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 | 771,037,244 | 15,742 | filimonov | Replicated
https://github.com/ClickHouse/ClickHouse/blob/4ee96869a2500fa3e75575c767631aa9fbfef732/src/Storages/StorageReplicatedMergeTree.cpp#L1889-L1913
vs non-replicated:
https://github.com/ClickHouse/ClickHouse/blob/8276a1c8d23190426cd34c4d5cf81c6fff56b098/src/Storages/StorageMergeTree.cpp#L1176-L1186 | 2021-02-01T17:51:36 | 2021-02-01T17:51:36 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 771,292,257 | 15,713 | ghost | @amosbird
Any news for the plan to develop the feature like `limit 5 by page 2 by browser 2 by os`?
If you don't mind I would like to develop the feature. | 2021-02-02T02:03:28 | 2021-02-02T02:04:04 | {} | NONE |
ClickHouse | ClickHouse | 771,353,002 | 19,958 | alexey-milovidov | > gcc version 4.8.5
We don't support gcc 4.8.5, the minimum version is 10. | 2021-02-02T04:36:33 | 2021-02-02T04:36:33 | {} | MEMBER |
ClickHouse | ClickHouse | 771,385,993 | 19,958 | leosunli | /root/shengjun/source/ClickHouse/build/CMakeFiles/CMakeError.log as follow:
1.8.2
[root@emr-dev-support011158132038 ninja-1.8.2]# cd ..
[root@emr-dev-support011158132038 shengjun]# cd source/ClickHouse/
[root@emr-dev-support011158132038 ClickHouse]# ll -rth
总用量 596K
-rw-r--r-- 1 root root 723 2月 1 10:54 PreLo... | 2021-02-02T05:59:27 | 2021-02-02T05:59:27 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 771,432,084 | 19,955 | gj-zhang | > Other caveats:
>
> * useless partitioning that leads to large number of partitions;
> * large number of tables with identical structure.
thanks
| 2021-02-02T07:33:22 | 2021-02-02T07:33:22 | {} | NONE |
ClickHouse | ClickHouse | 770,903,025 | 19,299 | alesapin | Failure in test keeper https://github.com/ClickHouse/ClickHouse/issues/18950.
| 2021-02-01T14:36:32 | 2021-02-01T14:36:32 | {} | MEMBER |
ClickHouse | ClickHouse | 770,918,861 | 19,473 | KochetovNicolai | Query: `SELECT (-2, 2), ccc FROM (SELECT NULL AS ccc UNION ALL SELECT * FROM (SELECT 2 AS ccc) ANY INNER JOIN (SELECT 7 AS ccc) USING (ccc)) WHERE ccc > NULL format TSV` (actually, it does not contain totals)
Graph.
 — fail: 1, passed: 2509, skipped: 230
flaky
>Integration tests (asan) — fail: 13, passed: 197, error: 9
Looks related, need to take a look | 2021-02-02T07:59:13 | 2021-02-02T07:59:13 | {} | MEMBER |
ClickHouse | ClickHouse | 771,454,389 | 19,770 | stigsb | @tavplubix is there anything else I need to do before this PR is ready for merge? | 2021-02-02T08:17:02 | 2021-02-02T08:17:02 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 770,993,788 | 19,920 | den-crane | docker: Docker version 17.12.0-ce, build c97c6d6
>time curl http://localhost:8123
is it curl inside container?
Do you have swap?
```
free -h
vmstat 1 10
``` | 2021-02-01T16:45:41 | 2021-02-01T16:45:41 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 771,031,896 | 19,907 | den-crane | it's a bug https://github.com/ClickHouse/ClickHouse/issues/16925 | 2021-02-01T17:42:48 | 2021-02-01T17:42:48 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 771,035,200 | 11,727 | azat | Actually it is still possible, due to some corruption (i.e. if server will be hard rebooted), but in recent version you can use fsync for [Distributed](https://clickhouse.tech/docs/en/engines/table-engines/special/distributed/) to avoid these:
```sql
Distributed(logs, default, hits[, sharding_key[, policy_name]])
... | 2021-02-01T17:48:02 | 2021-02-01T17:48:25 | {} | MEMBER |
ClickHouse | ClickHouse | 771,050,166 | 19,473 | chenziliang | This graph is very useful for triaging. May i ask if it is automatically generated or by manual ? Thanks | 2021-02-01T18:11:31 | 2021-02-01T18:11:31 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 767,613,030 | 18,342 | vitlibar | The test `01560_DateTime_and_DateTime64_comparision` failed:
```
2020-12-22 08:22:51 01560_DateTime_and_DateTime64_comparision: [ FAIL ] 0.24 sec. - result differs with reference:
2020-12-22 08:22:51 --- /usr/share/clickhouse-test/queries/0_stateless/01560_DateTime_and_DateTime64_compa... | 2021-01-26T15:17:46 | 2021-02-02T02:06:42 | {} | MEMBER |
ClickHouse | ClickHouse | 771,286,281 | 19,766 | ghost | It seems that the failed tests are not related to the PR.
1. test_distributed_ddl_parallel/test.py::test_all_in_parallel
> Code: 36. DB::Exception: Received from 172.18.0.6:9000. DB::Exception: There was an error on [n1:9000]: Code: 36, e.displayText() = DB::Exception: external dictionary 'slow_dict' not found (v... | 2021-02-02T01:46:51 | 2021-02-02T02:52:11 | {} | NONE |
ClickHouse | ClickHouse | 770,842,421 | 15,607 | buyology | I think this should be reopened: it seems unnecessary to me that you need to provide the password to the database you are already in. | 2021-02-01T13:05:09 | 2021-02-01T13:05:09 | {} | NONE |
ClickHouse | ClickHouse | 770,853,770 | 19,878 | alesapin | Hi!
Alter of the key columns in ClickHouse is quite limited. You can alter enums but only add new values. Removing old values can change the representation of the primary key.
```
:) ALTER TABLE t
MODIFY COLUMN `b` Enum8('a' = 1)
Received exception from server (version 21.2.1):
Code: 524. DB::Exception: ... | 2021-02-01T13:23:52 | 2021-02-01T13:23:52 | {} | MEMBER |
ClickHouse | ClickHouse | 770,879,790 | 19,878 | alexey-milovidov | > The same as on non-Replicated MergeTree
But how it's related to Replicated / non-Replicated MergeTree? | 2021-02-01T14:03:43 | 2021-02-01T14:03:43 | {} | MEMBER |
ClickHouse | ClickHouse | 771,021,583 | 12,481 | hanguyen6 | I no longer see the issue. Will reopen this ticket with stack trace if the issue happens again. Thanks | 2021-02-01T17:26:27 | 2021-02-01T17:26:27 | {} | NONE |
ClickHouse | ClickHouse | 771,267,638 | 19,919 | akuzm | The Arcadia error is unrelated to this PR. | 2021-02-02T00:53:50 | 2021-02-02T00:53:50 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 771,285,738 | 19,518 | bharatnc | > To run in CI - check https://github.com/ClickHouse/ClickHouse/blob/master/tests/performance/README.md
>
> But it's a bit problematic to collect the base line after the PR was merged, so you can just do smth like
>
> ```
> echo 'select toIPv6(...) from numbers(10000000)' | clickhouse-benchmark
> ```
Right, ... | 2021-02-02T01:45:11 | 2021-02-02T01:45:11 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 771,311,637 | 19,908 | vitlibar | `UNION DISTINCT` was not implemented in `20.12`, so the test failed:
```
2021-02-01 15:01:48 01680_predicate_pushdown_union_distinct_subquery: [ FAIL ] 0.07 sec. - return code 62
2021-02-01 15:01:48 Code: 62. DB::Exception: Syntax error: failed at position 50 ('DISTINCT'): DISTINCT SELECT 200... | 2021-02-02T02:50:02 | 2021-02-02T02:50:02 | {} | MEMBER |
ClickHouse | ClickHouse | 771,313,407 | 19,584 | alexey-milovidov | > AST fuzzer (UBSan) — ../src/Functions/DivisionUtils.h:103:40: runtime error: 9.22337e+22 is outside the r
#19876 | 2021-02-02T02:54:11 | 2021-02-02T02:54:11 | {} | MEMBER |
ClickHouse | ClickHouse | 771,432,186 | 19,969 | alexey-milovidov | What kind of integration do you mean? | 2021-02-02T07:33:33 | 2021-02-02T07:33:33 | {} | MEMBER |
ClickHouse | ClickHouse | 770,988,394 | 19,943 | den-crane | yes,
`select count(*) from table ` -- uses optimization and reads in-memory parts structure.
--optimize_trivial_count_query arg Process trivial 'SELECT count() FROM table' query from metadata.
`select * from table` -- also cached because CH relays on Linux file cache and does not... | 2021-02-01T16:37:55 | 2021-02-01T16:37:55 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 771,024,464 | 19,375 | alexey-milovidov | Yes, sure. | 2021-02-01T17:30:59 | 2021-02-01T17:30:59 | {} | MEMBER |
ClickHouse | ClickHouse | 771,377,043 | 19,473 | chenziliang | > @chenziliang this graph is automatically generated. You can get graphviz output using `explain pipeline graph = 1, compact = 0 select ...` (it won't have last output format, however).
Thanks a lot | 2021-02-02T05:38:56 | 2021-02-02T05:38:56 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 771,388,215 | 19,964 | alexey-milovidov | ```
mkdir build_coverity
cd build_coverity
CC=clang CXX=clang++ cmake -D ENABLE_CCACHE=0 ..
COVERITY_UNSUPPORTED_COMPILER_INVOCATION=1 cov-build --dir cov-int ninja clickhouse
``` | 2021-02-02T06:04:39 | 2021-02-02T06:04:39 | {} | MEMBER |
ClickHouse | ClickHouse | 771,414,652 | 19,964 | alexey-milovidov | ```
2021-02-02T06:58:25.017239Z|cov-build|2553417|info|> Build time (cov-build overall): 01:00:04.595690
2021-02-02T06:58:25.017239Z|cov-build|2553417|info|>
2021-02-02T06:58:25.055563Z|cov-build|2553417|info|> Build time (C/C++ emits total): 16:43:25.440091
2021-02-02T06:58:25.055563Z|cov-build|2553417|info|>
2... | 2021-02-02T07:00:04 | 2021-02-02T07:00:04 | {} | MEMBER |
ClickHouse | ClickHouse | 770,823,230 | 19,889 | KochetovNicolai | ```
$ addr2line -afiCe clickhouse.1 0x7f81ccd9c18b 0x7f81ccd7b859 0xe4407c4 0x1cea3b28 0x1cea3ab9 0xe1edc5b 0x19054469 0x1904e7ac 0x1aac5b52 0x1aac5189 0x1ad1eaf8 0x1ab11c79 0x1ab11b67 0x1ab10574 0x1ab0ee4a 0x1ab0e90b 0x1ab2130c 0x1ab21267 0x1ab21122 0xe242bff 0xe2468d6 0x7f81ccf51609 0x7f81cce78293
0x00007f81ccd9c18... | 2021-02-01T12:33:08 | 2021-02-01T12:33:08 | {} | MEMBER |
ClickHouse | ClickHouse | 771,013,832 | 16,616 | filimonov | Loosely coupled with https://github.com/ClickHouse/ClickHouse/issues/19228 (to load the dictionary data we may use either the account of DICTIONARY DEFINER either the current user). | 2021-02-01T17:14:41 | 2021-02-01T17:16:14 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 771,201,651 | 19,945 | hlinowes | It makes sense to me why the alias is useful/needed here, but I am confused why the query seems to succeed without aliasing but when the `ORDER BY` is removed. The following query succeeds:
```
select avg(abs(t.`score`)) as `sc`, t.`category`
from temp_table as t
ARRAY JOIN t.`score`, t.`category`
GROUP BY t.`cate... | 2021-02-01T22:23:33 | 2021-02-01T22:23:33 | {} | NONE |
ClickHouse | ClickHouse | 771,230,113 | 19,333 | vitlibar | Depends on the improvement https://github.com/ClickHouse/ClickHouse/pull/18896 which has not been merged to `20.12`. | 2021-02-01T23:22:09 | 2021-02-01T23:22:09 | {} | MEMBER |
ClickHouse | ClickHouse | 771,313,178 | 8,687 | zhuyoucui0909 | The workaround is not convenience for us to do auto sql generation.
Is there any plan that ClickHouse will support multiple array join officially? | 2021-02-02T02:53:49 | 2021-02-02T02:53:49 | {} | NONE |
ClickHouse | ClickHouse | 771,318,519 | 19,920 | gj-zhang | > docker: Docker version 17.12.0-ce, build c97c6d6
>
> > time curl http://localhost:8123
>
> is it curl inside container?
>
> Do you have swap?
>
> ```
> free -h
> vmstat 1 10
> ```
Question:
> is it curl inside container?
> outside
and I curl inside container slow too
```shell
[ck01 ~]$ free -h
... | 2021-02-02T03:07:09 | 2021-02-02T03:13:14 | {} | NONE |
ClickHouse | ClickHouse | 771,334,709 | 19,946 | alexey-milovidov | > @robot-clickhouse
Stress test (memory) — Sanitizer assert (in stderr.log)
#19957 | 2021-02-02T03:47:45 | 2021-02-02T03:47:45 | {} | MEMBER |
ClickHouse | ClickHouse | 771,335,606 | 19,584 | alexey-milovidov | Need to debug integration test: `test_dictionaries_dependency_xml`. | 2021-02-02T03:50:20 | 2021-02-02T03:50:20 | {} | MEMBER |
ClickHouse | ClickHouse | 771,429,864 | 19,955 | alexey-milovidov | Group INSERTs to larger batches to lower the amount of INSERT queries. | 2021-02-02T07:29:27 | 2021-02-02T07:29:27 | {} | MEMBER |
ClickHouse | ClickHouse | 770,891,453 | 19,785 | filimonov | May be we can just add smth like `ALTER TABLE MATERIALIZE COLUMN x` ? | 2021-02-01T14:19:54 | 2021-02-01T14:19:54 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 771,261,582 | 19,592 | olgarev | @CurtizJ Hi, will you review this, pls? | 2021-02-02T00:39:04 | 2021-02-02T00:39:04 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 771,337,438 | 13,962 | alexey-milovidov | @vitlibar Maybe you can help? | 2021-02-02T03:55:41 | 2021-02-02T03:55:41 | {} | MEMBER |
ClickHouse | ClickHouse | 771,388,017 | 18,508 | depressed-pho | Amended the patch so that the function rejects Nullable arguments. | 2021-02-02T06:04:13 | 2021-02-02T06:04:13 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 770,938,434 | 19,518 | rlazimi-dev | Will this function carry over the subnet mask from IPv4 to IPv6?
For example, the following statement:
> SELECT toIPv6('127.0.0.1/24')
returns:
> ::ffff:127.0.0.1/24 | 2021-02-01T15:25:49 | 2021-02-01T15:25:49 | {} | NONE |
ClickHouse | ClickHouse | 770,984,098 | 19,878 | Blackmorse | @alexey-milovidov
> But how it's related to Replicated / non-Replicated MergeTree?
Sorry, I was sure that the same operation works on simple MergeTree(). My bad | 2021-02-01T16:31:19 | 2021-02-01T16:31:19 | {} | NONE |
ClickHouse | ClickHouse | 771,104,126 | 19,473 | KochetovNicolai | @chenziliang this graph is automatically generated. You can get graphviz output using `explain pipeline graph = 1, compact = 0 select ...` (it won't have last output format, however). | 2021-02-01T19:34:51 | 2021-02-01T19:34:51 | {} | MEMBER |
ClickHouse | ClickHouse | 771,313,331 | 15,713 | amosbird | You need to describe the feature and the potential use cases. And you can also attach alternative implementations and list the pros/cons. It's not a formal process but it'll be appreciated to have the proposal well thought out. | 2021-02-02T02:54:01 | 2021-02-02T02:54:01 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 771,380,016 | 19,646 | gj-zhang | > Is your table Replicated ?
> What CH version?
yes
20.9.3.45 | 2021-02-02T05:45:45 | 2021-02-02T05:45:45 | {} | NONE |
ClickHouse | ClickHouse | 770,890,837 | 19,785 | filimonov | +1 often needed to backfill the materialized column.
That workaround may affect inserts (if the list of columns is not listed explicitly). | 2021-02-01T14:19:03 | 2021-02-01T14:19:03 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 771,017,545 | 4,910 | filimonov | @kitaisreal fixed? | 2021-02-01T17:20:29 | 2021-02-01T17:20:29 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 771,424,162 | 19,964 | alexey-milovidov | `Build successfully submitted.` | 2021-02-02T07:18:50 | 2021-02-02T07:18:50 | {} | MEMBER |
ClickHouse | ClickHouse | 771,439,405 | 19,580 | alesapin | Wow, test is flaky. node2 lost one part (`all_2_2_0` inserted on `node3`):
```
> assert node2.query("SELECT COUNT() FROM t2") == "510\n"
E AssertionError: assert '410\n' == '510\n'
E - 510
E ? ^
E + 410
E ? ^
```
For some reason, there are no messages about this part... | 2021-02-02T07:48:10 | 2021-02-02T07:52:30 | {} | MEMBER |
ClickHouse | ClickHouse | 771,220,771 | 19,949 | den-crane | the same with float
```
select JSONExtract('{"a":[3,3.14]}','a','Array(float)'),
JSONExtract('{"a":[3,3.14]}','a','Array(double)');
┌─JSONExtract('{"a":[3,3.14]}', 'a', 'Array(float)')─┬─JSONExtract('{"a":[3,3.14]}', 'a', 'Array(double)')─┐
│ [3,0] │ [3,3.14] ... | 2021-02-01T23:00:14 | 2021-02-01T23:00:14 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 771,241,585 | 19,519 | olgarev | @CurtizJ Hi, thank you for the review, I've made the changes. OK now? | 2021-02-01T23:50:04 | 2021-02-01T23:50:04 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 771,312,167 | 19,896 | depressed-pho | Updated my patch. It's now named `Encrypted(cipher)` where `cipher` has to be a literal `'AES-128-GCM-SIV'` for now. The code is guarded by `#if USE_SSL`, and has a stateless SQL test. | 2021-02-02T02:51:16 | 2021-02-02T02:51:38 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 771,336,734 | 19,952 | alexey-milovidov | Thank you! I have committed the fix to master. | 2021-02-02T03:53:40 | 2021-02-02T03:53:40 | {} | MEMBER |
ClickHouse | ClickHouse | 771,381,967 | 19,961 | yiguolei | Great feature !!! | 2021-02-02T05:50:02 | 2021-02-02T05:50:02 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 771,430,569 | 19,955 | alexey-milovidov | Other caveats:
- useless partitioning that leads to large number of partitions;
- large number of tables with identical structure. | 2021-02-02T07:30:44 | 2021-02-02T07:30:44 | {} | MEMBER |
ClickHouse | ClickHouse | 771,440,919 | 19,967 | sundy-li | ok, I'll fix that soon. | 2021-02-02T07:51:21 | 2021-02-02T07:51:21 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 770,869,295 | 15,450 | alesapin | All failures are known. | 2021-02-01T13:48:03 | 2021-02-01T13:48:03 | {} | MEMBER |
ClickHouse | ClickHouse | 770,928,899 | 19,893 | vitlibar | The bug was introduced in https://github.com/ClickHouse/ClickHouse/pull/17359, which hasn't been merged to `20.12` | 2021-02-01T15:13:15 | 2021-02-01T15:13:15 | {} | MEMBER |
ClickHouse | ClickHouse | 771,014,192 | 19,875 | alexey-milovidov | @azat FYI this appeared after a change that `MemoryTracker::BlockerInThread` stopped to block total memory tracker. | 2021-02-01T17:15:14 | 2021-02-01T17:15:14 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 771,118,083 | 19,518 | filimonov | To run in CI - check https://github.com/ClickHouse/ClickHouse/blob/master/tests/performance/README.md
But it's a bit problematic to collect the base line after the PR was merged, so you can just do smth like
```
echo 'select toIPv6(...) from numbers(10000000)' | clickhouse-benchmark
``` | 2021-02-01T19:58:11 | 2021-02-01T19:58:11 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 771,221,003 | 19,949 | den-crane | duplicate https://github.com/ClickHouse/ClickHouse/issues/13962 | 2021-02-01T23:00:47 | 2021-02-01T23:00:47 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 771,311,989 | 19,909 | vitlibar | `UNION DISTINCT` was not implemented in `20.11`, so the test failed:
```
2021-02-01 15:01:48 01680_predicate_pushdown_union_distinct_subquery: [ FAIL ] 0.07 sec. - return code 62
2021-02-01 15:01:48 Code: 62. DB::Exception: Syntax error: failed at position 50 ('DISTINCT'): DISTINCT SELECT 200... | 2021-02-02T02:50:49 | 2021-02-02T02:50:49 | {} | MEMBER |
ClickHouse | ClickHouse | 771,338,918 | 13,962 | alexey-milovidov | The reason is quite stupid: 0.1 in JSON is treat as double and Float64(0.1) is not representable in Float32. | 2021-02-02T04:00:05 | 2021-02-02T04:00:05 | {} | MEMBER |
ClickHouse | ClickHouse | 771,386,273 | 19,958 | leosunli | [root@emr-dev-support011158132038 ClickHouse]# gcc -v
使用内建 specs。
COLLECT_GCC=gcc
COLLECT_LTO_WRAPPER=/usr/local/libexec/gcc/x86_64-pc-linux-gnu/9.1.0/lto-wrapper
目标：x86_64-pc-linux-gnu
配置为：../configure -enable-checking=release -enable-languages=c,c++ -disable-multilib
线程模型：posix
gcc 版本 9.1.0 (GCC)
Thanks @alex... | 2021-02-02T06:00:10 | 2021-02-02T06:17:23 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 771,219,119 | 19,331 | vitlibar | Depends on the improvement https://github.com/ClickHouse/ClickHouse/pull/18896 which has not been merged to `20.12`. | 2021-02-01T22:56:27 | 2021-02-01T22:56:27 | {} | MEMBER |
ClickHouse | ClickHouse | 771,300,033 | 15,713 | amosbird | > If you don't mind I would like to develop the feature.
That's great! However, I didn't think about this feature thoroughly. Perhaps you can propose the feature first and here some feadbacks from the community. | 2021-02-02T02:20:45 | 2021-02-02T02:20:45 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 771,369,545 | 19,897 | alexey-milovidov | Is it possible that it is somehow related to `ccache`?
But `.bc` files are generated - it means that the build is at least not a no-op. | 2021-02-02T05:19:32 | 2021-02-02T05:19:32 | {} | MEMBER |
ClickHouse | ClickHouse | 771,456,777 | 19,896 | depressed-pho | Amended it again to fix a build failure occuring when !USE_SSL. | 2021-02-02T08:21:16 | 2021-02-02T08:21:16 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 770,988,821 | 19,890 | vitlibar | The bug was introduced in https://github.com/ClickHouse/ClickHouse/pull/17359, which hasn't been merged to `20.11` | 2021-02-01T16:38:26 | 2021-02-01T17:15:05 | {} | MEMBER |
ClickHouse | ClickHouse | 771,018,805 | 4,910 | kitaisreal | @filimonov currently supported for Flat, Hashed, ComplexKeyHashed, Direct, ComplexKeyDirect, RangeHashed.
Cache, ComplexKeyCache, SSDCache, SSDComplexKeyCache, Polygon, IPTrie are not yet supported. | 2021-02-01T17:22:24 | 2021-02-01T17:22:24 | {
"heart": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 771,032,703 | 19,907 | den-crane | Do you have a datasample / steps to reproduce it? | 2021-02-01T17:44:09 | 2021-02-01T17:44:09 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 771,300,704 | 19,943 | nautaa | So if I don't want to use the cache for testing disk io, can I use it `--optimize_trivial_count_query=0 --min_bytes_to_use_direct_io=1`, and why it is faster? @den-crane | 2021-02-02T02:22:06 | 2021-02-02T02:22:06 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 771,340,408 | 19,645 | JonasBollack | > 20.5.4.40 is obsolete.
>
> Please check against 21.1.2.15
>
> Please provide `show create table "A1"."B"` and `show create table "A1"."A" `
> 20.5.4.40 is obsolete.
>
> Please check against 21.1.2.15
>
> Please provide `show create table "A1"."B"` and `show create table "A1"."A" `
please help me... | 2021-02-02T04:03:46 | 2021-02-02T04:03:46 | {} | NONE |
ClickHouse | ClickHouse | 771,340,444 | 13,962 | alexey-milovidov | Also note that we do accurate conversions to integers. Example: `JSONExtract('{"v":1.1}', 'v', 'UInt64')` will return 0 and `JSONExtract('{"v":1.1}', 'v', 'Nullable(UInt64)')` will return NULL. | 2021-02-02T04:03:53 | 2021-02-02T04:03:53 | {} | MEMBER |
ClickHouse | ClickHouse | 771,415,878 | 19,964 | alexey-milovidov | ```
$ du -hc cov-int
5,7G cov-int
$ tar czvf clickhouse.tgz cov-int
``` | 2021-02-02T07:02:29 | 2021-02-02T07:02:29 | {} | MEMBER |
ClickHouse | ClickHouse | 771,420,747 | 19,960 | alexey-milovidov | > AST fuzzer (MSan) — SUMMARY: MemorySanitizer: use-of-uninitialized-value (/workspace/clickhouse+0x8d6ec9e) in memcmp
https://github.com/ClickHouse/ClickHouse/issues/19967 | 2021-02-02T07:12:10 | 2021-02-02T07:12:10 | {} | MEMBER |
ClickHouse | ClickHouse | 771,429,269 | 19,958 | alexey-milovidov | Version 20.3-lts is very old, the support will end in a month. I recommend to simply use version 21.1 or 20.8-lts (worse). | 2021-02-02T07:28:18 | 2021-02-02T07:28:18 | {} | MEMBER |
ClickHouse | ClickHouse | 771,444,207 | 19,584 | azat | >Need to debug integration test: test_dictionaries_dependency_xml.
Not related, but should be addressed by #19971 (that PR also includes https://github.com/ClickHouse/ClickHouse/pull/19584#issuecomment-771024679) | 2021-02-02T07:57:27 | 2021-02-02T07:58:08 | {} | MEMBER |
ClickHouse | ClickHouse | 770,834,641 | 19,889 | KochetovNicolai | ```
(gdb) disas 'DB::IDatabase::tryGetCreateTableQuery(std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > const&, DB::Context const&) const'
Dump of assembler code for function DB::IDatabase::tryGetCreateTableQuery(std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::... | 2021-02-01T12:52:32 | 2021-02-01T12:52:32 | {} | MEMBER |
ClickHouse | ClickHouse | 770,836,041 | 19,889 | KochetovNicolai | Hm, that's unlikely:
```cpp
void __ubsan::__ubsan_handle_type_mismatch_v1_abort(TypeMismatchData *Data, ValueHandle Pointer) {
GET_REPORT_OPTIONS(true);
handleTypeMismatchImpl(Data, Pointer, Opts);
Die();
}
``` | 2021-02-01T12:54:55 | 2021-02-01T12:54:55 | {} | MEMBER |
ClickHouse | ClickHouse | 770,900,618 | 18,950 | alesapin | ```
2021.02.01 00:28:15.659125 [ 9 ] {} <Debug> TestKeeperTCPHandler: Received close event with xid 1 for session id #0
2021.02.01 00:28:15.659282 [ 9 ] {} <Debug> TestKeeperTCPHandler: Session #0 successfully closed
2021.02.01 00:28:15.757905 [ 9 ] {} <Trace> TestKeeperTCPHandlerFactory: Test keeper request. Addres... | 2021-02-01T14:33:04 | 2021-02-01T14:37:43 | {} | MEMBER |
ClickHouse | ClickHouse | 770,981,207 | 19,878 | Blackmorse | @alesapin ,
> You can alter enums but only add new values
But documentation says:
> The Enum type can be changed without cost using ALTER, if only the set of values is changed. It is possible to both add and remove members of the Enum using ALTER (removing is safe only if the removed value has never been use... | 2021-02-01T16:26:59 | 2021-02-01T16:26:59 | {} | NONE |
ClickHouse | ClickHouse | 771,021,443 | 19,375 | Avogar | @alexey-milovidov All checks passed, can I merge it? | 2021-02-01T17:26:13 | 2021-02-01T17:26:13 | {} | MEMBER |
ClickHouse | ClickHouse | 771,024,679 | 19,584 | azat | >@azat Could you please take a look at the test_materialize_mysql_database?
@alexey-milovidov The test is flaky, apparently part of data came after SELECT query:
```
2021.01.31 14:04:43.615618 [ 54 ] {} <Debug> executeQuery: (internal) /*Materialize MySQL step 1: execute dump data*/ INSERT INTO t(a, b, _sign, _v... | 2021-02-01T17:31:19 | 2021-02-01T17:32:34 | {} | MEMBER |
ClickHouse | ClickHouse | 771,331,320 | 19,946 | alexey-milovidov | > AST fuzzer (UBSan) — ../src/Functions/GeoHash.cpp:299:24: runtime error: -5.96529e+07 is outside the range of representable values of type 'unsigned int'
#19956 | 2021-02-02T03:38:28 | 2021-02-02T03:38:28 | {} | MEMBER |
ClickHouse | ClickHouse | 771,437,793 | 19,969 | ys-zhaoyh | > What kind of integration do you mean?
I have a configuration information in the question
Sentry is configured in the cluster, mainly for Hive.The ACL is also configured.In the Hive default HDFS path(/user/hive/warehouse), the table I created only inserts data, and no query data is returned. Other directories ar... | 2021-02-02T07:45:01 | 2021-02-02T08:06:00 | {} | NONE |
ClickHouse | ClickHouse | 770,852,841 | 16,616 | alexey-milovidov | @vdimir Looks like user and password should not be set, the source should just go to current ClickHouse instance. | 2021-02-01T13:22:14 | 2021-02-01T13:22:14 | {
"heart": 1
} | MEMBER |
ClickHouse | ClickHouse | 770,903,349 | 19,896 | vzakaznikov | If we make it more generic, it would nice for the encryption parameters to the codec to match our current `encrypt` function (parameter names, order and values).
| 2021-02-01T14:37:02 | 2021-02-01T14:37:02 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 771,322,951 | 15,713 | ghost | I understand that there is no formal process.
I'll think about the feature carefully to propose it.
Thank you. | 2021-02-02T03:17:32 | 2021-02-02T03:17:32 | {} | NONE |
ClickHouse | ClickHouse | 771,422,859 | 19,967 | alexey-milovidov | @sundy-li You did not check types of keys (Field).
`key_column.type->serializeBinary(keys[i], wbs[i]);`
This is only legal if type of the field corresponds to the type of the column. | 2021-02-02T07:16:22 | 2021-02-02T07:16:22 | {} | MEMBER |
ClickHouse | ClickHouse | 757,371,926 | 18,842 | alexey-milovidov | Download of 6 million pages took 2 days on single server, the download speed is 5..10 MB/sec, the total size with logs is about 600 GB. | 2021-01-09T21:45:47 | 2021-01-09T21:45:47 | {} | MEMBER |
ClickHouse | ClickHouse | 757,393,101 | 18,549 | alexey-milovidov | @TszKitLo40 We can continue on this task... | 2021-01-10T01:00:35 | 2021-01-10T01:00:35 | {} | MEMBER |
ClickHouse | ClickHouse | 757,401,144 | 18,890 | UnamedRus | ```
SELECT count(1)
FROM system.numbers
Query id: 58f5ef7d-8122-4756-be5e-f60007bffc11
Cancelling query.
Ok.
Query was cancelled.
0 rows in set. Elapsed: 10.842 sec. Processed 52.30 billion rows, 418.39 GB (4.82 billion rows/s., 38.59 GB/s.)
```
It's probably one of the settings:
```
SELECT *
FROM... | 2021-01-10T02:21:53 | 2021-01-10T02:21:53 | {} | CONTRIBUTOR |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.