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 | 612,940,971 | 10,236 | filimonov | Related #9339 | 2020-04-13T15:12:08 | 2020-04-13T15:12:08 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 613,128,152 | 10,235 | alexey-milovidov | CI fix: https://github.com/ClickHouse/ClickHouse/pull/10239 | 2020-04-13T22:28:17 | 2020-04-13T22:28:17 | {} | MEMBER |
ClickHouse | ClickHouse | 613,264,492 | 9,990 | filimonov | > @filimonov
>
> For single regular expression we can add `extractGroups` that will return an array of captured groups starting from 1 in the first match. Also we can add `extractAllGroups` that will return two-dimensional array of groups from all matches.
Yep. Re2-based, based on extractAll , w/o 'multi-' prefix... | 2020-04-14T07:06:47 | 2020-04-14T07:06:47 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 613,344,605 | 9,985 | alesapin | Do not test until merged with master. | 2020-04-14T09:58:36 | 2020-04-14T09:58:36 | {} | MEMBER |
ClickHouse | ClickHouse | 613,418,087 | 9,043 | azat | Looks like it has been fixed (4a83e85893)?
```sql
CREATE TABLE merge AS system.one ENGINE = Merge(currentDatabase(), 'dist');
CREATE TABLE dist_1 AS system.one ENGINE = Distributed(test_shard_localhost, system, one);
CREATE TABLE dist_2 AS system.one ENGINE = Distributed(test_shard_localhost, system, one);
-- tr... | 2020-04-14T12:38:43 | 2020-04-14T12:38:43 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 613,540,304 | 10,260 | zzsmdfj | This is the first time for me to contribute to ClickHouse. If there is anything that does not meet the requirements, let me know. | 2020-04-14T16:20:35 | 2020-04-14T16:20:35 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 613,789,771 | 10,246 | alexey-milovidov | #10238 | 2020-04-15T03:08:43 | 2020-04-15T03:08:43 | {} | MEMBER |
ClickHouse | ClickHouse | 612,790,448 | 10,224 | filimonov | Because there is already too many running concurrent merges / mutations.
While some merge or mutation is running it decreases the maximum size for other mutations/mutations, which can be executed at the same time. That logic prevents ClickHouse from running several huge merges/mutations simultaneously.
Check sy... | 2020-04-13T07:37:59 | 2020-04-13T11:11:31 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 613,057,245 | 10,233 | CurtizJ | The same tests fail in 20.3 branch. | 2020-04-13T19:32:18 | 2020-04-13T19:32:18 | {} | MEMBER |
ClickHouse | ClickHouse | 613,366,866 | 9,439 | azat | But Distributed table should not have materialized columns.
Refs: https://github.com/ClickHouse/ClickHouse/issues/7365#issuecomment-543649137
Refs: #7377 | 2020-04-14T10:47:43 | 2020-04-14T10:47:43 | {} | MEMBER |
ClickHouse | ClickHouse | 613,456,316 | 9,990 | alexey-milovidov | @alex-zaitsev
```
[['abc', 111], ['def', 222], ['ghi', 333]]
``` | 2020-04-14T13:53:01 | 2020-04-14T13:53:01 | {} | MEMBER |
ClickHouse | ClickHouse | 613,518,423 | 10,259 | qoega | ``` sql
CREATE TABLE mass_table_701 ENGINE = MergeTree ORDER BY number AS SELECT number, arrayMap (x -> (x, [x], [[x]], (x, toString(x))), arrayMap(x -> range(x), range(number % 10))) AS value FROM system.numbers LIMIT 100000;
INSERT INTO mass_table_701 SELECT * FROM generateRandom('`number` UInt64,`value` Array(Tupl... | 2020-04-14T15:42:28 | 2020-04-14T15:42:28 | {} | MEMBER |
ClickHouse | ClickHouse | 612,790,955 | 10,225 | filimonov | duplicate #10224 | 2020-04-13T07:39:54 | 2020-04-13T07:39:54 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 612,938,283 | 8,183 | filimonov | ```
clickhouse-client --format=TSVRaw -q"select 'ALTER TABLE ' || database || '.' || table || ' ATTACH PARTITION ID \'' || partition_id || '\';\n' from system.detached_parts group by database, table, partition_id order by database, table, partition_id;" | clickhouse-client -mn
``` | 2020-04-13T15:06:14 | 2020-04-13T15:06:14 | {
"+1": 11
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 612,984,425 | 8,183 | alexey-milovidov | @filimonov Though your script is technically incorrect because it doesn't escape names properly. | 2020-04-13T16:50:55 | 2020-04-13T16:50:55 | {} | MEMBER |
ClickHouse | ClickHouse | 613,089,167 | 10,241 | den-crane | The performance impact is visible with a high number of MVs.
```
drop table if exists testX;
drop table if exists testXA;
drop table if exists testXB;
drop table if exists testXC;
drop table if exists testXD;
drop table if exists testXE;
drop table if exists testXF;
create table testX ( A Int64) engine=Mer... | 2020-04-13T20:44:50 | 2020-04-13T20:44:50 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 613,509,644 | 7,310 | alexey-milovidov | Superseded by #9983 | 2020-04-14T15:26:57 | 2020-04-14T15:26:57 | {} | MEMBER |
ClickHouse | ClickHouse | 612,820,545 | 10,229 | soloduev | Ok, thanks! But, why CH can't do auto down-cast to supported DateTime? | 2020-04-13T09:10:12 | 2020-04-13T09:10:12 | {} | NONE |
ClickHouse | ClickHouse | 612,986,987 | 8,167 | CurtizJ | No, automatical run `MATERIALIZE TTL` for `ALTER ... MODIFY TTL` is not implemented yet. But now it's easy to do. | 2020-04-13T16:56:40 | 2020-04-13T16:58:37 | {} | MEMBER |
ClickHouse | ClickHouse | 613,033,111 | 10,052 | alexey-milovidov | > Performance — 150 errors
- unknown issue. | 2020-04-13T18:37:36 | 2020-04-13T18:37:36 | {} | MEMBER |
ClickHouse | ClickHouse | 613,127,555 | 10,230 | alexey-milovidov | FreeBSD build for old versions most likely won't be fixed (unless someone will send a pull request to 19.14 branch). | 2020-04-13T22:26:25 | 2020-04-13T22:26:25 | {} | MEMBER |
ClickHouse | ClickHouse | 613,336,151 | 9,361 | jangorecki | @filimonov what does it mean? | 2020-04-14T09:39:36 | 2020-04-14T09:39:36 | {} | NONE |
ClickHouse | ClickHouse | 613,570,597 | 10,252 | vvchistyakov | Hi @den-crane
Yep, partitions with is_frozen flag have only read permission on filesystem on my server. However. I've checked the replica for this nodes. And I've got the same flags for same partitions. Nobody had made a backup for that. | 2020-04-14T17:18:10 | 2020-04-14T17:18:10 | {} | NONE |
ClickHouse | ClickHouse | 613,634,391 | 9,810 | 4ertus2 | Draft for fix constants from the left table #9950
Right ones fixed in #10207 | 2020-04-14T19:21:12 | 2020-04-14T19:22:18 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 613,751,179 | 10,269 | alexey-milovidov | Now it's no more than 1.2% slower than before. Actually there is no noticeable difference in performance. | 2020-04-15T00:41:44 | 2020-04-15T00:54:03 | {} | MEMBER |
ClickHouse | ClickHouse | 613,129,272 | 10,052 | alexey-milovidov | @qoega @Vxider You can add a performance test if you like. | 2020-04-13T22:31:46 | 2020-04-13T22:31:46 | {} | MEMBER |
ClickHouse | ClickHouse | 613,489,782 | 9,990 | alex-zaitsev | @alexey-milovidov , I prefer two arrays that maps better to nested structures. Also not sure it makes sense to perform type inference. The data may be dirty. | 2020-04-14T14:52:33 | 2020-04-14T14:52:33 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 612,932,582 | 9,339 | filimonov | Postgress commands:
```
psql
psql (9.2.23)
Type "help" for help.
postgres=# help
You are using psql, the command-line interface to PostgreSQL.
Type: \copyright for distribution terms
\h for help with SQL commands
\? for help with psql commands
\g or terminate with semicolon to execute... | 2020-04-13T14:53:30 | 2020-04-13T14:53:30 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 612,942,953 | 8,167 | qoega | ALTER ... MATERIALIZE TTL is implemented. | 2020-04-13T15:16:28 | 2020-04-13T15:16:28 | {} | MEMBER |
ClickHouse | ClickHouse | 612,982,480 | 8,167 | alexey-milovidov | @CurtizJ mentioned that TTL will be calculated for old parts automatically when adding TTLs, but I forget if this is already implemented or not. | 2020-04-13T16:46:20 | 2020-04-13T16:46:20 | {} | MEMBER |
ClickHouse | ClickHouse | 613,136,681 | 5,387 | alexey-milovidov | @akuzm This is still to do. | 2020-04-13T22:56:24 | 2020-04-13T22:56:24 | {} | MEMBER |
ClickHouse | ClickHouse | 613,358,520 | 9,499 | alesapin | Do not test until merge with master. | 2020-04-14T10:29:54 | 2020-04-14T10:29:54 | {} | MEMBER |
ClickHouse | ClickHouse | 613,459,154 | 10,254 | alexey-milovidov | > AND/OR/XOR
bitAnd, bitOr, bitXor, bitNot. | 2020-04-14T13:58:02 | 2020-04-14T13:58:02 | {} | MEMBER |
ClickHouse | ClickHouse | 613,503,108 | 6,802 | alexey-milovidov | "ASAP" is going to be nearest days. | 2020-04-14T15:15:17 | 2020-04-14T15:15:17 | {} | MEMBER |
ClickHouse | ClickHouse | 613,611,239 | 10,243 | drook | FreeBSD ports is a system where "ported" 3rd-party software, which is not a part of FreeBSD base system is stored and maintained. Process of "porting" is basically and merely a process of describing the software, the scheme for its building and the source for a tarball. It is a community-managed official source of all ... | 2020-04-14T18:36:53 | 2020-04-14T18:42:29 | {} | NONE |
ClickHouse | ClickHouse | 613,627,196 | 10,197 | 4ertus2 | No chance to fix it inside *Join.cpp. So we have to fix it outside. There we could auto-wrap keys into `toSupertype(t1.k) = toSupertype(t2.k)` in SyntaxAnalyzer. Then it should work. It's not very difficult I think. The main trouble is to get function name from known supertype. We could parse `CAST(x AS Supertype)` and... | 2020-04-14T19:06:57 | 2020-04-14T19:06:57 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 612,868,034 | 10,183 | abraithwaite | > But what an action do you expect on failure?
That's why I'm not sure about retries, but failing the script would be better than continuing. | 2020-04-13T11:50:55 | 2020-04-13T11:50:55 | {} | NONE |
ClickHouse | ClickHouse | 613,515,835 | 10,252 | den-crane | is_frozen -- is not related to TTL or to active.
is_frozen -- means that this part's folder has `-r--r--r--` filesystem flags instead of `-rw-rw-r--` it means that this part was fetched from other replica or a `freeze` command (backup) was executed against this part. | 2020-04-14T15:38:04 | 2020-04-14T15:38:56 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 612,793,663 | 2,805 | qoega | @zhang2014 do you plan to return to this issue? | 2020-04-13T07:48:59 | 2020-04-13T07:48:59 | {} | MEMBER |
ClickHouse | ClickHouse | 612,858,612 | 10,229 | soloduev | Now, if I understand correctly, day zero is set, if we do not fall into the range?
[Code line common/DateLUTImpl.h 667](https://github.com/ClickHouse/ClickHouse/blob/master/base/common/DateLUTImpl.h#L667)
Even here marked as _TODO_... | 2020-04-13T11:16:07 | 2020-04-13T11:18:09 | {} | NONE |
ClickHouse | ClickHouse | 613,129,501 | 10,237 | alexey-milovidov | Tests did not start in three hours. | 2020-04-13T22:32:33 | 2020-04-13T22:32:33 | {} | MEMBER |
ClickHouse | ClickHouse | 613,352,709 | 8,639 | alesapin | `02000_optimize_skip_unused_shards_type_mismatch` and `01120_join_constants` -- tests of features which are absent in 20.1.
`00804_test_delta_codec_compression` in unbundled mode -- test fixed in master.
`Integration test_server_initialization` -- ok, test fixed in master. | 2020-04-14T10:16:05 | 2020-04-14T10:16:05 | {} | MEMBER |
ClickHouse | ClickHouse | 613,479,612 | 10,256 | alexey-milovidov | @qoega We must also add a check for "AppleClang" compiler but I don't know what is their versioning scheme. | 2020-04-14T14:34:48 | 2020-04-14T14:34:48 | {} | MEMBER |
ClickHouse | ClickHouse | 613,550,023 | 9,950 | akuzm | > I'm mostly Ok with this solution.
>
> BTW, What will happen in query:
> `select cast(7 as String), * from (select number "'String'" FROM numbers(2));`
> ?
Works as expected:
```
:) select cast(7 as String), * from (select number "'String'" FROM numbers(2));
┌─CAST(7, 'String')─┬─'String'─┐
│ 7 ... | 2020-04-14T16:38:23 | 2020-04-14T16:38:23 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 613,627,791 | 10,252 | den-crane | >it means that this part was fetched from other replica
anyway it's not related to TTL. You can `unfreeze` them using chmod | 2020-04-14T19:08:09 | 2020-04-14T19:09:02 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 613,690,687 | 10,243 | alexey-milovidov | We did not provide source tarballs,
and tarballs available on GitHub are incomplete - they don't contain submodules that are essential for build.
| 2020-04-14T21:28:23 | 2020-04-14T21:28:40 | {} | MEMBER |
ClickHouse | ClickHouse | 612,815,941 | 10,212 | asb-72 | Also on system.query_log:
```
ALTER TABLE system.query_log
DROP PARTITION '2020-04-06'
Received exception from server (version 20.3.5):
Code: 72. DB::Exception: Received from 192.168.34.165:9000. DB::Exception: Unsigned type must not contain '-' symbol: Cannot parse UInt32 from String.
```
| 2020-04-13T08:57:32 | 2020-04-13T08:57:32 | {} | NONE |
ClickHouse | ClickHouse | 612,837,828 | 10,230 | drook | [build.log.gz](https://github.com/ClickHouse/ClickHouse/files/4469037/build.log.gz)
| 2020-04-13T10:04:43 | 2020-04-13T10:04:43 | {} | NONE |
ClickHouse | ClickHouse | 612,929,180 | 9,339 | filimonov | MySQL also has its own:
```
List of all MySQL commands:
Note that all text commands must be first on line and end with ';'
? (\?) Synonym for `help'.
clear (\c) Clear the current input statement.
connect (\r) Reconnect to the server. Optional arguments are db and host.
delimiter (\d) Set statement ... | 2020-04-13T14:45:30 | 2020-04-13T14:46:31 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 613,383,832 | 10,190 | BayoNet | In progress. | 2020-04-14T11:23:06 | 2020-04-14T11:23:06 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 613,462,049 | 10,243 | alexey-milovidov | ClickHouse should be build from git repository with submodules, not from ports. | 2020-04-14T14:03:16 | 2020-04-14T14:03:16 | {} | MEMBER |
ClickHouse | ClickHouse | 613,578,236 | 9,852 | vitlibar | Backport to `19.17` is impossible (because the implementation relies on the format change in https://github.com/ClickHouse/ClickHouse/pull/8044). | 2020-04-14T17:33:25 | 2020-04-14T17:33:25 | {} | MEMBER |
ClickHouse | ClickHouse | 612,795,784 | 5,323 | qoega | Probably fixed. I can't reproduce on 20.3 as I don't see OR -> IN optimizations in queries. | 2020-04-13T07:56:07 | 2020-04-13T07:56:07 | {} | MEMBER |
ClickHouse | ClickHouse | 612,876,763 | 9,334 | youssefqarri | I have the same issue, with a basic configuration :
- one partition
- CSV format (not json)
It works for few hours, and then, with any intervention or modification of the messages formats, it stops working. | 2020-04-13T12:22:49 | 2020-04-13T12:22:49 | {} | NONE |
ClickHouse | ClickHouse | 606,123,334 | 9,948 | alesapin | Status:
- [ ] doesn't work for compact parts.
- [ ] limited support for nested columns.
| 2020-03-30T17:05:40 | 2020-04-13T15:29:43 | {} | MEMBER |
ClickHouse | ClickHouse | 613,142,722 | 10,241 | SaltTan | The change happened in September 2019
19.11.11.57 -> 19.11.12.69
19.13.4.32 -> 19.13.5.44 | 2020-04-13T23:17:26 | 2020-04-13T23:17:26 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 613,514,941 | 7,430 | alexey-milovidov | This is still relevant and @vitlibar said that he will continue whenever he will have a time. | 2020-04-14T15:36:28 | 2020-04-14T15:36:28 | {} | MEMBER |
ClickHouse | ClickHouse | 613,736,893 | 10,268 | alexey-milovidov | @KochetovNicolai It triggers "Pipeline stuck". | 2020-04-14T23:51:10 | 2020-04-14T23:51:16 | {} | MEMBER |
ClickHouse | ClickHouse | 612,800,762 | 10,224 | org0000h | thank you | 2020-04-13T08:12:04 | 2020-04-13T08:12:04 | {} | NONE |
ClickHouse | ClickHouse | 612,815,728 | 9,624 | alesapin | `Build 14/17` -- ok, this branch doesn't support clang tidy build and builds for freebsd/arm.
`Unit tests relwithdebuginfo gcc-9`: `ThreadPool.ExceptionFromSchedule` -- test fixed in master
`Integration tests:` -- `test_server_initialization/test.py::test_partially_dropped_tables` -- test fixed in master.
`Integrat... | 2020-04-13T08:56:59 | 2020-04-13T08:56:59 | {} | MEMBER |
ClickHouse | ClickHouse | 612,817,946 | 10,229 | qoega | It is well documented https://clickhouse.tech/docs/en/sql_reference/data_types/datetime/
DateTime data type
```
Supported range of values: [1970-01-01 00:00:00, 2105-12-31 23:59:59].
```
C# MaxValue/MinValue has nothing in common with ClickHouse.
```sql
SELECT toDateTime('9999-12-31T20:59:59', 'UTC')
┌─toDa... | 2020-04-13T09:03:09 | 2020-04-13T09:03:09 | {} | MEMBER |
ClickHouse | ClickHouse | 611,469,848 | 10,116 | eiffel0311 | > What CH version do you use?
>
> Can you show 'show create table A' and 'show create table B'
tableB(from table tableB):
CREATE TABLE tutorial.dws_brand_ss (
`industry_id` UInt16,
`doc_id` UInt64,
`account_name` String,
`account_type_id` UInt8,
`collect_cnt` Nullable(UInt32),
`reply_cnt` UInt32,
`for... | 2020-04-09T11:06:53 | 2020-04-13T10:20:42 | {} | NONE |
ClickHouse | ClickHouse | 612,888,595 | 10,217 | alesapin | Fix only for processors. Cherry-pick in 20.1 is not required. | 2020-04-13T13:02:30 | 2020-04-13T13:02:30 | {} | MEMBER |
ClickHouse | ClickHouse | 613,046,258 | 10,052 | akuzm | > > Performance — 150 errors
>
> * unknown issue.
Should be fixed in master now (def6817ede8854d7c422429d2bbe9204c134ae25) | 2020-04-13T19:06:55 | 2020-04-13T19:06:55 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 613,251,519 | 5,837 | Stephane987 | Hi, I had the same question and found the answer that I posted here :
https://stackoverflow.com/questions/61120771/how-to-configure-nifi-invokehttp-to-post-insert-in-clickhouse | 2020-04-14T06:29:50 | 2020-04-14T06:30:16 | {} | NONE |
ClickHouse | ClickHouse | 613,289,877 | 8,183 | simPod | Do you plan to somehow make this operation easier on ClickHouse side? | 2020-04-14T08:05:17 | 2020-04-14T08:05:17 | {
"+1": 9,
"eyes": 3
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 613,734,109 | 4,385 | alexey-milovidov | Cannot merge with master. Will open a new PR. | 2020-04-14T23:40:51 | 2020-04-14T23:40:51 | {} | MEMBER |
ClickHouse | ClickHouse | 613,744,009 | 6,256 | alexey-milovidov | Continued in #10269
I have added your tests and slightly modified the code.
Let's look at performance tests... | 2020-04-15T00:15:07 | 2020-04-15T00:15:07 | {} | MEMBER |
ClickHouse | ClickHouse | 613,744,292 | 6,256 | alexey-milovidov | And sorry for the delay :joy_cat: | 2020-04-15T00:16:08 | 2020-04-15T00:16:08 | {} | MEMBER |
ClickHouse | ClickHouse | 613,744,653 | 10,269 | alexey-milovidov | The test of interest is `SELECT count() FROM hits_100m_single WHERE NOT ignore(path(URL))` | 2020-04-15T00:17:30 | 2020-04-15T00:17:30 | {} | MEMBER |
ClickHouse | ClickHouse | 613,784,378 | 10,216 | alexey-milovidov | Something is still wrong in performance test. | 2020-04-15T02:49:02 | 2020-04-15T02:49:02 | {} | MEMBER |
ClickHouse | ClickHouse | 612,791,425 | 4,222 | qoega | Reproducing in 20.4
**How to reproduce**
``` sql
SELECT
dummy,
x
FROM system.one
ANY LEFT JOIN
(
SELECT
dummy,
materialize(1) AS x
) AS t2 USING (dummy)
GROUP BY dummy
```
DB::Exception: **Not found column x in block. There are only columns: dummy.**
**Expected behavior**... | 2020-04-13T07:41:31 | 2020-04-13T07:41:31 | {} | MEMBER |
ClickHouse | ClickHouse | 612,949,886 | 9,950 | 4ertus2 | Another fix #10207 | 2020-04-13T15:31:53 | 2020-04-13T15:31:53 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 612,984,021 | 8,183 | alexey-milovidov | @den-crane Attaching parts one by one will lead to data duplication if there are old parts after merge along with merged parts. I don't recommend doing ATTACH PART in a loop.
@filimonov This is way better! | 2020-04-13T16:50:00 | 2020-04-13T16:51:17 | {} | MEMBER |
ClickHouse | ClickHouse | 613,552,701 | 10,198 | akuzm | Will backport it, because the test in bug fix https://github.com/ClickHouse/ClickHouse/pull/10207 depends on it, and the function is self-contained. | 2020-04-14T16:43:40 | 2020-04-14T16:44:02 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 613,032,964 | 10,052 | alexey-milovidov | `01231_distributed_aggregation_memory_efficient_mix_levels`
- already fixed. | 2020-04-13T18:37:14 | 2020-04-13T18:37:14 | {} | MEMBER |
ClickHouse | ClickHouse | 613,522,956 | 7,662 | alexey-milovidov | @akuzm said that this interface does not give performance benefits.
Although the change is not very complex, we will postpone it. | 2020-04-14T15:50:25 | 2020-04-14T15:50:25 | {} | MEMBER |
ClickHouse | ClickHouse | 613,531,284 | 9,940 | qoega | > @qoega Which are `CREATE` statements and are there any ways to reproduce that?
Here is complete scenario, but I'm sure is can be minimized
https://gist.github.com/qoega/64c15b57a96e602bd779e4ee115cba17
@excitoon | 2020-04-14T16:03:56 | 2020-04-14T16:04:22 | {} | MEMBER |
ClickHouse | ClickHouse | 613,793,279 | 1,839 | alexey-milovidov | `+inf` is not supported in all text formats including TSV, CSV, JSONEachRow. | 2020-04-15T03:22:18 | 2020-04-15T03:32:54 | {} | MEMBER |
ClickHouse | ClickHouse | 612,792,163 | 8,629 | blinkov | @ZonghaoWang, is this question still relevant? | 2020-04-13T07:44:00 | 2020-04-13T07:44:00 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 612,942,463 | 10,229 | filimonov | See https://github.com/ClickHouse/ClickHouse/pull/9404
You can check that / take a build from CI / help with QA. | 2020-04-13T15:15:19 | 2020-04-13T15:15:19 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 613,475,978 | 10,215 | qoega | Darwin build fails
```
[2975/4015] Linking CXX executable base/common/tests/date_lut2
FAILED: base/common/tests/date_lut2
: && /usr/bin/ccache /usr/bin/clang++-8 --target=x86_64-apple-darwin -std=c++17 -fdiagnostics-color=always -std=c++2a -fsized-deallocation -gdwarf-aranges -pipe -msse4.1 -msse4.2 -mpopcnt -... | 2020-04-14T14:28:13 | 2020-04-14T14:28:13 | {} | MEMBER |
ClickHouse | ClickHouse | 613,732,311 | 10,250 | alexey-milovidov | I've added a check, so the build won't even start on gcc-8. | 2020-04-14T23:34:37 | 2020-04-14T23:34:37 | {} | MEMBER |
ClickHouse | ClickHouse | 613,774,477 | 10,212 | alexey-milovidov | @asb-72 system.query_log has monthly partitions by default, not daily. | 2020-04-15T02:10:36 | 2020-04-15T02:10:36 | {} | MEMBER |
ClickHouse | ClickHouse | 613,791,667 | 4,101 | alexey-milovidov | > What I'm trying to understand at the moment, is why getMaxSourcePartsSize is implemented the way it is.
> Why bigger merges are preferred when thread pool is empty is smaller when thread pool is busy?
We have fixed number of slots in background processing pool and we don't want for long running merges to occupy... | 2020-04-15T03:16:08 | 2020-04-15T03:17:11 | {
"+1": 4
} | MEMBER |
ClickHouse | ClickHouse | 613,797,116 | 2,509 | alexey-milovidov | Now you can add -OrNull suffix to any aggregate function. For example, `minOrNull`.
We still don't have an option to make it default. | 2020-04-15T03:37:10 | 2020-04-15T03:37:10 | {} | MEMBER |
ClickHouse | ClickHouse | 612,799,784 | 9,361 | filimonov | You can try smth like.
```
insert_order DEFAULT bitOr(bitShiftLeft(toUInt64(now64()),24), rowNumberInAllBlocks())
```
| 2020-04-13T08:08:51 | 2020-04-13T08:08:51 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 613,034,488 | 10,052 | alexey-milovidov | But I need to figure out how INSERT SELECT appeared to be parallel for materialized view after this change... | 2020-04-13T18:40:26 | 2020-04-13T18:40:26 | {} | MEMBER |
ClickHouse | ClickHouse | 613,440,875 | 9,990 | alex-zaitsev | I like extractAllGroups. Taking the example above
```
SELECT extractAllGroups('abc=111, def=222, ghi=333', '("[^"]*?"|[\\w]*)=("[^"]*?"|[\\w]*)')
should return
[['abc','def','ghi'], ['111','333','333']]
```
| 2020-04-14T13:25:15 | 2020-04-14T13:25:25 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 613,463,098 | 10,250 | qoega | In your log there is `/usr/lib/ccache/g++-8` | 2020-04-14T14:05:11 | 2020-04-14T14:05:11 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 613,500,642 | 5,387 | alexey-milovidov | @akuzm will continue tomorrow. | 2020-04-14T15:11:14 | 2020-04-14T15:11:14 | {} | MEMBER |
ClickHouse | ClickHouse | 613,502,731 | 6,185 | alexey-milovidov | @KochetovNicolai is going to continue on this week. | 2020-04-14T15:14:38 | 2020-04-14T15:14:38 | {} | MEMBER |
ClickHouse | ClickHouse | 613,632,397 | 10,005 | 4ertus2 | The issue on picture is duplicate for #9810 | 2020-04-14T19:16:54 | 2020-04-14T19:16:54 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 613,128,922 | 10,052 | alexey-milovidov | > But I need to figure out how INSERT SELECT appeared to be parallel for materialized view after this change...
Ok. | 2020-04-13T22:30:41 | 2020-04-13T22:30:41 | {} | MEMBER |
ClickHouse | ClickHouse | 613,129,994 | 10,239 | alexey-milovidov | It's strange that the task for `Functional stateful tests (debug)` is created only at `14 April 2020, 01:02:12` - after 6 hours waiting. | 2020-04-13T22:34:03 | 2020-04-13T22:34:03 | {} | MEMBER |
ClickHouse | ClickHouse | 613,290,326 | 10,249 | dependabot-preview[bot] | Looks like jinja2 is up-to-date now, so this is no longer needed. | 2020-04-14T08:06:15 | 2020-04-14T08:06:15 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 613,458,674 | 10,254 | alexey-milovidov | Already done: #9091 | 2020-04-14T13:57:15 | 2020-04-14T13:57:15 | {} | MEMBER |
ClickHouse | ClickHouse | 613,691,228 | 10,243 | alexey-milovidov | You can obtain complete source tree with
git clone **--recursive** https://github.com/ClickHouse/ClickHouse.git | 2020-04-14T21:29:49 | 2020-04-14T21:29:49 | {} | MEMBER |
ClickHouse | ClickHouse | 624,218,406 | 10,486 | azat | @alexey-milovidov any news on this one? | 2020-05-05T18:08:45 | 2020-05-05T18:08:45 | {} | MEMBER |
ClickHouse | ClickHouse | 624,227,717 | 6,515 | alexey-milovidov | @fancyqlx
> The memory usage became normal after I used FreeThreadPool in AsynchronousBlockInputStream instead of GlobalThreadPool. I guess this kind of thread pool may hold memory in an unexpected way.
Yes, it may happen due to combination of usage of ThreadPool and specific behaviour of jemalloc.
If the threa... | 2020-05-05T18:26:18 | 2020-05-05T18:26:42 | {} | MEMBER |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.