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 | 610,110,728 | 9,836 | hustnn | > That partially loaded table = "inconsistancy". For example in replicated case it is not clear if the replica is healthy and in-sync unless you check the set of parts. Knowledge about replica health is important for other features, etc. Too many potential bad side effects.
Just one question. Does the replica consis... | 2020-04-07T00:47:54 | 2020-04-07T00:47:54 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 610,372,804 | 10,084 | filimonov | https://github.com/ClickHouse/ClickHouse/issues/8865
https://github.com/ClickHouse/ClickHouse/pull/8866
Also https://github.com/ClickHouse/ClickHouse/pull/9872 | 2020-04-07T13:01:35 | 2020-04-07T13:03:29 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 610,422,857 | 10,068 | CurtizJ | The problem was, that checksum of indexes was calculated as uncompressed files, but actually they are written compressed. Right? | 2020-04-07T14:34:01 | 2020-04-07T14:34:01 | {} | MEMBER |
ClickHouse | ClickHouse | 610,469,322 | 10,096 | filimonov | Related to https://github.com/ClickHouse/ClickHouse/pull/9412 | 2020-04-07T15:54:17 | 2020-04-07T15:54:17 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 610,558,917 | 10,079 | blinkov | Fixed. | 2020-04-07T18:49:15 | 2020-04-07T18:49:15 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 610,408,211 | 8,110 | blinkov | @achimbab have you tried `max_rows_to_read=N` with `read_overflow_mode=break`? | 2020-04-07T14:07:45 | 2020-04-07T14:07:45 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 610,611,984 | 10,101 | vitlibar | > Is it because I have large number of configuration files for dictionaries?
It was because you have large number of databases. Any database is attached to ExternalLoader as a repository of configurations of external dictionaries (even for databases which don't contain any dictionaries). On shutdown all databases ar... | 2020-04-07T20:47:02 | 2020-04-07T20:47:02 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 610,669,347 | 10,053 | alexey-milovidov | There is no assumed order in values of `type`.
It's only a coincidence that it looks useful. | 2020-04-07T23:29:05 | 2020-04-07T23:31:18 | {} | MEMBER |
ClickHouse | ClickHouse | 610,700,650 | 9,577 | alexey-milovidov | @excitoon This PR still requires changes. | 2020-04-08T01:26:35 | 2020-04-08T01:26:35 | {} | MEMBER |
ClickHouse | ClickHouse | 609,874,152 | 10,043 | filimonov | More related:
https://github.com/ClickHouse/ClickHouse/pull/9622
https://github.com/ClickHouse/ClickHouse/pull/7572 | 2020-04-06T15:44:46 | 2020-04-06T15:44:46 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 609,929,545 | 10,069 | blinkov | @vk26, could you add it to all languages? | 2020-04-06T17:25:49 | 2020-04-06T17:25:49 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 610,047,810 | 7,080 | alexey-milovidov | Almost safe. You only have to check that the replica is not active and do it within the same transaction.
Yes, only `/path/to/replica` has to be deleted. | 2020-04-06T21:30:58 | 2020-04-06T21:31:34 | {} | MEMBER |
ClickHouse | ClickHouse | 610,229,634 | 10,052 | qoega | Actually all integrational, stateless, stateful tests are available via docker. There are no tests for now that can show whether this change improves performance and there are no problems with it. | 2020-04-07T07:41:12 | 2020-04-07T07:41:12 | {} | MEMBER |
ClickHouse | ClickHouse | 610,286,712 | 9,998 | Jokser | @alexey-milovidov Everything is ok. Could you please merge it? | 2020-04-07T09:42:04 | 2020-04-07T09:42:04 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 610,362,776 | 10,088 | vanzi | Now that I think about this, the query is actually ambiguous, so this might not be a bug. | 2020-04-07T12:40:20 | 2020-04-07T12:40:20 | {} | NONE |
ClickHouse | ClickHouse | 610,373,940 | 6,954 | filimonov | That behavior changed in 19.16. See:
https://github.com/ClickHouse/ClickHouse/issues/10084
| 2020-04-07T13:03:56 | 2020-04-07T13:03:56 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 610,412,442 | 10,036 | qoega | We should fix all the translations as well | 2020-04-07T14:15:30 | 2020-04-07T14:15:30 | {} | MEMBER |
ClickHouse | ClickHouse | 610,603,643 | 10,101 | alexey-milovidov | Is it because I have large number of configuration files for dictionaries? | 2020-04-07T20:28:02 | 2020-04-07T20:28:02 | {} | MEMBER |
ClickHouse | ClickHouse | 610,718,934 | 10,089 | den-crane | there is no guidance. It is just safe-limiter against spontaneous resource utilization burst.
If your system is able to handle more than 100 you can set any number. | 2020-04-08T02:37:08 | 2020-04-08T02:37:08 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 610,725,972 | 10,087 | AlexanderJLiu | > Number of columns also impacts. Check this `select count() from system.parts_columns` for A and B
@den-crane I've checked, nodes in A have about `410,000` and in B have about `1,000,000`.
So the influence is the result of a combination of various factors?
>Do you use AWS EBS ? It can be that IOPS are limit... | 2020-04-08T03:04:46 | 2020-04-08T03:04:46 | {} | NONE |
ClickHouse | ClickHouse | 610,106,497 | 7,080 | cw9 | Thank you!
| 2020-04-07T00:30:45 | 2020-04-07T00:30:45 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 610,196,598 | 9,836 | filimonov | > Does the replica consistency check not working on partially loaded table?
I have 10 parts, you have 20. Does it mean i need to download 10 parts from you, or i still loading my own parts set? | 2020-04-07T06:15:38 | 2020-04-07T06:15:38 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 610,226,598 | 10,077 | qoega | Docs should be fixed as well. For example:
https://clickhouse.tech/docs/en/engines/database_engines/mysql/#creating-a-database | 2020-04-07T07:34:10 | 2020-04-07T07:34:25 | {} | MEMBER |
ClickHouse | ClickHouse | 610,701,437 | 9,642 | alexey-milovidov | Please don't worry about PR assigments by Ivan Blinkov. He has written a script that assigns pull request to random people. We decided that we will remove this script but Ivan has not stopped it yet. | 2020-04-08T01:29:55 | 2020-04-08T01:29:55 | {} | MEMBER |
ClickHouse | ClickHouse | 609,936,970 | 9,826 | KochetovNicolai | @lu-yuki-matsumoto thank you so much!
Your test perfectly reproduces the case you have described. | 2020-04-06T17:39:47 | 2020-04-06T17:39:47 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 610,035,522 | 7,080 | cw9 | Hi, is it safe to `rmr /path/to/replica` on ZK for this case? Is there anything else to be deleted? | 2020-04-06T21:03:22 | 2020-04-06T21:03:22 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 610,203,754 | 5,538 | peaksnail | hi
I create a Join table with multi join key, and It failed with tips:
joinGet
```
joinGet only supports StorageJoin containing exactly one key
```
join
```
DB::Exception: Unsupported JOIN keys in StorageJoin. Type: 10.
```
tables
```
CREATE TABLE stock_temp (`storeId` String, `warehouseId` String,... | 2020-04-07T06:36:50 | 2020-04-07T06:39:38 | {} | NONE |
ClickHouse | ClickHouse | 610,573,624 | 9,991 | den-crane | @garrettlish Except you are missing this:
create table XA (A DateTime, F Float) Engine=MergeTree order by A;
insert into XA select toDateTime('2020-01-01 00:00:00')+intDiv(number,10), 0 from numbers(10000000);
select count() from XA
│ 10000000 │
1 rows in set. Elapsed: 0.002 sec. Processed 10.00 million rows... | 2020-04-07T19:21:26 | 2020-04-07T19:22:28 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 610,673,105 | 10,053 | alexey-milovidov | Current solution does not look practical. | 2020-04-07T23:42:43 | 2020-04-07T23:42:43 | {} | MEMBER |
ClickHouse | ClickHouse | 610,674,443 | 10,053 | azat | > Current solution does not look practical.
Ok, how about making enum of the following possible values in settings?
- all (query + exception)
- query (query_start + query_finish)
- query_start
- query_finish
- exception (exception_before_start + exception_before_processing)
- exception_before_start
- except... | 2020-04-07T23:47:34 | 2020-04-07T23:47:34 | {} | MEMBER |
ClickHouse | ClickHouse | 610,699,301 | 9,967 | alexey-milovidov | There are two reasons why it's better to close this issue:
1. This is not a bugfix but a new feature. So, it should not be backported.
2. The lifetime of version 19.16 is over. | 2020-04-08T01:20:53 | 2020-04-08T01:20:53 | {} | MEMBER |
ClickHouse | ClickHouse | 610,817,470 | 10,113 | Sam--Shan | 
i find the original code | 2020-04-08T08:11:17 | 2020-04-08T08:11:17 | {} | NONE |
ClickHouse | ClickHouse | 609,856,140 | 9,826 | lu-yuki-matsumoto | @KochetovNicolai
I should apologize for you about mistake I made in my query.
Group column is in tableA.
Below query can reproduce only case of returning 7 values.
(I have not find artificial data to reproduce cases of returning 4-6 values yet.)
```
create table tableA (id UInt64, col1 UInt64, colDate Dat... | 2020-04-06T15:13:49 | 2020-04-06T15:27:09 | {} | NONE |
ClickHouse | ClickHouse | 609,918,749 | 10,011 | erste | We purge all data, recreate all databases and tables and issue still persist.
And ith simple database as in first message alter table also not working.
Another strange behavior we noticed recently. Our cluster have one leader for all replicas:
```
SELECT
database,
table,
is_leader,
can_become_... | 2020-04-06T17:05:02 | 2020-04-06T17:05:02 | {} | NONE |
ClickHouse | ClickHouse | 610,219,411 | 10,069 | vk26 | @blinkov Hi! I added to other locales too. | 2020-04-07T07:17:20 | 2020-04-07T07:17:20 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 610,277,830 | 7,572 | filimonov | I've just looked on the idea (really like that), and way of configuring the feature (i'm not so happy about that, check comments).
Didn't look on the code yet - the changeset involves moving pieces of code around, and need feally deep look to follow the changes.
@zhang2014 it is much simpler when bigger refacto... | 2020-04-07T09:23:15 | 2020-04-07T09:23:15 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 610,278,593 | 10,081 | xiashuijun | users.xml配置，src/Core/Settings.h | 2020-04-07T09:24:55 | 2020-04-07T09:24:55 | {} | NONE |
ClickHouse | ClickHouse | 610,603,401 | 10,101 | alexey-milovidov | I don't understant why there were huge number of messages on shutdown... | 2020-04-07T20:27:31 | 2020-04-07T20:27:31 | {} | MEMBER |
ClickHouse | ClickHouse | 610,685,135 | 10,049 | alexey-milovidov | PS. Please don't worry about PR assigments by Ivan Blinkov. He has written a script that assigns pull request to random people. We decided that we will remove this script but Ivan has not stopped it yet. | 2020-04-08T00:27:05 | 2020-04-08T00:27:05 | {} | MEMBER |
ClickHouse | ClickHouse | 610,721,510 | 9,811 | alexey-milovidov | @blinkov has written a script that assigns pull requests to random people. But it was by mistake and he will disable this script soon. | 2020-04-08T02:46:42 | 2020-04-08T02:46:42 | {} | MEMBER |
ClickHouse | ClickHouse | 610,779,759 | 10,111 | fangru07 | > 您是否检查过没有其他应用可以监听роrt9000？
no,don`t open remote connection can start,so i think port 9000 is no problem | 2020-04-08T06:39:27 | 2020-04-08T06:39:27 | {} | NONE |
ClickHouse | ClickHouse | 610,796,261 | 10,053 | azat | >If the user really wants to get this feature, they can use Null engine for system.query_log and create a meterialized view on top of it. Maybe it's enough?
For some uses cases - yes, but it will not allow to change the type without recreating the table.
Why I found this useful is:
- you can limit amount of info... | 2020-04-08T07:23:41 | 2020-04-08T07:23:41 | {} | MEMBER |
ClickHouse | ClickHouse | 610,801,850 | 10,113 | Sam--Shan | aaaCH is a 60 million data | 2020-04-08T07:37:01 | 2020-04-08T07:37:01 | {} | NONE |
ClickHouse | ClickHouse | 610,183,266 | 9,935 | ljr900723 | Could you please someone to help on this? | 2020-04-07T05:32:41 | 2020-04-07T05:32:41 | {} | NONE |
ClickHouse | ClickHouse | 610,684,651 | 10,049 | alexey-milovidov | Yes, it's Ok. Let's update the tests. | 2020-04-08T00:24:57 | 2020-04-08T00:24:57 | {} | MEMBER |
ClickHouse | ClickHouse | 610,697,504 | 10,105 | alexey-milovidov | This PR is Ok.
> Maybe it is better to add conversion?
Yes, it is better to add conversion.
Because the issue is only happens when inserting to local shard and when inserting to remote shards, conversion happens automatically. So, it's better to also add conversion for local shard.
| 2020-04-08T01:13:59 | 2020-04-08T01:13:59 | {} | MEMBER |
ClickHouse | ClickHouse | 610,701,786 | 6,826 | alexey-milovidov | Ivan Blinkov has assigned this PR by mistake.
He has written a script that assigns pull requests randomly but he has failed. | 2020-04-08T01:31:15 | 2020-04-08T01:31:15 | {} | MEMBER |
ClickHouse | ClickHouse | 610,007,698 | 10,011 | filimonov | @alesapin
Exception cames from this lines:
https://github.com/ClickHouse/ClickHouse/blob/3a1e375b8ab8a264b14bd646cc677149aa910a92/dbms/src/Storages/MergeTree/MergeTreeData.cpp#L689-L697
Maybe `getStoragePolicy` returns trash because global_context was not initialized (or badly / partially initialized) for DDLWork... | 2020-04-06T20:03:07 | 2020-04-06T20:04:18 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 610,336,900 | 10,037 | AlexanderJLiu | @alesapin Thanks~, I've solved this problem by copying `metadata` node. Considering that other people who want to upgrade clickhouse will also encounter this problem, I'll give the `python code` I've used. Also, `zkCli.sh` maybe work, but I don't think it's convenient to handle lots of replicated tables.
In pyth... | 2020-04-07T11:39:17 | 2020-04-07T11:39:17 | {} | NONE |
ClickHouse | ClickHouse | 610,390,810 | 9,643 | hellvesper | This is great, is there build instruction? I'm trying to build on Ubuntu 18.04 with clan8 and 9 without luck. | 2020-04-07T13:36:09 | 2020-04-07T13:36:09 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 610,587,333 | 10,077 | alexey-milovidov | All table functions:
- mysql
- odbc
- jdbc
- remote
- cluster
Should have consistent behaviour:
We should allow to specify database and table names without quotes.
The same for MySQL database engine. | 2020-04-07T19:51:45 | 2020-04-07T19:52:00 | {} | MEMBER |
ClickHouse | ClickHouse | 610,617,109 | 10,102 | den-crane | related https://github.com/ClickHouse/ClickHouse/issues/6287 https://github.com/ClickHouse/ClickHouse/issues/5131 | 2020-04-07T20:59:01 | 2020-04-07T20:59:59 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 610,700,832 | 9,577 | alexey-milovidov | Ugly hack was added without any explanation :( | 2020-04-08T01:27:16 | 2020-04-08T01:27:16 | {} | MEMBER |
ClickHouse | ClickHouse | 610,708,405 | 10,053 | alexey-milovidov | PS. Please don't worry about "spam" assigments by Ivan Blinkov. He has written a script that assigns pull request to random people. We decided that we will remove this script but Ivan has not stopped it yet. | 2020-04-08T01:57:07 | 2020-04-08T01:57:07 | {} | MEMBER |
ClickHouse | ClickHouse | 610,733,885 | 10,087 | den-crane | Yes. Probably it's because of the number of parts 44000 + 20 logical cores
CH iterates through 44000 parts using 20 cores.
Try to set mergetree parameter max_part_loading_threads (auto=20 cores) to 48.
config.xml
```
<merge_tree>
<max_part_loading_threads>48</max_part_loading_threads>
</merge... | 2020-04-08T03:37:13 | 2020-04-08T03:47:26 | {
"+1": 2,
"hooray": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 610,775,766 | 10,111 | qoega | have you checked that there is no other application that listen for роrt 9000? | 2020-04-08T06:27:15 | 2020-04-08T06:27:15 | {} | MEMBER |
ClickHouse | ClickHouse | 610,824,256 | 10,032 | phuongdm1987 | I found the solution for it. many thanks @crade
We need set collate for it, like below:
`ORDER BY field ASC COLLATE 'vi'` | 2020-04-08T08:26:13 | 2020-04-08T08:26:13 | {} | NONE |
ClickHouse | ClickHouse | 610,096,830 | 10,071 | filimonov | There were bug reports about that... | 2020-04-06T23:57:03 | 2020-04-06T23:57:03 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 610,672,937 | 10,105 | azat | Without the patch there will be `assert_cast` failure in debug build and `SIGSEGV` in release:
```
2020.04.08 02:40:32.766293 [ 1244895 ] {18f3904c-11c5-4f72-b017-cdd4d1bdb28d} <Error> executeQuery: Code: 368, e.displayText() = DB::Exception: Bad cast from type DB::ColumnVector<unsigned long> to DB::ColumnNullable ... | 2020-04-07T23:42:05 | 2020-04-07T23:42:05 | {} | MEMBER |
ClickHouse | ClickHouse | 610,277,397 | 10,068 | qoega | Need to rerun tests:
`2020.04.06 22:51:54.206687 [ 209 ] {} <Error> Application: File not found: /etc/clickhouse-server/users.d/access_management.xml` | 2020-04-07T09:22:21 | 2020-04-07T09:22:21 | {} | MEMBER |
ClickHouse | ClickHouse | 610,378,092 | 10,081 | filimonov | You need to put it info default profile in user.xml | 2020-04-07T13:12:16 | 2020-04-07T13:12:16 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 610,382,846 | 10,087 | filimonov | The question is number of parts in both systems. Check count in system.parts | 2020-04-07T13:21:15 | 2020-04-07T13:21:15 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 610,455,307 | 10,068 | alesapin | > The problem was, that checksum of indexes was calculated as uncompressed files, but actually they are written compressed. Right?
They both must be compared compressed and uncompressed. | 2020-04-07T15:29:52 | 2020-04-07T15:29:52 | {} | MEMBER |
ClickHouse | ClickHouse | 610,707,856 | 10,053 | alexey-milovidov | > Ok, how about making enum of the following possible values in settings?
Almost Ok. We have to made setting values consistent with Enum values in system.query_log (QueryFinish instead of query_finish).
But this solution looks too heavy for this task and it will be over complicated. Need to think more...
And it's ... | 2020-04-08T01:55:07 | 2020-04-08T01:55:46 | {} | MEMBER |
ClickHouse | ClickHouse | 610,718,464 | 10,094 | alexey-milovidov | PS. @blinkov has assigned wrong people by mistake. | 2020-04-08T02:35:19 | 2020-04-08T02:35:19 | {} | MEMBER |
ClickHouse | ClickHouse | 610,374,657 | 10,084 | filimonov | The issue happend only when you recreate table. In metadata it is written in flattened format.
On-disk represenatation did not change. | 2020-04-07T13:05:26 | 2020-04-07T13:05:26 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 610,604,265 | 10,101 | alexey-milovidov | But configuration is not changed on shutdown. So, what's the reason I have these messages at all? | 2020-04-07T20:29:23 | 2020-04-07T20:29:23 | {} | MEMBER |
ClickHouse | ClickHouse | 610,749,520 | 10,087 | AlexanderJLiu | > Try to set mergetree parameter max_part_loading_threads (auto=20 cores) to 48.
@den-crane Thanks~, I've tried this, and it does improve the starting speed to `7-8` minutes.
So, what's the `proper value` for this setting, does it depend on the both logical cores and system parts?
I also notice that when click... | 2020-04-08T04:49:55 | 2020-04-08T04:49:55 | {} | NONE |
ClickHouse | ClickHouse | 610,803,992 | 10,113 | Sam--Shan | 
| 2020-04-08T07:42:00 | 2020-04-08T07:42:00 | {} | NONE |
ClickHouse | ClickHouse | 609,733,961 | 10,055 | filimonov | Most probably the issue with root certs & paths.
```
apt-get install ca-certificates
```
and this one: https://github.com/ClickHouse/ClickHouse/issues/3983#issuecomment-452920726
(I was seeing issues with caConfig path few times) | 2020-04-06T11:21:16 | 2020-04-06T15:38:24 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 609,984,000 | 10,063 | alesapin | Can be merged according to previous tests run. | 2020-04-06T19:13:12 | 2020-04-06T19:13:12 | {} | MEMBER |
ClickHouse | ClickHouse | 610,274,638 | 10,079 | blinkov | @SButterfly we're working on this. As a workaround you can use docs from some recent stable release, https://clickhouse.tech/docs/v20.3/en/ | 2020-04-07T09:16:28 | 2020-04-07T09:16:28 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 610,356,479 | 10,077 | zhang2014 | Test failure irrelevant to the modification | 2020-04-07T12:26:28 | 2020-04-07T12:26:28 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 610,401,614 | 6,286 | stavrolia | Yep, we have disagreement about interface of this function. It steel need a lot of work, and I have not find any customer of this feature to have any discussion and feeling we need to spend time on it. | 2020-04-07T13:56:06 | 2020-04-07T13:56:06 | {
"confused": 4
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 610,555,149 | 10,052 | Vxider | I think this PR is beneficial in the INSERT SELECT query. I ran a simple test on my computer.
```
CREATE TABLE my_ints(number UInt64) ENGINE MergeTree order by tuple();
CREATE MATERIALIZED VIEW mv ENGINE MergeTree order by tuple() AS SELECT * FROM my_ints;
```
With the parallel insert:
```
INSERT INTO my_int... | 2020-04-07T18:41:46 | 2020-04-07T18:41:46 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 610,777,220 | 10,081 | BayoNet | It looks like the [Settings Introduction](http://localhost/docs/en/operations/settings/) topic also answers this question. | 2020-04-08T06:31:48 | 2020-04-08T06:31:48 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 610,082,637 | 10,041 | filimonov | > so there is 40501 free inode, that big no ?!!
If depends. You have table with 237 columns, so clickhouse with each insert will create 237*2 files + about 5 more.
So you need 480 files / inodes per insert. By looking on the part number i can deduct that you did about 200 inserts before the error. 480*200 = 96000... | 2020-04-06T23:09:20 | 2020-04-06T23:11:22 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 610,199,146 | 10,076 | filimonov | But simple count queries should not be issue. Reopened.
What is your ClickHouse version? Can you show the create statement of the table you're selecting from? | 2020-04-07T06:23:19 | 2020-04-07T06:23:19 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 610,484,146 | 10,037 | alesapin | Fix https://github.com/ClickHouse/ClickHouse/pull/10095. | 2020-04-07T16:20:10 | 2020-04-07T16:20:10 | {
"+1": 1,
"hooray": 1
} | MEMBER |
ClickHouse | ClickHouse | 610,666,496 | 10,067 | alexey-milovidov | @demo-99 Some test have failed, please look at the reports. | 2020-04-07T23:18:52 | 2020-04-07T23:18:52 | {} | MEMBER |
ClickHouse | ClickHouse | 610,698,336 | 10,026 | CurtizJ | I hope for this time perftest report will be prepared. | 2020-04-08T01:16:59 | 2020-04-08T01:16:59 | {} | MEMBER |
ClickHouse | ClickHouse | 610,701,199 | 9,545 | alexey-milovidov | @ludv1x Please don't worry about PR assigments by Ivan Blinkov. He has written a script that assigns pull request to random people. We decided that we will remove this script but Ivan has not stopped it yet. | 2020-04-08T01:28:47 | 2020-04-08T01:28:47 | {} | MEMBER |
ClickHouse | ClickHouse | 610,772,289 | 10,098 | AlexanderJLiu | Same error occur using versioin `20.3.5.21` of clickhouse.
**Reproduce**:
1. Just create a replicated table with new replica and do nothing with `index_granularity_bytes`.
2. Restart, and clickhouse-server can't be started. Error says
```
2020.04.08 14:06:51.814429 [ 13328 ] {} <Error> Application: DB::Except... | 2020-04-08T06:16:14 | 2020-04-08T06:16:14 | {} | NONE |
ClickHouse | ClickHouse | 610,789,437 | 10,111 | filimonov | Most probably you have <listen_host> listed twice, and clickhouse trys to listen same interface several times. | 2020-04-08T07:06:05 | 2020-04-08T07:06:05 | {
"+1": 2
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 610,806,158 | 1,784 | qoega | After successful run of
```
set optimize_throw_if_noop = 1;
OPTIMIZE TABLE tbl FINAL;
```
that was done after replication is done I would expect no duplicates should be guaranteed. | 2020-04-08T07:47:00 | 2020-04-08T07:47:00 | {
"+1": 2
} | MEMBER |
ClickHouse | ClickHouse | 610,297,638 | 4,190 | ekbfh | Up | 2020-04-07T10:05:55 | 2020-04-07T10:05:55 | {} | NONE |
ClickHouse | ClickHouse | 610,454,331 | 4,717 | SaltTan | I don't have questions, but the documentation has not been changed to reflect the new behaviour. | 2020-04-07T15:28:12 | 2020-04-07T15:28:12 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 610,718,371 | 10,094 | alexey-milovidov | One test have failed. | 2020-04-08T02:34:56 | 2020-04-08T02:34:56 | {} | MEMBER |
ClickHouse | ClickHouse | 610,697,943 | 10,020 | alexey-milovidov | It happens only on erroneous cases, so let it throw. | 2020-04-08T01:15:35 | 2020-04-08T01:15:35 | {} | MEMBER |
ClickHouse | ClickHouse | 610,698,817 | 9,967 | alexey-milovidov | All builds have failed. It looks like Poco submodule is referring incorrect commit. | 2020-04-08T01:18:47 | 2020-04-08T01:18:47 | {} | MEMBER |
ClickHouse | ClickHouse | 610,713,801 | 9,559 | alexey-milovidov | @blinkov has assigned Nikolay Kochetov by mistake.
This task is "work in progress" and is not ready to be assigned.
Let's forgive Ivan Blinkov for the mistake. | 2020-04-08T02:17:38 | 2020-04-08T02:17:38 | {} | MEMBER |
ClickHouse | ClickHouse | 610,720,363 | 10,087 | AlexanderJLiu | @filimonov I've checked the count of parts in these two clusters by `select count() from system.parts`.
Nodes in cluster A have about `16000` parts, and in B have about `44000` parts.
Most tables in A are partitioned by `Month`，and in B are partitioned by `day`. I think it's the reason leading to less parts in A... | 2020-04-08T02:42:25 | 2020-04-08T02:42:25 | {} | NONE |
ClickHouse | ClickHouse | 609,959,789 | 10,062 | grachov | Thanks again! 👍 | 2020-04-06T18:23:56 | 2020-04-06T18:23:56 | {} | NONE |
ClickHouse | ClickHouse | 610,198,315 | 10,076 | filimonov | That is a duplicate of #1576 | 2020-04-07T06:20:50 | 2020-04-07T06:20:50 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 610,276,892 | 9,991 | garrettlish | thanks @den-crane, seems that `KeyCondition` only supports few functions but not `toString`.
https://github.com/ClickHouse/ClickHouse/blob/ea6f90b4f2c3cc2f7d8b846c769b7f3e84907e47/src/Storages/MergeTree/KeyCondition.cpp#L109-L263 | 2020-04-07T09:21:20 | 2020-04-07T09:21:20 | {} | NONE |
ClickHouse | ClickHouse | 610,344,834 | 6,991 | compasses | Since it was closed, CH support UDF now? Or any docs about it? I believe it will be a great win if CH support extension, which user can define, mostly the high level user.
Recently i want to implement some udf, and found i have lots of things to do.
| 2020-04-07T11:59:04 | 2020-04-07T11:59:04 | {
"+1": 2
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 610,393,580 | 10,090 | qoega | OK for me | 2020-04-07T13:41:14 | 2020-04-07T13:41:14 | {} | MEMBER |
ClickHouse | ClickHouse | 610,722,213 | 10,080 | den-crane | https://github.com/ClickHouse/ClickHouse/issues/5197
It looks like it was fixed in #5133
19.6 is not supported.
You need to upgrade at least to 19.6.3.18 or better to LTS 19.14.10.16 https://repo.yandex.ru/clickhouse/deb/lts/main/
| 2020-04-08T02:49:29 | 2020-04-08T02:49:29 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 610,723,177 | 10,087 | den-crane | Number of columns also impacts. Check this `select count() from system.parts_columns` for A and B.
Do you use AWS EBS ? It can be that IOPS are limited by 500 (EBS GP2). | 2020-04-08T02:53:40 | 2020-04-08T02:57:36 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 599,458,510 | 9,142 | alexey-milovidov | Backported to 20.1. | 2020-03-16T10:28:09 | 2020-03-16T10:28:09 | {} | MEMBER |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.