Datasets:
open-index
/
open-github-issues

Dataset card Data Studio Files
xet
 Community
1
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
823,780,610
10,935
filimonov
Similar https://github.com/ClickHouse/ClickHouse/issues/23059#
2021-04-21T05:10:14
2021-04-21T05:10:35
{}
CONTRIBUTOR
ClickHouse
ClickHouse
823,958,563
23,037
mjh568412187
> > https://packagecloud.io/Altinity/clickhouse > > packagecloud/Altinity is unofficial ClickHouse project, it may lag behind on the latest patches and stable releases. > I recommend to install as on the website https://clickhouse.tech/#quick-start (CentOS or RedHat tab) I need your help https://github.com/Cli...
2021-04-21T10:32:27
2021-04-21T10:32:27
{}
NONE
ClickHouse
ClickHouse
823,965,317
23,425
tavplubix
Which file it cannot unlink? Is there stacktrace for the exception in server log?
2021-04-21T10:45:32
2021-04-21T10:45:32
{}
MEMBER
ClickHouse
ClickHouse
824,002,558
21,555
tavplubix
Duplicate of #21554
2021-04-21T11:56:08
2021-04-21T11:56:08
{}
MEMBER
ClickHouse
ClickHouse
823,423,145
22,165
alexey-milovidov
Testflows failed, no progress for the past two weeks.
2021-04-20T16:33:09
2021-04-20T16:33:09
{}
MEMBER
ClickHouse
ClickHouse
823,474,992
23,356
inakivb
related_number SimpleAggregateFunction(anyLast,Float32) should work if the numbers are ascending. This will keep the last record sent
2021-04-20T17:42:37
2021-04-20T17:43:45
{}
NONE
ClickHouse
ClickHouse
823,520,498
23,342
alexey-milovidov
Ok but see also #23275
2021-04-20T18:51:03
2021-04-20T18:51:03
{}
MEMBER
ClickHouse
ClickHouse
823,529,595
23,353
alexey-milovidov
Ok. I've uploaded the `hits_100m_obfuscated` table to new Playground: https://gh-api.clickhouse.tech/play?user=play#c2VsZWN0ICJ0MSIuIl9fZmNvbF85IiAiX19mY29sXzkiCmZyb20gKAooCnNlbGVjdCBudWxsICJfX2Zjb2xfOSIKZnJvbSBoaXRzXzEwMG1fb2JmdXNjYXRlZCAiVF8xQkM2QTkiCikKdW5pb24gYWxsICgKc2VsZWN0ICJ0MCIuIl9fZmNvbF85IiAiX19mY29sXzki...
2021-04-20T19:05:44
2021-04-20T19:05:44
{}
MEMBER
ClickHouse
ClickHouse
823,538,959
23,377
tavplubix
But `data/` and `metadata/` directories are still there: https://github.com/ClickHouse/ClickHouse/tree/508bfd0b3fb95b1314ea5f9d12c9d70290e77500/programs/server/metadata
2021-04-20T19:20:55
2021-04-20T19:20:55
{}
MEMBER
ClickHouse
ClickHouse
823,543,513
1,647
alexey-milovidov
Fixed.
2021-04-20T19:28:27
2021-04-20T19:28:27
{}
MEMBER
ClickHouse
ClickHouse
823,563,588
17,317
alexey-milovidov
> Maybe also related to greedy execution of alias columns. #2582 No, it's not the reason.
2021-04-20T20:02:02
2021-04-20T20:02:02
{}
MEMBER
ClickHouse
ClickHouse
823,575,803
23,298
alexey-milovidov
You can check different linkers: `lld`, `gold`. If needed, use the latest `lld` version. Currently we are using lld-11. The size of the `libdbms.a` on my machine is 2244M and it works with lld-11. If nothing helps, build without debug info (Release instead of RelWithDebInfo).
2021-04-20T20:22:51
2021-04-20T20:22:51
{}
MEMBER
ClickHouse
ClickHouse
823,816,029
23,356
esbenkolsbjerg
Sadly both `number` and `related_number` are not ascending or (increasing for that matter). The idea is that `related_number` fits together with `number`, hence we want them as a pair but we only wanna keep the record of the highest `number` as it is the important one. I know we can do this if we split the logic into t...
2021-04-21T06:37:20
2021-04-21T06:38:03
{}
NONE
ClickHouse
ClickHouse
823,330,451
21,309
nikitamikhaylov
@amosbird Why not to simply add expression from FILTER BY (or INDEX HINT) to WHERE clause? It will trigger KeyCondition in the same way.
2021-04-20T14:41:06
2021-04-20T14:41:06
{}
MEMBER
ClickHouse
ClickHouse
823,429,364
22,729
alexey-milovidov
@zvonand The tests take 848 seconds of time. They will run for every commit and every PR. It's too much for a single feature. Could you make these tests faster? I'm sure that most of the time these tests do nothing.
2021-04-20T16:42:32
2021-04-20T16:43:09
{}
MEMBER
ClickHouse
ClickHouse
824,001,771
21,974
tavplubix
Merges are executed regularly even when there are no writing queries (such as INSERT, ALTER or OPTIMIZE). There are several kinds of merges (ordinary merges, TTL merges, recompression merges) and a lot of heuristics are used to determine if some parts should be merged or not, so it's hard to tell how often background m...
2021-04-21T11:54:35
2021-04-21T11:54:35
{}
MEMBER
ClickHouse
ClickHouse
824,002,876
21,625
tavplubix
Duplicate of #21554 and #21555
2021-04-21T11:56:42
2021-04-21T11:56:42
{}
MEMBER
ClickHouse
ClickHouse
824,040,060
22,045
nikitamikhaylov
@OmarBazaraa See the `test_storage_mongodb` inside /tests/integration folder. MongoDB is created inside a container of ClickHouse node. Check the line `cluster.add_instance('node', with_mongo=True)`.
2021-04-21T12:59:45
2021-04-21T12:59:45
{}
MEMBER
ClickHouse
ClickHouse
824,044,701
23,294
tavplubix
ecc6256: Functional stateless tests (release, wide parts enabled) - 01602_max_distributed_connections - flaky Integration tests - test_hedged_requests* - was broken in master Performance - unrelated SQLancer test - `ClickHouse exception, code: 159, host: localhost, port: 8123; Read timed out`, but it's hard to unde...
2021-04-21T13:05:39
2021-04-21T13:05:39
{}
MEMBER
ClickHouse
ClickHouse
823,391,618
18,250
Enmk
## Minimal viable example On MySql side ``` $ docker run -d -e MYSQL_ROOT_PASSWORD=pass --network=host mysql:5.6 $ docker exec -it <<CONTAINER>> bash # in docker: $ MYSQL_PWD=pass mysql <<EOF CREATE DATABASE db; USE db; CREATE TABLE user_data( id INT NOT NULL AUTO_INCREMENT, date DATE, name VARC...
2021-04-20T15:51:23
2021-04-20T15:51:23
{}
CONTRIBUTOR
ClickHouse
ClickHouse
823,423,772
12,455
azat
>after creating a MergeTree table with nullable primary keys, ClickHouse won't be able to rollabck to previous versions. (potential backward incompatibility) @amosbird BTW right now if you will run newer clickhouse and then rollback to the version w/o this patch you will not get any errors, but simply index analysis...
2021-04-20T16:34:10
2021-04-20T16:34:10
{}
MEMBER
ClickHouse
ClickHouse
823,483,860
23,353
tavplubix
Fixed in #22063, let's backport it
2021-04-20T17:56:28
2021-04-20T17:56:28
{}
MEMBER
ClickHouse
ClickHouse
823,535,738
23,274
sevirov
@Avogar Please, review this PR.
2021-04-20T19:15:40
2021-04-20T19:15:40
{}
CONTRIBUTOR
ClickHouse
ClickHouse
823,757,320
23,037
mjh568412187
> > https://packagecloud.io/Altinity/clickhouse > > packagecloud/Altinity is unofficial ClickHouse project, it may lag behind on the latest patches and stable releases. > I recommend to install as on the website https://clickhouse.tech/#quick-start (CentOS or RedHat tab) I installed your suggestion and installed...
2021-04-21T03:52:57
2021-04-21T03:52:57
{}
NONE
ClickHouse
ClickHouse
823,966,037
23,423
tavplubix
It should work, recursive materialized views are supported
2021-04-21T10:46:54
2021-04-21T10:46:54
{}
MEMBER
ClickHouse
ClickHouse
823,970,134
23,312
vdimir
We got error because with join_algorithm = auto we save key column into sample_block https://github.com/ClickHouse/ClickHouse/blob/71c27a4dc03c53d10c63b6c7c38b908954faf014/src/Interpreters/HashJoin.cpp#L573-L575 But dictionary_reader always uses HashJoin and return block without it. So, it leads to wrong indices...
2021-04-21T10:54:59
2021-04-21T10:54:59
{}
MEMBER
ClickHouse
ClickHouse
824,005,671
21,541
tavplubix
Duplicate of #21554
2021-04-21T12:01:32
2021-04-21T12:01:32
{}
MEMBER
ClickHouse
ClickHouse
823,554,036
21,556
alexey-milovidov
@OmarBazaraa Did you install new version over existing data or uploaded data from scratch? In the latter case it's possible that parts distribution and column statistics are completely different. It can be also affected by existence of compact data parts. But if you just install a new version over existing data, ...
2021-04-20T19:45:57
2021-04-20T19:45:57
{}
MEMBER
ClickHouse
ClickHouse
823,571,180
22,080
alexey-milovidov
it's not fixed.
2021-04-20T20:15:09
2021-04-20T20:15:09
{}
MEMBER
ClickHouse
ClickHouse
823,576,672
23,298
alexey-milovidov
We never faced this issue by ourself, but I heard about a similar issue from my collegues: they statically link ClickHouse (a huge chunk of code) with another very large system, and the total size becomes so large, that the linker cannot do its job.
2021-04-20T20:24:31
2021-04-20T20:24:45
{}
MEMBER
ClickHouse
ClickHouse
823,611,818
23,362
alexey-milovidov
Duplicate #23360.
2021-04-20T21:29:15
2021-04-20T21:29:15
{}
MEMBER
ClickHouse
ClickHouse
823,861,296
23,283
bun4uk
I have a similar issue with `+` symbol. My object name is `s_user_name+0+0001188763.json.gz` In the exception, I see mention of `s_user_name%200%200001188763.json.gz` which decodes as `s_user_name 0 0001188763.json.gz`. So probably `+` somehow decodes as `space`
2021-04-21T07:58:16
2021-04-21T07:58:16
{}
CONTRIBUTOR
ClickHouse
ClickHouse
823,315,237
21,831
jorisgio
Hi, so basically it didn't reproduce anymore. It did happen when the servers were under very heavy load, and since the load reduced it didn't trigger. > In both cases, it looks like clickhouse just waits for some background operation or query. The thread was gone, even though the condition was not triggered to wak...
2021-04-20T14:21:49
2021-04-20T14:22:00
{}
CONTRIBUTOR
ClickHouse
ClickHouse
823,559,209
23,380
den-crane
it's a mis-usage and the minor issue. should be Tuple ` SimpleAggregateFunction(argMax, Tuple(String, DateTime)`
2021-04-20T19:54:37
2021-04-20T20:03:03
{}
CONTRIBUTOR
ClickHouse
ClickHouse
824,042,160
23,294
alexey-milovidov
@Mergifyio update
2021-04-21T13:02:55
2021-04-21T13:02:55
{ "+1": 1 }
MEMBER
ClickHouse
ClickHouse
823,421,335
23,319
alexey-milovidov
https://github.com/ClickHouse/ClickHouse/pull/23368
2021-04-20T16:30:24
2021-04-20T16:30:24
{}
MEMBER
ClickHouse
ClickHouse
823,431,177
23,348
alexey-milovidov
I will fix it.
2021-04-20T16:45:14
2021-04-20T16:45:14
{ "+1": 1 }
MEMBER
ClickHouse
ClickHouse
823,440,878
18,250
alexey-milovidov
@vzakaznikov Your sequence of commands contains the same mistake: ``` secret TEXT, ``` I will close this issue as incorrect.
2021-04-20T16:50:38
2021-04-20T16:53:04
{}
MEMBER
ClickHouse
ClickHouse
823,500,868
23,350
tavplubix
Looks like expected behavior. Replica does not become lost if it just turned of, so it's `/is_lost` should not be changed. It's changed only when other replicas consider inactive replica as staled according to its log pointer. As for `/is_active` node, it's ephemeral and it disappears when replica is turned off, but it...
2021-04-20T18:19:47
2021-04-20T18:19:47
{ "+1": 1 }
MEMBER
ClickHouse
ClickHouse
823,561,163
21,556
alexey-milovidov
My collegue @CurtizJ has found the source of this change.
2021-04-20T19:57:49
2021-04-20T19:57:49
{}
MEMBER
ClickHouse
ClickHouse
823,577,632
22,988
alexey-milovidov
We decided that exponential backoff is needed.
2021-04-20T20:26:14
2021-04-20T20:26:14
{}
MEMBER
ClickHouse
ClickHouse
823,723,315
23,180
hexiaoting
If you run updates in sequence in the same connection, then B is the final result. If you run `set mutations_sync = 1;`, then this will take effect on this connection. If you run `alter .... update .... settings mutations_sync=1;` this will take effect only in this SQL.
2021-04-21T02:15:42
2021-04-21T02:15:42
{}
CONTRIBUTOR
ClickHouse
ClickHouse
823,876,116
22,970
yx91490
How about check unique replica path when insert recoud to system.tables?
2021-04-21T08:20:10
2021-04-21T08:20:10
{}
NONE
ClickHouse
ClickHouse
823,969,782
23,411
tavplubix
Looks like duplicate of #23296
2021-04-21T10:54:21
2021-04-21T10:54:21
{}
MEMBER
ClickHouse
ClickHouse
824,013,894
23,432
tavplubix
Duplicate of #23353, fixed in #22063, the fix is backported to 21.3.8 (not released yet)
2021-04-21T12:15:45
2021-04-21T12:15:45
{}
MEMBER
ClickHouse
ClickHouse
823,305,905
23,140
amosbird
> > Fix multiple block insertion into distributed table with random_shard_insert = 1. This is a marginal feature. Mark as improvement. > > For changelog it is better to use name of the setting `insert_distributed_one_random_shard` Sure
2021-04-20T14:10:10
2021-04-20T14:10:10
{}
COLLABORATOR
ClickHouse
ClickHouse
823,409,390
18,250
alexey-milovidov
> secret TEXT, This is the mistake. BLOB should be used instead.
2021-04-20T16:12:34
2021-04-20T16:12:34
{}
MEMBER
ClickHouse
ClickHouse
823,422,622
18,250
vzakaznikov
The original test procedure can be easily reduced to sequence of commands. ``` user@user-host:~$ cat test.log | grep -e \\[clickhouse1\\] -e \\[mysql1\\] 1s 71ms [mysql1] bash# MYSQL_PWD=password mysql -D db -u user -e "DROP TABLE IF EXISTS user_data;" 1s 94ms ...
2021-04-20T16:32:14
2021-04-20T16:32:14
{}
CONTRIBUTOR
ClickHouse
ClickHouse
823,567,214
23,245
alexey-milovidov
One column is not enough. It is `hits_v1` from here: https://clickhouse.tech/docs/en/getting-started/example-datasets/metrica/ The reason for this behaviour is creation of ReadBuffers for every column (that is alright) that are not destroyed early. At the end of the query we have 100 * number of columns buffers, e...
2021-04-20T20:08:05
2021-04-20T20:08:05
{}
MEMBER
ClickHouse
ClickHouse
823,793,141
18,276
fuwhu
@CurtizJ I just fixed the conflict. Could you please help review this PR?
2021-04-21T05:44:35
2021-04-21T05:44:35
{}
CONTRIBUTOR
ClickHouse
ClickHouse
823,294,675
23,361
filimonov
Some details in telegram.
2021-04-20T13:55:48
2021-04-20T13:55:48
{}
CONTRIBUTOR
ClickHouse
ClickHouse
823,322,727
23,364
akuzm
Already fixed here https://github.com/ClickHouse/ClickHouse/pull/23363
2021-04-20T14:31:30
2021-04-20T14:31:30
{}
CONTRIBUTOR
ClickHouse
ClickHouse
823,442,051
12,455
amosbird
> > after creating a MergeTree table with nullable primary keys, ClickHouse won't be able to rollabck to previous versions. (potential backward incompatibility) > > @amosbird BTW right now if you will run newer clickhouse and then rollback to the version w/o this patch you will not get any errors, but simply index a...
2021-04-20T16:52:31
2021-04-20T16:52:31
{}
COLLABORATOR
ClickHouse
ClickHouse
823,781,000
23,059
filimonov
Then log spamming is the issue https://github.com/ClickHouse/ClickHouse/issues/10935#issuecomment-797360292
2021-04-21T05:11:26
2021-04-21T05:11:26
{}
CONTRIBUTOR
ClickHouse
ClickHouse
823,801,998
23,300
Sallery-X
> Do you mean that you want multiple replicas to share the same data in NFS? There is no way it can work, `execute_merges_on_single_replica_time_threshold` and other settings will not help. I try to modify some code to make it work
2021-04-21T06:05:55
2021-04-21T06:05:55
{}
CONTRIBUTOR
ClickHouse
ClickHouse
823,963,663
23,414
traceon
The error that is printed at `clickhouse-client` side is a generic one. The `clickhouse-server` should have more informative log (right before a similar `Authentication failed: ...` message.) Please, post that log message here. Also, try binding to that LDAP server from the same terminal session, using OpenLDAP's ...
2021-04-21T10:42:10
2021-04-21T10:42:10
{}
CONTRIBUTOR
ClickHouse
ClickHouse
824,005,265
23,392
alexey-milovidov
Test failures reviewed.
2021-04-21T12:00:50
2021-04-21T12:00:50
{}
MEMBER
ClickHouse
ClickHouse
823,424,128
21,087
alexey-milovidov
Testflows failed. No progress since Mar 7. @vzakaznikov Maybe it's better to implement it with functional tests, so it will use reliable testing framework?
2021-04-20T16:34:44
2021-04-20T16:34:44
{}
MEMBER
ClickHouse
ClickHouse
823,431,074
23,348
alexey-milovidov
> Fix LOGICAL_ERROR for max_distributed_connections=0 This is not a changelog entry for a user.
2021-04-20T16:45:05
2021-04-20T16:45:05
{}
MEMBER
ClickHouse
ClickHouse
823,446,492
18,250
alexey-milovidov
@vzakaznikov Everything works astonishingly good if we do it right: ``` milovidov@milovidov-desktop:~/work/ClickHouse$ docker run -d -e MYSQL_ROOT_PASSWORD=pass --network=host mysql:5.6 Unable to find image 'mysql:5.6' locally 5.6: Pulling from library/mysql 62deabe7a6db: Pull complete 81ba35f9cdf1: Pull c...
2021-04-20T16:59:32
2021-04-20T17:25:10
{}
MEMBER
ClickHouse
ClickHouse
823,508,867
2,582
alexey-milovidov
Fixed here: #16995
2021-04-20T18:32:37
2021-04-20T18:32:37
{}
MEMBER
ClickHouse
ClickHouse
823,619,235
21,417
alexey-milovidov
Это ошибка самого PostgreSQL (несовместимость версий). Вот здесь можно посмотреть подробнее: https://www.google.ru/search?q=relhasoids+does+not+exist
2021-04-20T21:42:50
2021-04-20T21:42:50
{}
MEMBER
ClickHouse
ClickHouse
823,778,648
2,582
filimonov
I think that exact test was fixed even earlier...
2021-04-21T05:05:00
2021-04-21T05:05:00
{}
CONTRIBUTOR
ClickHouse
ClickHouse
823,437,868
23,349
alexey-milovidov
> Respect insert_allow_materialized_columns for INSERT into Distributed() Also not a good enough changelog entry. I will fix it.
2021-04-20T16:49:01
2021-04-20T16:49:01
{}
MEMBER
ClickHouse
ClickHouse
823,449,947
23,200
tavplubix
ClickHouse build check - `Build failed without build log. It's inner CI problem.`
2021-04-20T17:05:01
2021-04-20T17:05:01
{}
MEMBER
ClickHouse
ClickHouse
823,616,088
22,304
alexey-milovidov
The logic is correct. Single ZooKeeper session must live for whole duration of INSERT query. INSERT query cannot be done if ZooKeeper session is expired. An expiration of ZooKeeper session (e.g. connection loss between ClickHouse and ZooKeeper) is normal if happens from time to time and it is inevitable. But if ...
2021-04-20T21:37:20
2021-04-20T21:37:20
{}
MEMBER
ClickHouse
ClickHouse
823,632,866
17,304
alexey-milovidov
@CurtizJ Shouldn't be possible to disambiguate them with parens?
2021-04-20T22:11:59
2021-04-20T22:12:09
{}
MEMBER
ClickHouse
ClickHouse
823,525,254
23,375
CurtizJ
Fast test: `executeQuery: Code: 344, e.displayText() = DB::Exception: Collations support is disabled, because ClickHouse was built without ICU library`
2021-04-20T18:58:56
2021-04-20T18:58:56
{}
MEMBER
ClickHouse
ClickHouse
823,557,295
21,556
alexey-milovidov
`MergeTreeWhereOptimizer: condition "(Playback = 1) AND (SessionStart >= '2021/03/03 00:00:00') AND (SessionStart <= '2021/03/08 00:00:00')" moved to PREWHERE` vs. `MergeTreeWhereOptimizer: condition "(Playback = 1)" moved to PREWHERE`
2021-04-20T19:51:16
2021-04-20T19:51:16
{}
MEMBER
ClickHouse
ClickHouse
823,563,464
17,317
alexey-milovidov
Still relevant in version 21.6 testing.
2021-04-20T20:01:49
2021-04-20T20:01:49
{}
MEMBER
ClickHouse
ClickHouse
823,596,886
21,188
alexey-milovidov
Docker is quite old: > Built: Tue Feb 27 22:17:54 2018 Update to the latest. For example, I have > Fri Dec 4 23:02:49 2020 on my machine and I'm able to run the tests. Most likely it will work on Ubuntu 20.04 or newer.
2021-04-20T20:59:25
2021-04-20T20:59:25
{}
MEMBER
ClickHouse
ClickHouse
824,004,375
21,974
tavplubix
Duplicate of #21554 and #21555 and #21625 Please, do not create duplicate issues.
2021-04-21T11:59:15
2021-04-21T11:59:15
{}
MEMBER
ClickHouse
ClickHouse
824,030,832
23,425
tavplubix
Hm, there is `dropIfEmpty()` in the stacktrace. It means that some error (probably related to ZooKeeper) occurred on table creation, than ClickHouse tried to cancel table creation by removing some files from disk, but another error (`Cannot unlink file`) happened and information about the first error was lost. Seem...
2021-04-21T12:44:51
2021-04-21T12:44:51
{}
MEMBER
ClickHouse
ClickHouse
823,422,434
21,659
alexey-milovidov
@zvonand These tests take 3210 seconds of time to run. Why it's needed?
2021-04-20T16:31:56
2021-04-20T16:31:56
{}
MEMBER
ClickHouse
ClickHouse
823,445,158
18,250
alexey-milovidov
@Enmk > SELECT name, hex(secret) FROM user_data'; Here is a mistake - extraneous quote character.
2021-04-20T16:57:19
2021-04-20T16:57:19
{}
MEMBER
ClickHouse
ClickHouse
823,447,288
23,202
nikitamikhaylov
I think everything should be Ok.
2021-04-20T17:00:51
2021-04-20T17:00:51
{}
MEMBER
ClickHouse
ClickHouse
823,452,707
21,659
zvonand
> @zvonand These tests take 3210 seconds of time to run. Why it's needed? This is completely CICD's fault They take about 30 seconds on all other runners, and one hour in yandex checks Trying to figure out why, but still no luck
2021-04-20T17:08:44
2021-04-20T17:08:44
{}
CONTRIBUTOR
ClickHouse
ClickHouse
823,459,413
23,353
tavplubix
Minimized query: ``` :) select null as c from system.one where c in ('aaa') Received exception from server (version 21.3.1): Code: 48. DB::Exception: Received from localhost:9000. DB::Exception: Serialization is not implemented for data type Nothing: while converting 'aaa' to Nothing: While processing (NULL AS ...
2021-04-20T17:16:51
2021-04-20T17:16:51
{}
MEMBER
ClickHouse
ClickHouse
823,809,632
23,180
SE2AI
@hexiaoting many thanks wenting, according to your answer, it seems A will never be the final result. so considering sequence , what is the side effects of `mutation_sync=0` for one connection (and the replicated table)? In our scenario, we have to keep execution sequences and we use distributed table with one r...
2021-04-21T06:23:08
2021-04-21T07:26:34
{}
NONE
ClickHouse
ClickHouse
824,045,138
23,294
alexey-milovidov
Then it's ready.
2021-04-21T13:06:11
2021-04-21T13:06:11
{}
MEMBER
ClickHouse
ClickHouse
823,343,268
21,309
amosbird
> @amosbird Why not to simply add expression from FILTER BY (or INDEX HINT) to WHERE clause? It will trigger KeyCondition in the same way. It's to bind this condition to a given table without intruding normal users. The condition might refer to `dictGet` functions so that we can toggle visibility in other ways.
2021-04-20T14:57:15
2021-04-20T14:57:15
{}
COLLABORATOR
ClickHouse
ClickHouse
823,523,005
23,059
alexey-milovidov
@sinmaker No, you should not be worried about this exception message. It happens when a client connects to ClickHouse and closes connection immediately. It may possibly indicate some minor issues with your infrastructure, but it is not an issue of ClickHouse itself.
2021-04-20T18:55:17
2021-04-20T18:55:17
{ "+1": 1 }
MEMBER
ClickHouse
ClickHouse
823,816,256
23,416
maver1ck
I think this could be duplicate of https://github.com/ClickHouse/ClickHouse/issues/17956
2021-04-21T06:37:51
2021-04-21T06:37:51
{}
NONE
ClickHouse
ClickHouse
823,874,700
22,970
yx91490
> It's not allowed to create different tables with the same zookeeper path and replica name, such CREATE query should fail with `Replica already exists` error. Could you please provide more information about how did you create the second table? Do you use auxiliary ZooKeepers configs? I remembered that I remove the ...
2021-04-21T08:18:02
2021-04-21T08:18:02
{}
NONE
ClickHouse
ClickHouse
823,996,503
23,425
GeraltXLi
> Which file it cannot unlink? Is there stacktrace for the exception in server log? Just like this. ![image](https://user-images.githubusercontent.com/13702295/115548278-e21bae80-a2d9-11eb-8079-cee8c617dad6.png)
2021-04-21T11:44:38
2021-04-21T11:44:38
{}
NONE
ClickHouse
ClickHouse
824,007,826
23,423
GeraltXLi
Thank you very much for your reply. But the Materialized view, which select from another Materialized view, does not have data. Did I write something wrong? ```sql /*source table*/ CREATE TABLE test_scene_info_local ( collect_time DateTime64(3), scene String, scene_start_time DateTime64(3), pss UInt64 )ENGIN...
2021-04-21T12:05:21
2021-04-21T12:08:08
{}
NONE
ClickHouse
ClickHouse
824,018,730
23,365
jorisgio
Thanks. Happy to share more information if you need, it is currently easy to reproduce
2021-04-21T12:24:31
2021-04-21T12:24:51
{}
CONTRIBUTOR
ClickHouse
ClickHouse
823,510,313
22,711
nikitamikhaylov
@stephane-moreau Please show the log of the copier.
2021-04-20T18:34:57
2021-04-20T18:34:57
{}
MEMBER
ClickHouse
ClickHouse
823,554,829
23,380
den-crane
The same ``` insert into A select arrayJoin([('a', '2020-01-05 00:00:00'), ('v','2020-01-05 00:00:00'), ('zx','2020-01-05 00:00:00')]); ```
2021-04-20T19:47:08
2021-04-20T19:47:08
{}
CONTRIBUTOR
ClickHouse
ClickHouse
823,585,045
23,377
alexey-milovidov
Ok, I will remove them also :)
2021-04-20T20:38:34
2021-04-20T20:38:34
{}
MEMBER
ClickHouse
ClickHouse
824,042,537
23,294
mergify[bot]
**Command `update`: success** > **Branch has been successfully updated** >
2021-04-21T13:03:29
2021-04-21T13:03:29
{}
CONTRIBUTOR
ClickHouse
ClickHouse
823,440,216
18,250
alexey-milovidov
@vzakaznikov It's better to post a sequence of commands so I can easily copy-paste it.
2021-04-20T16:50:07
2021-04-20T16:50:07
{}
MEMBER
ClickHouse
ClickHouse
823,539,755
23,316
sevirov
@tavplubix Please, review this PR.
2021-04-20T19:22:15
2021-04-20T19:22:15
{}
CONTRIBUTOR
ClickHouse
ClickHouse
823,634,193
17,304
CurtizJ
It works now. The issue was, that initially multiple TTLs with `GROUP BY` were not supported.
2021-04-20T22:15:07
2021-04-20T22:15:07
{}
MEMBER
ClickHouse
ClickHouse
823,785,529
19,854
fuwhu
@Avogar Sorry for late update. The test were added, please help review again. thanks a lot.
2021-04-21T05:24:34
2021-04-21T05:24:34
{}
CONTRIBUTOR
ClickHouse
ClickHouse
823,998,863
23,365
CurtizJ
Until it will be fixed, as workaround you can disable Vertical merges by setting of MergeTree `enable_vertical_merge_algorithm`.
2021-04-21T11:49:05
2021-04-21T11:49:05
{ "+1": 1 }
MEMBER
ClickHouse
ClickHouse
809,619,217
21,318
azat
>@azat Is it relevant? Yes, this is minor issue, but once it was fixed another one pops up, will get back to this later.
2021-03-29T18:40:49
2021-03-29T18:40:49
{}
MEMBER
ClickHouse
ClickHouse
809,905,060
22,260
filimonov
That can be also important for odbc-bridge
2021-03-30T04:49:06
2021-03-30T04:49:06
{}
CONTRIBUTOR
ClickHouse
ClickHouse
810,005,836
22,306
alesapin
Not important
2021-03-30T08:00:33
2021-03-30T08:00:33
{}
MEMBER
ClickHouse
ClickHouse
810,094,105
22,338
alexey-milovidov
I did not reproduce it - did not check that the fix has effect. Moreover, we disabled Integration tests with MSan in PRs, so we will have to merge blindly.
2021-03-30T10:07:01
2021-03-30T10:07:01
{}
MEMBER