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 | 437,167,000 | 3,499 | den-crane | fixed in 18.14.13 ?
Connected to ClickHouse server version 18.14.13 revision 54409.
SELECT 1 FROM xxxx WHERE row LIKE '%test:0%'
0 rows in set. Elapsed: 0.697 sec. Processed 100.00 thousand rows, 4.69 GB (143.42 thousand rows/s., 6.73 GB/s.) | 2018-11-08T21:44:22 | 2018-11-08T21:44:22 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 437,458,293 | 2,451 | ryn9 | Any better way to get developers attention on this matter? | 2018-11-09T18:51:25 | 2018-11-09T18:51:25 | {} | NONE |
ClickHouse | ClickHouse | 438,117,656 | 3,574 | Livenux | * OS: Ubuntu 16.04.5
* Clickhouse-client version: ClickHouse client version 18.14.13.
* Installation method: apt | 2018-11-13T03:19:00 | 2018-11-13T03:19:00 | {} | NONE |
ClickHouse | ClickHouse | 438,184,838 | 3,576 | Jessica4j | SELECT * FROM A WHERE A.COL IS NOT NULL
| 2018-11-13T08:49:58 | 2018-11-13T08:49:58 | {} | NONE |
ClickHouse | ClickHouse | 438,285,082 | 3,572 | abyss7 | You're right - it's a regression. Thanks for reporting. | 2018-11-13T14:28:37 | 2018-11-13T14:29:04 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 438,506,315 | 3,568 | champtar | Thanks @alexey-milovidov
Do you plan to cut a release this week ? | 2018-11-14T01:43:31 | 2018-11-14T01:43:31 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 438,636,081 | 3,507 | TimonKK | I have a payment end date in the analyst. And it can be more than 2105th year for some "eternal" tariffs. I did not come up with this, but I need to build reports on this parameter.
Thanks for advice about Int64, I will try it | 2018-11-14T11:50:42 | 2018-11-14T11:50:42 | {} | NONE |
ClickHouse | ClickHouse | 439,044,130 | 2,408 | SkeLLLa | I don't know if it still relevant to you, but as far as I can see both those clients for node.js that listed on official site have very poor quality. Also they use http interface that's not very fast. So as a result all performance that clickhouse has is lost in client libraries. | 2018-11-15T13:43:28 | 2018-11-15T13:43:28 | {} | NONE |
ClickHouse | ClickHouse | 436,999,342 | 3,536 | ztlpn | But `clearOldTemporaryDirectories` shouldn't delete temporary directories that have been touched during the last 24 hours. I guess you have changed (set to 0) the `temporary_directories_lifetime` MergeTree setting? | 2018-11-08T13:48:19 | 2018-11-08T13:48:19 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 437,035,125 | 1,655 | lamberken | hi, @alexey-milovidov @ztlpn, I also think `dynamic column` is useful for metric system.
Is there a plan to add the feature ?
| 2018-11-08T15:25:34 | 2018-11-08T15:25:34 | {
"+1": 4
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 437,639,567 | 3,561 | volfco | If I change my zookeeper task to the following (these are the same nodes as above):
```
<shard>
<weight>1</weight>
<replica>
<host>172.29.3.17</host><!-- aus-clickhouse05 -->
<port>9000</port>
</replica>
... | 2018-11-11T02:59:01 | 2018-11-11T04:09:36 | {} | NONE |
ClickHouse | ClickHouse | 438,383,705 | 3,565 | alexey-milovidov | And also you don't need a `unique_ptr`. | 2018-11-13T18:27:44 | 2018-11-13T18:27:44 | {} | MEMBER |
ClickHouse | ClickHouse | 439,011,180 | 2,402 | TH-HA | Hello
Cascading MV are not working , but can we have multiple MV on the same source to feed the input but also the aggregation ?
Typically can we feed the input and aggregated table directly from the Kafka table ?
Thanks | 2018-11-15T11:37:20 | 2018-11-15T11:37:20 | {} | NONE |
ClickHouse | ClickHouse | 439,394,314 | 2,601 | alex-zaitsev | Table has an index (tags_id, created_at), 100M rows. Prewhere on index columns improves performance 3 times.
```
:) SELECT sum(cityHash64(*)) a FROM benchmark_cpu where (tags_id, created_at) in (select tags_id, max(created_at) from benchmark_cpu group by tags_id);
1 rows in set. Elapsed: 2.132 sec. Processed 33... | 2018-11-16T13:34:31 | 2018-11-16T13:36:45 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 437,053,402 | 3,520 | igor-sh8 | Hi.
There was no user "default." or "default" in configs. At the same time, everything was fine in the logs too.
As a fix I just created user "default" on all nodes in the cluster with strict network limitation (only for cluster nodes). And It seems helps us. | 2018-11-08T16:11:56 | 2018-11-08T16:11:56 | {} | NONE |
ClickHouse | ClickHouse | 437,778,121 | 3,554 | hannamonnah | hi!
my query's filter is "`where toDate(DateTime_value) < today()-5`" and it has suspended for 5 days.
but when i writer filter as "`where toDate(DateTime_value) = '2018-11-12'` " it executed in 10 seconds.
in system.mutations table command column of my filter is shown like "`delete where toDate(DateTime_value) < '... | 2018-11-12T07:06:16 | 2018-11-12T07:06:16 | {} | NONE |
ClickHouse | ClickHouse | 438,392,609 | 3,565 | amosbird | thanks | 2018-11-13T18:53:24 | 2018-11-13T18:53:24 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 436,754,766 | 2,460 | giuliaScalaberni | Hi @blinkov,
I've figured out the problem.
In the ETL flow there were a loss of data before the creation of the CSV file.
So ClickHouse isn't involved... Thank you for support and guys you are doing a very great job! | 2018-11-07T19:50:21 | 2018-11-07T19:50:21 | {} | NONE |
ClickHouse | ClickHouse | 438,382,534 | 523 | lemmsh | from my experience user->table->where in the users.xml will blow up really fast and you'll get unmanageable hell in the config xml.
Anyway, you can have both authentication and authorization behind some interfaces with implementation laying in a dedicated module. And have the users.xml as default implementation. Th... | 2018-11-13T18:24:23 | 2018-11-13T18:24:23 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 438,395,508 | 3,565 | alexey-milovidov | Yes. | 2018-11-13T19:01:52 | 2018-11-13T19:01:52 | {} | MEMBER |
ClickHouse | ClickHouse | 438,390,712 | 3,578 | healiseu | ```sql
SELECT version()
┌─version()─┐
│ 18.12.17 │
└───────────┘
```
apt-cache policy clickhouse-client
clickhouse-client:
Installed: 18.12.17
apt-cache policy clickhouse-server
clickhouse-server:
Installed: 18.12.17
So the release I am running is two months old and it is installed on Linux Ubunt... | 2018-11-13T18:48:07 | 2018-11-13T19:46:11 | {} | NONE |
ClickHouse | ClickHouse | 438,494,350 | 3,578 | den-crane | seems already fixed even for enable_optimize_predicate_expression = 1 | 2018-11-14T00:43:23 | 2018-11-14T00:43:23 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 439,601,724 | 3,592 | kanterov | It looks like the build is failing for a reason not related to this PR:
> The job exceeded the maximum time limit for jobs, and has been terminated.
| 2018-11-17T08:57:23 | 2018-11-17T08:57:23 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 439,661,068 | 684 | xiaods | hi team, i am newcomer, i came again this issue, i have reading the user documents and surfing many more web stuffs, but not found any full examples for use case, i am very pain for it. hope anyone can maintain a example case site to guide me. thanks a lot for this cool db. | 2018-11-18T01:44:51 | 2018-11-18T01:44:51 | {} | NONE |
ClickHouse | ClickHouse | 439,691,779 | 3,600 | xiaods | @blinkov i found the root reason, the cluster name in system.clusters is remove one char 'c'. it let me confuse and not found the difference. when i correct the bitbook-service to bitbook-servie, it works like charm. thanks a lot. | 2018-11-18T13:10:07 | 2018-11-18T13:10:07 | {} | NONE |
ClickHouse | ClickHouse | 436,968,077 | 3,162 | ztlpn | Fix was released in v18.14.13 | 2018-11-08T11:47:44 | 2018-11-08T11:47:44 | {
"+1": 2
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 437,988,405 | 3,569 | alexey-milovidov | Confirmed. | 2018-11-12T18:46:20 | 2018-11-12T18:46:20 | {} | MEMBER |
ClickHouse | ClickHouse | 438,247,327 | 523 | alexey-milovidov | Now we have a demand for row-level security even from our internal users. We will implement it.
But our internal users propose some ad-hoc solutions like specifying user -> table -> where condition in users.xml.
| 2018-11-13T12:22:26 | 2018-11-13T12:22:26 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 438,587,107 | 3,562 | fancyqlx | > Well, that's a very good question. The case with multiple LowCardinality keys can actually be improved, but it' s more difficult than in case with single key and potential speedup is lower.
>
> There are two main optimizations for GROUP BY LowCardinality, and both of them optimize access to hash table (when for ag... | 2018-11-14T08:59:36 | 2018-11-14T08:59:36 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 439,453,323 | 3,595 | den-crane | `AND (partition = '\'2018-11-13\'')` | 2018-11-16T16:44:18 | 2018-11-16T16:44:33 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 437,908,917 | 3,554 | den-crane | @ztlpn seems a bug in alter delete
today()-5 ---> '((2018 - 11) - 7 )' | 2018-11-12T14:50:40 | 2018-11-12T14:50:40 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 438,255,918 | 3,576 | abyss7 | Very likely it was fixed in this PR: https://github.com/yandex/ClickHouse/pull/3534 | 2018-11-13T12:55:23 | 2018-11-13T12:55:23 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 438,493,277 | 3,578 | den-crane | this is because of enable_optimize_predicate_expression = 1
it was disabled in 18.14.11 by default
duplicate of #3149 | 2018-11-14T00:38:11 | 2018-11-14T00:38:19 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 438,869,942 | 3,583 | den-crane | The same with
`<source><file><path>/tmp/ttt.tsv</path> <format>TabSeparated</format></file></source>`
```
cat /tmp/ttt.tsv
0 2018-01-01 2018-01-01 1
0 2018-01-02 2018-01-02 2
0 2018-01-03 2018-01-03 3
``` | 2018-11-15T00:29:18 | 2018-11-15T00:29:48 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 439,415,426 | 3,587 | SkeLLLa | @blinkov thanks for reply.
I'm talking mostly about libraries that use http interface, and doesn't use odbc/jdbc.
Recently I reviewed couple of nodejs drivers. For example in this case time that lib adds is up to 2x comparing to time that clickhouse uses to perform query. So if clickhouse request takes 1s, then lib... | 2018-11-16T14:47:23 | 2018-11-16T14:47:23 | {
"+1": 1
} | NONE |
ClickHouse | ClickHouse | 437,638,532 | 3,561 | volfco | If I remove nodes 02 and 03, and just have destination_cluster list 05, data is being written as expected.
```
/var/local/zookeeper/zookeeclickhouse-copier --config-file=config.xml --task-path=/clickhouse/copytask16 --base-dir=./ --log-level=debug --safe-mode
Include not found: clickhouse_compression
Logging debu... | 2018-11-11T02:33:53 | 2018-11-11T02:33:53 | {} | NONE |
ClickHouse | ClickHouse | 437,741,250 | 3,536 | sundy-li | Currently we gave up running ClickHouse on NFS | 2018-11-12T03:26:33 | 2018-11-12T03:26:33 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 438,019,484 | 2,560 | silviucpp | I tested with master as well today and seems fixed !
@alexey-milovidov I think you can close it
Silviu | 2018-11-12T20:33:58 | 2018-11-12T20:33:58 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 438,160,542 | 3,554 | hannamonnah | i found out why this was occurred. i forgot quote with date.
when date is written like 2018-11-7, db transforms ((2018 - 11 ) - 7)
thanks. | 2018-11-13T07:10:15 | 2018-11-13T07:10:15 | {
"+1": 1
} | NONE |
ClickHouse | ClickHouse | 438,384,666 | 3,578 | abyss7 | Can you provide your CH version? I can't reproduce it on recent master at least. | 2018-11-13T18:30:29 | 2018-11-13T18:30:29 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 439,457,838 | 538 | kmatt | I started seeing similar issues when changing to a set of partition columns that created several hundred to a few thousand partitions instead of one per day - partitioning on the primary key to reduce query times. One possible solution is to use a lower cardinality partition scheme. | 2018-11-16T16:58:28 | 2018-11-16T16:58:28 | {} | NONE |
ClickHouse | ClickHouse | 439,693,292 | 3,602 | xiaods | my faults. close it. | 2018-11-18T13:35:51 | 2018-11-18T13:35:51 | {} | NONE |
ClickHouse | ClickHouse | 437,442,179 | 3,536 | ztlpn | > > But `clearOldTemporaryDirectories` shouldn't delete temporary directories that have been touched during the last 24 hours. I guess you have changed (set to 0) the `temporary_directories_lifetime` MergeTree setting?
>
> No, it just happens on NFS with no configure change.
@sundy-li Can you explain in more deta... | 2018-11-09T17:57:44 | 2018-11-09T17:58:15 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 437,501,637 | 3,554 | den-crane | it works in background and eventually will finish.
check
select * from system.mutations where parts_to_do>0; | 2018-11-09T21:32:07 | 2018-11-09T21:32:07 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 437,804,617 | 3,566 | 4ertus2 | It was fixed in master one ot two weeks ago. You need a newer release. | 2018-11-12T09:00:35 | 2018-11-12T09:00:35 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 438,388,142 | 3,565 | alexey-milovidov | ```
milovidov@milovidov-Pro-P30:~/work/ClickHouse$ cat test.cpp
#include <variant>
int main()
{
std::variant<int, float> x;
std::get<float>(x) = 1.1;
return std::get<float>(x);
}
milovidov@milovidov-Pro-P30:~/work/ClickHouse$ g++-7 -O0 -std=c++17 test.cpp
milovidov@milovidov-Pro-P30:~/work/Click... | 2018-11-13T18:40:54 | 2018-11-13T18:40:54 | {} | MEMBER |
ClickHouse | ClickHouse | 438,503,935 | 3,571 | alexey-milovidov | This is not a bug.
But for better usability we can throw an exception when `WITH TOTALS` is specified without any aggregate functions. | 2018-11-14T01:31:07 | 2018-11-14T01:31:07 | {
"+1": 2
} | MEMBER |
ClickHouse | ClickHouse | 438,504,435 | 2,560 | alexey-milovidov | Ok.
The root cause of this issue is more fundamental.
Query context is passed by reference, but sometimes copied to temporary.
Proper fix is still to do. | 2018-11-14T01:33:40 | 2018-11-14T01:33:40 | {} | MEMBER |
ClickHouse | ClickHouse | 438,976,279 | 2,402 | mvanderkroon | Thanks for the update!
| 2018-11-15T09:32:24 | 2018-11-15T09:32:24 | {} | NONE |
ClickHouse | ClickHouse | 439,077,726 | 1,661 | Slind14 | So, I revisited the possibility of collapsing and replacing MergeTrees because we are looking at the multi billion records a day.
Due to the amount of requests we receive we have a load balancing setup with a good amount of nodes. Because some requests (updates) are coming in with < 5ms inbetween it seems impossible... | 2018-11-15T15:23:00 | 2018-11-15T15:23:00 | {} | NONE |
ClickHouse | ClickHouse | 439,415,869 | 3,585 | abyss7 | Also if it's possible can you provide a sample of the most probable queries that may lead to segmentation fault? | 2018-11-16T14:48:53 | 2018-11-16T14:48:53 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 439,687,887 | 3,600 | xiaods | @blinkov thanks for your info. i search a test case to evaluate the clickhouse engine, came across another trap:
sample test sql:
```
create database test;
use test;
CREATE TABLE ontime_local (FlightDate Date,Year UInt16) ENGINE = MergeTree(FlightDate, (Year, FlightDate), 8192);
CREATE TABLE ontime_all... | 2018-11-18T12:06:15 | 2018-11-18T12:06:15 | {} | NONE |
ClickHouse | ClickHouse | 439,689,421 | 684 | xiaods | @blinkov thanks for your kind comments. i create a issue for my case: https://github.com/yandex/ClickHouse/issues/3600 | 2018-11-18T12:31:54 | 2018-11-18T12:31:54 | {} | NONE |
ClickHouse | ClickHouse | 437,045,040 | 3,536 | sundy-li | > But `clearOldTemporaryDirectories` shouldn't delete temporary directories that have been touched during the last 24 hours. I guess you have changed (set to 0) the `temporary_directories_lifetime` MergeTree setting?
No, it just happens on NFS with no configure change. | 2018-11-08T15:51:10 | 2018-11-08T15:51:10 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 437,105,977 | 3,553 | alexey-milovidov | Missing linuxCapability.h. | 2018-11-08T18:28:10 | 2018-11-08T18:28:10 | {} | MEMBER |
ClickHouse | ClickHouse | 438,183,844 | 3,333 | GSokol | Could I implement the new function which should return Float32? | 2018-11-13T08:46:28 | 2018-11-13T08:46:28 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 438,399,050 | 3,568 | champtar | @KochetovNicolai Are these failure false positive ? (I see Travis timeout) | 2018-11-13T19:12:13 | 2018-11-13T19:12:13 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 438,565,987 | 3,575 | yunchat | @abyss7 thanks .Yes, I meant the errors message: 'Too many simultaneous queries. Maxmum:xx'。the configuration set "max_concurrent_queries and max_connections" can't guaranteed write stability。Hope the feature like ' pools of read-only and read-write queries' in the future。
| 2018-11-14T07:34:46 | 2018-11-14T07:34:46 | {} | NONE |
ClickHouse | ClickHouse | 439,144,616 | 3,589 | ujiseven | Have same problem on my laptop
```
uname -a
Linux 3.19.0-25-generic #26~14.04.1-Ubuntu SMP Fri Jul 24 21:16:20 UTC 2015 x86_64 x86_64 x86_64 GNU/Linux
```
```
2018.11.15 21:30:33.254839 [ 99 ] {2642d9ca-49f1-4d0a-81f8-81e624d7a3e5} <Trace> ParallelAggregatingBlockInputStream: Aggregating
2018.11.15 21:30:33.28... | 2018-11-15T18:35:20 | 2018-11-15T18:38:34 | {} | NONE |
ClickHouse | ClickHouse | 439,729,466 | 3,601 | alexey-milovidov | > I guess it'd be a performance issue to use the whole public suffix list?
It should not if we will use carefully crafted hash table.
(Now it is not required.) | 2018-11-18T21:57:50 | 2018-11-18T21:58:52 | {} | MEMBER |
ClickHouse | ClickHouse | 437,169,736 | 3,499 | alexey-milovidov | Yes. https://github.com/yandex/ClickHouse/pull/3530 | 2018-11-08T21:54:09 | 2018-11-08T21:54:16 | {} | MEMBER |
ClickHouse | ClickHouse | 438,503,592 | 3,571 | alexey-milovidov | `WITH TOTALS` make sense only when you apply aggregation (specify at lease one aggregate function).
Totals row will contain total values for each aggregate column in a query and default values (zeros, NULLs, empty strings) for all other columns.
Example for system.parts table:
```
SELECT table, sum(bytes_on_dis... | 2018-11-14T01:29:42 | 2018-11-14T01:29:42 | {} | MEMBER |
ClickHouse | ClickHouse | 436,933,553 | 3,447 | BayoNet | I have corrected descriptions of the geo-functions.
https://github.com/yandex/ClickHouse/pull/3464/ | 2018-11-08T09:42:25 | 2018-11-08T09:42:25 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 437,745,118 | 2,508 | shangshujie365 | I have the same problem, stream_flush_interval_ms can't work | 2018-11-12T03:54:50 | 2018-11-12T03:54:50 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 438,204,769 | 2,402 | mvanderkroon | We have the same issue, any progress on this? | 2018-11-13T09:53:37 | 2018-11-13T09:53:37 | {} | NONE |
ClickHouse | ClickHouse | 438,250,083 | 3,198 | fillest | We don't use neither kafka nor docker (and we use just very basic features of Clickhouse) but I see lots of such messages (I also don't see out-of-file-handlers errors though):
```
kernel netlink: 'TCPHandler': attribute type 1 has an invalid length.
kernel netlink: 'TCPHandler': attribute type 1 has an invalid leng... | 2018-11-13T12:32:50 | 2018-11-13T12:34:52 | {} | NONE |
ClickHouse | ClickHouse | 438,382,905 | 3,565 | alexey-milovidov | You can also note that this code is out of style (double assignment in a single statement). | 2018-11-13T18:25:27 | 2018-11-13T18:25:27 | {} | MEMBER |
ClickHouse | ClickHouse | 438,490,932 | 3,568 | alexey-milovidov | Yes.
And errors in dictionary tests are unrelated. | 2018-11-14T00:26:51 | 2018-11-14T00:26:51 | {} | MEMBER |
ClickHouse | ClickHouse | 438,931,753 | 3,393 | shangshujie365 | but 18.10.3 has not this problem | 2018-11-15T06:24:16 | 2018-11-15T06:24:16 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 439,043,684 | 3,581 | abyss7 | I'm in progress of looking into it | 2018-11-15T13:41:56 | 2018-11-15T13:41:56 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 439,147,712 | 3,589 | ujiseven | when i set max_block_size = 100 its work on 18.12.13(18.14.13) but slowly
<Debug> MemoryTracker: Peak memory usage (for query): 537.18 MiB. | 2018-11-15T18:45:16 | 2018-11-15T19:02:47 | {} | NONE |
ClickHouse | ClickHouse | 437,826,675 | 3,562 | KochetovNicolai | Well, that's a very good question. The case with multiple LowCardinality keys can actually be improved, but it' s more difficult than in case with single key and potential speedup is lower.
There are two main optimizations for GROUP BY LowCardinality, and both of them optimize access to hash table (when for aggregat... | 2018-11-12T10:15:55 | 2018-11-12T10:15:55 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 438,366,675 | 3,574 | alexey-milovidov | We can implement it for compatibility with mysql CLI:
```
sudo mysql -uroot --host=127.0.0.1 --password
Enter password:
``` | 2018-11-13T17:41:14 | 2018-11-13T17:41:14 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 438,385,600 | 3,565 | amosbird | I don't follow. You mean we cannot use `std::variant` since we had that double assignment code existing? | 2018-11-13T18:33:16 | 2018-11-13T18:33:16 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 438,510,327 | 3,568 | alexey-milovidov | More likely next week. | 2018-11-14T02:05:08 | 2018-11-14T02:05:08 | {} | MEMBER |
ClickHouse | ClickHouse | 439,079,969 | 1,285 | Axelfoley85 | Ran into the same issue with version 1.1.54390
`Code: 32. DB::Exception: Attempt to read after eof: while receiving packet from clickhouse:9000, 172.19.0.2` | 2018-11-15T15:29:03 | 2018-11-15T15:29:03 | {} | NONE |
ClickHouse | ClickHouse | 439,404,120 | 3,579 | 4n70w4 | May be anyone have a working build kafka + clickhouse in docker-compose for transfer CSV/TSV data? | 2018-11-16T14:09:53 | 2018-11-16T14:09:53 | {} | NONE |
ClickHouse | ClickHouse | 439,673,835 | 3,600 | blinkov | You can create Distributed table on each server and insert new data to it on random server, which will then properly distribute this data across cluster. See https://clickhouse.yandex/docs/en/operations/table_engines/distributed/ | 2018-11-18T07:38:57 | 2018-11-18T07:38:57 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 438,167,779 | 3,574 | blinkov | This is indeed looks like incorrect command to run. You should either specify password value like `--password=12345` (which is often not a good idea since it'll be visible in tools like `ps` or `top`), or use `--ask-password` if you want to enter it interactively. | 2018-11-13T07:44:29 | 2018-11-13T07:44:29 | {
"hooray": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 438,263,685 | 3,575 | abyss7 | Right now I only can advise to update your current configuration settings (like, `max_concurrent_queries` and `max_connections`). Also, there is no such error message "too many process" - maybe you meant: "Too many simultaneous queries". The possible solution in the future is to implement separate configuration options... | 2018-11-13T13:22:31 | 2018-11-13T13:22:31 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 438,382,319 | 3,565 | alexey-milovidov | The code is absolutely wrong.
You're trying to get data from `std::variant` but you never set it to the proper type.
It was set by the following code:
```
const auto & null_value_ref = std::get<T>(attribute.null_values) =
null_value.get<typename NearestFieldType<T>::Type>();
```
It works for you only... | 2018-11-13T18:23:47 | 2018-11-13T18:23:47 | {} | MEMBER |
ClickHouse | ClickHouse | 438,599,108 | 3,350 | alex-krash | Seems to be finished.
Here are benchmark results (CMake Release profile):
Reference query:
```
SELECT count()
FROM
(
SELECT base64Encode(materialize('This is a long string to test ClickHouse base64 functions impl..'))
FROM numbers(20000000)
)
```
Without base64Encode() call (just materialize()):
`... | 2018-11-14T09:39:35 | 2018-11-14T09:39:35 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 439,018,907 | 3,585 | EvgenyVinogradov | By our investigation we encountering such error on intensive loading of 2 tables:
1st-table by JDBC batch inserting into local ReplicatedMergeTree on one replica of shard
2nd-table by insert ... select ... into local ReplicatedMergeTree on one replica of shard
If 2 insertions are on the same host
Cluster configu... | 2018-11-15T12:08:54 | 2018-11-15T15:01:30 | {} | NONE |
ClickHouse | ClickHouse | 439,376,905 | 3,585 | abyss7 | Is it a new load workflow or have you updated to the new CH version? Do you have any stable version number which doesn't reproduce the bug? | 2018-11-16T12:20:27 | 2018-11-16T12:20:45 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 439,407,079 | 3,587 | blinkov | @SkeLLLa there are no specific short-term plans, but which exactly libraries have you tried?
P.S. There's also JDBC driver for JVM-based products that is kind of official http://github.com/yandex/clickhouse-jdbc | 2018-11-16T14:19:28 | 2018-11-16T14:19:28 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 439,479,586 | 3,595 | Jack012a | It works now.
Thank you | 2018-11-16T18:13:40 | 2018-11-16T18:13:40 | {} | NONE |
ClickHouse | ClickHouse | 438,395,421 | 3,565 | amosbird | > And also you don't need a unique_ptr.
std::variant cannot get rid of all `unique_ptr` wrappers as ComplexKeyCacheDictionary uses `ContainerType = Value[]; ` which doesn't handle ownerships.
| 2018-11-13T19:01:38 | 2018-11-13T19:01:38 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 439,415,394 | 3,585 | abyss7 | Can you tell if you built the server from source or installed the .deb from public repo? | 2018-11-16T14:47:16 | 2018-11-16T14:47:16 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 439,417,289 | 3,531 | Jack012a | Hello,
I have tried the proposed method but it doesn't work for us.
```
:) alter table default.MyTable detach partition '2018-11-12'
ALTER TABLE default.MyTable
DETACH PARTITION '2018-11-12'
Ok.
0 rows in set. Elapsed: 0.002 sec.
```
Based on my experience, the detached partitions should be moved int... | 2018-11-16T14:53:28 | 2018-11-16T16:27:41 | {} | NONE |
ClickHouse | ClickHouse | 437,806,284 | 3,256 | KochetovNicolai | It's not fixed indeed. Sorry, have forgotten to check it. | 2018-11-12T09:06:42 | 2018-11-12T09:06:42 | {} | MEMBER |
ClickHouse | ClickHouse | 438,255,254 | 3,333 | abyss7 | Sure - it would be appreciated. | 2018-11-13T12:52:56 | 2018-11-13T12:52:56 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 438,300,697 | 2,402 | MyReserved | > We have the same issue, any progress on this?
Cascading Materialized views are not working at the moment.
KafkaEngine use one MV so you cannot use another MV on target table
I asked @alexey-milovidov last week | 2018-11-13T15:11:39 | 2018-11-13T15:11:39 | {} | NONE |
ClickHouse | ClickHouse | 438,517,277 | 2,298 | 4n70w4 | @alexey-milovidov
I have the same problem on the same data.
```
SELECT *
FROM weather_queue2
Received exception from server (version 18.14.13):
Code: 117. DB::Exception: Received from localhost:9000, 127.0.0.1. DB::Exception: Expected end of line: (at row 1)
Could not print diagnostic info because two las... | 2018-11-14T02:42:56 | 2018-11-14T02:42:56 | {} | NONE |
ClickHouse | ClickHouse | 439,064,288 | 3,585 | EvgenyVinogradov | Batch size 100000 records and loads by one thread on one shard
both tables has a same structure described below:
CREATE TABLE default.ONLINE_LOG_local ( RECORD_ID UInt64, ENV_ID Nullable(String), ENV_TIMEIN DateTime, SAF_PLAN_ID Nullable(String), TXN_SOURCE Nullable(String), TXN_MTI Nullable(String), TXN_F2 N... | 2018-11-15T14:45:47 | 2018-11-15T14:50:01 | {} | NONE |
ClickHouse | ClickHouse | 439,419,261 | 3,587 | alex-zaitsev | @SkeLLLa, have you looked at https://github.com/artpaul/clickhouse-cpp
Also, there is TCP protocol driver for Go which is used for high performance data ingestion: https://github.com/kshvakov/clickhouse
There is TCP protocol driver for .NET: https://github.com/killwort/ClickHouse-Net
And for Java: https://github.c... | 2018-11-16T14:59:26 | 2018-11-16T14:59:26 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 439,467,864 | 3,587 | SkeLLLa | @alex-zaitsev thanks for suggestions.
My main goal is to have some good official lib aka driver and build libs for other langs around it.
So as for JDBC - the main con is that it requires JVM to run, so it's not good to have such dependency if you're building a library for node.js or php. Dotnet client seems qui... | 2018-11-16T17:31:58 | 2018-11-16T17:31:58 | {
"+1": 1
} | NONE |
ClickHouse | ClickHouse | 439,673,672 | 684 | blinkov | @xiaods which exactly issue do you have and what kind of example do you need? | 2018-11-18T07:34:58 | 2018-11-18T07:34:58 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 439,729,572 | 3,601 | alexey-milovidov | You can also provide a performance test to test for performance degradation.
Look at the dbms/tests/performance directory. | 2018-11-18T21:59:28 | 2018-11-18T21:59:28 | {} | MEMBER |
ClickHouse | ClickHouse | 437,982,679 | 523 | lemmsh | a follow-up after more than a year using this functionality within our internal standalone clickhouse proxy.
1. Access rules can be arbitrary complex and must follow the typical development process with pull requests and CI. There is no way oracle-vpd-like solution (dba configures access rules with a DDL) can be con... | 2018-11-12T18:27:35 | 2018-11-12T18:27:35 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 438,853,371 | 863 | iameugenejo | Is there any progress on this? I'm on 18.5.1 and querying against a view that has union all scans the entire rows of all tables. | 2018-11-14T23:08:31 | 2018-11-14T23:08:31 | {
"+1": 2
} | NONE |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.