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 | 740,079,200 | 17,850 | den-crane | Atomic is default engine starting from 20.11 | 2020-12-07T17:55:32 | 2020-12-07T17:55:32 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 740,576,420 | 17,726 | alexey-milovidov | Tests have failed for reason. | 2020-12-08T11:57:50 | 2020-12-08T11:57:50 | {} | MEMBER |
ClickHouse | ClickHouse | 740,701,151 | 16,935 | ucasfl | > You made these changes under this `if`:
> https://github.com/ClickHouse/ClickHouse/blob/eb46951e75f49776a096073ef1f45699c10d7fb1/src/Interpreters/TreeRewriter.cpp#L124
>
>
> So, if I write something like sumIfOrNull(1, 0) we won't rewrite it and we will get 0 instead of desired `Null`
Yes! this is a problem,... | 2020-12-08T15:46:30 | 2020-12-08T15:46:30 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 740,745,402 | 9,260 | alexey-milovidov | @alex-zaitsev It is related to LowCardinality? | 2020-12-08T16:33:49 | 2020-12-08T16:33:49 | {} | MEMBER |
ClickHouse | ClickHouse | 739,689,762 | 17,852 | filimonov | It is release build with debug info. Just strip debug symbol (`strip --strip-debug executable`) if you need more compact binary. | 2020-12-07T06:06:18 | 2020-12-07T06:06:18 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 739,926,272 | 17,866 | dkxiaohei | Sorry, I currently have no authority to add/edit labels to an issue in this project. Is there someone who can give me a hand? Thanks in advance! | 2020-12-07T13:43:09 | 2020-12-07T13:43:09 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 739,977,174 | 17,861 | den-crane | WA (normalize join manually):
```sql
SELECT *
FROM `/t0`
INNER JOIN
(
SELECT *
FROM `/t1`
) AS `\t1` USING (a)
Query id: cd2ef341-cb94-4f9c-8a46-fae9be938713
┌─a─┬─b─┬─ 1.b─┐
│ 0 │ 0 │ 1 │
└───┴───┴─────────────┘
```
| 2020-12-07T15:08:54 | 2020-12-07T15:09:52 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 740,346,555 | 17,865 | Tasselmi | > `show create table dws_oversea.ads_anu_onmic_user_room_di_dis`
```
CREATE TABLE IF NOT EXISTS `dws_oversea`.`ads_anu_onmic_user_room_di_rep` ON CLUSTER `default` (
`dt` Date,
`uid` Int64,
`room_id` Int64,
`gender` Int32,
`is_new` Int32,
`app_id` Int32,
`vip_level` Int32,
`r... | 2020-12-08T03:15:21 | 2020-12-08T03:15:21 | {} | NONE |
ClickHouse | ClickHouse | 740,734,364 | 17,795 | igorepim | May be even a binary search? | 2020-12-08T16:22:57 | 2020-12-08T16:22:57 | {} | NONE |
ClickHouse | ClickHouse | 739,560,821 | 17,847 | fibersel | > Experimental codecs should also be protected by a setting:
> `allow_experimental_codecs`, so users will not start to use it without being aware of the status of this feature.
Well, I will account this condition before merging.
First step - implement and benchmark | 2020-12-06T20:43:15 | 2020-12-06T20:43:15 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 739,565,064 | 17,844 | l1t1 | I think he means the code part
localhost :) select count(*) from numbers_mt(100000)a,numbers_mt(100000)b;
```
SELECT count(*)
FROM numbers_mt(100000) AS a
, numbers_mt(100000) AS b
```
Query id: e6d36818-9d22-46ce-b156-5b9c2fef0117
┌─────count()─┐
│ 10000000000 │
└─────────────┘ | 2020-12-06T21:14:28 | 2020-12-06T21:16:44 | {} | NONE |
ClickHouse | ClickHouse | 739,707,205 | 17,852 | xhsky | thank you, filimonov. it works! | 2020-12-07T06:45:04 | 2020-12-07T06:45:04 | {} | NONE |
ClickHouse | ClickHouse | 739,585,530 | 17,807 | alexey-milovidov | `max_parallel_replicas` is an "expert level" setting, it is very difficult to use correctly and it is not supposed to work at all when non-replicated tables are involved. I suggest to remove this setting from docs. | 2020-12-06T23:37:28 | 2020-12-06T23:37:28 | {} | MEMBER |
ClickHouse | ClickHouse | 739,793,839 | 17,824 | alesapin | Cherry-picked manually. | 2020-12-07T09:29:14 | 2020-12-07T09:29:14 | {} | MEMBER |
ClickHouse | ClickHouse | 739,796,604 | 17,458 | alesapin | Join race seems unrelated, but requires investigation. | 2020-12-07T09:34:15 | 2020-12-07T09:34:15 | {} | MEMBER |
ClickHouse | ClickHouse | 739,840,981 | 11,298 | azat | >In what build configurations have you encountered them? I added suppressions to CMake files where appropriate, but still.
Ok, I see the problem, cached `HAS_SUGGEST_DESTRUCTOR_OVERRIDE` in cmake (for another compiler I guess, since I don't clean cmake cache usually)
After reset the cache for this variable everyt... | 2020-12-07T10:54:19 | 2020-12-07T10:54:19 | {} | MEMBER |
ClickHouse | ClickHouse | 739,969,849 | 17,853 | den-crane | I think https://github.com/ClickHouse/ClickHouse/blob/master/cmake/target.cmake#L30
these will be unavailable
set (ENABLE_PROTOBUF OFF CACHE INTERNAL "")
set (ENABLE_GRPC OFF CACHE INTERNAL "")
set (ENABLE_PARQUET OFF CACHE INTERNAL "")
set (ENABLE_ORC OFF CACH... | 2020-12-07T14:56:53 | 2020-12-07T14:56:53 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 740,395,657 | 17,884 | l1t1 | you need sudo | 2020-12-08T05:53:53 | 2020-12-08T05:53:53 | {
"confused": 1
} | NONE |
ClickHouse | ClickHouse | 739,580,605 | 17,748 | azat | >Performance — 2 errors, 5 faster, 7 slower, 24 unstable
```
DB::Exception: Unknown function firstSignificantSubdomainCustom. Maybe you meant: ['firstSignificantSubdomain','cutToFirstSignificantSubdomain']: While processing SELECT count() FROM hits_100m_single WHERE NOT ignore(firstSignificantSubdomainCustom(URL, '... | 2020-12-06T23:05:05 | 2020-12-06T23:05:05 | {} | MEMBER |
ClickHouse | ClickHouse | 739,641,230 | 17,788 | ghost | @den-crane
Thank you. | 2020-12-07T03:29:28 | 2020-12-07T03:29:28 | {} | NONE |
ClickHouse | ClickHouse | 740,014,280 | 17,709 | alesapin | `test_distributed_respect_user_timeouts` known flaky test | 2020-12-07T16:08:14 | 2020-12-07T16:08:14 | {} | MEMBER |
ClickHouse | ClickHouse | 740,355,202 | 11,298 | baibaichen | @azat Thank you.
I see the issue, it looks like we need update documentation [Build on Mac OS X](https://github.com/ClickHouse/ClickHouse/blob/master/docs/en/development/build-osx.md) to reflect this fact.
I will submit PR in this weekend. | 2020-12-08T03:43:46 | 2020-12-08T03:50:19 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 740,399,734 | 17,875 | l1t1 | ``` sql
localhost :) select arrayReduce('groupUniqArray', ['world','hello','hello']);
SELECT arrayReduce('groupUniqArray', ['world', 'hello', 'hello'])
┌─arrayReduce('groupUniqArray', ['world', 'hello', 'hello'])─┐
│ ['world','hello'] │
└────────────────────────────────... | 2020-12-08T06:03:52 | 2020-12-08T06:14:44 | {} | NONE |
ClickHouse | ClickHouse | 740,540,377 | 13,804 | Enmk | It looks like this issue was recently fixed with #16952 :
```SQL
SELECT
n,
toTypeName(dt64) AS dt64_typename,
'<',
dt64 < dt,
toDateTime(dt64) < dt,
dt64 < toDateTime64(dt, 1, 'UTC'),
'<=',
dt64 <= dt,
toDateTime(dt64) <= dt,
dt64 <= toDateTime64(dt, 1, 'UTC... | 2020-12-08T10:42:18 | 2020-12-08T10:54:36 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 740,004,086 | 17,865 | den-crane | ```
20.12.1.5183
create table set_test(test String , index x1(test) type set(2000) GRANULARITY 1)
Engine=MergeTree order by tuple();
insert into set_test select toString(number) from numbers(1000);
SELECT *
FROM set_test
WHERE (test IN ('1', '2')) AND (1 = 1)
LIMIT 10;
Query id: 3c9ba27e-449e-42b3-a13... | 2020-12-07T15:51:26 | 2020-12-07T15:51:50 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 740,114,397 | 17,642 | tavplubix | I think it should not be possible to create table with the same name as some permanently detached one has | 2020-12-07T18:58:57 | 2020-12-07T18:58:57 | {} | MEMBER |
ClickHouse | ClickHouse | 740,552,282 | 17,790 | taylor131415 | After changing 'where' to 'prewhere' in sql1, speed get much faster than before. Close issue. | 2020-12-08T11:05:38 | 2020-12-08T11:05:38 | {} | NONE |
ClickHouse | ClickHouse | 740,590,153 | 17,874 | nikitamikhaylov | > Psst, you changed function names but didn't update the documentation :D
Thanks! | 2020-12-08T12:27:15 | 2020-12-08T12:27:15 | {} | MEMBER |
ClickHouse | ClickHouse | 740,634,742 | 10,177 | filimonov | Was debugging similar issue today:
```
2020.12.08 03:19:56.018261 [ 1 ] {} <Error> Application: Caught exception while loading metadata: Code: 49, e.displayText() = DB::Exception: ALTER MODIFY ORDER BY of default-partitioned tables is not supported, Stack trace:
0. 0x3503c40 StackTrace::StackTrace() /usr/bin/clickho... | 2020-12-08T13:57:37 | 2020-12-08T13:57:37 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 740,667,395 | 17,901 | den-crane | a new feature "ALTER PRIMARY KEY" is planned on 2021
https://github.com/ClickHouse/ClickHouse/issues/17623 | 2020-12-08T14:53:20 | 2020-12-08T14:53:20 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 739,793,938 | 17,825 | alesapin | Cherry-picked manually. | 2020-12-07T09:29:25 | 2020-12-07T09:29:25 | {} | MEMBER |
ClickHouse | ClickHouse | 739,982,015 | 17,872 | nblagodarnyi | Everything is ok in ClickHouse server version 19.16.2 revision 54427.
clickhouse02 :) INSERT INTO tmp.test_hang_target_1 SELECT SUM( t1.ID + t1.N1 + t1.N2 + t1.N3 + t1.N4 + t1.N5 ) s from tmp.test_hang_source_1 t1;
INSERT INTO tmp.test_hang_target_1 SELECT SUM(((((t1.ID + t1.N1) + t1.N2) + t1.N3) + t1.N4) + t1.N... | 2020-12-07T15:16:56 | 2020-12-07T15:16:56 | {} | NONE |
ClickHouse | ClickHouse | 740,018,245 | 17,872 | den-crane | 20.13.1.5273
```sql
CREATE TABLE test (ID Int64) engine = Null;
INSERT INTO test
SELECT * FROM ( SELECT SUM( (materialize(number) as x) + materialize(x) + materialize(x+2) + materialize(x+3) + materialize(x+4) ) s from numbers_mt(100000000));
Elapsed: 0.037 sec.
INSERT INTO test
SELECT SUM( (mater... | 2020-12-07T16:14:26 | 2020-12-07T16:14:26 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 740,237,735 | 16,424 | alexey-milovidov | > If during that time period the chosen replica will finish the merge others will just download the result, otherwise, they will do the merge normally.
This criteria is questionable - it will work only for small merges. | 2020-12-07T23:13:34 | 2020-12-07T23:13:34 | {} | MEMBER |
ClickHouse | ClickHouse | 739,835,217 | 11,298 | abyss7 | > Introduces some `-Wsuggest-destructor-override` warnings
In what build configurations have you encountered them? I added suppressions to CMake files where appropriate, but still. | 2020-12-07T10:42:51 | 2020-12-07T10:42:51 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 739,985,665 | 17,870 | excitoon | This is both update of AWS SDK which we need to support so-called global S3 region, and attempt to solve ClickHouse-Extras/aws-sdk-cpp#4 in more compact way. | 2020-12-07T15:22:25 | 2020-12-07T15:22:25 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 740,259,727 | 17,849 | den-crane | JFYI: I don't have an opinion on this. I mentioned `arrayReduce` as a workaround.
Documentation process is hard. Hundreds of topics and features are not documented at all. | 2020-12-08T00:12:23 | 2020-12-08T00:12:23 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 739,794,022 | 17,827 | alesapin | Cherry-picked manually. | 2020-12-07T09:29:33 | 2020-12-07T09:29:33 | {} | MEMBER |
ClickHouse | ClickHouse | 739,827,416 | 17,269 | ucasfl | Build failed on gcc-9: `constructor is not user-provided because it is explicitly defaulted in the class body, ... and the implicitly-defined constructor does not initialize 'wide::integer<256, unsigned int>::base_type wide::integer<256, unsigned int>::items [4]`.
Seems this is a gcc bug [#80935](https://gcc.gnu.org... | 2020-12-07T10:28:42 | 2020-12-07T10:28:42 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 740,504,723 | 16,952 | KochetovNicolai | `01415_sticking_mutations` is flappy
yandex check is broken on master | 2020-12-08T09:38:08 | 2020-12-08T09:38:08 | {} | MEMBER |
ClickHouse | ClickHouse | 740,637,816 | 16,935 | Avogar | You made these changes under this `if`:
https://github.com/ClickHouse/ClickHouse/blob/eb46951e75f49776a096073ef1f45699c10d7fb1/src/Interpreters/TreeRewriter.cpp#L124
So, if I write something like sumIfOrNull(1, 0) we won't rewrite it and we will get 0 instead of desired `Null` | 2020-12-08T14:03:13 | 2020-12-08T14:03:13 | {} | MEMBER |
ClickHouse | ClickHouse | 740,664,736 | 17,884 | den-crane | @physili not related to CH.
Package management requires root privs. If your company does not provide a root password probably it's for a reason.
You can use CH without installation. Downoad server binary or tar with binary Linux tgz package clickhouse-server-20.13.1.5343.tgz https://github.com/ClickHouse/ClickHous... | 2020-12-08T14:48:59 | 2020-12-08T14:49:17 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 740,668,447 | 10,850 | alarin | @alexey-milovidov trying to use complex_key_direct and it's not working. What's wrong?
```
CREATE DICTIONARY targeting_service_mv.dict_account2wallet_redis
(
account_id String DEFAULT '',
wallet_uid String DEFAULT ''
)
PRIMARY KEY wallet_uid
SOURCE(REDIS(host 'localhost' port 6379 db_index 0 storage_t... | 2020-12-08T14:55:09 | 2020-12-08T14:55:09 | {} | NONE |
ClickHouse | ClickHouse | 740,673,584 | 16,877 | alesapin | `Make TestKeeperStorage a backend for TestKeeper (remove a bunch of duplicated code)`
Now it's not quite easy, Not going to do it in this PR. | 2020-12-08T15:03:43 | 2020-12-08T15:03:43 | {} | MEMBER |
ClickHouse | ClickHouse | 739,610,914 | 17,838 | weeds085490 | > BTW, it's a summary metric, why not use `system.parts`, it has more detail `metrics` (per database, table ...).
> And it's also a table in memory for statistics, clickhouse exporters (or custom HTTP metrics handler) work good with it.
When we are building the monitoring system of clickhouse, we find the number of... | 2020-12-07T01:27:19 | 2020-12-07T01:27:19 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 739,686,950 | 17,850 | JKeyZ | thank u, den-crane. it's work.
is default engine ordinary in old version(20.5.2.7), atomic in newly? | 2020-12-07T06:00:14 | 2020-12-07T06:00:14 | {} | NONE |
ClickHouse | ClickHouse | 739,862,693 | 17,666 | alexey-milovidov | It looks like the hypervisor is not configured correctly. | 2020-12-07T11:35:28 | 2020-12-07T11:35:28 | {} | MEMBER |
ClickHouse | ClickHouse | 739,895,853 | 15,607 | filimonov | BTW - now you can't create dictionaries in clickhouse-local due to that... | 2020-12-07T12:44:47 | 2020-12-07T12:44:47 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 739,947,762 | 17,731 | Farfaday | Hi all,
today I tested:
- shutting down clickhouse
- moving out all the wal_x_yyyyy.bin files out of the data/db/table directories
- setting in config.xml
```
<merge_tree>
<in_memory_parts_enable_wal>0</in_memory_parts_enable_wal>
</merge_tree>
```
Then the clickhouse starting time went ba... | 2020-12-07T14:20:17 | 2020-12-07T14:20:17 | {} | NONE |
ClickHouse | ClickHouse | 740,277,218 | 17,874 | depressed-pho | Psst, you changed function names but didn't update the documentation :D | 2020-12-08T01:00:20 | 2020-12-08T01:00:20 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 740,322,889 | 17,883 | vegastar002 | 它让你设置一下这个，在命令行下
set allow_experimental_multiple_joins_emulation | 2020-12-08T02:39:41 | 2020-12-08T02:39:41 | {} | NONE |
ClickHouse | ClickHouse | 739,975,775 | 17,860 | den-crane | Documentation never stated that virtual columns can be used for join.
And it does not have any sense in case of _part_index.
WA:
```sql
SELECT a
FROM
(
SELECT
*,
_part_index AS part_index
FROM t0
) AS t0
INNER JOIN
(
SELECT
*,
_part_index AS part_index
... | 2020-12-07T15:06:35 | 2020-12-07T15:06:35 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 740,058,332 | 17,807 | alexey-milovidov | Ok, it's better to leave it in docs.
So, the query with `max_parallel_replicas` > 1 will work correctly only if:
- it reads only from Replicated tables;
- these Replicated tables have the same SAMPLE key or GLOBAL modifier is used for IN/JOIN;
- if IN or JOIN is involved, either it joins by sample key or subkey o... | 2020-12-07T17:18:47 | 2020-12-07T17:22:11 | {} | MEMBER |
ClickHouse | ClickHouse | 740,090,814 | 17,348 | xjewer | committed changed submodule by mistake, fixed | 2020-12-07T18:16:51 | 2020-12-07T18:16:51 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 740,118,431 | 17,383 | BayoNet | Will be merged in #17859 | 2020-12-07T19:06:10 | 2020-12-07T19:06:10 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 740,382,527 | 11,428 | PhantomPhreak | Figured out the root cause.
We're not using RBAC yet, and limiting allowed database for a user, using `<allow_databases>` configuration parameter in `users.xml`. As described in a docs, this approach is still working: https://clickhouse.tech/docs/en/operations/access-rights/
There are `_temporary_and_external_tables`... | 2020-12-08T05:19:38 | 2020-12-08T05:20:31 | {
"+1": 1
} | NONE |
ClickHouse | ClickHouse | 740,453,158 | 17,886 | spz0507 | same problem | 2020-12-08T08:01:10 | 2020-12-08T08:01:10 | {} | NONE |
ClickHouse | ClickHouse | 740,544,402 | 17,885 | l1t1 | plz post the full sql | 2020-12-08T10:49:50 | 2020-12-08T10:49:50 | {} | NONE |
ClickHouse | ClickHouse | 740,573,936 | 17,630 | alexey-milovidov | Sandbox error:
> ResourceNotAvailable: {74f3cad3:8}: sandbox.agentr.errors.ResourceNotAvailable('Resource #1871059875 is not available at all sources [].',)
Will try again... | 2020-12-08T11:52:02 | 2020-12-08T11:52:02 | {} | MEMBER |
ClickHouse | ClickHouse | 740,064,857 | 17,656 | azat | >host String (or maybe two separate columns with host and port) - name of host, which should execute the query
`system.clusters` has `host_address`/`host_name` and `port`, so better to use the same naming.
>Maybe unfold arrays in table structure? I mean something like:
Completely agree with everything and unfo... | 2020-12-07T17:29:51 | 2020-12-07T17:29:51 | {} | MEMBER |
ClickHouse | ClickHouse | 740,299,868 | 17,862 | den-crane | The main difficulty is that CH collapses/evaluates AggregatingMergeTree during the merge process. But the merge merges only some part of data at a time.
Try TTL where.
It will not give pure result because it operates in a scope of parts in the merge. But eventually it will. | 2020-12-08T01:27:19 | 2020-12-08T01:32:07 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 740,659,930 | 17,885 | den-crane | Do not use groupArray for this.
```
create table temp(A Int64) Engine=Log;
insert into temp select * from numbers_mt(100000000);
select * from temp limit 10 offset 99000000;
10 rows in set. Elapsed: 0.124 sec. Processed 99.08 million rows,
``` | 2020-12-08T14:41:14 | 2020-12-08T14:41:14 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 740,660,794 | 17,642 | filimonov | > I think it should not be possible to create table with the same name as some permanently detached one has
+1. I think that should prevent that. https://github.com/ClickHouse/ClickHouse/pull/17642/commits/d9a710db774abd3c0b8b6c104829235866408b21
WIll add tests... | 2020-12-08T14:42:38 | 2020-12-08T14:42:38 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 739,584,349 | 17,071 | alexey-milovidov | Duplicate of #8482 | 2020-12-06T23:30:00 | 2020-12-06T23:30:00 | {} | MEMBER |
ClickHouse | ClickHouse | 739,612,957 | 17,838 | weeds085490 | > BTW, it's a summary metric, why not use `system.parts`, it has more detail `metrics` (per database, table ...).
> And it's also a table in memory for statistics, clickhouse exporters (or custom HTTP metrics handler) work good with it.
We can also get information about `merge` from `system.merges`, but we also nee... | 2020-12-07T01:36:57 | 2020-12-07T01:36:57 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 739,654,415 | 17,849 | den-crane | ```
SELECT
[1, 2] AS arr,
arrayReduce('min', arr),
arrayReduce('max', arr),
arrayReduce('avg', arr)
Query id: f41c09c9-6b7f-49be-ad98-6fb3b1c9840a
┌─arr───┬─arrayReduce('min', [1, 2])─┬─arrayReduce('max', [1, 2])─┬─arrayReduce('avg', [1, 2])─┐
│ [1,2] │ 1 │ ... | 2020-12-07T04:24:41 | 2020-12-07T04:24:41 | {
"+1": 3,
"hooray": 2
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 739,946,970 | 17,869 | alexey-milovidov | @nikitamikhaylov
> The source directory /home/jakalletti/ClickHouse/ClickHouse/contrib/dragonbox does not contain a CMakeLists.txt file.
But the file exists and CI successfully builds the project. | 2020-12-07T14:19:01 | 2020-12-07T14:19:01 | {} | MEMBER |
ClickHouse | ClickHouse | 739,979,295 | 17,873 | den-crane | duplicate https://github.com/ClickHouse/ClickHouse/issues/4222 | 2020-12-07T15:12:35 | 2020-12-07T15:12:35 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 740,028,105 | 17,630 | alexey-milovidov | We failed to suppress TSan on function level, now let's do it on target level:
`set_source_files_properties(file PROPERTIES COMPILE_FLAGS option)` | 2020-12-07T16:30:03 | 2020-12-07T16:30:03 | {} | MEMBER |
ClickHouse | ClickHouse | 740,399,495 | 17,884 | physili | Sudo asked me for root password。But the company does not provide a root password
| 2020-12-08T06:03:18 | 2020-12-08T06:03:18 | {} | NONE |
ClickHouse | ClickHouse | 740,587,554 | 147 | shadowDy | Hello!
Do you have any updates about it now? | 2020-12-08T12:21:29 | 2020-12-08T12:21:29 | {} | NONE |
ClickHouse | ClickHouse | 740,687,567 | 17,849 | rbeeli | Hi @den-crane
I can imagine, it is a lot of work. It's just unfortunate if the end user cannot find functionality that's already there and make use of it, like in my case with `arrayReduce`. If the shortcut alias functions I suggested above get added, I would gladly help with updating the documentation and create a p... | 2020-12-08T15:26:14 | 2020-12-08T15:26:14 | {} | NONE |
ClickHouse | ClickHouse | 740,708,038 | 17,642 | filimonov | > Also it would be great to support `DETACH TABLE ... PERMANENTLY` for `MySQL` database engine, because `DatabaseConnectionMySQL::dropTable(...)` actually detaches table permanently.
Indeed. Added that, I will NOT add tests (a bit cumbersome), and since now dropTable uses detachTablePermanently i guess we can consid... | 2020-12-08T15:57:10 | 2020-12-08T15:57:10 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 739,585,908 | 13,890 | alexey-milovidov | @ucasFL You can test that table size will be slightly lower (after OPTIMIZE) if min_compress_block_size is larger. | 2020-12-06T23:40:03 | 2020-12-06T23:40:03 | {} | MEMBER |
ClickHouse | ClickHouse | 739,586,154 | 17,623 | alexey-milovidov | @l1t1 We want to make releases more frequent than once a month. | 2020-12-06T23:41:54 | 2020-12-06T23:41:54 | {
"+1": 2,
"confused": 1
} | MEMBER |
ClickHouse | ClickHouse | 739,793,981 | 17,826 | alesapin | Cherry-picked manually. | 2020-12-07T09:29:28 | 2020-12-07T09:29:28 | {} | MEMBER |
ClickHouse | ClickHouse | 739,948,600 | 17,870 | alexey-milovidov | What exactly was fixed? Why "not for changelog"? | 2020-12-07T14:21:44 | 2020-12-07T14:21:44 | {} | MEMBER |
ClickHouse | ClickHouse | 740,156,489 | 17,642 | tavplubix | Also it would be great to support `DETACH TABLE ... PERMANENTLY` for `MySQL` database engine, because `DatabaseConnectionMySQL::dropTable(...)` actually detaches table permanently. | 2020-12-07T20:19:10 | 2020-12-07T20:19:10 | {} | MEMBER |
ClickHouse | ClickHouse | 740,563,544 | 17,900 | Akazz | No need in testing these changes | 2020-12-08T11:28:29 | 2020-12-08T11:28:29 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 740,652,140 | 17,889 | aubweb9 | Yes we do.
Thanks I will modify the variable (which has currently a default value of 86400)
👍
| 2020-12-08T14:28:07 | 2020-12-08T14:28:07 | {} | NONE |
ClickHouse | ClickHouse | 739,827,311 | 17,833 | nvartolomei | backport to 20.8 lts as well please? | 2020-12-07T10:28:32 | 2020-12-07T10:28:32 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 739,867,225 | 17,849 | rbeeli | Hi @den-crane
Great that there already exists a way of doing this, but unfortunately it's not documented very well, see [https://clickhouse.tech/docs/en/sql-reference/functions/array-functions/#arrayreduce](https://clickhouse.tech/docs/en/sql-reference/functions/array-functions/#arrayreduce). If I search for "min", ... | 2020-12-07T11:45:22 | 2020-12-07T11:45:22 | {} | NONE |
ClickHouse | ClickHouse | 740,028,726 | 17,637 | alexey-milovidov | > Special build check has failed because SymbolIndex is disabled on some platforms.
Still needs to be fixed. | 2020-12-07T16:31:02 | 2020-12-07T16:31:02 | {} | MEMBER |
ClickHouse | ClickHouse | 740,050,148 | 17,807 | edani | Thanks for the explanation! I would argue in favor of leaving the setting in the doc and expanding the documentation. Azat's short explanation makes it easy to understand the correct use cases. I can do a documentation PR if everyone is ok with it.
| 2020-12-07T17:04:31 | 2020-12-07T17:04:31 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 740,148,485 | 9,260 | alexey-milovidov | @alesapin It's not fixed. | 2020-12-07T20:03:13 | 2020-12-07T20:03:13 | {} | MEMBER |
ClickHouse | ClickHouse | 740,182,226 | 17,637 | alexey-milovidov | ```
/build/src/Interpreters/InterpreterSystemQuery.cpp:282:13: error: 'break' will never be executed [-Werror,-Wunreachable-code-break]
break;
^~~~~
```
On MacOS. | 2020-12-07T21:11:23 | 2020-12-07T21:11:23 | {} | MEMBER |
ClickHouse | ClickHouse | 740,302,321 | 17,866 | dkxiaohei | I have no access to 'https://a.yandex-team.ru/review/1552413' due to my network condition. So I am not sure why the 2 checks failed. Is there someone or some document that I can turn to? | 2020-12-08T01:35:06 | 2020-12-08T01:35:06 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 740,323,651 | 17,883 | glaryaugus | > 它让你设置一下这个，在命令行下
> set allow_experimental_multiple_joins_emulation
没用，本身allow_experimental_multiple_joins_emulation就是等于1。创建了视图不能查询这还有什么意义。 | 2020-12-08T02:42:19 | 2020-12-08T02:42:19 | {} | NONE |
ClickHouse | ClickHouse | 740,430,227 | 9,260 | alex-zaitsev | I've got the same problem during mutation in 20.11.5.18. There were no skipping indices involved. | 2020-12-08T07:12:13 | 2020-12-08T07:12:13 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 740,549,163 | 17,718 | pic-Nick | This error is reproducible on 20.8.8.2, 20.9.7.11 and 20.10.6.27. | 2020-12-08T10:59:38 | 2020-12-08T10:59:38 | {} | NONE |
ClickHouse | ClickHouse | 740,606,632 | 17,659 | nautaa | @alexey-milovidov PTAL. | 2020-12-08T13:04:05 | 2020-12-08T13:04:05 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 739,910,575 | 17,865 | Tasselmi | this is always broken:
```
SELECT *
FROM dws_oversea.ads_anu_onmic_user_room_di_dis
WHERE (dt = '2020-12-02') AND (channel_name IN ('af_Social_facebook', 'af_asm_440644258')) AND (1 = 1)
LIMIT 10
Received exception from server (version 20.3.10):
Code: 49. DB::Exception: Received from xxx.com:3306. DB::Exceptio... | 2020-12-07T13:14:07 | 2020-12-07T13:16:54 | {} | NONE |
ClickHouse | ClickHouse | 739,965,635 | 17,656 | tavplubix | Maybe unfold arrays in table structure? I mean something like:
```
┌─entry────────────┬─host───────┬─status──────┬─some_column─┐
│ query-0000000042 │ node1:9000 │ active │ ... │
│ query-0000000042 │ node2:9000 │ finished │ ... │
│ query-0000000042 │ node3:9000 │ unknown │ ... │
... | 2020-12-07T14:50:23 | 2020-12-07T14:50:23 | {
"+1": 2
} | MEMBER |
ClickHouse | ClickHouse | 740,000,391 | 17,865 | den-crane | `show create table dws_oversea.ads_anu_onmic_user_room_di_dis` | 2020-12-07T15:45:30 | 2020-12-07T15:45:41 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 740,057,729 | 17,848 | azat | CI checks looks unrelated (although I don't know what with the `Stress test (address)`, don't have access) | 2020-12-07T17:17:42 | 2020-12-07T17:17:42 | {} | MEMBER |
ClickHouse | ClickHouse | 740,078,007 | 17,656 | bharatnc | Thank you for the review @azat and @tavplubix let me take sometime to read the reviews more closer and also make these changes. | 2020-12-07T17:53:23 | 2020-12-07T17:53:23 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 740,641,228 | 17,889 | den-crane | Do you use mysql8?
https://github.com/ClickHouse/ClickHouse/issues/11475#issuecomment-643292140
Need to enable `set global information_schema_stats_expiry=0;` in mysql | 2020-12-08T14:09:12 | 2020-12-08T14:09:12 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 743,492,846 | 18,019 | alexey-milovidov | 20.9 no longer supported. | 2020-12-12T00:13:19 | 2020-12-12T00:13:19 | {} | MEMBER |
ClickHouse | ClickHouse | 743,770,315 | 17,847 | fibersel | > How to use it:
>
> 1. When you will integrate your codecs to `CompressionCodecFactory`, they will become available under symbolic names in clickhouse-compressor.
> 2. You can use clickhouse-compressor to test compression speed and ratio. Then the same for decompression.
How can I measure compression and decomp... | 2020-12-12T15:15:57 | 2020-12-12T15:15:57 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 743,780,073 | 18,026 | alexey-milovidov | This is normal.
1. `clickhouse.old` file is a backup of previous version. It allows to quickly switch to previous version if needed. This file can be safely removed to save space.
2. Minor changes in file ownership and access are insignificant. If you found something not convenient or worse please tell. | 2020-12-12T16:28:18 | 2020-12-12T16:28:18 | {} | MEMBER |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.