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 | 689,353,540 | 14,547 | alesapin | tests started, so I'll merge. | 2020-09-09T07:05:25 | 2020-09-09T07:05:25 | {} | MEMBER |
ClickHouse | ClickHouse | 689,474,204 | 13,859 | ssotik | Thanks. Helped for me. | 2020-09-09T10:29:26 | 2020-09-09T10:29:26 | {} | NONE |
ClickHouse | ClickHouse | 689,562,852 | 4,959 | muligolan | I'm experiencing the same problem:
<Error> executeQuery: Code: 33, e.displayText() = DB::Exception: Cannot read all data. Bytes read: 93974. Bytes expected: 764646. (version 19.11.4.24) (from 10.10.200.208:29780) (in query: INSERT INTO {table_name}, Stack trace:
0. /usr/bin/clickhouse-server(StackTrace::StackTrace(... | 2020-09-09T13:27:25 | 2020-09-09T14:06:10 | {} | NONE |
ClickHouse | ClickHouse | 689,610,670 | 14,494 | alesapin | Ok tests passed and current implementation already useful. Will add priorities in the next PRs. | 2020-09-09T14:45:41 | 2020-09-09T14:45:41 | {} | MEMBER |
ClickHouse | ClickHouse | 689,698,894 | 14,645 | alexey-milovidov | It was merged 40 minues ago, I will check again after pull. | 2020-09-09T17:08:16 | 2020-09-09T17:08:16 | {} | MEMBER |
ClickHouse | ClickHouse | 689,238,250 | 13,875 | den-crane | > 4.7g
quite small
Do you have some graphs? To see how RES (4.7) changes. | 2020-09-09T01:15:49 | 2020-09-09T01:16:14 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,331,911 | 14,576 | prakharkohq | Thanks @filimonov and @den-crane I will look into this. https://github.com/ClickHouse/ClickHouse/blob/master/utils/convert-month-partitioned-parts/main.cpp | 2020-09-09T06:16:37 | 2020-09-09T06:16:37 | {} | NONE |
ClickHouse | ClickHouse | 689,447,378 | 5,782 | ogbofjnr | @alexey-milovidov , why there is no uniqExactIf in documentation? https://clickhouse.tech/docs/en/sql-reference/aggregate-functions/reference/ | 2020-09-09T09:33:24 | 2020-09-09T09:33:24 | {} | NONE |
ClickHouse | ClickHouse | 689,544,194 | 14,398 | hustnn | @alexey-milovidov I created a PR https://github.com/ClickHouse/ClickHouse/pull/14626 by following our discussion. | 2020-09-09T12:55:33 | 2020-09-09T12:55:33 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,222,639 | 14,574 | den-crane | You need to check ulimits for a process, not the current shell.
```
ps -ef|grep click
...
clickho+ 32747 1 1 Sep08 ? 00:01:25 clickhouse-server --daemon --pid-file=/var/run/clickhouse-server/clickhouse-server.pid --config-file=/etc/clickhouse-server/config.xml
more /proc/32747/limits
Limit ... | 2020-09-09T00:31:25 | 2020-09-09T00:39:17 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,229,689 | 13,875 | den-crane | Check how much memory CH is uses -- RES htop/top
and check `free -h` how much memory is free/cached | 2020-09-09T00:46:06 | 2020-09-09T00:46:40 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,280,220 | 14,606 | iCloudFly | 前面的结果说反了，重新发一次。
更新一个表的两个列（selectFlag是不允许为null，downloadFlag允许为null）。
结果不允许为null的列更新的值是正确的，
但是允许为null的列更新的值永远是93147008
不知道这是为什么？是bug吗？还是我的配置不对？
| 2020-09-09T03:38:57 | 2020-09-09T03:38:57 | {} | NONE |
ClickHouse | ClickHouse | 689,283,762 | 14,573 | chang0391 | yes, if this is not supported, It's better to only have this table affected
accessing of other tables should not be affected, but I can not access other tables.
or is there any way to skip gtid number or event? | 2020-09-09T03:50:24 | 2020-09-09T03:50:51 | {} | NONE |
ClickHouse | ClickHouse | 689,338,010 | 14,595 | alexey-milovidov | https://github.com/ClickHouse/ClickHouse/pull/14607 | 2020-09-09T06:32:52 | 2020-09-09T06:32:52 | {} | MEMBER |
ClickHouse | ClickHouse | 689,452,268 | 14,546 | simPod | Ok, I missed this sentence
> The layout in this case must be complex_key_hashed or complex_key_cache. | 2020-09-09T09:43:12 | 2020-09-09T09:43:12 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,669,550 | 14,573 | zhang2014 | When synchronization failure(we stop sync for data consistency), we always want the users to see the exception info as much as possible(I think this reasonable for experimental features.)
FYI:
https://github.com/ClickHouse/ClickHouse/pull/10851#issuecomment-669324802
https://github.com/ClickHouse/ClickHouse/pull/1... | 2020-09-09T16:19:30 | 2020-09-09T16:19:30 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,686,158 | 14,645 | alexey-milovidov | `SELECT CAST(toNullable('hello'), 'Enum8(\'Hello\' = 1)')` | 2020-09-09T16:49:01 | 2020-09-09T16:49:01 | {} | MEMBER |
ClickHouse | ClickHouse | 689,699,209 | 14,521 | myrrc | > @alexey-milovidov
> I have added ILIKE for "SHOW DATABASES/TABLES".
>
> But is there a bug with ilike ???
> It is not work with system.databases
>
> ```
> dell123 :) SELECT name FROM system.databases ... | 2020-09-09T17:08:54 | 2020-09-09T17:08:54 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,237,802 | 14,568 | yd1261137568 | > > system.processes The table shows a large number of drop partiton jams,They can't be killed.Still stuck after restart
>
> Maybe they are waiting for somethin (check system.mutations & system.merges).
>
> If you want force removeing of those - you need to stop the node, clean the /clickhouse/task_queue in zooke... | 2020-09-09T01:14:20 | 2020-09-09T01:14:20 | {} | NONE |
ClickHouse | ClickHouse | 689,335,662 | 14,595 | alexey-milovidov | > worth mark as backward incompatible?
Not needed. It can possibly lead to incompatibilities - but only with some software that depends on whitespaces in query formatting - rare case. | 2020-09-09T06:26:43 | 2020-09-09T06:26:43 | {} | MEMBER |
ClickHouse | ClickHouse | 689,353,151 | 14,581 | alesapin | Integration tests ok | 2020-09-09T07:04:34 | 2020-09-09T07:04:34 | {} | MEMBER |
ClickHouse | ClickHouse | 689,470,839 | 14,195 | akuzm | > Are end users expected to periodically read off tables, convert data into Jaeger format and ingest it?
Haven't thought this through yet.
There are basically two models, we can pull data from ClickHouse using some external app, or make ClickHouse itself push the data.
Pull is the simplest to implement in CH -... | 2020-09-09T10:21:50 | 2020-09-09T10:21:50 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,557,700 | 14,603 | abyss7 | Integration test error comes from the master branch. | 2020-09-09T13:18:36 | 2020-09-09T13:18:36 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,575,712 | 14,130 | den-crane | @fibersel
These folders are for async inserts. Distributed table stores .bin files with data which should be inserted.
shard1_replica1 -- means that data in this folder should be inserted into the shard1.
`shard1_replica1,shard1_replica2` -- I am not sure how this folder happen, and it does not matter. The error ... | 2020-09-09T13:49:13 | 2020-09-09T13:50:43 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,686,421 | 14,646 | alesapin | Only some asserts and test added, but I'm going to fix it. | 2020-09-09T16:49:29 | 2020-09-09T16:49:29 | {} | MEMBER |
ClickHouse | ClickHouse | 689,698,076 | 2,485 | nikitamikhaylov | Fixed here #10188. | 2020-09-09T17:06:36 | 2020-09-09T17:06:36 | {} | MEMBER |
ClickHouse | ClickHouse | 689,230,323 | 14,573 | chang0391 | yes, CH supports create table R as L ，however, Engine of Materializemysql synchronize data from mysql to CH in real time, if I execute create table R like L in MySQL, CH should execute create table R as L , but CH show this issue " Cannot convert create like statement to ClickHouse SQL" | 2020-09-09T00:48:01 | 2020-09-09T02:16:22 | {} | NONE |
ClickHouse | ClickHouse | 689,303,232 | 14,007 | mjodeh | `select database, table, progress, partition_id, num_parts from system.merges` always have 16
system │ metric_log │ 0 │ 202009 │ 91
system │ trace_log │ 0.9829424307036247 │ 202009 │ 93
This is on 2 shards whereas the ot... | 2020-09-09T05:00:22 | 2020-09-09T05:00:22 | {} | NONE |
ClickHouse | ClickHouse | 689,426,110 | 11,234 | traceon | @jjtjiang , you mean some master/admin user and password? It is not needed for "Simple Bind" method. With that approach, the actual user's name and password, that is being authenticated by ClickHouse, are used to "bind" in LDAP server, and so they are checked that way.
Having said that, you really should not use pla... | 2020-09-09T08:53:57 | 2020-09-09T08:53:57 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,110,986 | 14,538 | simPod | Seems like it's finished, no query is running, replication queue is empty. Those dates stayed "null" though 🤔
```
delete_ttl_info_min delete_ttl_info_max
1970-01-01 00:00:00 1970-01-01 00:00:00
``` | 2020-09-08T20:13:44 | 2020-09-09T10:24:12 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,674,075 | 14,589 | fastio | Hi @filimonov, I‘m happy to accomplish this task. | 2020-09-09T16:27:17 | 2020-09-09T16:27:34 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,685,332 | 14,453 | alexey-milovidov | AST Fuzzer: https://github.com/ClickHouse/ClickHouse/issues/14645 | 2020-09-09T16:47:40 | 2020-09-09T16:47:40 | {} | MEMBER |
ClickHouse | ClickHouse | 689,256,175 | 14,584 | IVitamin-C | 
| 2020-09-09T02:17:55 | 2020-09-09T02:17:55 | {} | NONE |
ClickHouse | ClickHouse | 689,314,790 | 14,573 | chang0391 | @zhang2014
> yes, CH supports create table R as L ，however, Engine of Materializemysql synchronize data from mysql to CH in real time, if I execute create table R like L in MySQL, CH should execute create table R as L , but CH show this issue " Cannot convert create like statement to ClickHouse SQL"
>
... | 2020-09-09T05:33:15 | 2020-09-09T05:33:15 | {} | NONE |
ClickHouse | ClickHouse | 689,338,210 | 14,592 | simPod | @filimonov so the point is that I can run huge inserts that take even several seconds and it's fine while I keep 1 QPS pace? | 2020-09-09T06:33:22 | 2020-09-09T06:33:22 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,348,139 | 14,584 | filimonov | > may have hundreds of materialized views
Usually it's not the best idea - this way every insert fan-outs to hundred of inserts / hundred of new parts in derived tables.
New parts need to be merged...
You can also take a look on `parallel_view_processing`, but as @den-crane said the insert will fail once any of... | 2020-09-09T06:54:32 | 2020-09-09T06:54:32 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,473,717 | 13,869 | ssotik | Thank. Helped for me. | 2020-09-09T10:28:17 | 2020-09-09T10:28:17 | {} | NONE |
ClickHouse | ClickHouse | 689,277,291 | 8,282 | wanghaifen | Have you found a solution？ | 2020-09-09T03:28:59 | 2020-09-09T03:28:59 | {} | NONE |
ClickHouse | ClickHouse | 689,349,135 | 14,605 | filimonov | Duplicate of #5319 | 2020-09-09T06:56:32 | 2020-09-09T06:56:32 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,228,608 | 13,875 | mjodeh | Any suggestions to overcome those two errors? `Cannot quickly remove` and `Memory limit exceeded`?
I updated the merge config like this
`<merge_tree>
<max_suspicious_broken_parts>5</max_suspicious_broken_parts>
<parts_to_delay_insert>300</parts_to_delay_insert>
<parts_to_throw_insert>600</p... | 2020-09-09T00:42:38 | 2020-09-09T00:42:38 | {} | NONE |
ClickHouse | ClickHouse | 689,241,979 | 14,568 | yd1261137568 | Sorry, it seems that I can't upload pictures here. After executing on cluster, I will wait until DDL timeout
tjtx-89-77.58os.org :) drop table if exists data_usergroup.aaa on cluster cluster_shard_one;
DROP TABLE IF EXISTS data_usergroup.aaa ON CLUSTER cluster_shard_one
┌─host─────────┬─port─┬─status─┬─error─┬─n... | 2020-09-09T01:29:38 | 2020-09-09T01:29:38 | {} | NONE |
ClickHouse | ClickHouse | 689,276,517 | 14,261 | macEar | @CurtizJ thanks, it helps | 2020-09-09T03:26:10 | 2020-09-09T03:26:10 | {} | NONE |
ClickHouse | ClickHouse | 689,278,636 | 14,251 | liangli1213 | @BohuTANG , I found 20.9.1.4585 under https://repo.yandex.ru/clickhouse/rpm/prestable/x86_64/ and install it, I'm not sure if your commit has been involved.
And so far the issue can be reproduced by following steps:
1. Setup replication relationship between MySQL and ClickHouse
2. Keep data inserting into MySQL
3... | 2020-09-09T03:33:56 | 2020-09-09T03:33:56 | {} | NONE |
ClickHouse | ClickHouse | 689,320,356 | 14,572 | chang0391 | @zhang2014 @BohuTANG
这个问题应该是一个事务对应的数据量大，那么一个GTID所对应的SQL语句就多，似乎是buffer或者数组什么的不够造成的，另外还发现像这样用 procedure循环insert，也会出现这样的问题。
另外，如果主键是字符类型，而不是数字类型，也会报这个错误 | 2020-09-09T05:46:48 | 2020-09-09T05:55:54 | {} | NONE |
ClickHouse | ClickHouse | 689,680,554 | 14,619 | alesapin | Integration OK. | 2020-09-09T16:38:51 | 2020-09-09T16:38:51 | {} | MEMBER |
ClickHouse | ClickHouse | 689,683,554 | 14,572 | zhang2014 | Bad parse protocol. @BohuTANG is trying to fix it. | 2020-09-09T16:44:24 | 2020-09-09T16:44:24 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,203,676 | 13,875 | mjodeh | @den-crane I update to 20.3.17.173 to weeks ago and at the beginning it worked perfectly, and starting from yesterday we are getting `Cannot quickly remove` and `Memory limit exceeded` exceptions, and yesterday those brought the server down! and the file system was also corrupted!
now the insertion rate is 1 bulk inse... | 2020-09-09T00:14:36 | 2020-09-09T00:14:36 | {} | NONE |
ClickHouse | ClickHouse | 689,252,839 | 14,584 | IVitamin-C | Thanks.
Can it be set by parameters？Because a ReplicatedMergeTree table may have hundreds of materialized views, but because a materialized view has no data due to mismatching of field types, other materialized views will be affected. | 2020-09-09T02:05:32 | 2020-09-09T02:05:32 | {} | NONE |
ClickHouse | ClickHouse | 689,281,493 | 14,521 | hexiaoting | @alexey-milovidov
I have added ILIKE for "SHOW DATABASES/TABLES".
But is there a bug with ilike ???
It is not work with system.databases
```
dell123 :) SELECT name FROM system.databases ... | 2020-09-09T03:42:56 | 2020-09-09T03:42:56 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,572,242 | 14,619 | alesapin | `01193_metadata_loading` known flaky test. | 2020-09-09T13:43:24 | 2020-09-09T13:43:24 | {} | MEMBER |
ClickHouse | ClickHouse | 689,687,548 | 14,645 | alexey-milovidov | @azat @akuzm The bug was introduced here: https://github.com/ClickHouse/ClickHouse/pull/12745 | 2020-09-09T16:51:29 | 2020-09-09T16:51:29 | {} | MEMBER |
ClickHouse | ClickHouse | 689,700,528 | 14,645 | azat | >It was merged 40 minues ago, I will check again after pull.
FWIW I've already tested (just in case):
```sh
$ wget https://clickhouse-builds.s3.yandex.net/14530/4c3c1cdaf3e4064a5d65a40dea5383e522e8f2ee/clickhouse_build_check/gcc9_relwithdebuginfo_none_bundled_unsplitted_disable_False_binary/clickhouse
$ chmod +... | 2020-09-09T17:11:17 | 2020-09-09T17:11:59 | {} | MEMBER |
ClickHouse | ClickHouse | 689,195,655 | 14,403 | den-crane | Check `select refcount system.parts where not active` for these parts.
Only inactive parts with refcount=1 can be deleted.
_refcount (UInt32) – The number of places where the data part is used.
A value greater than 2 indicates that the data part is used in queries or merges._
Also check `select count() from sy... | 2020-09-08T23:47:06 | 2020-09-08T23:48:52 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,213,951 | 14,588 | alexey-milovidov | #14603
#14495 | 2020-09-09T00:25:43 | 2020-09-09T00:26:44 | {} | MEMBER |
ClickHouse | ClickHouse | 689,225,567 | 13,875 | mjodeh | > What do you mean by that? CH is a userlevel application. It cannot corrupt FS.
It might do it if the file cache is too big, I don't know I'm just guessing what happened
and the insertion rate was 20 bulk insertion per second and the bulk size did not exceed 2000 rows | 2020-09-09T00:33:29 | 2020-09-09T00:33:29 | {} | NONE |
ClickHouse | ClickHouse | 689,260,732 | 14,573 | zhang2014 | > yes, CH supports create table R as L ，however, Engine of Materializemysql synchronize data from mysql to CH in real time, if I execute create table R like L in MySQL, CH should execute create table R as L , but CH show this issue " Cannot convert create like statement to ClickHouse SQL"
This is not allowed, becaus... | 2020-09-09T02:33:59 | 2020-09-09T02:33:59 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,265,960 | 14,572 | chang0391 | @BohuTANG @zhang2014
I created the function as below, when I executed the SQL "call bulk_insert()" in mysql, I can got correct result in MySQL, but I got the same issue "Attempt to read after eof: while receiving packet from localhost:9000" in CH
by the way, the table have one hundred rows data
st... | 2020-09-09T02:48:41 | 2020-09-09T02:48:41 | {} | NONE |
ClickHouse | ClickHouse | 679,773,098 | 14,007 | mjodeh | > Do you still run optimize table manually ? If yes then don't do it.
No
> How many rows do you insert by one query?
between 1 and 1k rows every 10 seconds per database, it depends on the traffic
CREATE TABLE db.mainTableSingle (`userId` String, `eventTime` UInt32 CODEC(DoubleDelta), `eventName` LowCardinal... | 2020-08-25T05:51:58 | 2020-09-09T05:01:14 | {} | NONE |
ClickHouse | ClickHouse | 689,306,316 | 14,569 | filimonov | ClickHouse itself doesn't work on Windows, so smth on the client side probably.
If you think it clickhouse misbehaves - please prepare some isolated sample showing the problem using clickhouse-client or curl queries | 2020-09-09T05:10:05 | 2020-09-09T05:10:05 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,414,616 | 11,234 | jjtjiang | If there can config username and passoword when 'enable_tls'=no(Specify 'no' for plain text (ldap://) protocol)? | 2020-09-09T08:32:33 | 2020-09-09T08:32:33 | {} | NONE |
ClickHouse | ClickHouse | 689,448,574 | 11,215 | keyboardblue | I reproduce it at version 20.5.3.27 (official build), when I try to global left join on a same table, my sql like below,
```
select
...
table_A a1 global left join table_A a2 on ....
where
1=1
```
And I use below sql instead of `where 1=1`,
```
where
(a1.id = a2.id or a1.id <> a2.id)
```
It works for me. | 2020-09-09T09:35:48 | 2020-09-09T09:39:27 | {} | NONE |
ClickHouse | ClickHouse | 689,696,664 | 14,645 | azat | @alexey-milovidov it should be fixed in #14530 | 2020-09-09T17:03:50 | 2020-09-09T17:03:59 | {} | MEMBER |
ClickHouse | ClickHouse | 689,336,151 | 14,595 | alexey-milovidov | Two OOM in build - ignore. | 2020-09-09T06:27:55 | 2020-09-09T06:27:55 | {} | MEMBER |
ClickHouse | ClickHouse | 689,580,088 | 14,130 | fibersel | @den-crane
Oh, I've heard you.So, the correct behavior is to check shard existence and throw an exceptions is one is abscent.
Yes? | 2020-09-09T13:56:36 | 2020-09-09T14:00:11 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,622,219 | 14,130 | den-crane | I think the best behavior is to
1. move unexpected folders to $broken (not sure where).
or leave them as is if it's not make a mess for the distributed sender.
2. alert into logs with clear message that found some unexpected folder and moved
3. start successfully
| 2020-09-09T15:03:12 | 2020-09-09T15:03:12 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,204,899 | 13,875 | den-crane | > file system was also corrupted
What do you mean by that? CH is a userlevel application. It cannot corrupt FS.
What FS do you use, BTW ? | 2020-09-09T00:18:36 | 2020-09-09T00:18:36 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,244,967 | 14,251 | liangli1213 | @BohuTANG , still can't see the latest rpm, are you able to get more information? | 2020-09-09T01:38:54 | 2020-09-09T01:38:54 | {} | NONE |
ClickHouse | ClickHouse | 689,270,219 | 14,569 | guoweibo | > What is $(interface) in your query?
> Can you create isolated example?
>
> P.S. 19.4 is not supported anymore
val table="test" win10 work
| 2020-09-09T03:03:11 | 2020-09-09T03:03:11 | {} | NONE |
ClickHouse | ClickHouse | 689,353,922 | 14,606 | filimonov | What is your ClickHouse version?
Can you please put your CREATE TABLE statement, and other queries in text format as a comment, so we can try to reproduce that in a lab environment.
Screenshots with code/queries are useless.
Machine translation below.
机器翻译：
您的ClickHouse版本是什么？
您能否以文本格式将CREATE TABLE语句和... | 2020-09-09T07:06:13 | 2020-09-09T07:06:13 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,441,565 | 14,045 | radiatorius | > [ 2.596764] XFS (sda1): Mounting V5 Filesystem
The same issue also occurs on the _ext4_ file system. | 2020-09-09T09:22:05 | 2020-09-09T09:22:05 | {} | NONE |
ClickHouse | ClickHouse | 689,505,986 | 14,606 | den-crane | duplicate: https://github.com/ClickHouse/ClickHouse/issues/14227
https://github.com/ClickHouse/ClickHouse/issues/13634 | 2020-09-09T11:41:07 | 2020-09-09T11:41:07 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,626,944 | 14,517 | SaltTan | Duplicate #14115 | 2020-09-09T15:10:44 | 2020-09-09T15:10:44 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,643,286 | 14,635 | den-crane | CH 19.11.4.24 is out of support.
Supported versions are 20.8, 20.7, 20.6(stable) , 20.3(LTS).
Try to upgrade to supported version and check is it still relevant issue or not.
BTW, the latest version in 19.11 is **19.11.14.1** | 2020-09-09T15:36:40 | 2020-09-09T15:38:18 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,648,026 | 11,512 | zhang2014 | CC: @4ertus2
Can you review #14535 please ? It's also about decimal . | 2020-09-09T15:44:28 | 2020-09-09T15:44:28 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,674,531 | 14,563 | bharatnc | Also addresses #14115 | 2020-09-09T16:28:09 | 2020-09-09T16:28:09 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,241,278 | 13,875 | mjodeh | Unfortunately it was not added to the monitoring we will add it to track the memory usage of CH
But from log checking the following happened
1- Merges are too slow (for almost 1 hour)
then loops between
2- Cannot quickly remove directory
3- Memory limit exceeded
it looped for 3 hours then the server went down
th... | 2020-09-09T01:27:15 | 2020-09-09T01:27:15 | {} | NONE |
ClickHouse | ClickHouse | 689,571,988 | 14,130 | fibersel | As long as distributed table doesn't store any data, we can just skip checking shards and replicas across whole cluster and remove code from 217 to 236 lines, is it so?
Correct me if i'm wrong. | 2020-09-09T13:43:00 | 2020-09-09T13:44:02 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,610,829 | 14,494 | alesapin | AST Fuzzer fixed in https://github.com/ClickHouse/ClickHouse/pull/14632 | 2020-09-09T14:45:57 | 2020-09-09T14:45:57 | {} | MEMBER |
ClickHouse | ClickHouse | 689,616,999 | 14,609 | alesapin | Still have
```
FAILED: src/CMakeFiles/dbms.dir/Storages/MergeTree/MergeTreePartition.cpp.o
/usr/bin/ccache /usr/bin/g++-10 -DAWS_SDK_VERSION_MAJOR=1 -DAWS_SDK_VERSION_MINOR=7 -DAWS_SDK_VERSION_PATCH=231 -DBUNDLED_STATIC_JEMALLOC=1 -DCONFIGDIR=\"\" -DENABLE_OPENSSL_ENCRYPTION -DHAVE_CONFIG_H -DLIBSASL_EXPORTS=1 -D... | 2020-09-09T14:55:23 | 2020-09-09T14:55:23 | {} | MEMBER |
ClickHouse | ClickHouse | 689,711,836 | 14,645 | alexey-milovidov | Yes, it's fixed. | 2020-09-09T17:34:14 | 2020-09-09T17:34:14 | {} | MEMBER |
ClickHouse | ClickHouse | 689,241,664 | 13,875 | mjodeh | log sample https://github.com/ClickHouse/ClickHouse/issues/13875#issuecomment-679441776 | 2020-09-09T01:28:34 | 2020-09-09T01:28:34 | {} | NONE |
ClickHouse | ClickHouse | 689,462,332 | 14,545 | 4ertus2 | gcc OOM in build check | 2020-09-09T10:03:36 | 2020-09-09T10:03:36 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,484,464 | 5,782 | akuzm | > @alexey-milovidov , why there is no uniqExactIf in documentation? https://clickhouse.tech/docs/en/sql-reference/aggregate-functions/reference/
`-If` is a generic suffix that can be applied to any aggregate function (see the docs here https://clickhouse.tech/docs/en/sql-reference/aggregate-functions/combinators/#ag... | 2020-09-09T10:52:42 | 2020-09-09T10:52:42 | {
"+1": 2
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,697,809 | 14,645 | alexey-milovidov | @azat The query is reproducing debug assertion on master. | 2020-09-09T17:06:03 | 2020-09-09T17:06:03 | {} | MEMBER |
ClickHouse | ClickHouse | 689,265,893 | 14,584 | den-crane | There is no such parameter. An insert breaks on any error. | 2020-09-09T02:48:30 | 2020-09-09T02:48:30 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,339,207 | 14,590 | alexey-milovidov | OOM in build - #14607 | 2020-09-09T06:35:39 | 2020-09-09T06:35:39 | {} | MEMBER |
ClickHouse | ClickHouse | 689,340,640 | 14,590 | alexey-milovidov | Fuzzer: https://github.com/ClickHouse/ClickHouse/issues/14608 (unrelated). | 2020-09-09T06:38:52 | 2020-09-09T06:38:52 | {} | MEMBER |
ClickHouse | ClickHouse | 689,619,535 | 14,415 | dreamerfable | Detailed description for updates of aggregatingmergetree:
- add missing content，like description of SimpleAggregateFunction and Sorting Key
- fix some wrong links
- correct the hint level according to the en version
- Make some sentences more understandable | 2020-09-09T14:59:09 | 2020-09-09T14:59:09 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,218,325 | 13,875 | mjodeh | > What FS do you use, BTW ?
xfs
| 2020-09-09T00:28:36 | 2020-09-09T00:28:36 | {} | NONE |
ClickHouse | ClickHouse | 689,229,609 | 14,196 | CurtizJ | > we already restrict using of dots in column names
But actually we can use dots and everything by wrapping column name in backticks. BTW, it doesn't look like a big problem, because we can match column name from query with column names in storage definition, cut its prefix and correctly interpret subcolumn name. | 2020-09-09T00:45:49 | 2020-09-09T00:45:49 | {} | MEMBER |
ClickHouse | ClickHouse | 689,233,262 | 13,875 | mjodeh | > It might do it if the file cache is too big, I don't know I'm just guessing what happened
`from top`
` clickho+ 20 0 279.8g 4.7g 40932 S 0.3 13.4 1489:08 clickhouse-serv`
`free -h`
Mem: 35G 12G 8.8G 46M 13G 21G
Swap: 0B 0B ... | 2020-09-09T00:58:37 | 2020-09-09T00:58:37 | {} | NONE |
ClickHouse | ClickHouse | 689,267,552 | 6,227 | den-crane | https://github.com/ClickHouse/ClickHouse/pull/14049 ALTER TABLE compress_table MODIFY COLUMN value3 CODEC(Default); | 2020-09-09T02:53:38 | 2020-09-09T02:53:38 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,356,402 | 14,592 | filimonov | Yes.
Actually 1 QPS is not a 'hard rule'. Monitor the number of parts / merges during the inserts. If the number stays stable - you can push more inserts / more data. | 2020-09-09T07:11:13 | 2020-09-09T07:11:27 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,414,479 | 14,607 | alesapin | Failure is not related. Something wrong with repo copy? | 2020-09-09T08:32:16 | 2020-09-09T08:32:16 | {} | MEMBER |
ClickHouse | ClickHouse | 689,439,786 | 14,045 | radiatorius | There are no non-standard OS settings, configuration files (config.xml, users.xml) attached.
> [config.zip](https://github.com/ClickHouse/ClickHouse/files/5193702/config.zip)
| 2020-09-09T09:18:31 | 2020-09-09T09:18:31 | {} | NONE |
ClickHouse | ClickHouse | 689,494,644 | 14,613 | alesapin | You have to specify database in the first argument for dictionaries created with `DDL` queries. Probably in your case `SELECT dictGet('default.my_d', 'name', ...)`? | 2020-09-09T11:15:29 | 2020-09-09T11:16:09 | {} | MEMBER |
ClickHouse | ClickHouse | 689,673,010 | 14,573 | zhang2014 | > yes, CH supports create table R as L ，however, Engine of Materializemysql synchronize data from mysql to CH in real time, if I execute create table R like L in MySQL, CH should execute create table R as L , but CH show this issue " Cannot convert create like statement to ClickHouse SQL"
> This is not allowed, beca... | 2020-09-09T16:25:28 | 2020-09-09T16:25:28 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 703,107,701 | 15,372 | dependabot-preview[bot] | OK, I won't notify you again about this release, but will get in touch when a new version is available. If you'd rather skip all updates until the next major or minor version, let me know by commenting `@dependabot ignore this major version` or `@dependabot ignore this minor version`.
If you change your mind, just re-... | 2020-10-03T13:56:16 | 2020-10-03T13:56:16 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 703,192,933 | 15,578 | alexey-milovidov | > For example this will cancel remote queries on other shards as soon as the query fail on one of them
I thought it was already working this way. | 2020-10-04T02:54:26 | 2020-10-04T02:54:26 | {} | MEMBER |
ClickHouse | ClickHouse | 702,707,996 | 15,443 | ucasfl | I have no idea about how to write a test for this, because row number is not a part of query results when dump query results into a file. @Avogar | 2020-10-02T12:32:19 | 2020-10-02T12:32:19 | {} | COLLABORATOR |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.