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 | 442,105,414 | 3,674 | tsl-karlp | nevermind, fixed yesterday at 0dd5777c1c53f1a2ea5778b445a28965995c8bcf by @geniousphp :) . Thank you! | 2018-11-27T15:42:14 | 2018-11-27T15:42:25 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 442,167,145 | 3,592 | alexey-milovidov | @champtar
It looks natural to have this support for monotonic functions.
`datetime < x` implies `toStartOfMinute(datetime) <= toStartOfMinute(x)`,
and we already have all required metadata about function monotonicity that says that function `toStartOfMinute` is monotonic on whole domain.
Also we can implement ... | 2018-11-27T18:32:10 | 2018-11-27T18:32:10 | {} | MEMBER |
ClickHouse | ClickHouse | 442,823,062 | 3,697 | KochetovNicolai | Now I can see that the difference is in point coord. types. It works if I change Floar32 to Float64.
```SQL
CREATE TABLE test_pointinpolygon_1
(
POS_LON Float64,
POS_LAT Float64
)
ENGINE = MergeTree
ORDER BY tuple()
SETTINGS index_granularity = 8192
Ok.
0 rows in set. Elapsed: 0.040 sec.
I... | 2018-11-29T12:53:40 | 2018-11-29T12:53:40 | {} | MEMBER |
ClickHouse | ClickHouse | 443,107,029 | 2,314 | astudnev | I met with the same bug on
SELECT
dictGetString('address_annotation', 'text', tuple(toUInt32(1), taker)) AS taker_annotation
FROM (
SELECT
'0x0000000000000000000000000000000000000000' taker
)
Code: 9. DB::Exception: Received from localhost:9000, ::1. DB::Exception: Incorrect size of nested column in co... | 2018-11-30T06:38:56 | 2018-11-30T06:56:29 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 443,220,808 | 3,699 | alexey-milovidov | The test can be easily implemented with `blockSize` function. | 2018-11-30T14:32:34 | 2018-11-30T14:32:34 | {} | MEMBER |
ClickHouse | ClickHouse | 443,222,493 | 3,716 | KochetovNicolai | I think it works in master.
```SQL
:) insert into lc_table (event_time, partition_type)
:-] values (toDateTime('0000-00-00 00:00:00'), 'string');
INSERT INTO lc_table (event_time, partition_type) VALUES
Ok.
1 rows in set. Elapsed: 0.006 sec.
mtdev04t.yandex.ru :) select * from lc_table
SELECT *
FROM l... | 2018-11-30T14:37:54 | 2018-11-30T14:37:54 | {} | MEMBER |
ClickHouse | ClickHouse | 443,241,765 | 3,694 | ztlpn | #3719 | 2018-11-30T15:36:56 | 2018-11-30T15:36:56 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 443,372,865 | 3,483 | filimonov | Please, try to create some reproducible example. | 2018-11-30T23:40:25 | 2018-11-30T23:40:25 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 443,632,370 | 3,731 | lamberken | ok，I see | 2018-12-03T08:47:03 | 2018-12-03T08:47:03 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 443,652,783 | 3,617 | alesapin | @chenxing-xc Unfortunately I can't build it locally:
1) With gcc I get cmake error:
```
CMake Error at contrib/libhdfs3-cmake/CMake/Platform.cmake:19 (LIST):
list index: 1 out of range (-1, 0)
Call Stack (most recent call first):
contrib/libhdfs3-cmake/CMakeLists.txt:17 (include)
```
2) With clang cmake is... | 2018-12-03T09:58:57 | 2018-12-03T10:01:38 | {} | MEMBER |
ClickHouse | ClickHouse | 443,663,655 | 1,976 | blinkov | @SUHAR1K which ClickHouse release do you use?
As far as I can see, this kind of error (https://github.com/yandex/ClickHouse/blob/e4b4b5a4fcc430a4fc427cd313fa7e91c45da3dd/dbms/src/Formats/CSVRowInputStream.cpp#L50 ) became skippable by these settings only recently in 18.14.*** releases (https://github.com/yandex/Clic... | 2018-12-03T10:35:39 | 2018-12-03T10:35:39 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 443,219,321 | 3,597 | ztlpn | Hi @healiseu, sorry for the late reply.
The problem with your queries is that you have created your tables in the `DemoDB` database and made it the default database in your session. As mutations are executed outside of the user session context, the information about the default database is lost and the server tries ... | 2018-11-30T14:27:23 | 2018-11-30T14:27:23 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 443,533,910 | 3,731 | lamberken | Maybe I still need two weeks to finish it. | 2018-12-02T19:15:50 | 2018-12-02T19:15:50 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 443,595,418 | 2,586 | zhang2014 | @alexey-milovidov Sorry for the delay. I will continue the work now. It has too much conflict : ( | 2018-12-03T05:42:55 | 2018-12-03T05:42:55 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 444,047,739 | 3,742 | viongpanzi | @nicelulu
The following log is printed on the leader node:
```
0. /home/clickhouse/clickhouse-current/bin/clickhouse-server(StackTrace::StackTrace()+0x16) [0x5c1e746]
1. /home/clickhouse/clickhouse-current/bin/clickhouse-server(DB::Exception::Exception(std::__cxx11::basic_string<char, std::char_traits<char>, std... | 2018-12-04T10:18:14 | 2018-12-04T10:18:14 | {} | NONE |
ClickHouse | ClickHouse | 444,116,059 | 3,747 | blinkov | @zhang2014 @sundy-li could you help with updating zh version to match this patch? | 2018-12-04T14:21:28 | 2018-12-04T14:21:46 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 442,026,576 | 3,678 | dqminh | Replacing `notEmpty(ipv6)` with `notEmpty(toStringCutToZero(ipv6))` didnt segfault Clickhouse. Looks like `notEmpty` and `FixedString` didnt work well together ? | 2018-11-27T11:33:38 | 2018-11-27T11:33:38 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 440,235,298 | 3,607 | filimonov | YAML would be nice. But actually configuration management in ClickHouse is quite flexible and mature. It's not really obvious sometimes and sometimes you need to find proper 'ClickHouse-way' to solve the problems, but it's doable with XMLs.
ClickHouse has quite a lot of options and it's quite hard (and usually ther... | 2018-11-20T11:07:29 | 2018-11-28T13:22:36 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 442,837,408 | 3,697 | bicubic | A float32 latlong has an error of ~1.7 meters. Two float32s have a cumulative error of ~3.4 meters. Any intermediate arithmetic using float32 in the point in polygon test will further compound the error.
The point that's shown on the highway looks a bit extreme though. | 2018-11-29T13:41:47 | 2018-11-29T13:41:47 | {} | NONE |
ClickHouse | ClickHouse | 443,177,438 | 3,703 | alesapin | Failed build doesn't related to changes. | 2018-11-30T11:35:07 | 2018-11-30T11:35:07 | {} | MEMBER |
ClickHouse | ClickHouse | 442,780,145 | 3,675 | thyn | May it would be helpfull for others. Is this a right query?
I think I can't get the maximum number of consecutive triggered conditions from the chain within the sliding time window. Am I right?
`
SELECT
countIf(1,step1_time != toDateTime(0)) AS s1,
countIf(1,step1_time != toDateTime(0) and step2_time =... | 2018-11-29T10:16:24 | 2018-12-01T13:24:28 | {} | NONE |
ClickHouse | ClickHouse | 443,471,214 | 3,724 | alexey-milovidov | I've fixed the following:
- obsolete comment;
- possible integer division in every iteration of a loop;
- utilize "useDefaultImplementationForConstants" to avoid code for const-const case;
- allow any unsigned integer types for "size" argument;
- fixed error with exception message when type of "size" argument is w... | 2018-12-02T00:50:07 | 2018-12-02T00:50:07 | {} | MEMBER |
ClickHouse | ClickHouse | 443,658,906 | 1,993 | blinkov | Materialized views act more like triggers for inserts into table they are attached to, so it doesn't make sense to create one on top of table function call. | 2018-12-03T10:19:40 | 2018-12-03T10:19:40 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 444,030,277 | 3,742 | viongpanzi | @nicelulu
Thanks for reply.
I have added some log to trace the cause, I found that the reason is that we can't get the user and password. And the password is right in our cluser.
```
void StorageReplicatedMergeTree::dropPartition(const ASTPtr & query, const ASTPtr & partition, bool detach, const Context & con... | 2018-12-04T09:25:48 | 2018-12-04T09:25:48 | {} | NONE |
ClickHouse | ClickHouse | 444,141,723 | 3,741 | kmatt | Confirmed it was due to the number of partitions and file count resulting. | 2018-12-04T15:31:02 | 2018-12-04T15:31:12 | {} | NONE |
ClickHouse | ClickHouse | 442,153,062 | 3,592 | alexey-milovidov | @kanterov For functional test - simply add cluster to the `dbms/programs/server/config.xml`. | 2018-11-27T17:48:59 | 2018-11-27T17:48:59 | {} | MEMBER |
ClickHouse | ClickHouse | 442,587,578 | 3,695 | xzkostyan | As alternative port readiness check for the client can be performed in something like this:
```bash
while true; do
timeout 1 bash -c 'cat < /dev/null > /dev/tcp/localhost/9000' 2>1 /dev/null;
rv=$?
if [[ $rv = 0 ]]; then
break
fi
done
```
I don't know which approach is better sleep o... | 2018-11-28T20:10:08 | 2018-11-28T20:10:08 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 443,202,795 | 3,701 | KochetovNicolai | You can [detach](https://clickhouse.yandex/docs/en/single/#detach) table. But you won't be able to use it.
Or you can stop fetches for table.
```
SYSTEM STOP FETCHES table;
```
```
SYSTEM START FETCHES table;
``` | 2018-11-30T13:26:33 | 2018-11-30T13:26:33 | {} | MEMBER |
ClickHouse | ClickHouse | 443,597,309 | 3,198 | cyfdecyf | @splichy I've tried the `setcap` command as you mentioned, but still gets netlink log messages. | 2018-12-03T05:55:36 | 2018-12-03T05:55:36 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 443,757,359 | 3,734 | TH-HA | Thanks a lot , waiting for the release to test that.
Regards
| 2018-12-03T15:48:00 | 2018-12-03T15:48:00 | {} | NONE |
ClickHouse | ClickHouse | 443,819,311 | 3,735 | TrollStation | For example, i have couple of values like `000000000000000000000000000000000000000000000002b5e3af16b1880000`. It's 50000000000000000000 in dec which overflow UInt64 as maximum possible unsigned integer type. Should i try to use `bitmaskToArray()` or something like this while splitting this number by parts? | 2018-12-03T18:37:43 | 2018-12-03T18:45:25 | {} | NONE |
ClickHouse | ClickHouse | 444,076,513 | 3,744 | KochetovNicolai | 00281_compile_sizeof_packed | 2018-12-04T12:02:15 | 2018-12-04T12:02:15 | {} | MEMBER |
ClickHouse | ClickHouse | 442,037,204 | 3,598 | nicelulu | @alesapin
Thank you.
When I executed the runner, I encountered the following error.
```
[root@server50-ke integration]# ./runner --binary /data0/dev/ClickHouse/build/dbms/programs/clickhouse --configs-dir /usr/local/etc/clickhouse-server/ 'test_replace_partition -ss'
Start tests
============================= tes... | 2018-11-27T12:13:49 | 2018-11-27T12:13:49 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 442,895,997 | 3,693 | alesapin | MakeCheck -- gcc-8 segfault base64 functions. | 2018-11-29T16:21:23 | 2018-11-29T16:21:23 | {} | MEMBER |
ClickHouse | ClickHouse | 442,958,394 | 3,526 | ztlpn | @submergerock Yes, the data is not inserted the second time because of deduplication. `insert_deduplicate` is a user setting (as are all settings in Settings.h). You can set it:
1. In a user profile in `users.xml`
2. For a session using the `SET` command: `SET insert_deduplicate=0`
3. (for SELECT statements) in the ... | 2018-11-29T19:17:03 | 2018-11-29T19:17:03 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 443,360,512 | 3,702 | alexey-milovidov | Postinstallation scripts should use common code with debian packages, rather than copy paste.
Spec file should be placed inside the repo rather than submodules.
`debian` and `rpm` directories should be moved inside new directory `packages`.
Only minimal number of libraries should be installed for build (currently it... | 2018-11-30T22:35:50 | 2018-11-30T22:35:50 | {} | MEMBER |
ClickHouse | ClickHouse | 444,095,957 | 3,583 | abyss7 | #3751 fixed there | 2018-12-04T13:17:03 | 2018-12-04T13:17:03 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 442,833,603 | 3,672 | M1ha-Shvn | I've made inserts more rare and the problem is gone. Sorry for bothering. | 2018-11-29T13:29:18 | 2018-11-29T13:29:18 | {} | NONE |
ClickHouse | ClickHouse | 442,851,689 | 3,696 | den-crane | or use select * from t **final**;
BTH ReplacingMergeTree(gmt, (name), 10)
10 is extremely low, index can be huge and will not fit into memory | 2018-11-29T14:25:07 | 2018-11-29T14:54:23 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 443,532,850 | 3,730 | den-crane | workaround : select median(toFloat64(a)) | 2018-12-02T18:59:58 | 2018-12-02T18:59:58 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 443,645,985 | 3,734 | blinkov | @TH-HA Parquet is going to be just yet another supported serialization format (similar to [these](https://clickhouse.yandex/docs/en/interfaces/formats/)) and, as far as I can see, current PR for HDFS support (#3617) includes either table function and table engine. | 2018-12-03T09:35:40 | 2018-12-03T09:36:26 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 443,660,023 | 1,992 | blinkov | @maddhuapplicate no, they are not combineable, but they are extendable, so you could write a new one with logic you need in C++ and open a pull request. | 2018-12-03T10:23:36 | 2018-12-03T10:23:36 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 444,038,979 | 3,742 | nicelulu | This method is to send a request to the leader, maybe the username and password of the leader server should be used. Instead of executing the username and password of the current query server. | 2018-12-04T09:52:18 | 2018-12-04T09:52:18 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 442,299,187 | 3,677 | shangshujie365 | ok | 2018-11-28T02:48:42 | 2018-11-28T02:48:42 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 442,152,723 | 3,592 | alexey-milovidov | @abyss7 Will continue review.
We need to test an idea, that we can integrate required functionality into KeyCondition.
If we will do it, it will also be helpful for functional indices.
As an example, you create an index by `reverse(domain)`, but query `domain = 'test.ru'`.
KeyCondition should derive condition `... | 2018-11-27T17:48:00 | 2018-11-27T17:48:00 | {} | MEMBER |
ClickHouse | ClickHouse | 442,156,711 | 3,592 | champtar | @alexey-milovidov would this KeyCondition improvement help using index like `PARTITION BY date ORDER BY (date, toStartOfMinute(datetime), request_file, clientip)` with request like `datetime < x AND datetime > y` ? | 2018-11-27T17:59:41 | 2018-11-27T17:59:41 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 442,586,306 | 3,695 | xzkostyan | Solves #3319. | 2018-11-28T20:06:04 | 2018-11-28T20:06:04 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 443,147,105 | 3,704 | filimonov | I don't know the original motivation, but docs highlight that this is "by design" https://clickhouse.yandex/docs/en/query_language/syntax/#keywords
I think those standard SQL function should be changed to be case-insensitive.
But changing it now for all the functions - doesn't sound as a good idea, as in lot o... | 2018-11-30T09:38:55 | 2018-11-30T13:18:26 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 443,220,610 | 3,699 | alexey-milovidov | A test? | 2018-11-30T14:31:53 | 2018-11-30T14:31:53 | {} | MEMBER |
ClickHouse | ClickHouse | 443,471,361 | 3,683 | alexey-milovidov | The code does not compile. | 2018-12-02T00:53:45 | 2018-12-02T00:53:45 | {} | MEMBER |
ClickHouse | ClickHouse | 443,615,519 | 3,234 | dstruck | https://github.com/yandex/ClickHouse/pull/3599 fixed the issue. | 2018-12-03T07:38:03 | 2018-12-03T07:38:03 | {} | NONE |
ClickHouse | ClickHouse | 443,631,401 | 3,731 | blinkov | It's usually easier to review incrementally rather than one huge patch, but as you wish.
I'd actually merge this PR too because partial translation is better than no translation at all. | 2018-12-03T08:43:12 | 2018-12-03T08:43:12 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 443,726,966 | 3,445 | filimonov | about `netlink` errors, can you check if the problem disappears without CAP_NET_ADMIN capability?
(It is optional and used to get accurate per-thread stats from kernel)
It's added during setup like that, so you can reset it in a similar way.
https://github.com/yandex/ClickHouse/blob/f1791e94e209b238b28a89c25e3a5f... | 2018-12-03T14:23:43 | 2018-12-03T14:25:34 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 443,727,629 | 3,701 | nikk0 | Thanks.
It works. | 2018-12-03T14:25:43 | 2018-12-03T14:25:43 | {} | NONE |
ClickHouse | ClickHouse | 443,750,635 | 2,508 | teunhoogendoorn | Same problem here:
With a dynamic throughput (for example 10.000 messages/sec during the day and 100 messages/sec during the night) it is hard to determine correct values for max_insert_block_size and stream_flush_interval_ms to import data into ClickHouse with a predictable delay. So something like the suggested "... | 2018-12-03T15:30:02 | 2018-12-03T15:30:02 | {} | NONE |
ClickHouse | ClickHouse | 443,942,771 | 3,741 | kmatt | Additionally these same nodes take 10s of minutes to shut down, while continuing to emit
2018.12.04 02:11:01.548270 [ 35 ] {} <Trace> testdb.customer_shard (StorageReplicatedMergeTree::mergeSelectingTask): Execution took 856 ms. │
| 2018-12-04T02:07:52 | 2018-12-04T02:11:41 | {} | NONE |
ClickHouse | ClickHouse | 442,063,254 | 3,679 | den-crane | Также в конфиге могут быть указаны "подстановки". Если у элемента присутствует атрибут incl, то в качестве значения будет использована соответствующая подстановка из файла. По умолчанию, путь к файлу с подстановками - **/etc/metrika.xml**
пример https://gist.github.com/den-crane/d6e0f1c7849a0bc44c27eab2de51c911
net... | 2018-11-27T13:46:28 | 2018-11-27T13:46:28 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 442,704,638 | 3,691 | bicubic | Would be nice to see greatCircleDistance support array inputs like this too. | 2018-11-29T04:43:28 | 2018-11-29T04:43:28 | {} | NONE |
ClickHouse | ClickHouse | 443,147,505 | 3,683 | nicelulu | ```
CREATE TABLE db.table_all ( id String ) ENGINE = Distributed(cluster1, db, table, sipHash64(id))
ALTER TABLE db.table_all MODIFY COLUMN id LowCardinality(String)
INSERT INTO db.table_all (id) VALUES ('123')
DB::Exception: Sharding key expression does not evaluate to an integer type
```
The user uses ALTER ... | 2018-11-30T09:40:24 | 2018-11-30T09:40:24 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 443,162,665 | 3,704 | alexey-milovidov | All function that either:
- listed in SQL standard;
- widely used in popular implementations (MySQL, Postgres);
- generated by popular BI tools;
should be made case insensitive.
This is possible in our design, example: https://github.com/yandex/ClickHouse/blob/master/dbms/src/Functions/concat.cpp#L189
| 2018-11-30T10:36:20 | 2018-11-30T10:36:30 | {} | MEMBER |
ClickHouse | ClickHouse | 443,220,417 | 3,707 | alexey-milovidov | Cancelled due to https://gcc.gnu.org/bugzilla/show_bug.cgi?id=47275#c4 | 2018-11-30T14:31:16 | 2018-11-30T14:31:16 | {} | MEMBER |
ClickHouse | ClickHouse | 443,680,681 | 3,679 | blinkov | @qJkee получилось настроить в итоге? | 2018-12-03T11:37:36 | 2018-12-03T11:37:36 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 443,912,449 | 3,737 | hatarist | Sorry! | 2018-12-03T23:37:35 | 2018-12-03T23:37:35 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 444,011,686 | 3,483 | AlexanderJLiu | > Please, try to create some reproducible example.
@filimonov I've setup env as mentioned above, and start clickhouse-server using `service clickhouse-server start`.
Then
```
$ clickhouse-client
ClickHouse client version 18.14.15.
Connecting to localhost:9000.
Connected to ClickHouse server version 18.14.15 r... | 2018-12-04T08:22:21 | 2018-12-04T08:22:21 | {} | NONE |
ClickHouse | ClickHouse | 444,020,801 | 3,742 | nicelulu | https://github.com/yandex/ClickHouse/pull/3598
Maybe replica is different from the current server username and password. | 2018-12-04T08:54:22 | 2018-12-04T08:54:22 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 442,336,829 | 3,683 | alexey-milovidov | Looks reasonable.
As I understand, the case is when you shard by a function of LowCardinality data type.
But there is lack of test and comment. | 2018-11-28T06:38:19 | 2018-11-28T06:38:19 | {} | MEMBER |
ClickHouse | ClickHouse | 443,426,693 | 3,582 | gekola | Hello.
For me stable does not build with llvm 7 after dfe2d13 because of missing `dbms/programs/clang/Compiler-7.0.0/cc1gen_reproducer_main.cpp` file. It seems to be present in master. Should a separate issue be created for this? | 2018-12-01T13:34:16 | 2018-12-01T13:34:16 | {} | NONE |
ClickHouse | ClickHouse | 443,864,502 | 3,529 | kmatt | Seeing similar, v18.14.15, Zookeeper 3.4.13, 4 nodes, 4 inserters round robin:
Code: 999. DB::Exception: Received from (hostname):9000, 151.139.73.213. DB::Exception: Cannot allocate block number in ZooKeeper: Coordination::Exception: Connection loss, path: /clickhouse/tables/4/(database)/block_numbers/153835200... | 2018-12-03T20:48:25 | 2018-12-03T20:52:43 | {
"+1": 1
} | NONE |
ClickHouse | ClickHouse | 443,881,988 | 3,737 | den-crane | The same as other RDMSs
postgresql
```
select count(*)x;
x
---
1
(1 row)
select count(*)from foo;
count
-------
0
(1 row)
``` | 2018-12-03T21:45:03 | 2018-12-03T21:46:02 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 444,055,006 | 3,683 | nicelulu |
> Ok. Still waiting for a test and code comment.
Done. | 2018-12-04T10:41:45 | 2018-12-04T10:41:45 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 442,089,130 | 3,680 | alexey-milovidov | ```
==844973==ERROR: AddressSanitizer: heap-use-after-free on address 0x6060012dbca0 at pc 0x7f28eefb2d5d bp 0x7f2847a576b0 sp 0x7f2847a576a8
READ of size 8 at 0x6060012dbca0 thread T26528 (TCPHandler)
#0 0x7f28eefb2d5c in Poco::Timestamp::Timestamp(Poco::Timestamp const&) /opt/home/proller/ClickHouse3/build_cla... | 2018-11-27T15:00:15 | 2018-11-27T15:00:15 | {} | MEMBER |
ClickHouse | ClickHouse | 443,671,086 | 1,825 | blinkov | @BayoNet I don't see this in your internal backlog | 2018-12-03T11:01:23 | 2018-12-03T11:01:23 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 443,961,558 | 3,697 | vicchen7 | After updating the data type of latitude and longitude using Float64 and the result is correct. | 2018-12-04T03:53:42 | 2018-12-04T03:53:42 | {} | NONE |
ClickHouse | ClickHouse | 442,067,214 | 3,679 | qJkee | Ну изначально только пытался сконфигурировать только пользователя.
После неудачи - добавил эту строчку отсюда - https://clickhouse.yandex/docs/ru/operations/access_rights/
Т.е стоит описать всех пользователей в users.xml? | 2018-11-27T13:58:18 | 2018-11-27T13:58:18 | {} | NONE |
ClickHouse | ClickHouse | 442,819,246 | 3,698 | KochetovNicolai | Could you show your query, table structure, or create a simple test which can reproduce a problem? | 2018-11-29T12:39:39 | 2018-11-29T12:39:39 | {} | MEMBER |
ClickHouse | ClickHouse | 443,167,374 | 3,658 | Schaefer-Thomas | Hi Nicolai,
it's an odbc-dictionary (see definitions below), 227 rows, underlying data did not change. select distinct on one of the tables seems stable and correct. Join between two MergeTree-Tables seems to be ok, too.
The duplicates only appear when joining the dictionary table with MergeTree-Table. The duplic... | 2018-11-30T10:53:54 | 2018-11-30T10:56:43 | {} | NONE |
ClickHouse | ClickHouse | 443,180,577 | 3,705 | KochetovNicolai | ClickHouse has INTERVAL type which can help you. For example:
```SQL
:) select toDate('2018-02-20') + interval 1 month
SELECT toDate('2018-02-20') + toIntervalMonth(1)
┌─plus(toDate('2018-02-20'), toIntervalMonth(1))─┐
│ 2018-03-20 │
└───────────────────────────────────────... | 2018-11-30T11:49:10 | 2018-11-30T11:49:10 | {} | MEMBER |
ClickHouse | ClickHouse | 443,597,012 | 3,198 | cyfdecyf | @fillest I see similar dmesg output for a Ubuntu 16.04.5 server which have clickhouse-server 18.14.15 installed. We are also using basic features of ClickHouse.
Simply run `clickhouse-client` and connect to local host would generate such error messages in dmesg.
```
[Mon Dec 3 13:31:06 2018] netlink: 'TCPHandle... | 2018-12-03T05:53:44 | 2018-12-03T05:53:44 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 443,656,304 | 1,486 | blinkov | @shernupup is this issue still relevant? how much physical memory your server had? were you actually running queries that require much memory to execute? | 2018-12-03T10:10:56 | 2018-12-03T10:10:56 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 443,659,768 | 3,708 | filimonov | I see few options of implementation possible
1) the most straight ahead: two parameters first used as sorting key, second - containing data.
```sql
select groupArraySorted( timestamp, page_type ); -- unlimited
select groupArraySorted(1000)( timestamp, page_type ); -- limited to first 1000
select groupArraySorted(-... | 2018-12-03T10:22:41 | 2018-12-03T10:26:22 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 444,014,533 | 3,483 | AlexanderJLiu | @filimonov Maybe I've found the answer from `man service` :
> service runs a System V init script in as predictable environment as possible, **removing most environment variables** and with current working directory set to /.
As the bold part said. Is there any way to solve this problem? Because we always start c... | 2018-12-04T08:32:39 | 2018-12-04T08:32:53 | {} | NONE |
ClickHouse | ClickHouse | 443,197,975 | 3,705 | dfulachier | Thanks Nicolai for highlighting this ClickHouse syntax, it will help me.
But in my case, the query is generated by Tableau BI tool, so I cannot master it.
I will open a ticket on ClickHouse/ODBC GitHub, the transformation may be done by ODBC Driver, as it is a {fn ... } ODBC escape sequence.
Nevertheless, it may be ... | 2018-11-30T13:06:00 | 2018-11-30T13:06:00 | {} | NONE |
ClickHouse | ClickHouse | 442,070,914 | 3,679 | den-crane | > Ну изначально только пытался сконфигурировать только пользователя.
> После неудачи - добавил эту строчку отсюда - https://clickhouse.yandex/docs/ru/operations/access_rights/
Не нужна вам эта строчка, в чем была неудача?
> Т.е стоит описать всех пользователей в users.xml?
да нет, и так все будет работать (ка... | 2018-11-27T14:09:20 | 2018-11-27T14:09:20 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 442,444,538 | 3,607 | filimonov | > I know how to do it properly. And did use everything which you suggested. This would be a good advise, BTW, but the problem definitely in XML by itself.
I wasn't aware of your background, quite often newbies want to change something just because they don't know yet how to use the stuff which already exists. So @Fe... | 2018-11-28T13:22:05 | 2018-11-28T13:24:21 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 442,523,384 | 3,691 | alexey-milovidov | `SELECT sqrt(arraySum(arrayMap((x, y) -> (x - y) * (x - y), [1, 2, 3], [4, 5, 6])))` | 2018-11-28T17:00:46 | 2018-11-28T17:00:46 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 442,821,104 | 3,696 | KochetovNicolai | ClickHouse do not merge data from different partitions. Your table is partitioned by month. You can use [custom partitioning](https://clickhouse.yandex/docs/en/single/?query=Log#custom-partitioning-key) or manually select only rows with maximal version. | 2018-11-29T12:46:36 | 2018-11-29T12:46:36 | {} | MEMBER |
ClickHouse | ClickHouse | 443,063,858 | 3,698 | wuyanxing001 | version:
ClickHouse client version 18.14.13.
Connecting to localhost:9000.
Connected to ClickHouse server version 18.14.13 revision 54409.
i use DBeaver , SQL like this
SELECT
*
FROM
( SELECT
SFZ18, JYAQ, AJLBXS, AJBH, XM
FROM
test_qiankexinxi
WHERE XM LIKE '%张%'
OR SFZ18 LIKE '%张%'
... | 2018-11-30T02:02:42 | 2018-11-30T02:02:42 | {} | NONE |
ClickHouse | ClickHouse | 443,359,586 | 3,706 | alexey-milovidov | We must ensure that libc++ is linked statically in default configuration. | 2018-11-30T22:31:44 | 2018-11-30T22:31:51 | {} | MEMBER |
ClickHouse | ClickHouse | 443,363,951 | 3,683 | alexey-milovidov | Ok. Still waiting for a test and code comment. | 2018-11-30T22:52:14 | 2018-11-30T22:52:14 | {} | MEMBER |
ClickHouse | ClickHouse | 443,667,289 | 1,837 | blinkov | @sachinkademane can you provide any additional info? | 2018-12-03T10:48:01 | 2018-12-03T10:48:01 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 443,757,059 | 2,508 | TH-HA | May be putting a Buffer between Kafka and your target can solve your issue.
Buffer engine (https://clickhouse.yandex/docs/en/operations/table_engines/buffer/) allow to specify min/max time,rows and size before flushing to the target.
Regards
| 2018-12-03T15:47:10 | 2018-12-03T15:52:57 | {} | NONE |
ClickHouse | ClickHouse | 443,253,052 | 3,708 | amosbird | I'd like to share my two cents from several months ago. The following is a dialog with @alexey-milovidov
```
Amos Bird, [30.07.18 16:42]
Hi, I'm thinking of extending the groupArray function for a while. The idea is
to add an additional order by clause as a decorator to each groupArray function.
It's like a WI... | 2018-11-30T16:10:26 | 2018-11-30T16:10:26 | {
"+1": 1
} | COLLABORATOR |
ClickHouse | ClickHouse | 443,657,129 | 2,095 | blinkov | Sounds like it's fixed in #2100 | 2018-12-03T10:13:49 | 2018-12-03T10:13:49 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 443,658,108 | 2,034 | blinkov | @shahvishal8464 have you tried to reproduce this on fresh ClickHouse releases? | 2018-12-03T10:17:01 | 2018-12-03T10:17:01 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 443,675,171 | 2,658 | blinkov | Link related issues: #75 #1130 #1870 | 2018-12-03T11:16:25 | 2018-12-03T11:16:25 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 443,680,406 | 3,716 | blinkov | @tishkinis try upgrading to v18.14.17-stable | 2018-12-03T11:36:36 | 2018-12-03T11:36:36 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 444,033,622 | 3,741 | filimonov | Slow start problem is known, please see #2278
Usually, it happens if you have a lot of tables and/or partitions.
Some changes which should improve that situation are already in master and should be released in next versions (no concrete date yet, stay tuned).
Slow shutdown - the reason is 'graceful' finishing of ... | 2018-12-04T09:36:12 | 2018-12-04T09:36:12 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 443,107,232 | 3,483 | AlexanderJLiu | > Some notice if you run clickhouse-server as a system service you need to setup env in a way it will be visible for system service.
@filimonov I've setup env(SHARD and HOSTNAME) in `/etc/profile`, and using command `source /etc/profile` to make it effective. But they still invisible for clickhouse-server(system se... | 2018-11-30T06:40:06 | 2018-11-30T06:40:06 | {} | NONE |
ClickHouse | ClickHouse | 443,359,425 | 3,706 | alexey-milovidov | I have merged this PR but without "Fix presence of both libc++ and -static-libstdc++ link flags". | 2018-11-30T22:31:10 | 2018-11-30T22:31:10 | {} | MEMBER |
ClickHouse | ClickHouse | 443,959,403 | 3,598 | nicelulu | @alesapin
It looks like it is missing libreadline.so.6. | 2018-12-04T03:40:45 | 2018-12-04T10:46:59 | {} | CONTRIBUTOR |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.