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 | 338,501,813 | 1,387 | vavrusa | There's one outstanding problem with it when interpreting nested fields, for example if the schema looks like:
```
struct Message {
a @0 :A;
struct A {
b @0 :B;
struct B {
c @0 :UInt64;
}
}
}
```
Then the corresponding column name would be `a.b.c`, which Click... | 2017-10-22T19:17:07 | 2017-10-22T19:30:25 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 338,665,631 | 1,389 | filimonov | There is cpu effective lib for that: https://github.com/aklomp/base64
| 2017-10-23T13:51:29 | 2017-10-23T13:51:29 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 338,961,968 | 1,392 | ludv1x | > So the client should be schema-aware to parse the output
Yes, of course.
Plain CSV and TSV formats are not intended to be interpreted uniquely without schema info.
So, either you use plain formats and maintain the schema or you use complex JSON format with the built-in schema.
Also, there is an option to use `... | 2017-10-24T11:39:06 | 2017-10-24T11:39:14 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 339,007,290 | 1,259 | silviucpp | Hello guys,
Please feel free to close this in case is not an issue
Silviu | 2017-10-24T14:22:26 | 2017-10-24T14:22:26 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 339,145,495 | 1,399 | vavrusa | I think it's the same issue as https://github.com/pocoproject/poco/pull/1904 (the patch from @proller was reverted). | 2017-10-24T21:55:39 | 2017-10-24T21:55:39 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 339,149,169 | 956 | ipostelnik | I'm seeing a slightly different variation of this where table_name is defined as above:
```
SELECT count(),col_name,toUInt32(toDateTime(date)) as ts
FROM table_name
WHERE (zoneId = 56759608)
AND (date BETWEEN toDate('2017-04-27') AND toDate('2017-10-24'))
GROUP BY col_name,date
ORDER BY ts
DB::Exce... | 2017-10-24T22:07:12 | 2017-10-24T22:07:12 | {} | NONE |
ClickHouse | ClickHouse | 339,294,686 | 1,404 | silviucpp | <img width="1024" alt="ckhouse" src="https://user-images.githubusercontent.com/4814731/31995567-a2c5c862-b98d-11e7-82c5-066af6084f07.png">
Problem seems related to the tmp folder inside the distributed table. Once deleted node is starting properly.
Table definition is as follow:
```sql
CREATE TABLE f5c00ace47... | 2017-10-25T11:06:24 | 2017-10-25T11:11:11 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 339,447,785 | 1,404 | silviucpp | Hello @ludv1x I will try to come up with more details and debug closely . It's 3rd time when happens so for sure I can grab the data you need.
Silviu | 2017-10-25T19:42:03 | 2017-10-25T19:42:03 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 339,636,460 | 1,404 | silviucpp | Here you can find the steps to replicate (the tmp folder is created at first insert in the distributed table) and never deleted. Once server is restarted is never starting until you go and delete that folder:
```sql
CREATE database f5c00ace4777417f96b336d50a8c40f1;
CREATE TABLE test_nested5_src(
`created_at` D... | 2017-10-26T11:30:30 | 2017-10-26T15:51:09 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 340,411,846 | 1,423 | kratorman | Already fixed https://github.com/yandex/ClickHouse/commit/d8038cf7467596e7e9a209323fc0001d218c0257 | 2017-10-30T11:03:58 | 2017-10-30T11:03:58 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 340,778,593 | 208 | ztlpn | @Dimon70007 Включите настройку `send_progress_in_http_headers`. Пока не началась передача результата, ClickHouse будет отсылать заголовок вида
```
X-ClickHouse-Progress: {"read_rows":"7471104","read_bytes":"74711040","total_rows":"455532544"}
``` | 2017-10-31T14:26:03 | 2017-10-31T14:26:03 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 337,706,612 | 1,330 | vavrusa | @ludv1x Ah thanks. Can I see the test results when it finishes or is it internal? | 2017-10-18T19:45:51 | 2017-10-18T19:45:51 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 337,714,444 | 1,330 | ludv1x | @vavrusa You can see only final statistics, logs are not available publicly. | 2017-10-18T20:15:15 | 2017-10-18T20:15:15 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 339,239,240 | 1,402 | leoluk | @vavrusa
Thank you for implementing the Kafka engine! :) | 2017-10-25T07:24:38 | 2017-10-25T07:24:38 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 339,114,465 | 1,399 | vavrusa | https://github.com/pocoproject/poco/commit/1101439eb0 (@bobrik) | 2017-10-24T20:05:13 | 2017-10-24T20:05:13 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 339,284,803 | 820 | valyala | It looks like the issue has been fixed in the latest stable release `1.1.54304`. The previous stable release `1.1.54292` doesn't understand `ECDSA` certificates. | 2017-10-25T10:21:09 | 2017-10-25T10:21:09 | {} | NONE |
ClickHouse | ClickHouse | 339,404,451 | 1,405 | ludv1x | The problem was fixed in `ClickHouse 1.1.54304`.
BTW `Page Down` led to Segfault in `1.1.54292` also -) | 2017-10-25T17:19:49 | 2017-10-25T17:19:49 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 339,444,790 | 1,409 | alexey-milovidov | Alex @ztlpn is fixing right now. | 2017-10-25T19:31:24 | 2017-10-25T19:31:24 | {} | MEMBER |
ClickHouse | ClickHouse | 339,459,120 | 1,402 | silviucpp | This PR was merged. Bug should be closed | 2017-10-25T20:22:30 | 2017-10-25T20:22:30 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 337,795,980 | 1,367 | vavrusa | @ludv1x thanks Vitaliy, that was super helpful, I fixed the mistake. | 2017-10-19T04:25:32 | 2017-10-19T04:25:32 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 337,944,093 | 1,282 | ztlpn | #1314 | 2017-10-19T15:26:56 | 2017-10-19T15:26:56 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 338,982,831 | 1,392 | filimonov | Ok, that's clear. I'll look in JDBC driver source.
About quoting - for example MySQL doesn't have such a restrictions, and it easier to make correct syntax when inserting data.
It's not a priority, but may be such a 'loose' mode for inserting (with automatic typecasing) like in Mysql can be useful in some situ... | 2017-10-24T13:04:46 | 2017-10-24T13:05:26 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 339,966,019 | 1,404 | silviucpp | Can you try also with <internal_replication>true</internal_replication> (this is what my cluster it's using)
Let me know if using this you can replicate if not I will provide you my full config.
My cluster is defined as follow:
```xml
<clickhouse_remote_servers>
<cluster_local>
<shard>
<internal_... | 2017-10-27T13:06:35 | 2017-10-27T13:07:22 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 339,981,561 | 1,417 | f4llou7 | Тоже самое. Но, если сделать подзапрос отдельной mv и вложить в table_mv, то корретно посчитает
| 2017-10-27T14:07:13 | 2017-10-27T14:07:13 | {} | NONE |
ClickHouse | ClickHouse | 340,760,463 | 208 | Dimon70007 | Привет. Также интересует возможность получать информацию о ходе выполнения запроса через http. | 2017-10-31T13:23:21 | 2017-10-31T13:23:21 | {} | NONE |
ClickHouse | ClickHouse | 340,872,166 | 1,298 | silviucpp | @alexey-milovidov can confirm that this was fixed | 2017-10-31T18:58:19 | 2017-10-31T18:58:19 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 339,490,066 | 1,411 | vavrusa | Table was created on v1.1.54290 | 2017-10-25T22:17:14 | 2017-10-25T22:17:14 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 339,681,630 | 1,412 | alexey-milovidov | ... withBinarySuffix. | 2017-10-26T14:19:59 | 2017-10-26T14:19:59 | {} | MEMBER |
ClickHouse | ClickHouse | 339,721,110 | 1,411 | ipostelnik | Restarting the affected node corrected the problem. Still I would like to know how it got into bad state. | 2017-10-26T16:23:17 | 2017-10-26T16:23:17 | {} | NONE |
ClickHouse | ClickHouse | 339,961,721 | 1,404 | ludv1x | I can't reproduce the problem. I do the following steps:
```
CREATE database IF NOT EXISTS f5c00ace4777417f96b336d50a8c40f1;
USE f5c00ace4777417f96b336d50a8c40f1;
DROP TABLE IF EXISTS test_nested5_src;
DROP TABLE IF EXISTS test_nested5;
CREATE TABLE test_nested5_src(
`created_at` DateTime,
`partition` Dat... | 2017-10-27T12:47:18 | 2017-10-27T14:02:22 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 337,752,728 | 1,367 | ludv1x | @vavrusa
stateless.00155_merges
```
Received exception from server:
Code: 9. DB::Exception: Received from localhost:9000, ::1. DB::Exception: Sizes of columns doesn't match: d: 2, s: 0.
Received exception from server:
Code: 9. DB::Exception: Received from localhost:9000, ::1. DB::Exception: Sizes of columns doe... | 2017-10-18T23:03:55 | 2017-10-18T23:04:26 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 338,024,444 | 812 | ztlpn | @alex09x Which ClickHouse version do you use? Can you try with the latest stable (1.1.54292)?
PS. How is your issue related to the original one? The error is different... | 2017-10-19T20:13:57 | 2017-10-19T20:15:35 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 338,588,131 | 1,370 | bacher | Yes, different column name fixes this problem.
I disagree about special design. I especially use an alias for table name **`t`** and use column name with table alias name (**t.payments_amount**). I think If Clickhouse engine supports table aliases, It has to support them in all cases. | 2017-10-23T08:40:25 | 2017-10-23T08:40:40 | {} | NONE |
ClickHouse | ClickHouse | 339,702,814 | 1,409 | ztlpn | @vavrusa @bobrik Can you verify that the problem goes away with the latest master (ec963e33974958c3fba7259fad897105f779baef)? | 2017-10-26T15:24:21 | 2017-10-26T15:24:21 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 339,973,433 | 1,415 | StarWix | Also this case doesn't work:
```
CREATE TABLE test
(
x Nullable(Int32)
) ENGINE = Log
INSERT INTO test VALUES(1), (0), (null)
SELECT * FROM test
┌──x─┐
│ 1 │
│ 0 │
│ \N │
└────┘
SELECT if(x != 0, modulo(10, x), null) FROM test
Received exception from server:
Code: 153. DB::Exception: Receiv... | 2017-10-27T13:36:46 | 2017-10-27T13:54:41 | {} | NONE |
ClickHouse | ClickHouse | 340,476,731 | 1,421 | ztlpn | Great suggestion. I'll try to improve it a bit. | 2017-10-30T15:16:59 | 2017-10-30T15:16:59 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 340,884,102 | 1,298 | alexey-milovidov | Now we have support for timezones with fractional offset from UTC.
* Known minor issue: if you have timezone with fractional offset, and the fraction was changed sometimes (there are not much: Lord Howe Island, Pacific/Pitcairn), then the function `toRelativeHourNum` will give incorrect result.
We still don't hav... | 2017-10-31T19:41:19 | 2017-10-31T19:41:19 | {} | MEMBER |
ClickHouse | ClickHouse | 338,052,931 | 812 | gnauq3mg6m64rs2n | @ztlpn
Connected to ClickHouse server version 1.1.54231 | 2017-10-19T22:13:45 | 2017-10-19T22:13:45 | {} | NONE |
ClickHouse | ClickHouse | 338,532,931 | 1,376 | domsia | Thank you for solving the problem. I have started my ch server by hands. I deleted the table's metadata SQL and started the server. After that, I set the `experimental_merge_tree_allow_custom_partitions` and attached table by hands. It works too. | 2017-10-23T02:37:25 | 2017-10-23T02:37:25 | {} | NONE |
ClickHouse | ClickHouse | 339,097,313 | 1,397 | alexey-milovidov | Unclear. | 2017-10-24T19:03:45 | 2017-10-24T19:03:45 | {} | MEMBER |
ClickHouse | ClickHouse | 339,124,906 | 1,400 | PsychodelEKS | `visitParamExtractInt` работает как надо, кстати. | 2017-10-24T20:43:10 | 2017-10-24T20:43:10 | {} | NONE |
ClickHouse | ClickHouse | 340,067,425 | 1,416 | ludv1x | Operator `notEqual` returns Nullable(UInt8) type
```
:) SELECT x, x != 0, isNull(x != 0) FROM test
SELECT
x,
x != 0,
isNull(x != 0)
FROM test
┌──x─┬─notEquals(x, 0)─┬─isNull(notEquals(x, 0))─┐
│ 1 │ 1 │ 0 │
│ 0 │ 0 │ ... | 2017-10-27T19:43:26 | 2017-10-27T19:43:57 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 337,756,669 | 1,367 | vavrusa | @ludv1x is there a script to run these manually or as part of the build? | 2017-10-18T23:28:27 | 2017-10-18T23:28:27 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 338,162,833 | 812 | filimonov | @alex09x did you saw that: https://groups.google.com/d/msg/clickhouse/bZSSvV1O5lA/NTCdV1dgBAAJ ?
BTW, if it's the same case as described in google group - probably better error informing could be useful. | 2017-10-20T09:59:29 | 2017-10-20T10:00:35 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 338,431,346 | 1,383 | vavrusa | ```
$ /usr/local/bin/python2 ../dbms/tests/clickhouse-test -b "$(pwd)/clickhouse-server" -c "$(pwd)/clickhouse-client" --no-zookeeper --no-shard -q $(pwd)/../dbms/tests/queries "(material|creat|attach)"
Running stateless tests.
00508_materialized_view_to: [ OK ]
00472_... | 2017-10-21T20:48:15 | 2017-10-21T20:48:15 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 338,621,045 | 1,365 | ludv1x | Strange behaviour detected https://paste.yandex-team.ru/312056
It looks like it ignores `return false;` | 2017-10-23T10:48:50 | 2017-10-23T10:48:50 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 338,741,255 | 1,384 | alexey-milovidov | We expect it in next release in 1..3 weeks. | 2017-10-23T17:48:10 | 2017-10-23T17:48:10 | {} | MEMBER |
ClickHouse | ClickHouse | 339,007,066 | 1,260 | silviucpp | Hello guys,
Please feel free to close this in case is not an issue
Silviu | 2017-10-24T14:21:46 | 2017-10-24T14:21:46 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 339,594,786 | 1,257 | felixendres | I have never used clickhouse, just found this issue in a web search, but for what it's worth:
I get
> src/tcmalloc.cc:281] Attempt to realloc invalid pointer 0xbaa900
> Fatal error 6: Aborted
when starting _emacs_ with LD_PRELOAD=/usr/lib/libtcmalloc_minimal.so.4.1.2
This suggests it's a bug in tcmalloc, howeve... | 2017-10-26T08:41:01 | 2017-10-26T08:41:22 | {} | NONE |
ClickHouse | ClickHouse | 337,702,122 | 1,367 | ludv1x | autotests | 2017-10-18T19:28:39 | 2017-10-18T19:28:39 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 337,716,925 | 1,367 | vavrusa | @ludv1x can you repost which tests have failed? | 2017-10-18T20:24:53 | 2017-10-18T20:24:53 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 338,544,537 | 1,370 | domsia | @Bacher I think this is a special design in CH that different from Mysql etc. Fields in **WHERE** that was from results table columns. This makes query pithier. Also, this made some problems. Specify a different table column name is the easiest way to solve the problem. | 2017-10-23T04:24:19 | 2017-10-23T04:24:19 | {} | NONE |
ClickHouse | ClickHouse | 339,102,749 | 1,365 | alexey-milovidov | We decided to postpone, because it is relevant only on totally broken network (which require NOC attention), and the modification will lead to busy loop and delay of replica repair. | 2017-10-24T19:23:13 | 2017-10-24T19:24:24 | {} | MEMBER |
ClickHouse | ClickHouse | 339,729,038 | 1,399 | vavrusa | @proller We had this error when testing v1.1.54300-testing (which had the libpoco from git submodules). But you merged in a new version 2 days ago (https://github.com/yandex/ClickHouse/commit/00c995bffce75c3ba2ac71459494ac8b96b27abb) which contains https://github.com/ClickHouse-Extras/poco/commit/3fb4c34161d5cf0c3ac1ca... | 2017-10-26T16:50:31 | 2017-10-26T16:50:31 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 337,703,878 | 1,330 | ludv1x | @vavrusa Yes, we have CI and autotests for each PR, but they are run only for people from the whitelist.
So, you are not in the whitelist and your PRs are not tested.
I think we have to add you into it-)
I made special comment here to force tests running https://github.com/yandex/ClickHouse/pull/1367#issuecomment-... | 2017-10-18T19:35:28 | 2017-10-18T19:35:28 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 338,757,133 | 1,387 | vavrusa | Thanks, having first class support for nested types would be the best, instead of relying on delimiters in column name. | 2017-10-23T18:41:35 | 2017-10-23T18:41:35 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 338,992,421 | 1,028 | silviucpp | Hello,
This function `groupUniqArray` seems also not working for fields of type : Array(Nullable(Int64)).
I receive an error as well:
Code: 48, e.displayText() = DB::Exception: Method getDataAt is not supported for ColumnNullable(ColumnVector<Int64>)
Is this known issue ? | 2017-10-24T13:36:46 | 2017-10-24T13:36:46 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 340,213,708 | 1,420 | filimonov | you've messed something in the steps to reproduce. You provided create statement for table
metrika_hits_local (incorrect one, it has two BrowserEngine columns). And you are trying to insert into table metrika_visits_local with other fields.
That quoting rules are not too clear, but you can always look on the click... | 2017-10-28T19:19:02 | 2017-10-29T06:47:15 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 340,458,855 | 1,361 | proller | Tried on 245 and current version with this test:
```
#!/usr/bin/env bash
clickhouse-client -q 'DROP TABLE IF EXISTS test.oa_copyright_music_scid_y_local;'
clickhouse-client -q 'DROP TABLE IF EXISTS test.oa_copyright_music_scid_y;'
clickhouse-client -q 'DROP TABLE IF EXISTS test.oa_copyright_music_scid_d;'
cli... | 2017-10-30T14:22:16 | 2017-10-30T14:22:27 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 340,850,106 | 208 | ztlpn | Это пользовательская настройка (https://clickhouse.yandex/docs/ru/operations/settings/index.html). Их часто путают с параметрами, задаваемыми в config.xml (наверно, добавим проверку, что они не перепутаны).
В вашем случае достаточно добавить `&send_progress_in_http_headers=1` в HTTP-запрос и должно заработать. Или н... | 2017-10-31T18:00:54 | 2017-10-31T18:01:10 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 338,946,634 | 1,368 | ludv1x | We have fixed SIGSEGV related to using of distributed DDL in 1.1.54289.
Could you check this behaviour in the newest version?
Also, could you provide stacktrace with SIGSEGV from `clickhouse-server.log`, please (it starts with `###`)? | 2017-10-24T10:30:02 | 2017-10-24T10:30:02 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 340,534,570 | 167 | alexey-milovidov | In first step, we will implement "batch updates", probably under a special syntax like "ALTER ... UPDATE/DELETE", that will perform updates rather suboptimal but without any sacrifice of SELECT performance.
In second step, we will apply some techniques to support short-running updates with little (if any) impact on ... | 2017-10-30T18:09:22 | 2017-10-30T18:09:22 | {} | MEMBER |
ClickHouse | ClickHouse | 340,872,048 | 1,298 | silviucpp | This seems fixed in last master in commit 4236bcbfa2365faf569b12cde64e6bf62d17517a | 2017-10-31T18:57:55 | 2017-10-31T18:57:55 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 338,630,584 | 1,384 | tobad357 | @vavrusa when do you expect kafka support to land in a stable version of clickhouse?
Or would we need to build it ourselves?
Also is there any examples I can check? | 2017-10-23T11:34:13 | 2017-10-23T11:34:13 | {} | NONE |
ClickHouse | ClickHouse | 338,964,932 | 1,393 | ludv1x | @YiuRULE Could you look at this problem, please? | 2017-10-24T11:52:21 | 2017-10-24T11:52:21 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 340,275,982 | 1,420 | Slach | i'm update my gist
https://gist.github.com/Slach/58bd50dd4f7789e981fa3cfb7cdea1fe
and error still reproduce
i add Vagrantfile
just run following command for reproduce
```
git clone https://gist.github.com/Slach/58bd50dd4f7789e981fa3cfb7cdea1fe
cd 58bd50dd4f7789e981fa3cfb7cdea1fe
vagrant up --provision
``` | 2017-10-29T16:44:01 | 2017-10-29T16:44:01 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 340,285,711 | 1,420 | filimonov | Yes. With that source file i get the error - reproducable without vagrant and virtualization.
But, it's not only ClickHouse see the error in that position of your input file.
You have there something like that:
78 Russia [] [] 0 [] [,] []
If you will replace [,] to ['',''] it will work. ;-) | 2017-10-29T19:01:46 | 2017-10-29T19:01:46 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 337,858,284 | 1,367 | ludv1x | > is there a script to run these manually or as part of the build?
`./release` the script only builds a deb package, so you should run `./clickhouse-test` manually. | 2017-10-19T09:51:26 | 2017-10-19T09:51:26 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 338,638,958 | 1,371 | silviucpp | Good catch for me on osx if I compile in debug always crashes in tryGet. If I apply this patch is fine | 2017-10-23T12:11:50 | 2017-10-23T12:11:50 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 338,646,786 | 812 | gnauq3mg6m64rs2n | @ztlpn
ClickHouse server version 1.1.54292
2017.10.23 15:41:58.829718 [ 4 ] <Error> void DB::BackgroundProcessingPool::threadFunction(): Code: 173, e.displayText() = DB::Exception: Allocator: Cannot mmap., errno: 12, strerror: Cannot allocate memory: (while reading column _value): (while reading from part /data/... | 2017-10-23T12:43:55 | 2017-10-23T12:43:55 | {} | NONE |
ClickHouse | ClickHouse | 338,741,240 | 1,384 | vavrusa | @tobad357 I don't know really. It should get to a stable release tag eventually.
@leoluk Can you create a separate ticket for it? | 2017-10-23T17:48:08 | 2017-10-23T17:48:08 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 339,403,183 | 1,404 | ludv1x | Hi!
Somebody has already reported the same problem.
Unfortunately, in that case, we were not able to investigate the problem due to lack of data, so your information will be helpful for the investigation.
Could you show the creation time of `tmp` directory?
Is this time similar to the shutdown or startup time of ... | 2017-10-25T17:15:35 | 2017-10-25T17:15:35 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 339,454,089 | 1,409 | ztlpn | #1410 | 2017-10-25T20:04:18 | 2017-10-25T20:04:18 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 339,830,208 | 1,414 | alexey-milovidov | This is dirty and could lead to silent build of incorrect code. | 2017-10-26T23:33:26 | 2017-10-26T23:33:26 | {} | MEMBER |
ClickHouse | ClickHouse | 340,871,871 | 1,225 | silviucpp | This seems fixed in last master in commit 4236bcbfa2365faf569b12cde64e6bf62d17517a | 2017-10-31T18:57:17 | 2017-10-31T18:57:17 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 337,955,048 | 1,362 | ztlpn | Yes, this is a duplicate of #1282
Please open a separate issue concerning the default values. | 2017-10-19T16:01:18 | 2017-10-19T16:01:18 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 338,020,560 | 360 | silviucpp | Hello,
The last master suppose to compile on Mac OS ? I have the same problem that @aadim faced in the past and which I see was fixed in the past:
ls -lah /usr/local/include/sql*
```h
lrwxr-xr-x 1 silviu.caragea admin 38B Jun 27 2016 /usr/local/include/sql.h -> ../Cellar/unixodbc/2.3.4/include/sql.h
lr... | 2017-10-19T20:00:35 | 2017-10-19T20:00:35 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 338,690,875 | 1,371 | alexey-milovidov | https://github.com/yandex/ClickHouse/commit/d77686685a56dcc5ac9c6874bdfbc1c590b32b2a | 2017-10-23T15:09:27 | 2017-10-23T15:09:27 | {} | MEMBER |
ClickHouse | ClickHouse | 338,843,287 | 1,389 | IvanJobs | I think it's good to consider this suggestion. | 2017-10-24T01:25:19 | 2017-10-24T01:25:19 | {} | NONE |
ClickHouse | ClickHouse | 339,590,874 | 1,402 | leoluk | No, this is a separate feature request. | 2017-10-26T08:26:39 | 2017-10-26T08:26:39 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 339,726,660 | 1,399 | proller | Not same.
all old clickhouse's including 1.1.54304 have very old poco in contrib.
Place from my patch already rewritten in poco and updated in submodule in current clickhouse master.
So it never been released. | 2017-10-26T16:42:17 | 2017-10-26T16:42:17 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 339,743,623 | 822 | 3manuek | @dorantor, I think it will be better for visibility to open a new issue as this is not directly related with the reported bug. If I'm not mistaken, there must be work being done to fix these things in long term.
Yeah, unfortunately you need to recreate the table with the new options :( | 2017-10-26T17:42:36 | 2017-10-26T17:42:36 | {} | NONE |
ClickHouse | ClickHouse | 340,237,511 | 1,420 | Slach | i update CREATE TABLE on my gist
> That quoting rules are not too clear, but you can always look on the clickhouse output to undersatnd what he expect as an input.
hmm ;) but clickhouse-client output look like [''] for empty Array(String) field
pls see https://gist.github.com/Slach/58bd50dd4f7789e981fa3cfb7cde... | 2017-10-29T04:50:04 | 2017-10-29T04:50:15 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 340,241,367 | 1,420 | filimonov | > but clickhouse-client output look like [''] for empty Array(String) field
Yes, you're right, sorry.
Now the table definition is ok, but i can't reproduce the bug. Tryed 1.1.54289 and 1.1.54304 and both work correctly with your sample. :-/
Could you try to reproduce it once again in your environment accura... | 2017-10-29T06:46:04 | 2017-10-29T06:56:22 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 340,500,357 | 167 | meshpaul | Is there any update on this feature... is this 2018 or will it make it in 2017 Q4.. We are looking for limited capability example would be a look up table but to update just the record without drop, rename whole table data. | 2017-10-30T16:25:06 | 2017-10-30T16:25:06 | {} | NONE |
ClickHouse | ClickHouse | 340,709,185 | 79 | sunsingerus | Hello!
It looks like Data type is still have issues with default values while reading from CSV file.
Have the following situation right now:
1. Log table with the following fields
```
date_1 Nullable(Date)
datetime_1 Nullable(DateTime)
timestamp_1 Nullable(DateTime)`
```
CSV file
```
date_1,datetime_1,ti... | 2017-10-31T09:38:46 | 2017-10-31T09:38:46 | {} | NONE |
ClickHouse | ClickHouse | 338,072,976 | 1,367 | vavrusa | Brilliant, this is great, thanks @ludv1x
```
$ /usr/local/bin/python2 ../dbms/tests/clickhouse-test -b "$(pwd)/clickhouse-server" -c "$(pwd)/clickhouse-client" --no-zookeeper --no-shard -q $(pwd)/../dbms/tests/queries "(sum|merges)"
Running stateless tests.
00507_sumwithoverflow: ... | 2017-10-20T00:22:12 | 2017-10-20T00:22:12 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 338,682,188 | 1,371 | alexey-milovidov | The standard doesn't guarantee that we can write into vector's buffer past the size but before capacity.
Obviously it works fine. But I doubt, it can introduce difficulties, for example, when running under UBSan.
Though resize is less efficient because it does useless memset. And 1 MB memset is not small... Maybe w... | 2017-10-23T14:43:29 | 2017-10-23T14:44:04 | {} | MEMBER |
ClickHouse | ClickHouse | 338,740,324 | 1,371 | vavrusa | Thanks @alexey-milovidov ! | 2017-10-23T17:44:54 | 2017-10-23T17:44:54 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 338,755,239 | 1,387 | alexey-milovidov | `ultimately implement Nested type as a proper type instead of recovering it from column name`
I have a branch with better support of serde of non-trivial data types:
https://github.com/yandex/ClickHouse/compare/unify-data-types-that-serialized-with-multiple-streams#diff-40ce183fab5c903c45be0c3dc6d28540R76
It wil... | 2017-10-23T18:35:08 | 2017-10-23T18:35:08 | {} | MEMBER |
ClickHouse | ClickHouse | 338,948,728 | 1,369 | ludv1x | Confirmed.
There is no any reason not to check it.
It has to be fixed. | 2017-10-24T10:39:12 | 2017-10-24T10:39:12 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 339,036,740 | 1,367 | vavrusa | @alexey-milovidov can you take a look? This should fix autotests again. | 2017-10-24T15:47:39 | 2017-10-24T15:47:39 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 338,648,822 | 1,384 | leoluk | Edit:
-> #1402 | 2017-10-23T12:52:08 | 2017-10-25T07:23:06 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 338,172,642 | 1,376 | ztlpn | Hi @domsia ! First, I must repeat the warning found in the source code: custom partitions are not ready for production yet! The data and metadata formats will change in the future and the future server versions WILL NOT be able to load tables created with the current version.
With that in mind, I would be very grate... | 2017-10-20T10:45:18 | 2017-10-20T10:45:18 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 338,442,682 | 1,287 | vavrusa | So my understanding is that only ALTER queries for replicated tables are problematic, since some require you to execute the query only on leader replica, and some require you to execute the query on any single replica.
For CREATE | ATTACH | DROP | DETACH it should be executed on distinct server instances regardless ... | 2017-10-22T00:52:40 | 2017-10-22T01:20:54 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 338,464,723 | 1,385 | alexey-milovidov | Можно полностью убрать ссылку на старую. И можешь перефразировать, так как слово "минималистичный" уже не актуально :) Также надо изменить в en версии документации. | 2017-10-22T09:44:19 | 2017-10-22T09:44:19 | {} | MEMBER |
ClickHouse | ClickHouse | 339,033,862 | 1,396 | ludv1x | Aliases have higher priority than column names.
At least MySQL does the same http://sqlfiddle.com/#!9/a6c585/17907 | 2017-10-24T15:39:00 | 2017-10-24T15:39:00 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 339,679,993 | 1,412 | ludv1x | Whats about `formatReadableSizeWithDecimalSuffix` ? | 2017-10-26T14:14:48 | 2017-10-26T14:14:48 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 339,830,432 | 1,414 | alexey-milovidov | Every unusual build scenarios should be enabled only explicitly.
Missing libraries, submodules, etc. should lead to fatal error with default settings. | 2017-10-26T23:34:46 | 2017-10-26T23:34:46 | {} | MEMBER |
ClickHouse | ClickHouse | 340,717,603 | 167 | meshpaul | @alexey-milovidov класс спасибо | 2017-10-31T10:13:24 | 2017-10-31T10:13:24 | {} | NONE |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.