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 | 811,860,683 | 22,437 | den-crane | Check the real memory consuming of Clickhouse in linux.
top / htop -- RES column.
20.8 calculates memory usage incorrectly.
Just set max_server_memory_usage_to_ram_ratio=10 or 100
never merged https://github.com/ClickHouse/ClickHouse/pull/16121 into 20.8
duplicate of https://github.com/ClickHouse/ClickHouse/is... | 2021-04-01T12:04:53 | 2021-04-01T12:04:53 | {
"+1": 1,
"-1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 811,603,633 | 22,008 | alexey-milovidov | Looks like there are two issues:
- query context is not attached to the parallel formatting threads;
- parallel formatting may want to allocate excessive amounts of memory. | 2021-04-01T03:04:59 | 2021-04-01T03:04:59 | {} | MEMBER |
ClickHouse | ClickHouse | 811,640,502 | 7,554 | alexey-milovidov | @qiuwei Unfortunately it was unfinished. Don't worry, the student defended the work with a good grade.
Now someone else has to take this task... | 2021-04-01T05:05:28 | 2021-04-01T05:05:28 | {
"eyes": 2,
"laugh": 4
} | MEMBER |
ClickHouse | ClickHouse | 811,775,565 | 22,384 | amosbird | I still don't understand why we need `argMax` with tuple arguments. Isn't `max` semantically equivalent? | 2021-04-01T09:18:43 | 2021-04-01T09:18:43 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 811,861,881 | 22,384 | ildus | @amosbird interesting, I guess `max` with reversed arguments could do the same, I don't see any differences on top of my mind. | 2021-04-01T12:07:22 | 2021-04-01T12:11:46 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 811,879,692 | 21,852 | nikitamikhaylov | @excitoon Please, fix build and style checks. Also it is better to rebase on master | 2021-04-01T12:40:36 | 2021-04-01T12:40:36 | {} | MEMBER |
ClickHouse | ClickHouse | 811,943,694 | 22,420 | den-crane | lineorder_demo -- Distributed?
Because stop merges is not applicable to Distributed Engine | 2021-04-01T14:23:43 | 2021-04-01T14:23:43 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 811,552,764 | 22,417 | alexey-milovidov | > Docs check — Docs check passed | 2021-04-01T00:26:32 | 2021-04-01T00:26:32 | {} | MEMBER |
ClickHouse | ClickHouse | 811,638,828 | 1,469 | alexey-milovidov | It's supported under experimental flag in the recent releases. | 2021-04-01T05:01:24 | 2021-04-01T05:01:24 | {} | MEMBER |
ClickHouse | ClickHouse | 811,655,122 | 22,381 | filimonov | > it means that too many merges are not scheduled even in presense of free pool
I can conclude from that that it works quite good now - do we really need to touch it? Risks to brake smth are quite high, the advantage is not very clear. | 2021-04-01T05:37:28 | 2021-04-01T05:38:46 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 811,657,202 | 22,377 | filimonov | Btw: you can translate it and send PR ;) | 2021-04-01T05:41:54 | 2021-04-01T05:41:54 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 811,807,104 | 22,384 | amosbird | ```
create temporary table t1 (i int, j String);
insert into t1 values (1, 'x'), (2, 'y'), (3, 'z');
with max((j, i)) as m, argMax((i, j)) as am
select
(m.2, m.1) max,
am argmax
from t1;
create temporary table t2 (i int, j String);
insert into t2 values (1, 'c'), (2, 'b'), (3, 'a');
with max((j, i)) as ... | 2021-04-01T10:14:15 | 2021-04-01T10:14:15 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 811,825,448 | 21,904 | traceon | @Tasselmi, addind to what @afiskon said: comment out `keeper_server` section in your `config.xml`. | 2021-04-01T10:50:53 | 2021-04-01T10:50:53 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 811,846,356 | 21,370 | vdimir | Because we remove GROUP_BY from query to `Merge` engine, we can clean `aggregation` from `TreeRewriterResult`. But for `WHERE` we still need accurate removing expressions containing identifiers from second table. Before that we removed `WHERE` completely that leads to unused primary or partition keys for left table wit... | 2021-04-01T11:35:08 | 2021-04-01T11:36:03 | {} | MEMBER |
ClickHouse | ClickHouse | 811,865,570 | 22,384 | ildus | and yes, `max` should be slower since it works on the outer tuple column, and `argMax` will work with inner columns directly. | 2021-04-01T12:15:11 | 2021-04-01T12:17:39 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 811,946,597 | 22,411 | nvartolomei | Thank you, @alexey-milovidov. | 2021-04-01T14:28:05 | 2021-04-01T14:28:05 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 811,647,489 | 22,410 | yishanchuan | I use the metrika.xml all the time. It's ok until this issue. | 2021-04-01T05:22:24 | 2021-04-01T05:22:34 | {} | NONE |
ClickHouse | ClickHouse | 811,754,420 | 22,397 | alexey-milovidov | I recommend to use version 21.3, not older. | 2021-04-01T08:45:36 | 2021-04-01T08:45:36 | {} | MEMBER |
ClickHouse | ClickHouse | 811,908,031 | 19,242 | abyss7 | > @abyss7 Is it abandoned?
It's not. I'll finish it. Just got out of scope. | 2021-04-01T13:28:39 | 2021-04-01T13:28:39 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 811,909,766 | 18,101 | alesapin | I'll close this temporarily. If you want to continue just let me know. | 2021-04-01T13:31:15 | 2021-04-01T13:31:15 | {} | MEMBER |
ClickHouse | ClickHouse | 811,943,826 | 18,101 | alexey-milovidov | The feature makes sense but better not on AST level. | 2021-04-01T14:23:56 | 2021-04-01T14:23:56 | {} | MEMBER |
ClickHouse | ClickHouse | 811,947,962 | 22,398 | alexey-milovidov | AMPLab benchmark: https://amplab.cs.berkeley.edu/benchmark/
Outdated and abandoned, does not include ClickHouse. | 2021-04-01T14:30:09 | 2021-04-01T14:30:09 | {} | MEMBER |
ClickHouse | ClickHouse | 811,738,516 | 21,437 | vdimir | @tangxueming I answered in the PR https://github.com/ClickHouse/ClickHouse/pull/21438#issuecomment-811734762
> This fix backported to 21.1 and will available since next update after 21.1.7, it will available soon.
But let's continue discussion here if you will have any questions. | 2021-04-01T08:20:41 | 2021-04-01T08:20:41 | {} | MEMBER |
ClickHouse | ClickHouse | 811,881,594 | 22,438 | KochetovNicolai | Assumption:
Exception from allocator, thrown when we copy strings here:
https://github.com/ClickHouse/ClickHouse/blob/4d49d2c671f4fccc2ad67c5afec7f1b04307a6b4/src/Common/ZooKeeper/ZooKeeper.cpp#L224
prevent callback from notifying event. | 2021-04-01T12:44:19 | 2021-04-01T12:44:19 | {} | MEMBER |
ClickHouse | ClickHouse | 811,916,040 | 22,192 | alesapin | No new failures. | 2021-04-01T13:40:28 | 2021-04-01T13:40:28 | {} | MEMBER |
ClickHouse | ClickHouse | 810,879,502 | 22,399 | Zhile | I think this is a bug. I tried 21.2.7.11, it works fine.
For 21.2.7.11 there's still another bug about JOIN on expression
https://github.com/ClickHouse/ClickHouse/issues/20193
| 2021-03-31T08:25:37 | 2021-04-01T05:08:46 | {} | NONE |
ClickHouse | ClickHouse | 811,785,075 | 22,384 | ildus | @amosbird no, argMax with tuple has same behavior as with two arguments, but it returns a tuple so can be used in SimpleAggregateFunction | 2021-04-01T09:34:51 | 2021-04-01T09:34:51 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 811,803,199 | 22,024 | afiskon | If between optimization and support of one of the platforms the ClickHouse community chooses optimization, this PR can be closed. | 2021-04-01T10:07:00 | 2021-04-01T10:07:00 | {} | NONE |
ClickHouse | ClickHouse | 811,546,218 | 22,207 | alexey-milovidov | There is a strange recursive submodule somewhere:
```
../../wjakob/clang-cindex-python3
``` | 2021-04-01T00:11:53 | 2021-04-01T00:11:53 | {} | MEMBER |
ClickHouse | ClickHouse | 811,598,727 | 22,008 | alexey-milovidov | > And do you mean these operator overloading under the word 'hook'?
Yes.
> BTW, It is only disabled for gcc compiler
I see, there is a case when jemalloc is disabled:
```
malloc_usable_size(ptr)
```
It's more complicated...
| 2021-04-01T02:49:55 | 2021-04-01T02:49:55 | {} | MEMBER |
ClickHouse | ClickHouse | 811,644,450 | 22,410 | filimonov | > There are 4 unfinished hosts (**0 of them are currently active**)
Maybe something wrong with your cluster definition? Some reconfigurations (also dns) since restart? | 2021-04-01T05:15:07 | 2021-04-01T05:15:28 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 811,651,709 | 22,381 | filimonov | > I observe that in most production scenarios the number of concurrently executed merges is less than the pool size (it's around 1 on Yandex.Metrica servers)
That just means your system is well designed and servers are not overloaded.
Imagine the system with 1000 tables and realtime inserts in all of them. Or jus... | 2021-04-01T05:30:33 | 2021-04-01T05:32:50 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 811,677,156 | 17,058 | Mr-KIDBK | I don't understand why it was designed this way... | 2021-04-01T06:27:01 | 2021-04-01T06:27:01 | {
"+1": 1
} | NONE |
ClickHouse | ClickHouse | 811,821,144 | 1,655 | l1t1 | it this function realized? | 2021-04-01T10:41:35 | 2021-04-01T10:41:35 | {} | NONE |
ClickHouse | ClickHouse | 811,881,718 | 22,380 | tavplubix | AST fuzzer (ASan) - OOM
Functional stateless tests - Timeout - known issue, all tests actually passed
Integration tests (thread) - known issue
Performance - not related | 2021-04-01T12:44:35 | 2021-04-01T12:44:35 | {} | MEMBER |
ClickHouse | ClickHouse | 811,947,085 | 22,398 | alexey-milovidov | h2o.ai dataframe like benchmark: https://h2oai.github.io/db-benchmark/ | 2021-04-01T14:28:50 | 2021-04-01T14:28:50 | {} | MEMBER |
ClickHouse | ClickHouse | 811,579,517 | 22,419 | alexey-milovidov | Version 21.3 is the LTS version which means that it comes with one year support.
Whenever an issue is found and fixed, the patch is released.
These patches are numbered sequentially: 21.3.2, 21.3.3, 21.3.4, etc...
You should prefer the version with latest patches in production.
When a new patch comes, there are... | 2021-04-01T01:51:16 | 2021-04-01T01:51:16 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 811,582,452 | 22,408 | alexey-milovidov | Could you please provide some reasoning - how comes that the T-Digest algorithm should return a better answer?
Your example looks like the natural behaviour of the underlying approximate algorithm.
But I cannot quickly quantify if it's as inacurrate as it should be or it can be better. | 2021-04-01T02:00:25 | 2021-04-01T02:00:25 | {} | MEMBER |
ClickHouse | ClickHouse | 811,617,701 | 21,589 | dmalkr |
> Could you please do a couple of other things:
>
> * add some docs here https://github.com/ClickHouse/ClickHouse/blob/master/docs/en/sql-reference/functions/array-functions.md
> The description and one example is enough.
>
> * add a performance test. It would be interesting to compare to `arrayR... | 2021-04-01T03:50:17 | 2021-04-01T03:50:17 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 811,657,986 | 22,379 | filimonov | optimize_read_in_order? | 2021-04-01T05:43:29 | 2021-04-01T05:43:29 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 811,809,764 | 22,384 | UnamedRus | ```
SELECT
argMax(key_b, key_a) AS arg,
argMax((key_b, key_a)) AS arg_tiple,
max((key_a, key_b)) AS mx
FROM
(
SELECT
(arrayJoin([(1, 'str'), (2, NULL)]) AS x).1 AS key_a,
x.2 AS key_b
)
Query id: 75ea828b-191c-432f-8f39-b8de99a7cd52
┌─arg─┬─arg_tiple─┬─mx───────┐
│ str │... | 2021-04-01T10:19:20 | 2021-04-01T10:22:17 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 811,817,839 | 22,384 | UnamedRus | @ildus
> I'm struggling to find the point of -SimpleState with argMax for two arguments, since the result value doesn't contain enough data to use it further. I'll look into how to fix that for tuple case.
> result value doesn't contain enough data to use it further. I'll look into how to fix that for tuple case... | 2021-04-01T10:35:09 | 2021-04-01T10:37:23 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 811,819,652 | 22,416 | l1t1 | thanks | 2021-04-01T10:38:24 | 2021-04-01T10:38:24 | {} | NONE |
ClickHouse | ClickHouse | 811,874,101 | 20,499 | abyss7 | Static build looks fine. | 2021-04-01T12:32:06 | 2021-04-01T12:32:06 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 811,908,696 | 22,285 | alesapin | Race in Join, but not related to changes. | 2021-04-01T13:29:44 | 2021-04-01T13:29:44 | {} | MEMBER |
ClickHouse | ClickHouse | 765,131,136 | 19,028 | ElapsedSoul | To compile with unenable the ldap and ssl support with command:
```
export CC=clang;export CXX=clang++;rm -rf build;mkdir -p build && cd build;cmake .. -DENABLE_LDAP=0 -DENABLE_SSL=OFF -DENABLE_TCMALLOC=OFF -DENABLE_JEMALLOC=OFF -DCMAKE_INSTALL_PREFIX=/opt/clickhouse
```
It works.
By using the newest version, I h... | 2021-01-22T05:18:49 | 2021-04-01T01:01:20 | {} | NONE |
ClickHouse | ClickHouse | 811,753,456 | 22,384 | ildus | @alexey-milovidov yes, the expectation is to get tuple if argument is tuple too, so it can be used in SimpleAggregateFunction | 2021-04-01T08:44:10 | 2021-04-01T08:44:10 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 811,758,918 | 22,404 | lxswbr | @KochetovNicolai
Hi!
According to your method, all the following information can be found. Can you help me interpret what it means? Thanks
Row 600:
────────
arrayStringConcat(arrayMap(lambda(tuple(x), demangle(addressToSymbol(x))), trace), '
'): __lll_lock_wait
_L_lock_883
__GI___pthread_mutex_lock
std::__... | 2021-04-01T08:52:01 | 2021-04-01T08:52:01 | {} | NONE |
ClickHouse | ClickHouse | 811,815,765 | 22,384 | UnamedRus | ```
WITH
max((number, rand())) AS mx,
argMax(rand(), number) AS arg,
argMax((rand(), number)) AS arg_tuple
SELECT arg
FROM numbers(1000000000)
Query id: 1d764dc2-0542-491d-8a6d-09aeec60e37d
┌────────arg─┐
│ 3762245095 │
└────────────┘
1 rows in set. Elapsed: 1.545 sec. Processed 1.00 billio... | 2021-04-01T10:31:10 | 2021-04-01T10:39:54 | {
"rocket": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 811,879,010 | 21,772 | nshah14285 | @alexey-milovidov
I'll provide the details once I run the above mentioned command provided by you. | 2021-04-01T12:39:22 | 2021-04-01T12:39:22 | {} | NONE |
ClickHouse | ClickHouse | 811,899,039 | 22,432 | ctapu4ok | > > Имеется Materialized база
>
> Что это? MaterializedMysql ?
Да | 2021-04-01T13:13:55 | 2021-04-01T13:13:55 | {} | NONE |
ClickHouse | ClickHouse | 811,935,937 | 22,430 | alexey-milovidov | Builds are ok - merge. | 2021-04-01T14:11:29 | 2021-04-01T14:11:29 | {} | MEMBER |
ClickHouse | ClickHouse | 811,946,416 | 22,422 | KochetovNicolai | Hi!
Integration tests try to start it's own docker container with zookeeper.
I suspect that if you already has running zookeeper on your server, it can prevent running zookeeper from docker somehow.
Also, why do you have 3 zookeeper processes running on the same server? | 2021-04-01T14:27:48 | 2021-04-01T14:27:48 | {} | MEMBER |
ClickHouse | ClickHouse | 811,946,626 | 22,398 | alexey-milovidov | https://fivetran.com/blog/warehouse-benchmark (without ClickHouse) | 2021-04-01T14:28:07 | 2021-04-01T14:28:07 | {} | MEMBER |
ClickHouse | ClickHouse | 811,547,908 | 22,207 | alexey-milovidov | I will make a fork to ClickHouse-Extras and remove the recursive submodule. | 2021-04-01T00:14:50 | 2021-04-01T00:14:50 | {} | MEMBER |
ClickHouse | ClickHouse | 811,552,059 | 22,381 | alexey-milovidov | > Is it possible that huge merges get starving to death?
It can be possible, need to experiment. | 2021-04-01T00:24:17 | 2021-04-01T00:24:17 | {} | MEMBER |
ClickHouse | ClickHouse | 811,599,031 | 22,008 | alexey-milovidov | BTW, the case from @CurtizJ may be not related to new/delete. | 2021-04-01T02:51:03 | 2021-04-01T02:51:03 | {} | MEMBER |
ClickHouse | ClickHouse | 811,602,961 | 22,008 | alexey-milovidov | @nikitamikhaylov This is the issue with parallel formatting. | 2021-04-01T03:03:09 | 2021-04-01T03:03:09 | {} | MEMBER |
ClickHouse | ClickHouse | 811,637,487 | 22,416 | filimonov | Mysql protocol don't support custom formats. | 2021-04-01T04:57:08 | 2021-04-01T04:57:08 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 811,770,951 | 22,384 | ildus | I'm struggling to find the point of `-SimpleState` with argMax for two arguments, since the result value doesn't contain enough data to use it further. I'll look into how to fix that for tuple case. | 2021-04-01T09:10:57 | 2021-04-01T09:10:57 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 811,800,627 | 21,904 | afiskon | @Tasselmi this is a configuration issue. NuRaft is not supported on MacOS but it is enabled in the config file you are using. | 2021-04-01T10:02:35 | 2021-04-01T10:02:35 | {} | NONE |
ClickHouse | ClickHouse | 811,756,448 | 21,437 | ilejn | @vdimir , thank you for backporting!
| 2021-04-01T08:48:25 | 2021-04-01T08:48:25 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 811,764,195 | 22,384 | Avogar | Looks like `argMax` doesn't work with `SimpleState` at all:
```
:) SELECT argMaxSimpleState(1, 1)
SELECT argMaxSimpleState(1, 1)
Query id: d30b6d47-c21b-4989-b58c-a6771207598f
0 rows in set. Elapsed: 0.001 sec.
Received exception from server (version 21.4.1):
Code: 42. DB::Exception: Received from loc... | 2021-04-01T09:00:15 | 2021-04-01T09:03:55 | {} | MEMBER |
ClickHouse | ClickHouse | 811,861,815 | 22,435 | zeekling | The article maybe can help you : https://www.bookstack.cn/read/clickhouse-20.10-en/d91b58242743006c.md
I add `-DCMAKE_BUILD_TYPE="Release"` after `cmake .. `
The clickhouser will be about 400MB. | 2021-04-01T12:07:16 | 2021-04-01T12:07:16 | {} | NONE |
ClickHouse | ClickHouse | 811,872,418 | 22,432 | den-crane | >Имеется Materialized база
Что это? MaterializedMysql ? | 2021-04-01T12:29:01 | 2021-04-01T12:29:01 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 811,576,485 | 21,438 | tangxueming | Hi! Does it work in version 21.1 with a Avro formatted message only single record？
Do I back to 20.3 version or running git cherry-pick locally #21919 ? | 2021-04-01T01:40:55 | 2021-04-01T01:40:55 | {} | NONE |
ClickHouse | ClickHouse | 811,579,387 | 21,437 | tangxueming | Hi! Does it work in version 21.1 with a Avro formatted message only single record？
Do I back to 20.3 version or running git cherry-pick locally #21919 ? | 2021-04-01T01:50:48 | 2021-04-01T01:50:48 | {} | NONE |
ClickHouse | ClickHouse | 811,727,109 | 22,217 | kssenii | @UnamedRus, it depends on what behaviour you expect. It should work, but the addresses might be cached. | 2021-04-01T08:02:19 | 2021-04-01T08:02:19 | {} | MEMBER |
ClickHouse | ClickHouse | 811,736,051 | 22,217 | UnamedRus | Clickhouse would understand that each IP address is different replica, so it would be able to use all replicas for LA and HA purposes. | 2021-04-01T08:16:39 | 2021-04-01T08:16:39 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 811,906,171 | 22,404 | KochetovNicolai | As I see, there should be at lease 601 rows. Can I see them all? :)
Those two just wait on mutex, which is ok.
Also it would be interesting to see the query indeed.
| 2021-04-01T13:25:33 | 2021-04-01T13:25:33 | {} | MEMBER |
ClickHouse | ClickHouse | 811,908,053 | 22,427 | alexey-milovidov | > Performance — 4 slower
Either not related or tolerable. | 2021-04-01T13:28:41 | 2021-04-01T13:28:41 | {} | MEMBER |
ClickHouse | ClickHouse | 811,940,818 | 22,420 | KochetovNicolai | Could you try to run `SYSTEM STOP MERGES` before `insert`? Also, if table is replicated, you should run this query for every replica. | 2021-04-01T14:19:20 | 2021-04-01T14:19:20 | {} | MEMBER |
ClickHouse | ClickHouse | 811,603,312 | 22,008 | alexey-milovidov | ```
curl 'http://localhost:8123/' -d "SELECT range(65535) FROM system.one ARRAY JOIN range(65536) AS number" | pv > /dev/null
```
```
gdb
br MemoryTracker::logMemoryUsage
```
```
Thread 142 "Formatter" hit Breakpoint 1, MemoryTracker::logMemoryUsage (this=this@entry=0x121aa020 <total_memory_tracker>, curren... | 2021-04-01T03:04:09 | 2021-04-01T03:04:09 | {} | MEMBER |
ClickHouse | ClickHouse | 811,645,657 | 22,408 | filimonov | Did you check tdunning/t-digest#154 ? | 2021-04-01T05:17:55 | 2021-04-01T05:17:55 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 811,646,274 | 22,404 | filimonov | Also: provide a sample of the query which can not be killed. | 2021-04-01T05:19:27 | 2021-04-01T05:19:27 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 811,740,080 | 22,431 | UnamedRus | Can't you just do the following:
```
INSERT INTO demo (....,dt) SELECT *, toDate('20200310') as dt ....
``` | 2021-04-01T08:23:32 | 2021-04-01T08:23:32 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 811,934,572 | 22,439 | UnamedRus | ```
ALTER TABLE ren UPDATE agg=agg WHERE 1;
``` | 2021-04-01T14:09:13 | 2021-04-01T14:09:13 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 811,628,115 | 21,431 | alexey-milovidov | Ok, I removed compression as we cannot use it for now.
Let's validate the changes in performance.
Previous run showed -0.3% in average, it can be a noise. | 2021-04-01T04:24:51 | 2021-04-01T04:24:51 | {} | MEMBER |
ClickHouse | ClickHouse | 811,636,091 | 19,262 | filimonov | Some extra context https://bugs.chromium.org/p/chromium/issues/detail?id=795089 | 2021-04-01T04:52:06 | 2021-04-01T04:52:06 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 811,784,398 | 21,437 | tangxueming | @ilejn @vdimir @alexey-milovidov
Thanks, I test on 20.11 version. It works well!
| 2021-04-01T09:33:41 | 2021-04-01T09:33:41 | {
"+1": 1
} | NONE |
ClickHouse | ClickHouse | 811,851,525 | 20,499 | abyss7 | I've managed to build server locally, but it has problems with PHDR cache code (debug+shared+ASAN build):
> ==42067==ERROR: AddressSanitizer: global-buffer-overflow on address 0x0000005034fc at pc 0x0000004ae8a7 bp 0x7fff598bbbf0 sp 0x7fff598bb3b8
READ of size 8 at 0x0000005034fc thread T0
#0 0x4ae8a6 in __asa... | 2021-04-01T11:45:59 | 2021-04-01T11:45:59 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 811,551,515 | 17,495 | alexey-milovidov | @simPod https://github.com/ClickHouse/ClickHouse/issues/15090#issuecomment-810180249
@Maokaman1 The chance that this integration with Etcd started to work was neglible. | 2021-04-01T00:22:43 | 2021-04-01T00:22:43 | {
"+1": 3
} | MEMBER |
ClickHouse | ClickHouse | 811,566,609 | 9,589 | alexey-milovidov | @mastertheknife This is wonderful!
Looking forward for patches.
And we will add the build to CI for support in all newer versions. | 2021-04-01T01:10:32 | 2021-04-01T01:10:32 | {} | MEMBER |
ClickHouse | ClickHouse | 811,579,841 | 21,589 | dmalkr | > So basically it's `arrayReduce`, but it takes a lambda instead of an aggregate function, right?
Yes, you are right. And this lambda function requires additional efforts :( | 2021-04-01T01:52:21 | 2021-04-01T01:52:21 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 811,646,884 | 22,410 | yishanchuan | > > There are 4 unfinished hosts (**0 of them are currently active**)
>
> Maybe something wrong with your cluster definition? Some reconfigurations (also dns) since restart?
cluster config :
```
<yandex>
<clickhouse_remote_servers>
<cluster_2shards_1replicas>
<shard>
... | 2021-04-01T05:21:06 | 2021-04-01T05:21:06 | {} | NONE |
ClickHouse | ClickHouse | 811,648,442 | 22,397 | filimonov | Please update to the latest 20.8 or newer and retry. Most probably fixed in https://github.com/ClickHouse/ClickHouse/pull/11807 | 2021-04-01T05:24:23 | 2021-04-01T05:24:23 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 811,734,762 | 21,438 | vdimir | > Does it work in version 21.1 with a Avro formatted message only single record？
@tangxueming
This fix backported to 21.1 and will available since next update after 21.1.7, it will available soon.
| 2021-04-01T08:14:36 | 2021-04-01T08:14:36 | {} | MEMBER |
ClickHouse | ClickHouse | 811,754,689 | 22,408 | alexey-milovidov | No. | 2021-04-01T08:45:57 | 2021-04-01T08:45:57 | {} | MEMBER |
ClickHouse | ClickHouse | 811,779,267 | 21,904 | Tasselmi | > > and compile it successfully, but run the clickhouse binary it shows:
> > ```
> > <jemalloc>: ../contrib/jemalloc/include/jemalloc/internal/tsd_generic.h:145: Failed assertion: "tsd_booted"
> > [1] 81768 abort ./clickhouse --vresion
> > ```
>
> This is something recent, but can be worked around by spe... | 2021-04-01T09:24:46 | 2021-04-01T09:24:46 | {} | NONE |
ClickHouse | ClickHouse | 811,824,057 | 20,499 | abyss7 | > It looks like lld-12 is selected for build with clang-11 and it does not work: https://clickhouse-builds.s3.yandex.net/20499/b79680b8711753a0def52617c74e75202584ed44/clickhouse_build_check/build_log_896821990_1613383537.txt
What problem was inside this log? | 2021-04-01T10:47:54 | 2021-04-01T10:47:54 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 811,908,368 | 22,287 | alesapin | not related to changes | 2021-04-01T13:29:13 | 2021-04-01T13:29:13 | {} | MEMBER |
ClickHouse | ClickHouse | 811,578,636 | 22,406 | depressed-pho | I got the same result If I set both `input_format_parallel_parsing` and `input_format_with_names_use_header` to `0`. That is, rows were still not inserted. | 2021-04-01T01:48:14 | 2021-04-01T01:48:14 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 811,614,811 | 22,426 | alexey-milovidov | @akuzm this is major usability issue:
```
$ clickhouse-client --compress 0 -n < 01783_parallel_formatting_memory.sql
^C^C^C^C^C
^C^C^C^C^C^C^C^C^C^C^C^C^C^C^C^C^C^C^C^C^C^C^C^C^C^C^C
``` | 2021-04-01T03:39:57 | 2021-04-01T03:40:13 | {} | MEMBER |
ClickHouse | ClickHouse | 811,618,428 | 21,361 | cyberhuman | 21.3.2.5 | 2021-04-01T03:52:57 | 2021-04-01T03:52:57 | {} | NONE |
ClickHouse | ClickHouse | 811,789,663 | 22,372 | otrazhenia | @vitlibar | 2021-04-01T09:42:50 | 2021-04-01T09:42:50 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 811,851,110 | 22,373 | alesapin | Found deadlock https://github.com/eBay/NuRaft/issues/190. | 2021-04-01T11:45:13 | 2021-04-01T11:45:13 | {} | MEMBER |
ClickHouse | ClickHouse | 823,497,689 | 22,352 | sevirov | Internal documentation ticket: DOCSUP-8737. | 2021-04-20T18:15:05 | 2021-04-20T18:15:05 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 823,555,650 | 21,556 | alexey-milovidov | https://github.com/ClickHouse/ClickHouse/commit/066a4daf5fbd44460caaaee6f18c5c27d2b5067e | 2021-04-20T19:48:31 | 2021-04-20T19:48:31 | {} | MEMBER |
ClickHouse | ClickHouse | 823,620,477 | 21,417 | alexey-milovidov | Хорошая новость - в версии ClickHouse начиная от 21.2 появилась родная поддержка PostgreSQL источника для словарей. Теперь не надо возиться с ODBC драйверами, и подобной ошибки тоже не будет.
https://github.com/ClickHouse/ClickHouse/blob/master/CHANGELOG.md#new-feature-2
https://clickhouse.tech/docs/en/sql-referenc... | 2021-04-20T21:45:28 | 2021-04-20T21:45:28 | {} | MEMBER |
ClickHouse | ClickHouse | 823,672,434 | 23,398 | den-crane | This feature has been implemented in
ClickHouse release v20.12.3.3-stable, 2020-12-13
Remove empty parts after they were pruned by TTL, mutation, or collapsing merge algorithm. #16895 (Anton Popov).
https://github.com/ClickHouse/ClickHouse/blob/master/docs/en/whats-new/changelog/2020.md | 2021-04-20T23:49:05 | 2021-04-20T23:49:32 | {
"+1": 1
} | CONTRIBUTOR |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.