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 | 698,362,848 | 15,233 | den-crane | Can you show DDL. `show create table my_table` Need to see an engine and column types. | 2020-09-24T13:59:10 | 2020-09-24T13:59:10 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,482,275 | 15,003 | tavplubix | Functional stateless tests (release, polymorphic parts enabled) — fail: 3, passed: 2271, skipped: 22:
```
2020-09-24 17:58:54 [22883b8487e4] 2020.09.24 17:58:53.993944 [ 46410 ] {2f7f8e82-6c66-4efe-a943-f02f916bd55e} <Error> executeQuery: Code: 57, e.displayText() = DB::Exception: Directory for table data store/68c/... | 2020-09-24T17:28:04 | 2020-09-24T17:28:04 | {} | MEMBER |
ClickHouse | ClickHouse | 698,579,878 | 15,255 | siradjev | Seems this one is fixed for 20.7.3.7 https://t.me/clickhouse_ru/183597
But it would be good to have test cases, given how frequently `toDate(DateTimeColumn)=toDate('YYYY-MM-DD')` construct breaks...
| 2020-09-24T20:46:40 | 2020-09-24T20:46:40 | {} | NONE |
ClickHouse | ClickHouse | 697,763,190 | 15,123 | alexey-milovidov | There is a feature request to allow to convert partition key from
date or toDate(datetime)
to
toMonday(date) or toStartOfMonth(date) or similar. | 2020-09-23T17:38:08 | 2020-09-23T17:38:08 | {} | MEMBER |
ClickHouse | ClickHouse | 697,835,535 | 15,201 | KochetovNicolai | ```
:) explain optimize = 1, actions = 1 select number from (select number from numbers(100000000) order by number desc) limit 10
EXPLAIN optimize = 1, actions = 1
SELECT number
FROM
(
SELECT number
FROM numbers(100000000)
ORDER BY number DESC
)
LIMIT 10
┌─explain────────────────────────────... | 2020-09-23T18:22:34 | 2020-09-23T18:22:34 | {} | MEMBER |
ClickHouse | ClickHouse | 698,091,955 | 15,185 | cmsxbc | > @cmsxbc I am not offer that solution to you. I just tie tickets for developers to get understanding what is going on.
>
> For you I can offer to "rename" index column to _index. But there is no easy way. You can add a new column and name it for example _index and place default as index `alter table xxxx add column... | 2020-09-24T03:38:58 | 2020-09-24T03:38:58 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,249,733 | 15,143 | 4ertus2 | Tests need #15002 backported first | 2020-09-24T10:08:13 | 2020-09-24T10:08:13 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,249,818 | 15,140 | 4ertus2 | Tests need #15002 backported first | 2020-09-24T10:08:22 | 2020-09-24T10:08:22 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,463,880 | 15,227 | alexey-milovidov | > It is a part of what, then?
Totals and extremes are separate (secondary) resultsets.
BTW, in older ClickHouse versions they were output to client even in INSERT SELECT query.
> Should WITH TOTALS be prohibited for INSERT SELECT?
Yes, we can do it. | 2020-09-24T16:52:04 | 2020-09-24T16:52:09 | {} | MEMBER |
ClickHouse | ClickHouse | 698,640,995 | 7,948 | den-crane | related: https://github.com/ClickHouse/ClickHouse/issues/15255
```
20.10.1.4704.
SELECT count() FROM test_partition_filtering WHERE toDate(toStartOfDay(timestamp)) = today() and zoneId=42;
Selected 1 parts by date, 1 parts by key, 1 marks by primary key, 1 marks to read from 1 ranges
Processed 8.19 thousand rows... | 2020-09-24T23:37:28 | 2020-09-24T23:39:32 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 697,726,588 | 13,613 | Slach | @grmorozov unfortunately `dictionaries_lazy_load` doesn't work for me on 20.3-lts version ;( | 2020-09-23T17:16:18 | 2020-09-23T17:16:18 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,226,993 | 15,144 | akuzm | Finally it works, and we have much less unstable queries than in master (only 12 compared to ~100):
https://clickhouse-test-reports.s3.yandex.net/15144/21f9cc65c99e8210d7b14a4bef5969710451bc5e/performance_comparison/report.html#fail1
After jemalloc bug is fixed, we can enable this in master. Or maybe we can just di... | 2020-09-24T09:22:53 | 2020-09-24T09:22:53 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,279,373 | 15,228 | filimonov | https://github.com/ClickHouse/ClickHouse/pull/13679 ? | 2020-09-24T11:15:42 | 2020-09-24T11:15:42 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,327,844 | 15,229 | den-crane | >I have a replication queue full on a replica (700 items+)
check network. Use `ping -s 1500 ` to ping among replicas back and forth.
check `system.replication_queue last_exception` column. It could be a lost of a part.
| 2020-09-24T12:59:21 | 2020-09-24T12:59:33 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,465,810 | 14,937 | alexey-milovidov | Actually it's not a bugfix.
But backporting it should be easy. We only have to fix some whitespaces in the test reference... | 2020-09-24T16:55:38 | 2020-09-24T16:55:38 | {} | MEMBER |
ClickHouse | ClickHouse | 698,592,869 | 15,255 | UnamedRus | Still broken in 20.7, but it seems fixed in 20.8, probably that [pr](https://github.com/ClickHouse/ClickHouse/pull/13497)
```
CREATE TABLE part_prunning_test
(
`ts` DateTime,
`dt` Date DEFAULT toDate(ts),
`db_time` DateTime DEFAULT ts
)
ENGINE = MergeTree()
PARTITION BY toDate(ts)
ORDER BY tuple... | 2020-09-24T21:16:22 | 2020-09-24T21:17:47 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,383,646 | 15,227 | akuzm | It is a part of what, then?
Should WITH TOTALS be prohibited for INSERT SELECT?
A query that works differently depending on context is a usability failure. | 2020-09-24T14:32:43 | 2020-09-24T14:32:43 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 697,847,293 | 15,200 | den-crane | hm,
```
`tag.key`
```
works
```sql
SELECT 1
FROM main_table
ALL INNER JOIN join_table USING (id)
WHERE empty(`tag.key`) = 0
``` | 2020-09-23T18:30:02 | 2020-09-23T18:30:02 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,158,897 | 15,181 | alesapin | Any ideas why it's processing so slow? Which queries you have in DDL queue? | 2020-09-24T07:07:11 | 2020-09-24T07:07:11 | {} | MEMBER |
ClickHouse | ClickHouse | 698,408,746 | 15,234 | danielkurniadi | I see. Thanks! | 2020-09-24T15:13:02 | 2020-09-24T15:13:02 | {} | NONE |
ClickHouse | ClickHouse | 698,584,733 | 15,233 | durchgedreht | Hi,
I can confirm the mentioned behaviour, your sample code works here as well! Stripping down our statement line by line I found that we have a dependency to another field that is created by IF statement and is even using the same alias and a join column as well. Even this works ok when I try to add that to your samp... | 2020-09-24T20:57:47 | 2020-09-24T20:57:47 | {} | NONE |
ClickHouse | ClickHouse | 697,757,409 | 9,336 | maqroll | Most prefab queries use just UNION. MySQL assumes DISTINCT by default. CK fails. Wouldn't be nice to have a setting to establish default behaviour of UNION to UNION ALL (and set it by default on mysql sessions)? | 2020-09-23T17:34:37 | 2020-09-23T17:34:37 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,079,438 | 14,298 | Awezome | I think /etc/cron.d/clickhouse-server also needs to be upgraded for systemctl . I used systemctl to start the server and still turned on the cron.d then got pid conflict because of the different pid . | 2020-09-24T02:50:48 | 2020-09-24T02:50:48 | {} | NONE |
ClickHouse | ClickHouse | 698,351,490 | 15,229 | den-crane | hmm, then maybe it a lack of background jobs because they all are busy and do merges / mutations / kafka ...
```
SELECT
name,
value
FROM system.settings
WHERE name LIKE '%pool%'
┌─name─────────────────────────────────────────┬─value─┐
│ connection_pool_max_wait_ms │ 0 │
│ distr... | 2020-09-24T13:41:16 | 2020-09-24T13:42:19 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,364,107 | 15,233 | den-crane | > 20.3.4.10
very old release.
And check that issue exists in the latest 20.3.19.4 | 2020-09-24T14:00:50 | 2020-09-24T14:00:50 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,409,044 | 14,414 | azat | >docker: invalid reference format: repository name must be lowercase
And this also breaks integration tests completely, I guess the problem is missing space before `-e`:
> `-e DOCKER_BASE_TAG=15007-371eafea11b0f80e8fca46013f043713aefb329b-e `
```
Running pytest container as: 'docker run --net=host --rm --nam... | 2020-09-24T15:13:31 | 2020-09-24T15:13:31 | {} | MEMBER |
ClickHouse | ClickHouse | 698,636,018 | 54 | alexey-milovidov | There is one more argument that we should throw an exception: #15249 | 2020-09-24T23:19:10 | 2020-09-24T23:19:10 | {} | MEMBER |
ClickHouse | ClickHouse | 697,863,554 | 14,731 | azat | >Conflicting files
>src/Storages/IStorage.h
rebased against upstream/master | 2020-09-23T18:41:22 | 2020-09-23T18:41:22 | {} | MEMBER |
ClickHouse | ClickHouse | 698,056,978 | 15,073 | alexey-milovidov | It will be better to allow arbitrary set of columns and PRIMARY KEY specification.
Then primary key columns and the rest columns will be serialized and deserialized with IDataType::serialize/deserializeBinary into single key/value blobs.
It will greater improve the usability of this feature! | 2020-09-24T01:25:15 | 2020-09-24T01:25:38 | {
"+1": 3
} | MEMBER |
ClickHouse | ClickHouse | 698,062,754 | 11,871 | alexey-milovidov | Geometry data types should be parametrized by "metric".
Metric can be: `Euclidean` and `LocalWGS84`.
Example: `Point('LocalWGS84')`.
When we do any calculation on geometry type, we should do the following:
- compute the bounding box of all geometries participating in the calculation (for all function arguments ... | 2020-09-24T01:46:08 | 2020-09-24T01:48:25 | {} | MEMBER |
ClickHouse | ClickHouse | 698,166,080 | 15,210 | yiguolei | @filimonov I found there is a cluster definition in config.xml.
`
<test_cluster_two_shards_localhost>
<shard>
<replica>
<host>localhost</host>
<port>9000</port>
</replica>
</shard>
<shar... | 2020-09-24T07:22:49 | 2020-09-24T07:23:30 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,235,621 | 15,189 | 4ertus2 | Fuzzer error `DB::Exception: Incorrect CREATE query: required list of column descriptions or AS section or SELECT. (version 20.10.1.4736) (from [::1]:35130) (in query: ATTACH TABLE test_01190.dict)` | 2020-09-24T09:39:44 | 2020-09-24T09:39:44 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,246,983 | 15,141 | 4ertus2 | Failed test test_complex_dictionaries is not related to PR changes | 2020-09-24T10:02:46 | 2020-09-24T10:02:46 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,295,005 | 15,228 | 4ertus2 | There's no `ID` in table `table`. Merge pattern '^table' should combine both `table` and `table_2`, `GROUP BY ID, key` cannot find `ID` in one of combined tables. It was wrong if it worked before. | 2020-09-24T11:51:47 | 2020-09-24T11:51:47 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,298,499 | 15,211 | l1t1 | can it support sync the views of mysql table, such as some columns,rows of a table? | 2020-09-24T11:59:33 | 2020-09-24T11:59:33 | {} | NONE |
ClickHouse | ClickHouse | 698,324,735 | 15,210 | den-crane | @yiguolei it's for tests in CH developing. The same table (the same data) appears twice in a query result. | 2020-09-24T12:53:25 | 2020-09-24T12:53:25 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,087,222 | 15,185 | den-crane | @cmsxbc I am not offering that solution to you. I just tie tickets for developers to get understanding what is going on.
For you I can offer to "rename" index column to _index. But there is no easy way. You can add a new column and name it for example _index and place default as index `alter table xxxx add column _... | 2020-09-24T03:20:14 | 2020-09-24T13:03:25 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,635,184 | 14,937 | alexey-milovidov | I will try to merge these backports. | 2020-09-24T23:16:11 | 2020-09-24T23:16:11 | {} | MEMBER |
ClickHouse | ClickHouse | 697,754,486 | 15,178 | alexey-milovidov | > test_s3_with_proxy
Flaky test, @Jokser | 2020-09-23T17:32:52 | 2020-09-23T17:32:52 | {} | MEMBER |
ClickHouse | ClickHouse | 698,270,978 | 15,191 | alesapin | Seems like test failures related to changes. | 2020-09-24T10:55:37 | 2020-09-24T10:55:37 | {} | MEMBER |
ClickHouse | ClickHouse | 698,369,296 | 11,298 | abyss7 | Status for now (0_stateless): `1369 failed, 456 passed, 1 skipped in 325.03s (0:05:25)` | 2020-09-24T14:09:20 | 2020-09-24T14:09:20 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,370,062 | 15,234 | den-crane | No such feature as index rename
Just create a new index with a new name and drop the old one.
https://clickhouse.tech/docs/en/sql-reference/statements/alter/index/ | 2020-09-24T14:10:31 | 2020-09-24T14:10:31 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,373,400 | 15,215 | alesapin | Waiting for fuzzer and at least one stress test. | 2020-09-24T14:15:50 | 2020-09-24T14:15:50 | {} | MEMBER |
ClickHouse | ClickHouse | 698,592,831 | 15,255 | siradjev | or 20.9... https://t.me/clickhouse_ru/183624 | 2020-09-24T21:16:17 | 2020-09-24T21:16:17 | {} | NONE |
ClickHouse | ClickHouse | 698,083,787 | 15,185 | cmsxbc | > related [#4390 (comment)](https://github.com/ClickHouse/ClickHouse/issues/4390#issuecomment-463623903)
The solution seems not work for copier.... the sql got from older server is not escaped yet.... | 2020-09-24T03:07:27 | 2020-09-24T03:07:27 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,144,784 | 15,187 | zhang2014 | Are you using mysql prefix indexes(not implement, we should support it)? | 2020-09-24T06:34:45 | 2020-09-24T06:35:09 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,147,354 | 15,179 | zhang2014 | I will trying to fix it. | 2020-09-24T06:40:41 | 2020-09-24T06:40:41 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,328,010 | 15,227 | alexey-milovidov | Please make some edits to the documentation. | 2020-09-24T12:59:39 | 2020-09-24T12:59:39 | {} | MEMBER |
ClickHouse | ClickHouse | 698,329,815 | 15,229 | simPod | Ping is 0.1 ms and
```sql
SELECT *
FROM system.replication_queue
WHERE table = 't1'
AND last_exception <> '';
```
is empty 🤔 I have active monitoring on broken items in replication queue, it's without exceptions the whole time.
All are postponed 0 times, `last_attempt_time` is `1970-01-01 00:00:00`. No... | 2020-09-24T13:02:50 | 2020-09-24T13:04:28 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,336,818 | 14,265 | tavplubix | Reverted because it breaks integration tests: `Integration tests (asan) — fail: 0, passed: 0, error: 0 `
| 2020-09-24T13:15:28 | 2020-09-24T13:15:28 | {} | MEMBER |
ClickHouse | ClickHouse | 698,348,315 | 15,233 | den-crane | try to remove alias `fact.`
and query columns with table `my_table.sa_type` or just `sa_type` | 2020-09-24T13:35:43 | 2020-09-24T13:35:43 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,440,371 | 15,144 | akuzm | > Maybe we can also try binding the two instances to different CPU nodes. It might have unfair CPU frequency but less cache contentions. It might also be good to leave out CPU 0 due to system interruption handling
I'm afraid of introducing some systemic difference by binding to different nodes, but I'll try it as we... | 2020-09-24T16:07:15 | 2020-09-24T16:07:15 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,479,715 | 14,879 | nikitamikhaylov | Too easy, nothing to review. | 2020-09-24T17:23:22 | 2020-09-24T17:23:22 | {} | MEMBER |
ClickHouse | ClickHouse | 698,619,897 | 14,414 | tavplubix | MemorySanitizer - use-of-uninitialized-value in `MergedBlockOutputStream::writeWithPermutation(...)`
https://clickhouse-test-reports.s3.yandex.net/15240/dce6a436f37713224c242092446e1a9fcf8b4a4a/stress_test_(memory)/stderr.log | 2020-09-24T22:25:04 | 2020-09-24T22:25:04 | {} | MEMBER |
ClickHouse | ClickHouse | 698,635,593 | 15,249 | alexey-milovidov | Duplicate of https://github.com/ClickHouse/ClickHouse/issues/54 | 2020-09-24T23:17:41 | 2020-09-24T23:17:41 | {} | MEMBER |
ClickHouse | ClickHouse | 698,166,137 | 15,192 | alesapin | Performance not affected, other tests ok. | 2020-09-24T07:22:55 | 2020-09-24T07:22:55 | {} | MEMBER |
ClickHouse | ClickHouse | 698,347,121 | 14,937 | crakjie | I made a PR to backport this to 20.3 https://github.com/ClickHouse/ClickHouse/pull/15231
| 2020-09-24T13:33:43 | 2020-09-24T13:33:43 | {} | NONE |
ClickHouse | ClickHouse | 698,399,459 | 15,236 | alesapin | Just checking what will happen. | 2020-09-24T14:57:51 | 2020-09-24T14:57:51 | {} | MEMBER |
ClickHouse | ClickHouse | 698,404,734 | 15,228 | 4ertus2 | Also I've made a mistake in query and found that we know `ID` as required and available column:
```
SELECT ID FROM m INNER JOIN b ON a.key = b.key GROUP BY b.ID;
Code: 47. DB::Exception: Received from localhost:9000. DB::Exception: Missing columns: 'a.key' while processing query: 'SELECT ID FROM m ALL INNER JOIN (... | 2020-09-24T15:06:13 | 2020-09-24T15:06:54 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,535,323 | 15,233 | den-crane | I could not reproduce it in 20.10
```sql
CREATE TABLE my_table (delivery_date DateTime, event_records Int64, sa_type Nullable(String))
ENGINE = MergeTree ORDER BY tuple()
insert into my_table select today(), number, 'a' from numbers(1000);
insert into my_table select today(), number, 'c' from numbers(1000);... | 2020-09-24T19:12:30 | 2020-09-24T19:12:30 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 697,737,171 | 6,953 | alexey-milovidov | Assigned to @ucasFL | 2020-09-23T17:22:38 | 2020-09-23T17:22:38 | {} | MEMBER |
ClickHouse | ClickHouse | 697,857,284 | 15,177 | azat | >In #15035 @azat said that wait() call us needed, however I do not know why.
Because [ThreadPool::wait() waits until scheduled_jobs](https://github.com/ClickHouse/ClickHouse/blob/3bc09bb867c7206e6ec4f1fb7a658b924f3d8276/src/Common/ThreadPool.cpp#L161) will be decremented, and this guarantee that the [job](https://gi... | 2020-09-23T18:36:55 | 2020-09-23T18:37:45 | {} | MEMBER |
ClickHouse | ClickHouse | 697,949,972 | 14,731 | alexey-milovidov | > @alexey-milovidov any news?
No news so far. | 2020-09-23T20:18:50 | 2020-09-23T20:18:50 | {} | MEMBER |
ClickHouse | ClickHouse | 698,096,472 | 15,210 | yiguolei | In this case, I have to re-shard data if I scale my cluster from 1 node to 2 node. And then have to re-shard again when I scale my cluster from 2 node to 4 node ..... And I have to stop importing data during re-sharding process.
Any better ideas?
| 2020-09-24T03:56:53 | 2020-09-24T03:57:10 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,138,500 | 15,210 | filimonov | With docker (or with Kubernetes) you can run as many instances on a single node as you want, and later move them to a dedicated nodes. But as @den-crane - having 100 shards on a single node sounds like a very bad idea (forget about performance), but i know people doing smth similar with 3 or 6 shards per node (on relat... | 2020-09-24T06:18:58 | 2020-09-24T06:20:27 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,301,416 | 15,228 | UnamedRus | > Merge pattern '^table' should combine both table and table_2
It's probably bad naming in my example.
Doesn't work either.
```
CREATE TABLE other_table(key UInt32, ID UInt32) ENGINE = MergeTree PARTITION BY tuple() ORDER BY key;
CREATE TABLE merge_table (key UInt32) ENGINE=Merge(default,'^table');
CREATE TAB... | 2020-09-24T12:06:08 | 2020-09-24T12:09:19 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,088,224 | 15,210 | den-crane | I strongly advice not to do it. 100 tables will kill your single node.
Much wiser is to use `clickhouse-copier` and re-shard your table when you get 100 nodes. | 2020-09-24T03:24:02 | 2020-09-24T03:24:02 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,161,656 | 15,157 | javisantana | Makes sense, thanks for the info, I didn't realize about dictGet and StorageSet. In any case, I think join tables and dictionaries would solve most of the issues (not at all them but looking at the bug reports seems like most of them come from these kind of tables) | 2020-09-24T07:13:17 | 2020-09-24T07:13:17 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,214,882 | 15,181 | simPod | I disabled the sync script and the queue is not processing anymore. It was probably somehow being overridden by trying to add new items to it. | 2020-09-24T09:00:06 | 2020-09-24T09:00:06 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,270,311 | 14,962 | akuzm | Yandex check fixed, and other checks passed before, can merge now. | 2020-09-24T10:54:06 | 2020-09-24T10:54:06 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,305,808 | 13,997 | 4ertus2 | ```
remove unneeded columns from right table subquery.
remove unneeded columns from subqueries.
```
These cases should be done at next layer. It's not rewriter case. If it does not work it's an issue itself. Nothing for rewriter here.
```
remove unnecessary repetitions of WHERE clause (probably will work only f... | 2020-09-24T12:15:24 | 2020-09-24T12:15:24 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,341,928 | 14,492 | vitlibar | The failed tests [test_rename_column/test.py::test_rename_distributed_parallel_insert_and_select](https://clickhouse-test-reports.s3.yandex.net/14492/a28867486792552775855d16404b1596c99e9d19/integration_tests_(asan).html#fail1) and [test_s3_with_proxy/test.py::test_s3_with_proxy_list[s3]](https://clickhouse-test-report... | 2020-09-24T13:24:31 | 2020-09-24T13:25:10 | {} | MEMBER |
ClickHouse | ClickHouse | 698,620,247 | 15,240 | tavplubix | 01502_long_log_tinylog_deadlock_race - flacky
01249_bad_arguments_for_bloom_filter - wrong skip_list, already fixed
Stress test (memory) — Killed by signal - not related, reported to CI Fest
Yandex synchronization check - sync doesn't work in master | 2020-09-24T22:26:11 | 2020-09-24T22:26:11 | {} | MEMBER |
ClickHouse | ClickHouse | 698,160,172 | 15,181 | simPod | Basically all are based on sync script: `CREATE TABLE IF NOT EXISTS ...` while it always exists as I'm not adding new tables.
CPU load is 30 % and error log is empty. There seems to be no reason. | 2020-09-24T07:09:59 | 2020-09-24T07:09:59 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,199,216 | 14,476 | javisantana | Do you have any clue so it can help to fix the issue (or try to debug it further) | 2020-09-24T08:30:09 | 2020-09-24T08:30:09 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,204,511 | 15,174 | l1t1 | where could I find the benchmark of the AArch64 vs x86 system? | 2020-09-24T08:40:25 | 2020-09-24T08:40:25 | {} | NONE |
ClickHouse | ClickHouse | 698,299,755 | 14,847 | crakjie | Can this fix be ported to 20.3? | 2020-09-24T12:02:25 | 2020-09-24T12:02:25 | {} | NONE |
ClickHouse | ClickHouse | 698,327,562 | 15,227 | alexey-milovidov | Totals is not a part of the resultset, in contrast to ROLLUP/CUBE. | 2020-09-24T12:58:48 | 2020-09-24T12:58:48 | {} | MEMBER |
ClickHouse | ClickHouse | 698,638,245 | 14,414 | alexey-milovidov | @tavplubix https://github.com/ClickHouse/ClickHouse/pull/15258 | 2020-09-24T23:26:53 | 2020-09-24T23:26:53 | {} | MEMBER |
ClickHouse | ClickHouse | 698,638,537 | 15,240 | alexey-milovidov | MSan: https://github.com/ClickHouse/ClickHouse/pull/15258 | 2020-09-24T23:28:01 | 2020-09-24T23:28:01 | {} | MEMBER |
ClickHouse | ClickHouse | 698,297,367 | 15,228 | 4ertus2 | There's inconsistency with `WHERE` that works
```
SELECT ID, key FROM merge_table INNER JOIN table_2 USING(key) WHERE ID = 0;
```
Any case there's nothing ralated to JOIN. | 2020-09-24T11:56:59 | 2020-09-24T11:56:59 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,358,431 | 15,181 | simPod | Ok maybe I might have misinterpreted what I see in ZK. Select from `system.zookeeper` returns items in DDL queue but that should not be an issue. I checked nodes in ZK where there are `active` and `finished` nodes where `active` has 0 items in it and `finished` has `n of items = n of replicas`.
So maybe the queue is... | 2020-09-24T13:51:53 | 2020-09-24T13:51:53 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,375,019 | 15,233 | durchgedreht | FYI: I removed a bunch of fields in the ... section. But the most important stuff is there.
```
CREATE TABLE my_db.my_table (`delivery_date` DateTime, `event_records` Int64, `sa_type` Nullable(String), ...) ENGINE = MergeTree ORDER BY (delivery_date, account_id, channel_id, s3_hash) SETTINGS index_granularity = 819... | 2020-09-24T14:18:19 | 2020-09-24T14:18:19 | {} | NONE |
ClickHouse | ClickHouse | 698,383,229 | 15,144 | amosbird | Maybe we can also try binding the two instances to different CPU nodes. It might have unfair CPU frequency but less cache contentions. It might also be good to leave out CPU 0 due to system interruption handling | 2020-09-24T14:32:03 | 2020-09-24T14:32:46 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 698,396,065 | 15,228 | 4ertus2 | Reproduced in master
```
CREATE TABLE a (key UInt32) ENGINE = MergeTree ORDER BY key;
CREATE TABLE b (key UInt32, ID UInt32) ENGINE = MergeTree ORDER BY key;
CREATE TABLE m (key UInt32) ENGINE = Merge(default, 'a');
INSERT INTO a VALUES (0);
INSERT INTO b VALUES (0, 1);
SELECT * FROM m INNER JOIN b USING(key... | 2020-09-24T14:52:30 | 2020-09-24T15:04:12 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,455,866 | 15,149 | akuzm | Stress test is failing in 20.7 | 2020-09-24T16:36:12 | 2020-09-24T16:36:12 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,638,772 | 14,962 | alexey-milovidov | Too early...
https://clickhouse-test-reports.s3.yandex.net/15240/dce6a436f37713224c242092446e1a9fcf8b4a4a/functional_stateless_tests_(release)/test_run.txt.out.log | 2020-09-24T23:28:53 | 2020-09-24T23:28:53 | {} | MEMBER |
ClickHouse | ClickHouse | 697,590,486 | 14,711 | alesapin | Ok, docs check passed. | 2020-09-23T15:56:28 | 2020-09-23T15:56:28 | {} | MEMBER |
ClickHouse | ClickHouse | 697,734,730 | 15,065 | alexey-milovidov | @ucasFL Yes, I will reserve it for you. | 2020-09-23T17:21:08 | 2020-09-23T17:21:08 | {} | MEMBER |
ClickHouse | ClickHouse | 697,971,621 | 15,180 | SaltTan | #5491 | 2020-09-23T21:03:49 | 2020-09-23T21:03:49 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 697,339,432 | 15,191 | fastio | > Seems like it will not work for replicated tables because we commit info in zookeeper separately?
@alesapin Thanks. I have tested it in my env. It works for ReplicatedMergeTree engine.
| 2020-09-23T12:43:46 | 2020-09-24T00:45:35 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 698,120,638 | 14,414 | azat | >docker: invalid reference format: repository name must be lowercase
Does not report as tests failure - https://clickhouse-test-reports.s3.yandex.net/15007/1bb547422a46ee0d5d211666e99e22cb079da573/integration_tests_(asan).html | 2020-09-24T05:27:33 | 2020-09-24T05:27:33 | {} | MEMBER |
ClickHouse | ClickHouse | 698,168,263 | 15,089 | alesapin | AST fuzzer repoted https://github.com/ClickHouse/ClickHouse/issues/15214
Integration test -- known flaky test
Performance broken. | 2020-09-24T07:27:47 | 2020-09-24T07:27:47 | {} | MEMBER |
ClickHouse | ClickHouse | 698,273,878 | 15,003 | tavplubix | Stress test (address) — Killed by signal - race in `Exception::extendedMessage(...)`
01320_create_sync_race_condition_zookeeper - fixed | 2020-09-24T11:02:31 | 2020-09-24T11:02:31 | {} | MEMBER |
ClickHouse | ClickHouse | 698,353,874 | 15,233 | durchgedreht | No, doesn't work. Can't get any data out of it, even Select ```count(*) from my_table``` leads to:
```
SQL Error [171]: ClickHouse exception, code: 171, host: localhost, port: 8003; Code: 171, e.displayText() = DB::Exception: Block structure mismatch in QueryPipeline stream: different columns:
sa_type Nullable(Not... | 2020-09-24T13:45:23 | 2020-09-24T13:45:23 | {} | NONE |
ClickHouse | ClickHouse | 698,409,127 | 15,215 | alesapin | Fuzzer OK, stress failed, but for these changes it's ok. | 2020-09-24T15:13:37 | 2020-09-24T15:48:11 | {} | MEMBER |
ClickHouse | ClickHouse | 698,634,098 | 15,239 | alexey-milovidov | It does not work but the error looks different.
Maybe we need to disable ThinLTO for tools like `llvm-tblgen`? | 2020-09-24T23:12:10 | 2020-09-24T23:12:10 | {} | MEMBER |
ClickHouse | ClickHouse | 698,144,359 | 15,211 | zhang2014 | Yes, it's very useful.
There are two ways implement this:
- [ ] Support MaterializeMySQL table engine(It will be implemented in #14541. I understand that this approach is not suitable for synchronizing MySQL ddl, because this may cause the behavior to be inconsistent with other table engines.)
- [ ] Support filter t... | 2020-09-24T06:33:37 | 2020-09-24T06:33:37 | {
"+1": 7
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 692,680,243 | 13,617 | alexey-milovidov | ERROR: type should be string, got "https://clickhouse-test-reports.s3.yandex.net/13617/3b7beada2232a67fafa2f7f8a0ecea44e9ccdfac/performance_comparison/report.html#fail1\r\n\r\n`Total: +1.211x`\r\nIt is 21% slower that is unbelievably slow.\r\n\r\nI hope there are a few simple reasons why it's slower, like https://github.com/ClickHouse/ClickHouse/pull/12278/files#diff-59b750fae16a0cf9bd548fa8b0b277e4R207" | 2020-09-15T12:24:53 | 2020-09-15T12:53:01 | {} | MEMBER |
ClickHouse | ClickHouse | 692,270,498 | 14,747 | den-crane | What is the table DDL ? Show create table ...
Do you have some cron-jobs at these time? | 2020-09-14T19:38:55 | 2020-09-14T19:38:55 | {} | CONTRIBUTOR |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.