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 | 836,450,473 | 23,962 | azat | >Functional stateless tests (release, DatabaseReplicated) — fail: 1, passed: 2920, skipped: 108
>Integration tests (thread) — Timeout, fail: 1, passed: 945, flaky: 3
Unrelated
>Yandex third-party checks (only for Yandex employees)
Broken in upstream/master | 2021-05-10T09:18:18 | 2021-05-10T09:18:18 | {} | MEMBER |
ClickHouse | ClickHouse | 837,150,316 | 23,942 | vzakaznikov | Started to see the following fails
```
May 10,2021 13:59:15 ⟥ And executing command
docker-compose --ansi never --project-directory "/ClickHouse/tests/testflows/map_type/map_type_env" --file "/ClickHouse/tests/testflows/map_type/map_type_env/docker-compose... | 2021-05-10T18:56:58 | 2021-05-10T18:56:58 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 837,396,339 | 23,979 | alexey-milovidov | @kitaisreal I'm ok adding this flag. | 2021-05-10T21:46:29 | 2021-05-10T21:46:29 | {} | MEMBER |
ClickHouse | ClickHouse | 838,763,739 | 23,746 | bnaecker | @alexey-milovidov Thanks for the reminder. I'm still waiting on two PRs against ClickHouse extra repos, one to [RocksDB](https://github.com/ClickHouse-Extras/rocksdb/pull/6) and one to [gRPC](https://github.com/ClickHouse-Extras/grpc/pull/6). Once those are integrated, I'll add another patch to `.gitmodules` pointing t... | 2021-05-11T16:12:28 | 2021-05-11T16:12:28 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 838,736,016 | 20,202 | KochetovNicolai | Currently, most of mentioned in https://github.com/ClickHouse/ClickHouse/issues/14730#issue-698305277 is implemented. Exceptions are: `WHERE` in projection definition and everything from `Projection Analysis and Settings`.
This feature is experimental. Enable `allow_experimental_projection_optimization` to use proje... | 2021-05-11T15:57:56 | 2021-05-11T16:18:54 | {} | MEMBER |
ClickHouse | ClickHouse | 838,837,929 | 24,017 | alesapin | SQLancer reported https://github.com/ClickHouse/ClickHouse/issues/24020 | 2021-05-11T17:04:47 | 2021-05-11T17:04:47 | {} | MEMBER |
ClickHouse | ClickHouse | 839,108,005 | 21,598 | iameugenejo | We're having the same issue with version `21.3.4.25` | 2021-05-11T20:08:51 | 2021-05-11T20:08:51 | {
"+1": 1
} | NONE |
ClickHouse | ClickHouse | 839,232,951 | 23,311 | alexey-milovidov | Some queries have to be adjusted... | 2021-05-11T22:23:12 | 2021-05-11T22:23:12 | {} | MEMBER |
ClickHouse | ClickHouse | 839,374,772 | 23,205 | naylinn69 | I think this is a legit bug since `\'xxx` is a valid CSV format and Clickhouse should be still able to parse it correctly.
I understand you could work around it with `INSERT INTO test FORMAT CSV '\''xxx'` but it would be nice not having to surround it with quotes. | 2021-05-12T02:01:08 | 2021-05-12T02:01:08 | {} | NONE |
ClickHouse | ClickHouse | 836,761,069 | 23,988 | alexey-milovidov | I've added changelog entry. | 2021-05-10T14:21:20 | 2021-05-10T14:21:20 | {} | MEMBER |
ClickHouse | ClickHouse | 837,394,520 | 23,349 | alexey-milovidov | The user only see the structure of Distributed table - if it can successfully insert data into it, materialized columns in remote table should not prevent sending the data. | 2021-05-10T21:44:50 | 2021-05-10T21:44:50 | {} | MEMBER |
ClickHouse | ClickHouse | 838,156,353 | 21,468 | alesapin | Sad, but I couldn't reproduce this behavior. We have a lot of tests for drop partition and a lot of clients use drop partition. There should be something very specific configuration... or something like this... | 2021-05-11T09:32:27 | 2021-05-11T09:32:27 | {} | MEMBER |
ClickHouse | ClickHouse | 838,406,053 | 23,844 | alexey-milovidov | @mergifyio update | 2021-05-11T12:41:41 | 2021-05-11T12:41:41 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 837,019,985 | 23,993 | ghost | Sure, I've redacted some info, but kept most of it.
```
filter: (&(ObjectCategory=Person)(userPrincipalName=serpa@company.com))
requesting: All userApplication attributes
# extended LDIF
#
# LDAPv3
# base <OU=users,DC=company,DC=com> with scope subtree
# filter: (&(ObjectCategory=Person)(userPrincipalName=ser... | 2021-05-10T17:36:08 | 2021-05-10T17:36:08 | {} | NONE |
ClickHouse | ClickHouse | 837,379,424 | 23,979 | azat | >Let's try to replace google::sparse_hash_map to something else instead of merging this PR.
When I added sparsehash (back in #6894), the main goal was to keep memory footprint as low as possible.
Also note that prellocation make difference even for the builtin hashtable (same 30% percent +-).
> Maybe robin hood ... | 2021-05-10T21:32:08 | 2021-05-10T21:32:08 | {} | MEMBER |
ClickHouse | ClickHouse | 838,036,831 | 23,631 | alexey-milovidov | > Stress test (thread) — Killed by signal (in clickhouse-server.log)
https://github.com/ClickHouse/ClickHouse/issues/24010 | 2021-05-11T08:07:20 | 2021-05-11T08:07:20 | {} | MEMBER |
ClickHouse | ClickHouse | 838,181,664 | 23,311 | alexey-milovidov | @mergifyio update | 2021-05-11T09:50:46 | 2021-05-11T09:50:46 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 838,182,039 | 23,311 | mergify[bot] | **Command `update`: success**
> **Branch has been successfully updated**
>
| 2021-05-11T09:51:04 | 2021-05-11T09:51:04 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 838,837,408 | 24,020 | alesapin | Same here https://clickhouse-test-reports.s3.yandex.net/24017/591abbadede68ef5eda4d0153332983954c69edf/sqlancer_test.html#fail1 | 2021-05-11T17:04:21 | 2021-05-11T17:04:35 | {} | MEMBER |
ClickHouse | ClickHouse | 839,293,717 | 23,942 | vzakaznikov | Trying to check test stability in CI/CD by executing all modules 4 times back to back. | 2021-05-11T23:49:03 | 2021-05-11T23:49:03 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 839,371,418 | 23,496 | dengc367 | Yeah, I need this feature too, do you have a plan to do this feature? @alexey-milovidov | 2021-05-12T01:50:28 | 2021-05-12T01:50:28 | {} | NONE |
ClickHouse | ClickHouse | 837,047,591 | 23,726 | akuzm | @mergifyio update | 2021-05-10T17:54:09 | 2021-05-10T17:54:09 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 837,047,888 | 23,726 | mergify[bot] | **Command `update`: success**
> **Branch has been successfully updated**
>
| 2021-05-10T17:54:21 | 2021-05-10T17:54:21 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 837,944,689 | 21,594 | sergey-gt-ekb | I will try to get core dump and traceback at weekend. | 2021-05-11T06:53:13 | 2021-05-11T06:53:13 | {} | NONE |
ClickHouse | ClickHouse | 838,241,371 | 24,010 | alexey-milovidov | @vdimir Most likely ThreadStatus is not being passed into thread. | 2021-05-11T10:35:25 | 2021-05-11T10:35:25 | {} | MEMBER |
ClickHouse | ClickHouse | 838,298,402 | 24,014 | l1t1 | select arrayMap(x,y->arrayPushBack(x,y), [[1,2],[3,4]],[5,6]) | 2021-05-11T11:20:03 | 2021-05-11T11:20:03 | {} | NONE |
ClickHouse | ClickHouse | 838,786,917 | 23,746 | bnaecker | No problem, I'll put together a patch to the submodules today. Thanks! | 2021-05-11T16:27:55 | 2021-05-11T16:27:55 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 837,090,178 | 17,769 | mergify[bot] | **Command `update`: success**
> **Branch has been successfully updated**
>
| 2021-05-10T18:20:55 | 2021-05-10T18:20:55 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 838,299,226 | 24,012 | amosbird | I planned to implement index analysis for row policies, but am not sure about the consequences. | 2021-05-11T11:20:42 | 2021-05-11T11:20:42 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 838,389,083 | 24,006 | gelezayka | @alexey-milovidov fantastic :-(
in code carbon-clickhouse following code
query:
insert into graphite.data_sharded format RowBinary
and write row
wb.WriteBytes(name)
wb.WriteFloat64(value)
wb.WriteUint32(timestamp)
wb.WriteUint16(TimestampToDays(timestamp))
wb.Write... | 2021-05-11T12:28:33 | 2021-05-11T12:41:39 | {} | NONE |
ClickHouse | ClickHouse | 838,783,067 | 20,393 | qoega | >SQLancer test
new one
https://github.com/ClickHouse/ClickHouse/issues/24020 | 2021-05-11T16:25:16 | 2021-05-11T16:25:27 | {} | MEMBER |
ClickHouse | ClickHouse | 836,440,146 | 23,979 | azat | >How do you have one hour loading time?
>If dictionary occupies 100 GB, it means just about 30 MB/sec - very slow even for google::sparse_hash_map.
I have even smaller for sparse_hash_map, ~6MB/s (for 2.28 billion of rows)
But this depends from the item size, size of value is 2 bytes, so it is ~0.64e6 keys per se... | 2021-05-10T09:08:58 | 2021-05-10T09:15:52 | {} | MEMBER |
ClickHouse | ClickHouse | 838,185,985 | 23,953 | alesapin | Seems like you are using an unofficial build? The exception in the server log doesn't look OK. Maybe you can provide some additional details like table schema and example data parts? There are a lot of reasons why client may hang, for example, because of the extensive load on the filesystem. | 2021-05-11T09:54:05 | 2021-05-11T09:54:05 | {} | MEMBER |
ClickHouse | ClickHouse | 838,191,805 | 23,780 | MaxWk | > It's not yet implemented but already planned in the roadmap. It will be done in Q2 or Q3.
thanks for you reply | 2021-05-11T09:58:22 | 2021-05-11T09:58:22 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 838,286,708 | 23,993 | ghost | Just to add a bit of more info.
We've been able to make it work if we use the "LDAP External Authenticator" only.
We've changed the config to:
```
<yandex>
<ldap_servers>
<my_ldap_server>
<host>ldap.company.com</host>
<port>636</port>
<bind_dn>{user_name}</bind_dn... | 2021-05-11T11:10:55 | 2021-05-11T11:19:40 | {} | NONE |
ClickHouse | ClickHouse | 838,357,947 | 23,793 | fuwhu | > @fuwhu Minor changes needed.
@alexey-milovidov
sorry for late reply.
Already updated per your suggestion, please help review again, thank you so much. | 2021-05-11T12:06:53 | 2021-05-11T12:06:53 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 838,109,896 | 19,225 | alesapin | Sorry, it's very hard to guess without additional info :( | 2021-05-11T08:59:42 | 2021-05-11T14:25:43 | {} | MEMBER |
ClickHouse | ClickHouse | 838,861,100 | 24,015 | amosbird | It won't work and this method will be deprecated since glibc 2.34, as they start to provide a new libc entrypoint. We have to ship a specific version of glibc sysroot as blob and build with that in near future. | 2021-05-11T17:20:42 | 2021-05-11T17:21:09 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 839,020,975 | 23,978 | kitaisreal | @Realist007 please add more tests for functionality.
1. Need to check how function will behave with constant arguments.
For example CREATE FUNCTION custom_plus as (a, b) -> a + b;
SELECT isConstant(custom_plus(1, 2)) should be true.
2. With non constant arguments.
3. Also need more tests to check for errors, it wo... | 2021-05-11T19:10:00 | 2021-05-11T19:10:00 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 836,759,975 | 23,988 | alexey-milovidov | Ok but it worth adding to changelog. | 2021-05-10T14:20:23 | 2021-05-10T14:20:23 | {} | MEMBER |
ClickHouse | ClickHouse | 837,003,373 | 23,993 | traceon | Can you please post the list of attributes with values for that user in LDAP? | 2021-05-10T17:24:40 | 2021-05-10T17:26:16 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 837,013,117 | 23,349 | mergify[bot] | **Command `update`: success**
> **Branch has been successfully updated**
>
| 2021-05-10T17:31:32 | 2021-05-10T17:31:32 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 837,346,605 | 23,349 | azat | >@azat Do you think, maybe Distributed table should always insert into materialized columns?
This an option, but why? Especially if there is `insert_allow_materialized_columns` setting.
Plus underlying table will still require this setting, and the INSERT will fail | 2021-05-10T21:09:18 | 2021-05-10T21:09:18 | {} | MEMBER |
ClickHouse | ClickHouse | 837,504,614 | 24,006 | alexey-milovidov | Remove partition key from your table (it is totally fine to not having partition key at all) or make it less granular (e.g. partitions by months, not by days). | 2021-05-10T23:41:12 | 2021-05-10T23:41:12 | {} | MEMBER |
ClickHouse | ClickHouse | 838,066,590 | 24,011 | filimonov | Confirmed. Also needs setting:
```
set joined_subquery_requires_alias=0
```
| 2021-05-11T08:28:28 | 2021-05-11T08:28:28 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 838,090,174 | 24,006 | gelezayka | i create new table like this:
CREATE TABLE graphite.data_sharded on cluster tss
(
`Path` String,
`Value` Float64,
`Time` UInt32,
`Date` Date,
`Timestamp` UInt32
)
ENGINE = ReplicatedGraphiteMergeTree('/clickhouse/tables/{layer}-{shard}/graphite.data_sharded', '{replica}', 'graphi... | 2021-05-11T08:45:33 | 2021-05-11T08:45:33 | {} | NONE |
ClickHouse | ClickHouse | 838,178,730 | 24,006 | alexey-milovidov | @gelezayka It must be an error in inserted data - more than 500 distinct months don't look realistic. | 2021-05-11T09:48:42 | 2021-05-11T09:48:42 | {} | MEMBER |
ClickHouse | ClickHouse | 838,287,809 | 24,014 | l1t1 | arrayMap(x,y->arrayConcat(x,[y]), [[1,2],[3,4]],[5,6]) | 2021-05-11T11:11:49 | 2021-05-11T11:13:09 | {} | NONE |
ClickHouse | ClickHouse | 838,406,379 | 23,844 | mergify[bot] | **Command `update`: failure**
> **Pull request can't be updated with latest base branch changes**
> Mergify needs the author permission to update the base branch of the pull request.
> excitoon-favorites needs to [authorize modification on its head branch](https://help.github.com/articles/allowing-changes-to-a-pull-req... | 2021-05-11T12:41:55 | 2021-05-11T12:41:55 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 838,521,101 | 23,647 | alesapin | Yes, CONNECTIONLOSS should be much better. | 2021-05-11T13:57:21 | 2021-05-11T13:57:21 | {} | MEMBER |
ClickHouse | ClickHouse | 838,784,862 | 23,746 | alexey-milovidov | Ok. I've missed the notifications somehow. | 2021-05-11T16:26:29 | 2021-05-11T16:26:29 | {} | MEMBER |
ClickHouse | ClickHouse | 836,553,348 | 21,594 | Maximus43 | > BTW can you also provide:
>
> * linux kernel version
> * linux kernel config
> * `sysctl -a`
> * `/proc/meminfo`
I have migrated Clickhouse VM to full virtualisation (KVM), it works perfect.
Previously Clickhouse was installed on the following instance:
`Linux 5.4.78-2-pve #1 SMP PVE 5.4.78-2 (Thu, 03 Dec ... | 2021-05-10T10:59:07 | 2021-05-10T10:59:07 | {} | NONE |
ClickHouse | ClickHouse | 837,008,382 | 23,349 | alexey-milovidov | @azat Do you think, maybe Distributed table should always insert into materialized columns? | 2021-05-10T17:28:15 | 2021-05-10T17:28:15 | {} | MEMBER |
ClickHouse | ClickHouse | 838,030,781 | 23,769 | vdimir | Seems not related
_Functional stateless tests (debug)_ https://clickhouse-test-reports.s3.yandex.net/23769/cb89a2be9783de646c95ee8c424ce6a664c7dffc/functional_stateless_tests_(debug).html#fail1
```
2021-05-10 19:04:25 01526_initial_query_id: [ FAIL ] 26.40 sec. - ... | 2021-05-11T08:02:57 | 2021-05-11T08:03:13 | {} | MEMBER |
ClickHouse | ClickHouse | 838,609,502 | 23,940 | sevirov | @vitlibar Hi. Please, review this PR. | 2021-05-11T14:48:27 | 2021-05-11T14:48:27 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 838,717,040 | 23,548 | kssenii | btw, `Functional stateless tests flaky check (address) — fail: 93, passed: 7` fails because need to add it to list of tests, which cannot be run in parallel (https://github.com/ClickHouse/ClickHouse/blob/master/tests/queries/skip_list.json#L404). | 2021-05-11T15:46:34 | 2021-05-11T15:46:34 | {} | MEMBER |
ClickHouse | ClickHouse | 836,630,424 | 20,539 | myrrc | Own format reduced testing time to about 3h 23 minutes. | 2021-05-10T12:21:46 | 2021-05-10T12:21:46 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 836,967,775 | 23,555 | genzgd | This issue repeated, again in conjunction with "memory limit exceeded" errors | 2021-05-10T16:58:59 | 2021-05-10T16:58:59 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 837,013,268 | 17,769 | mergify[bot] | **Command `update`: success**
> **Branch has been successfully updated**
>
| 2021-05-10T17:31:38 | 2021-05-10T17:31:38 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 837,178,950 | 22,021 | alexey-milovidov | @magvan Это как-то связано с Postgres? Может быть есть пример такого же поведения с ClickHouse словарём, чтобы легче было написать тест? | 2021-05-10T19:15:46 | 2021-05-10T19:15:46 | {} | MEMBER |
ClickHouse | ClickHouse | 838,073,734 | 24,011 | PhantomPhreak | Yes, with `joined_subquery_requires_alias=1` it's not crashing
```
:) select * FROM (SELECT 1), (SELECT 1), (SELECT 1);
SELECT *
FROM
(
SELECT 1
)
,
(
SELECT 1
)
,
(
SELECT 1
)
Query id: 00d48d73-781f-483d-8b2a-329d207e7865
0 rows in set. Elapsed: 0.120 sec.
Received exceptio... | 2021-05-11T08:33:34 | 2021-05-11T08:37:13 | {} | NONE |
ClickHouse | ClickHouse | 838,633,349 | 479 | socketpair | https://github.com/eBay/NuRaft | 2021-05-11T15:01:08 | 2021-05-11T15:01:08 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 837,010,508 | 17,769 | alexey-milovidov | @mergifyio update | 2021-05-10T17:29:44 | 2021-05-10T17:29:44 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 837,047,065 | 23,726 | mergify[bot] | Sorry but I didn't understand the command. Please consult [the commands documentation](https://docs.mergify.io/commands.html) 📚. | 2021-05-10T17:53:50 | 2021-05-10T17:53:50 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 838,176,091 | 24,010 | vdimir | Not sure how to investigate missing query id | 2021-05-11T09:46:53 | 2021-05-11T09:46:53 | {} | MEMBER |
ClickHouse | ClickHouse | 838,397,311 | 18,275 | fuwhu | > P.S.
You mean if we use `uniqExplicitMerge(user_id, sum)`, it will automatically push down the uniq operator onto each node and merge the finalized result, without setting `distributed_group_by_no_merge ` ?
if so, we may need to make the function name express the meaning of pushing down to each node for execution... | 2021-05-11T12:35:05 | 2021-05-11T12:35:43 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 839,232,174 | 17,769 | alexey-milovidov | Need to investigate all the failed tests, even if they are unrelated... | 2021-05-11T22:22:02 | 2021-05-11T22:22:02 | {} | MEMBER |
ClickHouse | ClickHouse | 839,377,850 | 23,496 | alexey-milovidov | INTERSECT and EXCEPT operators will be implemented not earlier than summer this year.
Right now you can use IN and NOT IN operators for this purpose. | 2021-05-12T02:08:55 | 2021-05-12T02:08:55 | {
"+1": 2
} | MEMBER |
ClickHouse | ClickHouse | 836,983,711 | 23,793 | alexey-milovidov | @fuwhu Minor changes needed. | 2021-05-10T17:10:21 | 2021-05-10T17:10:21 | {} | MEMBER |
ClickHouse | ClickHouse | 837,983,988 | 24,008 | filimonov | Actually, I'm not sure if it should be fixed or not: the new behavior seems more correct.
But there are also compatibility reasons...
Anyway it should be covered by tests in CI. | 2021-05-11T07:26:31 | 2021-05-11T07:26:31 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 838,171,312 | 23,992 | alexey-milovidov | @MaxWk The tests are located in `tests/queries/0_stateless` directory. | 2021-05-11T09:43:31 | 2021-05-11T09:43:31 | {} | MEMBER |
ClickHouse | ClickHouse | 839,159,849 | 24,016 | CurtizJ | Your test failed. Seems it's needed to specify timezone.
```
-1970-01-01 08:16:40.000 original
2021-05-11 20:23:08 -1970-01-01 08:33:20.000
2021-05-11 20:23:08 -1970-01-01 08:50:00.000
2021-05-11 20:23:08 -1970-01-01 09:06:40.000 original
2021-05-11 20:23:08 -1970-01-01 09:23:20.000
2021-05-11 20:23:08 -197... | 2021-05-11T20:55:17 | 2021-05-11T20:55:17 | {} | MEMBER |
ClickHouse | ClickHouse | 836,683,855 | 10,368 | amosbird | I've encountered similar issue twice. In both cases it happened when all replicas were restarting while there were many active inserts. Then some table in some shard failed to merge anymore with bloated replication_queue (> 1000 entries) | 2021-05-10T13:13:25 | 2021-05-10T13:13:25 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 837,005,732 | 23,349 | alexey-milovidov | @mergifyio update | 2021-05-10T17:26:24 | 2021-05-10T17:26:24 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 837,639,594 | 23,934 | hexiaoting | > @hexiaoting yes, it will be best solution for us, because if we merge it as is, there will be issues that some tool that cannot be updated is commited to our system tables column names.
OK, I will rewrite the implementation. | 2021-05-11T02:03:39 | 2021-05-11T02:03:39 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 838,111,384 | 23,843 | MaxWk | > @MaxWk please add documentation, fix style check, and add function test for this new function.
@kitaisreal could you help me check it again ,thanks | 2021-05-11T09:00:41 | 2021-05-11T09:00:41 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 836,647,187 | 23,991 | den-crane | you need to upgrade to 20.12.8.5
or set read_backoff_min_latency_ms=0
duplicate https://github.com/ClickHouse/ClickHouse/issues/18137 | 2021-05-10T12:39:12 | 2021-05-10T12:39:12 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 836,734,385 | 7,666 | SaltTan | This is broken again in 21.3
```
select version(), uptime() from cluster(segmented,system,one)
version() | uptime()
----------+---------
21.4.3.21 | 344985
20.9.7.11 | 344985
21.4.3.21 | 344985
20.9.7.11 | 344985
21.4.3.21 | 344985
``` | 2021-05-10T13:57:34 | 2021-05-10T13:57:34 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 836,975,377 | 23,979 | alexey-milovidov | Let's try to replace google::sparse_hash_map to something else instead of merging this PR.
Maybe robin hood hash table with 95% fill factor will be good for this use case. | 2021-05-10T17:04:18 | 2021-05-10T17:04:18 | {} | MEMBER |
ClickHouse | ClickHouse | 837,183,631 | 23,448 | alexey-milovidov | It is possible but requires moderate changes in query engine.
In current implementation, constant arguments of aggregate functions are processed in the same way as non-constant arguments. But the constant parameters may alter the implementation of aggregate function - in fact, different aggregate functions can be ru... | 2021-05-10T19:18:51 | 2021-05-10T19:19:26 | {} | MEMBER |
ClickHouse | ClickHouse | 837,510,425 | 24,006 | alexey-milovidov | About the setting - you probably put it in a wrong place (it is the query level setting, set up for a user profile, see users.xml).
Anyway, don't bother trying to change this setting. Just make less number of partitions:
> Large number of partitions is a common misconception. | 2021-05-10T23:47:44 | 2021-05-10T23:48:33 | {} | MEMBER |
ClickHouse | ClickHouse | 838,359,097 | 23,992 | MaxWk | > @MaxWk The tests are located in `tests/queries/0_stateless` directory.
thanks ,i added sometests on it | 2021-05-11T12:07:42 | 2021-05-11T12:07:42 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 838,490,922 | 23,647 | alesapin | Investigated this issue:
In the first case:
```
2021.04.26 14:14:04.312935 [ 578 ] {} <Trace> RaftInstance: cancel existing timer
2021.04.26 14:14:04.312967 [ 578 ] {} <Trace> RaftInstance: re-schedule election timer
2021.04.26 14:14:04.312981 [ 578 ] {} <Trace> RaftInstance: batch size hint: 0 bytes
2021.04.26 1... | 2021-05-11T13:37:37 | 2021-05-11T13:37:37 | {} | MEMBER |
ClickHouse | ClickHouse | 838,750,565 | 23,993 | traceon | This looks like an AD schema.
> So, with this we can understand that the correct bind_dn is just {user_name}
`COMPANY.COM\{user_name}` Bind DN should also work, I believe.
As well as `sAMAccountName={user_name},OU=users,DC=company,DC=com` with `serpa` user name.
Regarding the role mapping, this PR addresses... | 2021-05-11T16:07:38 | 2021-05-11T16:07:38 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 838,932,540 | 23,882 | amosbird | I was misled by mysql, https://www.db-fiddle.com/f/sRbD7EJBsozq84K3rPhBC8/0 . It should work | 2021-05-11T18:10:01 | 2021-05-11T18:10:01 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 838,981,583 | 21,992 | amosbird | https://github.com/ClickHouse/ClickHouse/issues/23539#issuecomment-829864824 can help | 2021-05-11T18:42:57 | 2021-05-11T18:42:57 | {
"+1": 1
} | COLLABORATOR |
ClickHouse | ClickHouse | 839,234,801 | 24,029 | alexey-milovidov | Only relevant to 21.6 (master). | 2021-05-11T22:25:48 | 2021-05-11T22:25:48 | {} | MEMBER |
ClickHouse | ClickHouse | 837,185,092 | 23,448 | alexey-milovidov | This task cannot be implemented earlier than #12552. | 2021-05-10T19:19:48 | 2021-05-10T19:19:48 | {} | MEMBER |
ClickHouse | ClickHouse | 838,156,527 | 23,978 | alexey-milovidov | What is missing:
- persistent storage of functions (in the `functions` directory as a list of `.sql` files);
- DROP FUNCTION query;
- access rights for functions manipulation. | 2021-05-11T09:32:34 | 2021-05-11T09:32:34 | {} | MEMBER |
ClickHouse | ClickHouse | 838,172,363 | 23,555 | filimonov | Yes. The fix should exists in 21.4.4 (while anyway I would recommend to update to 21.4.6.55).
If you will face that issue again - please attach the result of those queries.
```
WITH arrayMap(x -> demangle(addressToSymbol(x)), trace) AS all SELECT thread_id, query_id, arrayStringConcat(arrayMap(x -> '0x' || lower... | 2021-05-11T09:44:19 | 2021-05-11T09:48:58 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 838,306,329 | 479 | amosbird | Looks like it's superseded by nuraft as we've already done for etcd. | 2021-05-11T11:26:21 | 2021-05-11T11:32:03 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 836,972,808 | 21,594 | sergey-gt-ekb | > BTW can you also provide:
>
> * linux kernel version
> * linux [kernel_config.txt](https://github.com/ClickHouse/ClickHouse/files/6453721/kernel_config.txt)
> * sysctl -a
> * `/proc/meminfo`
Its a docker, inside open vz vps(host)
kernel 3.10.0-1062.4.2.vz7.116.7
kernel config attached(host)
[kernel_conf... | 2021-05-10T17:02:26 | 2021-05-10T17:02:26 | {} | NONE |
ClickHouse | ClickHouse | 838,166,557 | 23,746 | alexey-milovidov | @bnaecker Also we need a pull request that updates every affected submodule to commits with changes. | 2021-05-11T09:39:59 | 2021-05-11T09:39:59 | {} | MEMBER |
ClickHouse | ClickHouse | 838,754,211 | 23,548 | ucasfl | > btw, `Functional stateless tests flaky check (address) — fail: 93, passed: 7` fails because need to add it to list of tests, which cannot be run in parallel (https://github.com/ClickHouse/ClickHouse/blob/master/tests/queries/skip_list.json#L404).
Ok! | 2021-05-11T16:08:59 | 2021-05-11T16:08:59 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 838,818,945 | 23,746 | bnaecker | @alexey-milovidov Oops, I forgot about [this PR](https://github.com/ClickHouse-Extras/boringssl/pull/7) against BoringSSL as well! | 2021-05-11T16:51:00 | 2021-05-11T16:51:00 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 838,971,453 | 23,865 | amosbird | It works in master. | 2021-05-11T18:36:27 | 2021-05-11T18:36:36 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 841,082,854 | 24,074 | qoega | >SELECT * FROM ( SELECT a, b as "CAST(a, Int32)" FROM t_ab1 ) t_ab1 FULL JOIN t_ab2 ON (t_ab1.a == t_ab2.a); -- { serverError 49 }
I guess this case is caused by bad query. Probably some changes are not needed? | 2021-05-14T07:59:19 | 2021-05-14T07:59:19 | {} | MEMBER |
ClickHouse | ClickHouse | 841,113,654 | 23,958 | kssenii | @electrical, the fix is in 21.3. | 2021-05-14T08:59:43 | 2021-05-14T08:59:43 | {} | MEMBER |
ClickHouse | ClickHouse | 841,115,849 | 20,873 | llwdd | You can try 'SimpleAggregateFunction'.. | 2021-05-14T09:03:47 | 2021-05-14T09:03:47 | {} | NONE |
ClickHouse | ClickHouse | 841,523,674 | 24,131 | alexey-milovidov | Our CI system guarantees that ClickHouse builds successfully with clang-12.
More details can be found in "ClickHouse build check" in every commit. | 2021-05-14T21:59:50 | 2021-05-14T21:59:50 | {} | MEMBER |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.