | MEMBER |
ClickHouse | ClickHouse | 497,906,802 | 5,507 | ramazanpolat | Deleting because further improvement | 2019-06-01T02:59:11 | 2019-06-01T02:59:11 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 498,080,575 | 5,294 | alexey-milovidov | ? | 2019-06-03T00:50:04 | 2019-06-03T00:50:04 | {} | MEMBER |
ClickHouse | ClickHouse | 498,113,631 | 5,479 | yingfeng | Following query can also lead to same segmentation fault:
```
SELECT count(*)
FROM trips
WHERE bitmapHasAny(
bitmapBuild(CAST([1,2,3], 'Array(UInt64)'))
, bitmapBuild([id]))
``` | 2019-06-03T05:02:22 | 2019-06-03T05:02:22 | {} | NONE |
ClickHouse | ClickHouse | 498,195,252 | 4,002 | blinkov | At the moment "official" JDBC driver is supported by Yandex.Metrica team who develops it and uses it in production. This is the main reason for it's status, performance has nothing to do with it.
Other users are free to choose whatever implementation better fits their needs, this might include performance of select/... | 2019-06-03T10:05:16 | 2019-06-03T10:05:16 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 498,254,585 | 4,538 | 4ertus2 | It's error-prone to place both z-func and KeyConditions changes in one PR. It's hard to review it well. There's too much changes. | 2019-06-03T13:23:18 | 2019-06-03T13:23:18 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 498,825,853 | 3,271 | iameugenejo | @den-crane
* select timezone();
```UTC```
* select toTypeName(timestamp) from sample.log_stats_20181212 limit 1;
```DateTime```
* select toTypeName(ingest_timestamp) from sample.logs_local_20181212 limit 1;
```DateTime```
The clickhouse version is now `19.6.2.11`, and it's still showing the discrepancy ... | 2019-06-04T20:15:08 | 2019-06-04T20:15:21 | {} | NONE |
ClickHouse | ClickHouse | 499,009,103 | 5,383 | alesapin | This solution a little hacky, but doesn't disable DNS cache https://github.com/yandex/ClickHouse/pull/5534. | 2019-06-05T09:27:34 | 2019-06-05T09:27:34 | {} | MEMBER |
ClickHouse | ClickHouse | 497,564,942 | 5,495 | alexey-milovidov | To perform valgrind check, you should disable jemalloc at the build time.
(There is something incompatible by default.)
Also you will have to add some defines for RE2 to run successfully with valgrind.
Read this article about why: https://research.swtch.com/sparse
And also you will have to suppress false positi... | 2019-05-31T03:47:33 | 2019-05-31T03:47:33 | {} | MEMBER |
ClickHouse | ClickHouse | 497,716,860 | 5,208 | abyss7 | Will close for now. | 2019-05-31T13:53:21 | 2019-05-31T13:53:21 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 497,751,828 | 5,479 | yingfeng | Some extra information after compiling using debug options:
```
Thread 51 "AsyncBlockInput" received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x7fff5e9e1700 (LWP 19657)]
0x000000000bbc1a98 in DB::RoaringBitmapWithSmallSet<unsigned long, (unsigned char)32>::isSmall (this=0x0)
at ../dbms/src/Ag... | 2019-05-31T15:28:33 | 2019-05-31T15:28:33 | {} | NONE |
ClickHouse | ClickHouse | 498,552,439 | 5,519 | Chocorean | Here's a csv file which leads to a segfault. I've also joined my `users.xml` and `config.xml` files.
[files.zip](https://github.com/yandex/ClickHouse/files/3250987/files.zip)
| 2019-06-04T07:18:19 | 2019-06-04T07:18:19 | {
"+1": 1
} | NONE |
ClickHouse | ClickHouse | 497,572,316 | 5,294 | alexey-milovidov | ? | 2019-05-31T04:37:18 | 2019-05-31T04:37:18 | {} | MEMBER |
ClickHouse | ClickHouse | 497,711,057 | 4,248 | ajardan | As mentioned we cannot reproduce the issue on 19.8.1 revision 54420
Thanks a lot for your help @abyss7 ! | 2019-05-31T13:36:08 | 2019-05-31T13:36:08 | {} | NONE |
ClickHouse | ClickHouse | 498,373,055 | 5,417 | akuzm | I see some formatting changes in the lines that are not otherwise touched by this pull request. This leads to some problems:
- it complicates the review, because the reviewer has to filter out these changes,
- it makes `git blame` less useful,
- it might introduce unnecessary conflicts with other pull requests or ol... | 2019-06-03T18:30:22 | 2019-06-03T18:30:22 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 498,589,874 | 5,528 | andyyzh | Similar logic exists in FunctionBitmap and needs to be modified, thanks. | 2019-06-04T09:11:47 | 2019-06-04T09:11:47 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 498,730,516 | 5,512 | den-crane | ```
SELECT exp2(arraySum(x -> log2(x), [1, 4, 6, 8]))
┌─exp2(arraySum(lambda(tuple(x), log2(x)), [1, 4, 6, 8]))─┐
│ 192 │
└─────────────────────────────────────────────────────────┘
``` | 2019-06-04T15:49:53 | 2019-06-04T15:52:17 | {
"+1": 2
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 498,940,949 | 5,430 | alexey-milovidov | Unfinished. | 2019-06-05T05:11:28 | 2019-06-05T05:11:28 | {} | MEMBER |
ClickHouse | ClickHouse | 497,698,282 | 5,473 | 4ertus2 | #5503 | 2019-05-31T12:54:56 | 2019-05-31T12:54:56 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 497,839,517 | 4,247 | laplab | Keep in mind: https://github.com/gperftools/gperftools/wiki/gperftools'-stacktrace-capturing-methods-and-their-issues | 2019-05-31T19:51:27 | 2019-05-31T19:51:27 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 497,858,440 | 2,120 | chhaya-sharma | Hi @lamber-ken,
Have you able to solve this issue? I am facing the same issue. Please revert. | 2019-05-31T20:59:47 | 2019-05-31T20:59:47 | {} | NONE |
ClickHouse | ClickHouse | 497,908,983 | 5,479 | yingfeng |
Corresponding codes from `executeIntType` in class `FunctionBitmapCardinality` has not considered situations where arguments come from scalar subquery:
```
for (size_t i = 0; i < input_rows_count; ++i)
{
const AggregateFunctionGroupBitmapData<T> & bd1
= *reinterpret_cast<const AggregateFunctionGroupB... | 2019-06-01T03:40:00 | 2019-06-01T03:40:00 | {} | NONE |
ClickHouse | ClickHouse | 497,719,408 | 5,498 | proller | boost has to be updated as a whole, will do | 2019-05-31T14:00:55 | 2019-06-01T18:21:15 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 498,100,800 | 5,479 | andyyzh | ``` sql
CREATE TABLE trips ( \
id UInt64, \
vendor_id String, \
pickup_datetime DateTime \
) engine=MergeTree() \
PARTITION BY toYYYYMM(pickup_datetime) \
ORDER BY (pickup_datetime, id);
INSERT INTO trips SELECT number, 'nj', '2019-01-01 00:00:00' FROM numbers(1,5);
-- fine
SELECT count(*)
FROM tri... | 2019-06-03T03:27:53 | 2019-06-03T03:27:53 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 498,213,522 | 5,519 | Chocorean | This is after [#5379](https://github.com/yandex/ClickHouse/issues/5379): I had reinstall, downgraded and tested multiple inserts. | 2019-06-03T11:08:10 | 2019-06-03T11:08:10 | {} | NONE |
ClickHouse | ClickHouse | 495,617,797 | 3,174 | simPod | @WeiGangqiang also check for mutations `is_done=0` | 2019-05-24T13:03:42 | 2019-06-03T13:59:19 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 498,684,661 | 5,530 | den-crane | Replicated* Engines does not use `<remote_servers> `
/etc/clickhouse-server/config.xml
```
<!-- Hostname that is used by other replicas to request this server.
If not specified, than it is determined analoguous to 'hostname -f' command.
This setting could be used to switch replication to an... | 2019-06-04T14:01:42 | 2019-06-04T14:01:42 | {
"+1": 3
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 498,900,641 | 5,528 | yuzhichang | @alexey-milovidov I don't think the failing tests are caused by my change. On my machine, they are failing without my change. What's the test setup on github? | 2019-06-05T01:14:38 | 2019-06-05T01:14:38 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 498,043,813 | 5,514 | peaksnail | there has any(date) as date in the view and I use date='xxx' as condition , maybe the any(date) is found in WHERE? | 2019-06-02T16:05:04 | 2019-06-02T16:05:04 | {} | NONE |
ClickHouse | ClickHouse | 498,237,741 | 4,456 | k0st1an | On version 19.7.3.9 (official build) with same symptoms | 2019-06-03T12:34:25 | 2019-06-03T12:34:25 | {} | NONE |
ClickHouse | ClickHouse | 498,322,570 | 4,062 | simPod | @vasyaabr How did you find it in logs? My mutations hang on `parts_to_do` 1 too but so far I have not found a way to check whether it's doing something or not. | 2019-06-03T16:09:41 | 2019-06-03T16:09:41 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 498,337,103 | 4,631 | abyss7 | Can you tell if the initial number of consumers (`kafka_num_consumers`) exceeds at any point of time the total number of partitions for all consumed topics? | 2019-06-03T16:49:50 | 2019-06-03T16:49:50 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 498,530,092 | 5,384 | blinkov | @manish53 `git pull git@github.com:yandex/ClickHouse.git master` | 2019-06-04T05:46:21 | 2019-06-04T05:46:21 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 498,685,300 | 5,413 | blinkov | also there are some merge issues | 2019-06-04T14:03:08 | 2019-06-04T14:03:08 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 498,695,762 | 5,530 | wildermesser | Thank you, @den-crane it works like a charm!
It was not clear from the documentation.
Is HTTP used only for nodes communication? Data replication is going by TCP? | 2019-06-04T14:27:43 | 2019-06-04T14:27:43 | {} | NONE |
ClickHouse | ClickHouse | 498,034,174 | 5,511 | liaorui | @proller done! thanks. | 2019-06-02T14:04:04 | 2019-06-02T14:04:04 | {} | NONE |
ClickHouse | ClickHouse | 498,067,308 | 5,514 | den-crane | ```
select no, any(fee) as fee,any(date) as date from (
select * from bill order by date desc
) group by no
```
it is the same as `select * from bill FINAL`
another option is `select no, argMax(fee, date) as fee, max(date) as date from bill group by no` | 2019-06-02T21:34:18 | 2019-06-02T21:34:18 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 498,342,310 | 5,523 | orivej | Please test the build. | 2019-06-03T17:04:35 | 2019-06-03T17:04:35 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 498,354,348 | 5,506 | alesapin | Test would be fixed in master after docker push. | 2019-06-03T17:37:52 | 2019-06-03T17:37:52 | {} | MEMBER |
ClickHouse | ClickHouse | 498,377,832 | 5,384 | blinkov | @manish53 try to merge with upstream master and push again | 2019-06-03T18:44:12 | 2019-06-03T18:44:12 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 497,819,735 | 5,429 | vitlibar | > Any thoughts on this?
Looks good.
I've simplified your solution slightly and I'll merge it after passing automatic checks. | 2019-05-31T18:43:52 | 2019-05-31T18:44:33 | {} | MEMBER |
ClickHouse | ClickHouse | 498,213,269 | 5,513 | citrin | This patch fixes build for me:
https://termbin.com/4tlk
Shall I make a PR? | 2019-06-03T11:07:13 | 2019-06-03T11:07:13 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 498,347,302 | 5,417 | akuzm | @amosbird I'm seeing a ubsan failure on functional tests, could you look into it?
```
../dbms/src/Common/Arena.h:245:21: runtime error: null pointer passed as argument 2, which is declared to never be null
/usr/include/string.h:43:28: note: nonnull attribute specified here
#0 0x16ad4f82 in DB::Arena::insert(... | 2019-06-03T17:18:17 | 2019-06-03T17:18:17 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 498,481,542 | 5,479 | yingfeng | @vitlibar Could you please help with this issue since this feature is important to satisfy our underlying requirements. Thank you~ | 2019-06-04T01:13:27 | 2019-06-04T01:13:27 | {} | NONE |
ClickHouse | ClickHouse | 498,641,391 | 5,200 | BayoNet | @hczhcz
Hello :-).
I'm trying to add documentation for these functions, and I need some help. I have made a rough draft in my fork for these functions description.
Would you please to help me to fill it with valuable information? I am sure you can tell the users much more than I did. For example, I don't unde... | 2019-06-04T11:59:40 | 2019-06-04T12:01:21 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 490,709,622 | 5,220 | andyyzh | run:
``` sql
select bitmapAndCardinality(bitmapBuild(cast([3,19,47] as Array(UInt32))),
bitmapBuild(cast([19,7] as Array(UInt32))))
from test_bitmap
```
is fine.
I guess is between column and expressions to calculate problem.
Let me try to solve it. | 2019-05-09T01:33:26 | 2019-05-31T12:25:36 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 497,909,352 | 4,207 | yingfeng | Corruption happened if arguments come from scalar subqueries, seen from [issue 5479](https://github.com/yandex/ClickHouse/issues/5479)
| 2019-06-01T03:47:42 | 2019-06-01T03:47:42 | {} | NONE |
ClickHouse | ClickHouse | 497,929,518 | 5,479 | andyyzh | I'm trying to solve this problem.
Exec below sql, it's works fine.
``` sql
CREATE TABLE trips ( \
id UInt64, \
vendor_id String, \
pickup_datetime DateTime \
) engine=MergeTree() \
PARTITION BY toYYYYMM(pickup_datetime) \
ORDER BY (pickup_datetime, id);
INSERT INTO trips SELECT number, 'nj', '2019-... | 2019-06-01T09:37:40 | 2019-06-01T09:37:40 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 497,930,784 | 5,479 | yingfeng | Hi, Andy,
You are right, your SQL works fine, while if I change it into
```
SELECT count(*)
FROM trips
WHERE bitmapHasAny(
(
SELECT groupBitmapState(id)
FROM trips
), bitmapBuild(CAST([id], 'Array(UInt64)')));
```
The corruption happens. | 2019-06-01T09:56:37 | 2019-06-01T09:56:37 | {} | NONE |
ClickHouse | ClickHouse | 498,081,700 | 5,042 | alexey-milovidov | Stll unfinished. The tests don't cover all the cases (order by desc), and the code does not work correctly. | 2019-06-03T01:01:56 | 2019-06-03T01:01:56 | {} | MEMBER |
ClickHouse | ClickHouse | 498,160,342 | 5,497 | alesapin | Ok, but also we need to add Cap'n'proto, Protobuf and Parquet. | 2019-06-03T08:21:30 | 2019-06-03T08:21:30 | {} | MEMBER |
ClickHouse | ClickHouse | 498,907,900 | 5,528 | yuzhichang | @andyyzh FunctionBitmap::executeBitmapData is bit different with FunctionBitmapCardinality::executeIntType. executeBitmapData() uses the first bitmap as the result. So if the first bitmap is ColumnConst, the server will segment fault.
```
-- error, server segment fault
SELECT count(*) FROM trips WHERE bitmapCardinal... | 2019-06-05T01:52:31 | 2019-06-05T02:52:35 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 499,006,480 | 4,736 | wawaka | can anyone confirm that the problem is fixed in 19.7? My test shows that data is still being lost in 19.7 | 2019-06-05T09:19:20 | 2019-06-05T09:19:20 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 497,909,220 | 5,279 | yingfeng | Corruption happened if arguments come from scalar subqueries, seen from [issue 5479](https://github.com/yandex/ClickHouse/issues/5479) | 2019-06-01T03:44:38 | 2019-06-01T03:44:38 | {} | NONE |
ClickHouse | ClickHouse | 498,023,550 | 5,511 | liaorui | @proller , after `git submodule update --init --recursive`, there is another error:
`../dbms/src/IO/tests/gtest_aio_seek_back_after_eof.cpp: In member function 'virtual void ReadBufferAIOTest_TestReadAfterAIO_Test::TestBody()':
../dbms/src/IO/tests/gtest_aio_seek_back_after_eof.cpp:42:21: error: 'DB' is not a name... | 2019-06-02T11:44:21 | 2019-06-02T11:44:21 | {} | NONE |
ClickHouse | ClickHouse | 498,087,926 | 5,514 | peaksnail | @den-crane thanks
If I create table bill (no String, fee Int64 , updateDate Date ,createdate Date) engine = ReplacingMergeTree(createdate,(no,createdate),8192) with adding a updateDate field, but the updateDate field is not set in the primary key.
and the data in table will be the same except updateDate field,
... | 2019-06-03T01:54:42 | 2019-06-03T01:54:42 | {} | NONE |
ClickHouse | ClickHouse | 498,042,151 | 5,509 | blinkov | cc @zhang2014 @sundy-li | 2019-06-02T15:44:11 | 2019-06-02T15:44:11 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 498,106,874 | 5,479 | yingfeng | It seems the implementation of `executeIntType` in `FunctionBitmap.h` is quite different from corresponding `executeImpl` in `hasAllAny.h`, which has similar query semantics with bitmap functions.
```
void executeIntType(
Block & block, const ColumnNumbers & arguments, size_t input_rows_count, typename Column... | 2019-06-03T04:13:39 | 2019-06-03T04:13:39 | {} | NONE |
ClickHouse | ClickHouse | 498,200,830 | 5,196 | blinkov | @zhang2014 any updates? | 2019-06-03T10:22:40 | 2019-06-03T10:22:40 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 499,025,226 | 5,196 | zhang2014 | @blinkov The translation has been completed and can be reviewed @sundy-li
I did not translate the content of the `Working with Yandex.Metrica Dictionaries` document | 2019-06-05T10:16:24 | 2019-06-05T10:42:13 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 497,599,429 | 5,485 | svladykin | @den-crane Thanks for the explanation, this makes total sense.
So, if I have a table with web page visits like `visits(visitor, date)` and want to know how many total unique users and new unique users came each date then it should look like the following?
```sql
select date, total_uniq_visitors, runningDifferenceSt... | 2019-05-31T07:00:16 | 2019-05-31T07:00:16 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 498,242,510 | 5,519 | alesapin | Can you attach `.csv` file to reproduce error? | 2019-06-03T12:49:29 | 2019-06-03T12:49:29 | {} | MEMBER |
ClickHouse | ClickHouse | 498,939,347 | 5,528 | alexey-milovidov | > I don't think the failing tests are caused by my change. On my machine, they are failing without my change. What's the test setup on github?
It was temporarily issue with our CI environment. | 2019-06-05T05:01:44 | 2019-06-05T05:01:44 | {} | MEMBER |
ClickHouse | ClickHouse | 497,567,378 | 5,445 | alexey-milovidov | ? | 2019-05-31T04:04:50 | 2019-05-31T04:04:50 | {} | MEMBER |
ClickHouse | ClickHouse | 497,792,534 | 5,500 | den-crane | You can reduce number of the merges using Buffer tables settings (make larger batches).
You can set smaller max_bytes_to_merge_at_max_space_in_pool for these plain tables and reduce number of final merges. https://github.com/yandex/ClickHouse/issues/2607
You can set weaker compression for the first level (small) ... | 2019-05-31T17:23:06 | 2019-05-31T17:25:24 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 496,394,121 | 4,631 | hasnat | try adding following config to your kafka brokers `group.initial.rebalance.delay.ms=0` This would possibly sort out errors like
```
[thrd:GroupCoordinator]: GroupCoordinator/2: Timed out LeaveGroupRequest in flight (after 5940ms, timeout #0): possibly held back by preceeding blocking JoinGroupRequest with timeout in ... | 2019-05-28T07:14:06 | 2019-06-01T07:09:19 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 498,817,243 | 3,271 | den-crane | @iameugenejo can you show the output
select timezone();
select toTypeName(timestamp) from sample.log_stats_20181212 limit 1;
select toTypeName(ingest_timestamp) from sample.logs_local_20181212 limit 1; | 2019-06-04T19:50:35 | 2019-06-04T19:50:35 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 497,928,909 | 5,429 | vitlibar | > I guess that this will resolve table overlapping, just ignore my question
Yes, we want to run as many tests as possible in parallel. | 2019-06-01T09:28:15 | 2019-06-01T09:52:21 | {} | MEMBER |
ClickHouse | ClickHouse | 498,188,835 | 5,384 | blinkov | @manish53 there are still some conflicts, are you going to fix them? | 2019-06-03T09:46:15 | 2019-06-03T09:46:15 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 498,346,084 | 5,417 | akuzm | > @akuzm akuzm requested a review from alexey-milovidov an hour ago
Oops, disregard this, must've pressed something by accident. | 2019-06-03T17:14:56 | 2019-06-03T17:14:56 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 498,523,931 | 5,384 | manish53 | > @manish53 try to merge with upstream master and push again
Should i merge those files with alesapin-master? I am new in github so please ellaborate more what should i do to make this PR usefull. | 2019-06-04T05:13:02 | 2019-06-04T05:13:02 | {} | NONE |
ClickHouse | ClickHouse | 498,547,854 | 5,526 | hodgesrm | Erroneous entry caused by Github problem. | 2019-06-04T07:02:04 | 2019-06-04T07:02:04 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 498,865,422 | 3,271 | iameugenejo | @den-crane It's MergeTree | 2019-06-04T22:22:24 | 2019-06-04T22:22:24 | {} | NONE |
ClickHouse | ClickHouse | 498,985,113 | 5,464 | alesapin | We will return it in two next releases. | 2019-06-05T08:16:13 | 2019-06-05T08:16:13 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 499,132,727 | 5,534 | alesapin | Performance fail is not connected with changes. | 2019-06-05T15:27:54 | 2019-06-05T15:28:14 | {} | MEMBER |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.