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 | 696,325,236 | 15,027 | akuzm | > @akuzm Could you please validate if new profile runs are collected to flame graphs?
It's very simple to do: download the 'Test Output' and open the flame graph files. You would've noticed that they have about ten samples each (both CPU and Real variety), which means the setting didn't work and the queries use the ... | 2020-09-21T19:31:41 | 2020-09-21T19:31:41 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,493,747 | 15,115 | alexey-milovidov | 1 billion dataset:
```
The dataset size is: 1000000000
Will perform benchmark. Results:
[0.027, 0.002, 0.002],
[0.128, 0.060, 0.059],
[1.225, 0.119, 0.118],
[4.611, 0.144, 0.132],
[4.733, 0.286, 0.267],
[7.382, 0.666, 0.662],
[0.147, 0.111, 0.110],
[0.094, 0.058, 0.059],
[6.565, 1.255, 1.282],
[7.... | 2020-09-22T03:47:27 | 2020-09-22T03:47:27 | {} | MEMBER |
ClickHouse | ClickHouse | 696,511,457 | 15,116 | ucasfl | ok, thanks! | 2020-09-22T05:12:03 | 2020-09-22T05:12:03 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 696,542,977 | 15,112 | filimonov | And one more candidate:
https://github.com/elliotchance/sqltest
see also https://github.com/ClickHouse/ClickHouse/issues/10294
P.S. is there a way to run sqlancer from CI? Maybe it should be executed from time to time (for example when new release is forked)? Same proposition for coverage test. | 2020-09-22T06:55:00 | 2020-09-22T08:35:55 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,652,939 | 15,129 | amosbird | Thanks! It's because I forgot to clear the children... | 2020-09-22T11:07:20 | 2020-09-22T11:07:20 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 696,262,783 | 15,089 | alesapin | Doesn't work for compact parts :( | 2020-09-21T17:35:55 | 2020-09-21T17:36:06 | {} | MEMBER |
ClickHouse | ClickHouse | 696,348,928 | 14,414 | azat | test_s3_with_proxy - https://clickhouse-test-reports.s3.yandex.net/14731/6657be90a1e9621fa60321d891a30266174f8525/integration_tests_(release).html#fail1 | 2020-09-21T20:16:24 | 2020-09-21T20:16:24 | {} | MEMBER |
ClickHouse | ClickHouse | 696,420,524 | 15,007 | alexey-milovidov | Stopeed after 4.5 hours,
https://drive.google.com/file/d/1PNrqvhYkJV8pbgnVDUuM3VeGIAcRNJJY/view?usp=sharing
| 2020-09-21T22:51:39 | 2020-09-21T22:51:39 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 696,565,949 | 14,538 | simPod | Create table is
```sql
CREATE TABLE default.t1
(
`fetched_at` DateTime,
`a` UUID,
`b` UUID,
`c` UUID,
`d` UUID,
`e` UUID,
`in` UInt64,
`out` UInt64,
`in_s` Nullable(UInt64),
`out_s` Nullable(UInt64),
`link_type` UInt8,
`type` LowCardinality(String)
)
E... | 2020-09-22T07:53:09 | 2020-09-22T07:53:09 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,631,345 | 14,962 | akuzm | > #14570
>
> Is it going to fix this issue as well?
It does fix it. | 2020-09-22T10:15:12 | 2020-09-22T10:15:12 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,647,834 | 14,771 | akuzm | The `clone()` method of `ASTWithElement` is broken -- the cloned node references some source nodes. Apparently the reason is that the `IAST::children` is not cleared after copy construction, so it references both the old and the new nodes. As far as I understand, the general model for ASTs should be that `children` own... | 2020-09-22T10:55:03 | 2020-09-22T10:55:03 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,706,734 | 15,128 | simPod | Probably only corrupted table, my edge case I guess. | 2020-09-22T13:03:38 | 2020-09-22T13:03:38 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,266,873 | 15,060 | alexey-milovidov | Builds Ok :+1: | 2020-09-21T17:43:58 | 2020-09-21T17:43:58 | {} | MEMBER |
ClickHouse | ClickHouse | 696,269,492 | 15,076 | alesapin | ` Yandex third-party checks (only for Yandex employees) ` -- Flaky
` Performance ` -- not related
` AST fuzzer ` -- reported
` Integration tests (asan) — fail: 2, passed: 839, error: 0 ` Known failures. | 2020-09-21T17:49:03 | 2020-09-21T17:49:03 | {} | MEMBER |
ClickHouse | ClickHouse | 696,375,257 | 15,063 | azat | >Seems like now we have a lot of stderr in thread sanitizer build https://clickhouse-test-reports.s3.yandex.net/15063/5c58b9b1a75d59f242c50b5dd720b52c0bf29ad9/functional_stateless_tests_(thread)/test_run.txt.out.log. Probably need to add separate suppression for it.
This is because of TSAN verbosity, this is superfl... | 2020-09-21T21:03:37 | 2020-09-21T21:03:37 | {} | MEMBER |
ClickHouse | ClickHouse | 696,480,625 | 15,115 | alexey-milovidov | It works but sometimes `axel` may download file partially. | 2020-09-22T02:44:15 | 2020-09-22T02:44:15 | {} | MEMBER |
ClickHouse | ClickHouse | 696,702,255 | 15,126 | den-crane | It will be implemented in DateTime64 eventually. Stay tuned.
https://github.com/ClickHouse/ClickHouse/pull/9404 | 2020-09-22T12:55:29 | 2020-09-22T12:55:29 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,112,786 | 14,796 | 4ertus2 | Probably related `Data compressed with different methods`. Affects `join_algorithm = 'auto'` only.
```
SET join_algorithm = 'auto';
SET max_bytes_in_join = 100;
CREATE TABLE t (`x` UInt32, `s` LowCardinality(String)) ENGINE = Memory;
CREATE TABLE nr (`x` Nullable(UInt32), `s` Nullable(String)) ENGINE = Memory;
... | 2020-09-21T13:25:01 | 2020-09-21T13:25:01 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,277,647 | 15,052 | alexey-milovidov | Pipeline stuck in 20.9, @KochetovNicolai
But it's not related to this backport. | 2020-09-21T18:03:51 | 2020-09-21T18:03:51 | {} | MEMBER |
ClickHouse | ClickHouse | 696,335,285 | 15,101 | den-crane | WA:
```
SELECT *
FROM left
LEFT JOIN
(
SELECT
id,
cast(batch, 'LowCardinality(Nullable(String))')
FROM right
) AS right ON left.id = right.id
┌─id─┬─right.id─┬─cast(batch, 'LowCardinality(Nullable(String))')─┐
│ 0 │ 0 │ test │
│... | 2020-09-21T19:50:45 | 2020-09-21T19:50:45 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,416,784 | 15,092 | alexey-milovidov | Test Ok. | 2020-09-21T22:40:46 | 2020-09-21T22:40:46 | {} | MEMBER |
ClickHouse | ClickHouse | 696,493,654 | 15,116 | ucasfl | even if just made a small change, it need more than 20 minitues to link, especially for linking `zookeeper-create-entry-to-download-part`, need almost 15 minutes. | 2020-09-22T03:47:02 | 2020-09-22T03:47:02 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 696,588,445 | 15,125 | l1t1 | stderr.log
```
[root@centos clickhouse-server]# pwd
/var/log/clickhouse-server
[root@centos clickhouse-server]# tail stderr.log
Saved preprocessed configuration to '/var/lib/clickhouse//preprocessed_configs/users.xml'.
Logging trace to /var/log/clickhouse-server/clickhouse-server.log
Logging errors to /var/log/c... | 2020-09-22T08:43:07 | 2020-09-22T08:43:07 | {} | NONE |
ClickHouse | ClickHouse | 696,119,259 | 15,086 | den-crane | Please show your select query | 2020-09-21T13:36:17 | 2020-09-21T13:36:17 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,278,860 | 15,071 | amosbird | The I guess it's gcc 9.3.0's fault. gcc 10 compiles just fine. | 2020-09-21T18:06:03 | 2020-09-21T18:06:03 | {
"+1": 1
} | COLLABORATOR |
ClickHouse | ClickHouse | 696,611,223 | 15,021 | liangli1213 | @zhang2014 @BohuTANG , I found PARTITION are set as below after doing MaterializeMySQL convert, so how the number comes? is this by desired?
activity.sql:PARTITION BY intDiv(id, 4294967)
checkins.sql:PARTITION BY intDiv(id, 18446744073709551)
checkouts_kpit.sql:PARTITION BY intDiv(id, 18446744073709551)
| 2020-09-22T09:30:47 | 2020-09-22T09:30:47 | {} | NONE |
ClickHouse | ClickHouse | 696,612,407 | 15,021 | liangli1213 | And I have another question, if I delete records from MySQL, will the records also be deleted in ClickHouse? | 2020-09-22T09:33:32 | 2020-09-22T09:33:32 | {} | NONE |
ClickHouse | ClickHouse | 696,647,905 | 15,129 | akuzm | From the PR:
>The `clone()` method of `ASTWithElement` is broken -- the cloned node references some source nodes. Apparently the reason is that the `IAST::children` is not cleared after copy construction, so it references both the old and the new nodes. As far as I understand, the general model for ASTs should be that... | 2020-09-22T10:55:14 | 2020-09-22T10:55:14 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,655,597 | 12,631 | azat | >Using numactl leads to mystery segfaults all over the place. I couldn't find out the reason, closing until better times.
This may be because of jemalloc issue (but only if there are more than one cpu node on ci), see #15035 (and referenced PR in jemalloc), you can try disable percpu_arena | 2020-09-22T11:13:27 | 2020-09-22T11:19:14 | {} | MEMBER |
ClickHouse | ClickHouse | 696,705,769 | 15,123 | den-crane | No. It's not possible. You can only create a new table and copy data there using `insert into new select from old` | 2020-09-22T13:01:50 | 2020-09-22T13:01:50 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,099,810 | 14,877 | nvartolomei | If I'm not mistaken this only happens when static linking is disabled. | 2020-09-21T13:03:38 | 2020-09-21T13:03:38 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,169,592 | 15,007 | nikitamikhaylov | > > Looking forward to merge this PR when it's complete.
>
> @abyss7 BTW I'm open to any contributions here, since this should be merged ASAP, otherwise there will be more and more conflicts
@azat I will probably help with some of integration tests. | 2020-09-21T14:56:30 | 2020-09-21T14:56:30 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 696,278,339 | 15,049 | alexey-milovidov | RabbitMQ - bad test. | 2020-09-21T18:05:08 | 2020-09-21T18:05:08 | {} | MEMBER |
ClickHouse | ClickHouse | 696,376,620 | 15,007 | azat | Conflicts with #15110 | 2020-09-21T21:05:59 | 2020-09-21T21:05:59 | {} | MEMBER |
ClickHouse | ClickHouse | 696,410,157 | 10,136 | vitlibar | I'll finish the work in https://github.com/ClickHouse/ClickHouse/pull/15111 | 2020-09-21T22:22:20 | 2020-09-21T22:22:20 | {} | MEMBER |
ClickHouse | ClickHouse | 696,646,065 | 12,771 | filimonov | It looks like it supersedes https://github.com/ClickHouse/ClickHouse/pull/5253
And most probably librdkafka don't need libgsasl anymore. | 2020-09-22T10:50:40 | 2020-09-22T10:50:40 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,672,419 | 14,796 | 4ertus2 | Need reproduction with #15088 | 2020-09-22T11:53:38 | 2020-09-22T11:53:38 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,686,845 | 15,087 | fastio | @sundy-li Thanks.
> 1. How to change the current zookeeper cluster to another one smoothly? We can't use zk copy to migrate the znodes. The only way I can think of is to copy all data from old tables to the new tables.
True, we can't copy data of zookeeper cluster to another one. If the multi zookeeper clusters s... | 2020-09-22T12:24:37 | 2020-09-22T12:25:14 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,111,860 | 15,081 | den-crane | Maybe it's your outdated client does ?
Check `max((date, rate)) ` vs `toString(max((date, rate)) )`
```
SELECT max((date, rate)) AS last_rate
FROM (select today() date, 0 rate, 'y' pair)
WHERE pair = 'x';
┌─last_rate────────┐
│ ('1970-01-01',0) │
└──────────────────┘
``` | 2020-09-21T13:23:24 | 2020-09-21T13:23:24 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,119,801 | 15,086 | yishanchuan | select `type`,query,`exception`,written_rows,event_date,event_time from `system`.query_log ql ORDER BY event_time DESC;; | 2020-09-21T13:37:11 | 2020-09-21T13:37:11 | {} | NONE |
ClickHouse | ClickHouse | 696,267,019 | 15,060 | alexey-milovidov | Only builds matter. | 2020-09-21T17:44:17 | 2020-09-21T17:44:17 | {} | MEMBER |
ClickHouse | ClickHouse | 696,327,367 | 15,027 | akuzm | > > @akuzm Could you please validate if new profile runs are collected to flame graphs?
>
> It's very simple to do: download the 'Test Output' and open the flame graph files. You would've noticed that they have about ten samples each (both CPU and Real variety), which means the setting didn't work and the queries us... | 2020-09-21T19:35:31 | 2020-09-21T19:35:31 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,377,987 | 15,101 | alexey-milovidov | @4ertus2 but LowCardinality of Nullable is Ok. | 2020-09-21T21:08:39 | 2020-09-21T21:08:39 | {} | MEMBER |
ClickHouse | ClickHouse | 696,579,953 | 7,690 | qoega | You can check on build from master(for example via testing release) or wait 20.10 release. | 2020-09-22T08:24:51 | 2020-09-22T08:24:51 | {
"hooray": 1
} | MEMBER |
ClickHouse | ClickHouse | 696,618,941 | 15,061 | akuzm | What about `AST fuzzer — Fuzzer failed (134). See the logs`? It's a bug.https://github.com/ClickHouse/ClickHouse/issues/15129 | 2020-09-22T09:48:11 | 2020-09-22T09:48:11 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,701,804 | 15,090 | alexey-milovidov | We will first try NuRaft for the following reasons:
- less dependencies (`braft` has dependency on `brpc`);
- library is small and neat;
- documentation (`braft` only has chinese);
- examples (allows quick start).
But there is a chance that it will not work as expected, then we will continue with alternatives. | 2020-09-22T12:54:38 | 2020-09-22T12:54:44 | {
"+1": 6
} | MEMBER |
ClickHouse | ClickHouse | 696,703,677 | 15,122 | den-crane | changelog is not ready yet. Usually it delays for 2-4 weeks after release. | 2020-09-22T12:58:09 | 2020-09-22T12:58:29 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,704,674 | 15,125 | den-crane | It seems you're using systemd based distro.
Use only systemctl command
sudo systemctl start clickhouse-server
sudo systemctl stop clickhouse-server
sudo systemctl status clickhouse-server
And don't use service clickhouse-server or /etc/init.d/clickhouse-service
#14298 https://github.com/ClickHouse/ClickHo... | 2020-09-22T13:00:02 | 2020-09-22T13:00:02 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,092,370 | 15,081 | den-crane | AFAIK for compatibility with foreign DBs.
For example CH can work on PostgreSQL wire protocol, and it breaks PostgreSQL clients because 0000-00-00 is invalid date for them.
Also Java does not support 0000-00-00 as date and it creates huge confusion a lot of questions from users. They see Null dates from NotNul... | 2020-09-21T12:49:55 | 2020-09-21T12:51:06 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,135,858 | 15,086 | yishanchuan | Previous error is run by other tool by HTTP protocol.
When I remove the "ORDER BY event_time DESC" clause and run by clickhouse-client (TCP)，the strange response appear:
The result is partially correct,... | 2020-09-21T14:04:05 | 2020-09-21T14:11:42 | {} | NONE |
ClickHouse | ClickHouse | 696,265,350 | 15,071 | alexey-milovidov | Basically it's the same motivation as for std::array to be uninitialized by default. | 2020-09-21T17:40:54 | 2020-09-21T17:40:54 | {} | MEMBER |
ClickHouse | ClickHouse | 696,536,159 | 14,731 | azat | >Need manual editing of third-party code, I will do...
@alexey-milovidov any news? (I'm curious because this patch set can cause conflicts or some new storage will be added and it will break compilation) | 2020-09-22T06:35:01 | 2020-09-22T06:35:01 | {} | MEMBER |
ClickHouse | ClickHouse | 696,664,982 | 15,090 | fastio | How about this proj [braft](https://github.com/baidu/braft) ? | 2020-09-22T11:35:36 | 2020-09-22T11:35:36 | {
"+1": 2
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,677,699 | 12,513 | adrian17 | Still reproducible on 20.9.2.
| 2020-09-22T12:05:12 | 2020-09-22T12:05:12 | {} | NONE |
ClickHouse | ClickHouse | 696,706,395 | 15,122 | simPod | Aight, thanks for info! | 2020-09-22T13:03:01 | 2020-09-22T13:03:01 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,237,533 | 14,791 | tavplubix | > > RENAME TABLE t1 TO t2; -- Works fine, but path it still '/ch/tables/t1'
>
> Try to restart CH / or reattach the table.
I mean that macros unfolding will not solve the issue with table renaming, because we cannot rename path in ZooKeeper. We can unfold macros on table creation and restart/reattach after `REN... | 2020-09-21T16:49:18 | 2020-09-21T16:49:18 | {} | MEMBER |
ClickHouse | ClickHouse | 696,265,535 | 15,062 | akuzm | `test_rename_column/test.py::test_rename_distributed_parallel_insert_and_select` in CI Fest https://github.com/ClickHouse/ClickHouse/issues/14414 | 2020-09-21T17:41:14 | 2020-09-21T17:41:14 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,485,366 | 15,115 | alexey-milovidov | Yandex.Cloud Intel Cascade Lake 80 vCPU, 160 GB RAM, 4 TB Network SSD:
```
Starting clickhouse-server
Waiting for clickhouse-server to start
The dataset size is: 100000000
Will perform benchmark. Results:
[0.024, 0.002, 0.002],
[0.067, 0.012, 0.012],
[0.104, 0.017, 0.017],
[0.411, 0.020, 0.021],
[0.5... | 2020-09-22T03:08:12 | 2020-09-22T03:08:12 | {} | MEMBER |
ClickHouse | ClickHouse | 695,978,645 | 15,065 | filimonov | ### More array/aggregate functions
* fold / reduce and arrayFold / arrayReduce - to build some custom folding/reducing logic (see scala)
* maxN / minN / maxUniqN / minUniqN and same for arrays - collect several biggest / smalles elements from column / array without full sorting (complexity should be `N*log(number ... | 2020-09-21T08:33:57 | 2020-09-22T08:37:37 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,606,399 | 15,125 | l1t1 | if i run `clickhouse-server start`, it is ok
```
[root@centos clickhouse-server]# clickhouse-server start
Processing configuration file 'config.xml'.
There is no file 'config.xml', will use embedded config.
Include not found: networks
Logging trace to console
2020.09.22 17:12:15.380094 [ 12304 ] {} <Information>... | 2020-09-22T09:20:33 | 2020-09-22T09:20:33 | {} | NONE |
ClickHouse | ClickHouse | 696,628,091 | 15,128 | simPod | To start the replica I had to remove table sql from metadata folder and data folder. | 2020-09-22T10:08:02 | 2020-09-22T10:08:02 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,101,218 | 15,081 | simPod | Thanks for explanation.
Maybe just note that this still returns `0000-00-00` as max when no row is found.
```sql
SELECT max((date, rate)) AS last_rate
FROM exchange_rates
WHERE pair = 'x';
``` | 2020-09-21T13:06:10 | 2020-09-21T13:06:21 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,339,115 | 15,101 | 4ertus2 | Some of types are not allowed inside Nullable yet. LC - one of them. There's also tuples and arrays. You may check it with
```
SELECT toTypeName(toNullable(something))
```
For LC it returns
```
SELECT toTypeName(toNullable(toLowCardinality('c')))
┌─toTypeName(toNullable(toLowCardinality('c')))─┐
│ LowCardinal... | 2020-09-21T19:57:59 | 2020-09-21T20:04:01 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,644,817 | 3,997 | filimonov | Related #12771
/cc @ilejn | 2020-09-22T10:47:37 | 2020-09-22T10:47:37 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,227,397 | 14,791 | filimonov | > RENAME TABLE t1 TO t2; -- Works fine, but path it still '/ch/tables/t1'
Try to restart CH / or reattach the table.
> Maybe forbid renaming (throw exception) if these macros are used?
Sometimes renames are necessary (like change sorting key by recreating the table).
Due to that `{database}` and `{table}... | 2020-09-21T16:29:42 | 2020-09-21T16:30:18 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,264,440 | 15,071 | alexey-milovidov | That's the same as for builtin types.
They are not default initialized, explicit zero initialization possible, e.g.:
```
uint64_t x; // uninitialized
uint64_t x{}; // zero init
``` | 2020-09-21T17:39:03 | 2020-09-21T17:39:03 | {} | MEMBER |
ClickHouse | ClickHouse | 696,349,294 | 14,731 | azat | >Integration tests (asan) — fail: 1, passed: 839, error: 0
>Integration tests (release) — fail: 2, passed: 838, error: 0
Does not looks related | 2020-09-21T20:17:04 | 2020-09-21T20:17:04 | {} | MEMBER |
ClickHouse | ClickHouse | 696,473,898 | 15,073 | sundy-li | How to make `fast test` check run with the new added submodule? | 2020-09-22T02:12:19 | 2020-09-22T02:12:19 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,480,790 | 15,115 | alexey-milovidov | AWS c5.metal, 300 GB g.p. SSD:
```
./hardware.sh
~/clickhouse-benchmark ~
Starting clickhouse-server
Waiting for clickhouse-server to start
The dataset size is: 100000000
Will perform benchmark. Results:
[0.012, 0.002, 0.002],
[0.066, 0.018, 0.018],
[0.066, 0.028, 0.027],
[0.186, 0.033, 0.031],
[0... | 2020-09-22T02:45:06 | 2020-09-22T02:45:06 | {} | MEMBER |
ClickHouse | ClickHouse | 696,488,122 | 15,114 | den-crane | SMT like that
```sql
create table target (A Int64, SL String, SR String) Engine=MergeTree order by tuple();
create table left (A Int64, S String) Engine=MergeTree order by tuple();
create table right (A Int64, S String) Engine=MergeTree order by tuple();
create materialized view left_mv to target
as sel... | 2020-09-22T03:20:56 | 2020-09-22T03:20:56 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,539,896 | 14,414 | azat | 01370_client_autocomplete_word_break_characters - https://clickhouse-test-reports.s3.yandex.net/15063/3fd85430f2431af8b8ef11a1c80966a0820be6e5/functional_stateless_tests_(thread).html#fail1 | 2020-09-22T06:46:19 | 2020-09-22T06:46:19 | {} | MEMBER |
ClickHouse | ClickHouse | 696,563,272 | 15,120 | zhang2014 | Will be implemented in https://github.com/ClickHouse/ClickHouse/pull/12656 | 2020-09-22T07:46:59 | 2020-09-22T07:46:59 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,644,489 | 2,551 | filimonov | Related https://github.com/ClickHouse/ClickHouse/pull/12771 | 2020-09-22T10:46:47 | 2020-09-22T10:46:47 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,089,761 | 15,083 | handgunman | понятно, то есть настройка подключения влияет на настройки команды create view.. | 2020-09-21T12:44:48 | 2020-09-21T12:44:48 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,191,612 | 15,073 | filimonov | https://github.com/ClickHouse/ClickHouse/issues/14863
https://github.com/ClickHouse/ClickHouse/pull/8624 | 2020-09-21T15:30:14 | 2020-09-21T15:30:14 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,270,465 | 15,035 | alexey-milovidov | Merge with master and check again? | 2020-09-21T17:50:52 | 2020-09-21T17:50:52 | {} | MEMBER |
ClickHouse | ClickHouse | 696,488,301 | 15,116 | alexey-milovidov | Do you have some numbers for comparison? | 2020-09-22T03:21:51 | 2020-09-22T03:21:51 | {} | MEMBER |
ClickHouse | ClickHouse | 696,539,579 | 15,073 | filimonov | The script which is executed for fast builds is here:
https://github.com/ClickHouse/ClickHouse/blob/master/docker/test/fasttest/run.sh
If cmake for Rocksdb will respect `ENABLE_LIBRARIES` as a default value - it will just work automatically
https://github.com/ClickHouse/ClickHouse/blob/4cd7de175e9402c97cf686a85a... | 2020-09-22T06:45:25 | 2020-09-22T06:46:10 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,661,345 | 15,135 | akuzm | The original PR was not released yet, so we don't have to backport it or mention in changelog. | 2020-09-22T11:26:34 | 2020-09-22T11:26:34 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,112,419 | 14,265 | qoega | I've added your image name in integration tests run. Now it should pass it to integration tests parameters.
You've got an error as `:latest` tag is pushed only from master. Your image was build correctly but it has id of current PR.
https://clickhouse-test-reports.s3.yandex.net/14265/624a4c4d55f64b67cacbab9fed4e30b... | 2020-09-21T13:24:19 | 2020-09-21T13:24:19 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 696,278,088 | 15,051 | alexey-milovidov | > Stress test (memory) — Server failed to start
That's an old infrastructure issue. | 2020-09-21T18:04:41 | 2020-09-21T18:04:41 | {} | MEMBER |
ClickHouse | ClickHouse | 696,481,080 | 15,115 | alexey-milovidov | Yandex.Cloud Intel Cascade Lake 32 vCPU, 128 GB RAM, 300 GB Network SSD:
```
$ ./hardware.sh
~/clickhouse-benchmark ~
Starting clickhouse-server
Waiting for clickhouse-server to start
The dataset size is: 100000000
Will perform benchmark. Results:
[0.021, 0.001, 0.001],
[0.051, 0.011, 0.010],
[0.396... | 2020-09-22T02:46:38 | 2020-09-22T03:07:48 | {} | MEMBER |
ClickHouse | ClickHouse | 696,643,601 | 14,843 | akuzm | I still haven't written the integration test. Will merge as is to make progress, because this is a bug that affects some users. | 2020-09-22T10:44:31 | 2020-09-22T10:44:31 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,644,572 | 12,631 | akuzm | Using `numactl` leads to mystery segfaults all over the place. I couldn't find out the reason, closing until better times. | 2020-09-22T10:47:01 | 2020-09-22T10:47:01 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,121,685 | 15,081 | simPod | You're right, my test just caught it.
I have to rework all queries that rely on nil value `'0000-00-00 00:00:00'` now. | 2020-09-21T13:40:17 | 2020-09-21T13:40:17 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,264,843 | 15,071 | alexey-milovidov | PS. Our collegues has developed `wide_integer` as a proposal for C++ working group, so it it as consistent with standard as possible. | 2020-09-21T17:39:52 | 2020-09-21T17:39:52 | {} | MEMBER |
ClickHouse | ClickHouse | 696,264,469 | 15,066 | den-crane |
1. Do see any difference if you add `identity` (for a test only) ` toYYYYMMDDhhmmss(identity(date_time))=20200920214000` | 2020-09-21T17:39:07 | 2020-09-21T17:40:14 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,265,767 | 15,064 | alexey-milovidov | 00463_long_sessions_in_http_interface - flaky test
> test_rename_column
- recently broken, maybe here: #11913 | 2020-09-21T17:41:42 | 2020-09-21T17:41:42 | {} | MEMBER |
ClickHouse | ClickHouse | 696,277,146 | 15,058 | alexey-milovidov | RabbitMQ flaky test. | 2020-09-21T18:02:58 | 2020-09-21T18:02:58 | {} | MEMBER |
ClickHouse | ClickHouse | 696,280,007 | 15,024 | alexey-milovidov | > test_s3_with_proxy
- flaky test.
> test_rename_column
- recently broken, maybe here: #11913
> Performance
Ok. | 2020-09-21T18:08:09 | 2020-09-21T18:08:09 | {} | MEMBER |
ClickHouse | ClickHouse | 696,380,864 | 15,007 | azat | Integration tests for 9f891c42ec7fd6a31668d125cc874493a3c72e22 commit definitely hanged, can someone obtain logs from them? | 2020-09-21T21:14:18 | 2020-09-21T21:14:18 | {} | MEMBER |
ClickHouse | ClickHouse | 696,494,232 | 15,116 | alexey-milovidov | It's better to use `lld` for linking (in case if you are using `gold` linker).
If it's IO bound, you can switch to shared linking, example: https://github.com/ClickHouse/ClickHouse/pull/14711/files#diff-dd7c954046dc6b55d3bc37bc642e2bbeR3 | 2020-09-22T03:49:40 | 2020-09-22T03:49:40 | {} | MEMBER |
ClickHouse | ClickHouse | 696,114,902 | 15,087 | fastio | In production environment, hundreds of machines hold thousands of ReplicatedMergeTree tables.
The only zookeeper cluster is seriously overloaded even good performance machine used.
In some scenarios, we want to change the current zookeeper cluster to another one smoothly. | 2020-09-21T13:28:47 | 2020-09-21T13:28:47 | {
"+1": 7
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,137,521 | 15,082 | KochetovNicolai | It has decreased build time for `GatherUtils` from 3.5 to 2.5 min for me. | 2020-09-21T14:07:01 | 2020-09-21T14:07:20 | {} | MEMBER |
ClickHouse | ClickHouse | 696,279,463 | 15,027 | alexey-milovidov | @akuzm If profile data does not work alright please make another PR.
At least now we can compare clang vs. gcc in fair way. | 2020-09-21T18:07:07 | 2020-09-21T18:07:07 | {} | MEMBER |
ClickHouse | ClickHouse | 696,353,560 | 15,110 | nikitamikhaylov | A lot of copy-paste (I tried to reduce it as much as possible), but now all sources are isolated. | 2020-09-21T20:24:27 | 2020-09-21T20:25:49 | {} | MEMBER |
ClickHouse | ClickHouse | 696,494,522 | 15,116 | alexey-milovidov | Also there is usually no need to link all targets. If you only need clickhouse (server, client), write
```
ninja clickhouse
``` | 2020-09-22T03:51:13 | 2020-09-22T03:51:13 | {} | MEMBER |
ClickHouse | ClickHouse | 696,631,449 | 15,088 | 4ertus2 | test_s3_with_proxy/test.py::test_s3_with_proxy_list[s3] flap in integration tests
perf errors are not related to JOIN or LC too. | 2020-09-22T10:15:27 | 2020-09-22T10:15:27 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 697,760,675 | 12,567 | alexey-milovidov | It's intended to pass arbitrary query as is to several remote servers via composition of `remote`/`cluster` and `view` functions. | 2020-09-23T17:36:39 | 2020-09-23T17:36:39 | {} | MEMBER |
ClickHouse | ClickHouse | 698,354,194 | 15,233 | durchgedreht | That's without the fact already ;-) | 2020-09-24T13:45:52 | 2020-09-24T13:45:52 | {} | NONE |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.