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 | 695,365,296 | 11,913 | alexey-milovidov | @CurtizJ I suspect that this PR has broke integration tests:
https://github.com/ClickHouse/ClickHouse/commit/eb9ee723c37b0c616cd7d81a82982adb0fd94499 (look at this commit in master) | 2020-09-19T23:06:37 | 2020-09-19T23:06:37 | {} | MEMBER |
ClickHouse | ClickHouse | 695,543,418 | 14,979 | liangli1213 | @BohuTANG , as you mentioned Materialized View, and I have a requirement, to count the unique user name on daily basis for using a product, the usage data is generated randomly by time series, as per the description of Materialized View table, it only process increment data, so Materialized View seems not meet my requi... | 2020-09-20T02:39:04 | 2020-09-20T02:39:04 | {} | NONE |
ClickHouse | ClickHouse | 695,803,603 | 14,731 | alexey-milovidov | > Yandex third-party checks (only for Yandex employees)
Need manual editing of third-party code, I will do... | 2020-09-20T16:01:26 | 2020-09-20T16:01:26 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 695,805,088 | 15,027 | alexey-milovidov | @akuzm Could you please validate if new profile runs are collected to flame graphs? | 2020-09-20T16:13:49 | 2020-09-20T16:13:49 | {} | MEMBER |
ClickHouse | ClickHouse | 695,830,426 | 14,508 | alexey-milovidov | The setting allow to *keep nullable* if the data type is already nullable. | 2020-09-20T20:10:22 | 2020-09-20T20:10:22 | {} | MEMBER |
ClickHouse | ClickHouse | 695,836,617 | 14,791 | filimonov | Using {database} and {table} macros in ZooKeeper path lead to serious issues with table renaming.
How it will distinguish the situation "new table" vs "joining replication of existing table" for atomic case?
If we will start adding some uniqueness automatically for non Atomic database - the same problem will po... | 2020-09-20T21:08:20 | 2020-09-20T21:08:20 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 695,999,884 | 14,735 | DBA-Ray | Is the time data type not included in the plan? | 2020-09-21T09:21:05 | 2020-09-21T09:21:05 | {} | NONE |
ClickHouse | ClickHouse | 696,046,537 | 14,266 | keerthana-kannan-ch | In addition to my previous observation, if I change the rollup from max() to avg(), this is the result I get:
```
clickhouse-server1 :) select timestamp, avg(requests) from derived_metrics group by timestamp order by timestamp;
SELECT
timestamp,
avg(requests)
FROM derived_metrics
GROUP BY timestamp
... | 2020-09-21T11:06:20 | 2020-09-21T11:09:04 | {} | NONE |
ClickHouse | ClickHouse | 695,219,858 | 7,774 | alexey-milovidov | This is viable feature - to support queries with parameter substitution of identifiers, like
```
SELECT * FROM {table:Identifier}
```
But are you going to continue or we can close this? | 2020-09-19T14:26:43 | 2020-09-19T14:26:43 | {} | MEMBER |
ClickHouse | ClickHouse | 695,343,246 | 14,627 | alexey-milovidov | Another case: https://clickhouse-test-reports.s3.yandex.net/14962/194c8da69a302f5cff694ea22d84cacc1db53fa3/fuzzer/report.html#fail1
| 2020-09-19T18:44:43 | 2020-09-19T18:44:43 | {} | MEMBER |
ClickHouse | ClickHouse | 695,838,089 | 9,621 | alexey-milovidov | @CurtizJ as 19.14 no longer supported, we can remove compatibility code. | 2020-09-20T21:21:35 | 2020-09-20T21:21:46 | {} | MEMBER |
ClickHouse | ClickHouse | 695,861,643 | 14,979 | bohutang | Will the user name be deleted or updated? | 2020-09-21T01:00:20 | 2020-09-21T01:00:20 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,071,343 | 15,075 | handgunman | c настройкой join_use_nulls = 1 к CH подключается в частности DataLens
Речь не идет о хранении, речь идет о том, что при создании view поверх селекта CH по умолчанию делает ненулевым поля которые по логике должны быть nullable.
когда в поле могут быть любые допустимые значения, без nullable не обойтись. | 2020-09-21T12:05:57 | 2020-09-21T12:05:57 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,079,043 | 15,075 | handgunman | select test.id, test.val, test2.val2
from test
left join test2 using id
подключаюсь с настройкой join_use_nulls = 1
выполняю запрос - работает
создаю view
create or replace view tst as
select test.id, test.val, test2.val2
from test
left join test2 using id
делаю
select * from tst
Code: 349. DB::E... | 2020-09-21T12:22:27 | 2020-09-21T12:22:27 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 695,299,868 | 14,886 | bharatnc | > Azat already started working on it: #15007
@alexey-milovidov I did start work on it, but looks like @azat has made more progress on this. You can probably assign this task to him? Thanks! | 2020-09-19T15:57:11 | 2020-09-19T15:57:11 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 695,364,654 | 15,026 | alexey-milovidov | Performance Ok. | 2020-09-19T22:59:19 | 2020-09-19T22:59:19 | {} | MEMBER |
ClickHouse | ClickHouse | 695,808,679 | 14,884 | alexey-milovidov | We don't guarantee the order of keys in the result of *Map functions. | 2020-09-20T16:47:18 | 2020-09-20T16:47:18 | {} | MEMBER |
ClickHouse | ClickHouse | 695,834,723 | 15,035 | azat | >Functional stateless tests (thread) — fail: 1, passed: 2270, skipped: 12
01502_jemalloc_percpu_arena | FAIL | 6.29
-- | -- | --
Does not looks related to this PR, but interesting, TSAN bails, should be fixed by #15061
| 2020-09-20T20:50:15 | 2020-09-20T20:50:15 | {} | MEMBER |
ClickHouse | ClickHouse | 696,066,303 | 15,081 | den-crane | https://github.com/ClickHouse/ClickHouse/blob/master/CHANGELOG.md#clickhouse-release-207
Deprecate special printing of zero Date/DateTime values as 0000-00-00 and 0000-00-00 00:00:00. #12442 (alexey-milovidov).
https://github.com/ClickHouse/ClickHouse/pull/12442 | 2020-09-21T11:55:03 | 2020-09-21T11:55:03 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 695,216,636 | 13,405 | alexey-milovidov | Simply don't use these methods `#if defined(ARCADIA_BUILD)` | 2020-09-19T13:56:51 | 2020-09-19T13:56:51 | {} | MEMBER |
ClickHouse | ClickHouse | 695,218,006 | 11,235 | alexey-milovidov | We can implement it again in the following way:
- the function with OrderBy combinator will have state like `groupArray` and collect values along with order by fields;
- only at the finalization step the collected array will be fed to the transformed aggregate function. | 2020-09-19T14:09:31 | 2020-09-19T14:10:04 | {} | MEMBER |
ClickHouse | ClickHouse | 695,791,346 | 15,007 | azat | Conflicts with #15032 (since it has few bits for converting to python3 too) | 2020-09-20T14:08:17 | 2020-09-20T14:08:17 | {} | MEMBER |
ClickHouse | ClickHouse | 695,840,579 | 12,713 | alexey-milovidov | But if you do it, the Buffer table will stuck because data will be in the buffer but cannot be inserted into the destination table.
Buffer table does not have the info about how many partitions will be created at the moment when you INSERT data into Buffer,
so there is no simple way to solve this issue.
| 2020-09-20T21:48:13 | 2020-09-20T21:48:13 | {
"-1": 1,
"confused": 2
} | MEMBER |
ClickHouse | ClickHouse | 695,841,708 | 11,092 | alexey-milovidov | @amosbird It should be very easy to fix - just don't optimize if we have `parallel_replicas_count` != 0. | 2020-09-20T22:00:15 | 2020-09-20T22:00:15 | {} | MEMBER |
ClickHouse | ClickHouse | 695,862,316 | 14,829 | bohutang | Hello @vegastar002
We've fixed it in this patch https://github.com/ClickHouse/ClickHouse/pull/14852
You can use the latest master build and check it again:
https://clickhouse-builds.s3.yandex.net/0/3e0509af691d715747c8f6e6eb8e0f3e30942fbf/clickhouse_build_check/report.html | 2020-09-21T01:04:29 | 2020-09-21T01:04:29 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 695,863,331 | 10,473 | alexey-milovidov | @huang--yijian You can use `clickhouse-local` to generate MergeTree partitions.
Just CREATE TABLE and do INSERT SELECT.
The data will appear in /tmp directory,
alternatively you can specify path with `-- --path` parameter in command line
or provide a config file similar to server's config.xml | 2020-09-21T01:11:08 | 2020-09-21T01:11:08 | {} | MEMBER |
ClickHouse | ClickHouse | 696,075,792 | 15,075 | den-crane | >Речь не идет о хранении, речь идет о том, что при создании view поверх
> селекта CH по умолчанию делает ненулевым поля которые по логике должны быть nullable.
я ничего не понял вообще. Пример есть?
>когда в поле могут быть любые допустимые значения, без nullable не обойтись.
это конечно так.
! **Но у тысяч... | 2020-09-21T12:15:41 | 2020-09-21T12:20:40 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 695,320,386 | 13,617 | alexey-milovidov | @akuzm Simply recompiling libunwind with gcc does not help.
It happens because clang generates different frame info (that appeared to be more difficult to parse).
| 2020-09-19T16:14:07 | 2020-09-19T16:14:07 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 695,840,704 | 10,882 | alexey-milovidov | This feature is hardly ever needed. | 2020-09-20T21:49:41 | 2020-09-20T21:49:41 | {} | MEMBER |
ClickHouse | ClickHouse | 695,791,058 | 14,731 | azat | @alexey-milovidov can you take a look? (Maybe there is some public team, with alias, i.e. `@maintainers` that is preferable over personal ping?) | 2020-09-20T14:05:24 | 2020-09-20T14:05:24 | {} | MEMBER |
ClickHouse | ClickHouse | 696,045,024 | 15,069 | alesapin | Wait for integration tests with asan. | 2020-09-21T11:02:49 | 2020-09-21T11:02:49 | {} | MEMBER |
ClickHouse | ClickHouse | 696,087,763 | 15,083 | den-crane | ожидаемое поведение. Или создавайте view при join_use_nulls = 1 либо используйте join_use_nulls = 1 в settings запроса.
```sql
create or replace view tst as select test.id, test.val, test2.val2
from (select 1 id, 1 val) test
left join (select 2 id, 2 val2) test2 using id settings join_use_nulls = 1;
``` | 2020-09-21T12:41:01 | 2020-09-21T12:41:01 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 695,219,740 | 14,492 | vitlibar | Applied fix of testflows tests https://gist.github.com/vzakaznikov/80cd5715d2d59b76cbfce7432481a03f | 2020-09-19T14:25:38 | 2020-09-19T14:25:38 | {} | MEMBER |
ClickHouse | ClickHouse | 695,268,087 | 13,617 | alexey-milovidov | ```
clickhouse-benchmark <<< "SELECT ignore(SHA256(toString(1000000000+number))) FROM numbers_mt(10000000) FORMAT Null"
```
gcc: 0.050 sec.
clang: 0.052 sec.
```
clickhouse-benchmark --query_profiler_real_time_period_ns 10000000 <<< "SELECT ignore(SHA256(toString(1000000000+number))) FROM numbers_mt(10000000)... | 2020-09-19T15:31:32 | 2020-09-19T15:31:32 | {} | MEMBER |
ClickHouse | ClickHouse | 695,349,576 | 15,027 | alexey-milovidov | ```
Removing intermediate container bdbfff847862
Step 13/14 : COPY build.sh /
COPY failed: stat /place/sandbox-data/tasks/8/4/776403748/docker_root/tmp/docker-builder438288327/build.sh: no such file or directory
``` | 2020-09-19T19:51:19 | 2020-09-19T19:51:19 | {} | MEMBER |
ClickHouse | ClickHouse | 695,769,376 | 13,334 | ghost | @akuzm
I will fix these codes as soon as possible. Thank you. | 2020-09-20T10:12:30 | 2020-09-20T10:12:30 | {} | NONE |
ClickHouse | ClickHouse | 695,839,028 | 7,536 | alexey-milovidov | We cannot start using new ZooKeeper connection without full reinitialization of the state in memory (it is what RestartingThread is doing). Whenever ZooKeeper session has expired we should assume that our knowledge about anything from ZooKeeper is obsolete.
Maybe in some cases it isn't required but it's difficult to... | 2020-09-20T21:31:17 | 2020-09-20T21:31:17 | {} | MEMBER |
ClickHouse | ClickHouse | 695,844,936 | 14,791 | alexey-milovidov | Maybe forbid renaming (throw exception) if these macros are used? | 2020-09-20T22:35:40 | 2020-09-20T22:35:40 | {} | MEMBER |
ClickHouse | ClickHouse | 695,956,294 | 15,061 | alesapin | `> llvm-symbolizer does not work in the test env`
Fixing in https://github.com/ClickHouse/ClickHouse/pull/15069 | 2020-09-21T07:44:20 | 2020-09-21T07:44:20 | {} | MEMBER |
ClickHouse | ClickHouse | 695,994,172 | 14,791 | tavplubix | I think it's ok to forbid renaming. Also we can unfold such macros (#6917), but it doesn't solve the root case - table name can become inconsistent with path in ZooKeeper and it may cause unexpected behaviour:
```
CREATE TABLE t1 ... ENGINE=ReplicatedMergeTree('/ch/tables/{table}', 'r1') ... -- Unfolded to '/ch/tabl... | 2020-09-21T09:09:00 | 2020-09-21T09:09:00 | {} | MEMBER |
ClickHouse | ClickHouse | 696,069,373 | 15,075 | den-crane | 1. Изначально CH не умел null в join. Полноценная поддержка null в join полностью совместимая с другими СУБД и ANSI стандартом появилась около года назад. Большинство продакшин систем у пользователей работают (и моя тоже) работают в режиме join_use_nulls = 0 , если установить join_use_nulls = 1 все просто сломается п... | 2020-09-21T12:01:20 | 2020-09-21T12:01:20 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 695,758,598 | 12,709 | citrin | > You can edit the file `ClickHouse/src/Functions/CRC.cpp` and disable CRC function if you build with system zlib.
Faced the same problem, what exactly should be changed in `CRC.cpp`? | 2020-09-20T08:07:25 | 2020-09-20T08:07:25 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 695,808,007 | 15,034 | alexey-milovidov | > 00463_long_sessions_in_http_interface
Flaky test :(
> test_s3_with_proxy
Flaky test.
> test_rename_column
- recently broken, maybe here: #11913 | 2020-09-20T16:41:05 | 2020-09-20T16:41:05 | {} | MEMBER |
ClickHouse | ClickHouse | 694,161,551 | 14,925 | alexey-milovidov | How to pass it to every compiler invocation?
Do we need a shell script wrapper
```
#!/bin/sh
ulimit -m 10000000 -t 600
clang $@
```
or there is something more simple?
Upd: it won't work. | 2020-09-17T11:09:03 | 2020-09-20T20:31:34 | {} | MEMBER |
ClickHouse | ClickHouse | 696,053,600 | 14,685 | akuzm | > Now we have only 176 unstable queries (was 418), so maybe it has helped
> (or other recent attempts to lower binary size have helped).
I think it was this one https://github.com/ClickHouse/ClickHouse/pull/14883
(I increased instability thresholds for 30 tests, based on historical data) | 2020-09-21T11:22:53 | 2020-09-21T11:22:53 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 695,789,777 | 14,957 | azat | @alexey-milovidov BTW if we will ignore that fact that the fix was incorrect, why it was not backported? AFAIU this can be triggered on 20.5+ (and in release build it will not even hit assert) | 2020-09-20T13:52:20 | 2020-09-20T13:52:20 | {} | MEMBER |
ClickHouse | ClickHouse | 695,838,267 | 4,476 | alexey-milovidov | 100% fixed. | 2020-09-20T21:23:06 | 2020-09-20T21:23:06 | {} | MEMBER |
ClickHouse | ClickHouse | 695,842,016 | 10,524 | alexey-milovidov | Does it reproduce on version 20.8? | 2020-09-20T22:03:32 | 2020-09-20T22:03:32 | {} | MEMBER |
ClickHouse | ClickHouse | 695,871,563 | 14,979 | liangli1213 | There is always new record coming into table, but the username is unpredictable, it maybe appeared in same day, or the first time to appear today, the result I need is the unique username's count on daily basis. | 2020-09-21T02:05:26 | 2020-09-21T02:05:26 | {} | NONE |
ClickHouse | ClickHouse | 695,880,633 | 14,735 | DBA-Ray | Hello! I tested the version 20.10.1.4709 (official build),and i found the time data type is not supported.
2020.09.21 10:57:16.373628 [ 19505 ] {} <Error> MaterializeMySQLSyncThread: Code: 50, e.displayText() = DB::Exception: Unknown data type family: time, Stack trace (when copying this message, always include the ... | 2020-09-21T03:02:00 | 2020-09-21T03:02:00 | {} | NONE |
ClickHouse | ClickHouse | 695,976,470 | 15,007 | azat | >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 | 2020-09-21T08:29:46 | 2020-09-21T08:29:46 | {} | MEMBER |
ClickHouse | ClickHouse | 696,040,833 | 15,065 | filimonov | This one seems nice / not too small / isolated: https://github.com/ClickHouse/ClickHouse/issues/4420 | 2020-09-21T10:53:45 | 2020-09-21T10:53:45 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,043,935 | 15,069 | alesapin | Good news: it works https://clickhouse-test-reports.s3.yandex.net/15069/1bd76c820c035a7802b8448d0dd495469500e673/stress_test_(thread)/stderr.log
Bad news: it's a real race condition.
| 2020-09-21T11:00:23 | 2020-09-21T11:00:23 | {} | MEMBER |
ClickHouse | ClickHouse | 696,082,482 | 15,075 | den-crane | это ожидаемое поведение view. Используйте settings join_use_nulls = 1
create or replace view tst as select test.id, test.val, test2.val2
from (select 1 id, 1 val) test
left join (select 2 id, 2 val2) test2 using id settings join_use_nulls = 1; | 2020-09-21T12:29:44 | 2020-09-21T12:29:44 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 695,337,507 | 14,227 | den-crane | Fixed in https://github.com/ClickHouse/ClickHouse/pull/14646 | 2020-09-19T17:49:59 | 2020-09-19T17:49:59 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 695,808,181 | 15,034 | alexey-milovidov | Performance test has successfully confirmed the speedup. | 2020-09-20T16:42:35 | 2020-09-20T16:42:35 | {} | MEMBER |
ClickHouse | ClickHouse | 695,836,946 | 4,901 | filimonov | @sxpsxp12 the issue was incorrect, bahaviour was caused by misconfiguration. | 2020-09-20T21:11:36 | 2020-09-20T21:11:36 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 695,967,310 | 13,574 | alesapin | Oh sorry, I didn't notice when I've closed this issue. Just accident missclick. | 2020-09-21T08:10:11 | 2020-09-21T08:10:38 | {} | MEMBER |
ClickHouse | ClickHouse | 696,081,455 | 15,075 | handgunman | если у них нет null то ничего и не сломается. их же нет :)
у запроса и у представления не должно быть разного поведения, но запрос работает, спокойно возвращает null, а представление нет. | 2020-09-21T12:27:33 | 2020-09-21T12:27:33 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 695,362,522 | 14,265 | ilejn | The error
_Pulling kafka_kerberos (yandex/clickhouse-kerberos-kdc:latest)...
manifest for yandex/clickhouse-kerberos-kdc:latest not found_
The docker image originates from https://github.com/arenadata/docker_kdc | 2020-09-19T22:29:11 | 2020-09-19T22:29:11 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 695,760,899 | 12,709 | citrin | I've just changed `crc32_z` to `crc32` - should fix the build with zlib < 1.2.9. It is highly unlikely that in our setup it would be used for strings >= 4Gb. | 2020-09-20T08:35:09 | 2020-09-20T08:35:09 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 695,838,388 | 15,007 | abyss7 | Great effort! Looking forward to merge this PR when it's complete. | 2020-09-20T21:24:18 | 2020-09-20T21:24:18 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 695,838,523 | 15,007 | abyss7 | > Conflicts with #15032 (since it has few bits for converting to python3 too)
BTW, my changes apply to pytest scripts which are unused for now in CI. | 2020-09-20T21:25:44 | 2020-09-20T21:25:44 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 695,842,145 | 10,401 | alexey-milovidov | @4ertus2 The issue is still relevant. | 2020-09-20T22:04:58 | 2020-09-20T22:04:58 | {} | MEMBER |
ClickHouse | ClickHouse | 695,845,295 | 14,977 | ghost | Failed test-cases doesn't seem to be related to this PR.
If there is any other problem, please let me know.
* Failed test-cases
- test_rename_column/test.py::test_rename_with_parallel_insert
- test_rename_column/test.py::test_rename_distributed_parallel_insert_and_select
| 2020-09-20T22:39:20 | 2020-09-21T03:12:42 | {} | NONE |
ClickHouse | ClickHouse | 696,070,065 | 15,081 | simPod | I wonder what is the reason for the change 🤔 | 2020-09-21T12:03:03 | 2020-09-21T12:03:03 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,076,977 | 14,265 | ilejn | @qoega , could you assist in placing the image yandex/clickhouse-kerberos-kdc and/or let me know what I've done wrong. | 2020-09-21T12:18:19 | 2020-09-21T12:18:19 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 695,336,866 | 14,606 | den-crane | Fixed in https://github.com/ClickHouse/ClickHouse/pull/14646 | 2020-09-19T17:43:54 | 2020-09-19T17:43:54 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 695,341,934 | 15,025 | alexey-milovidov | Wait only builds. | 2020-09-19T18:31:15 | 2020-09-19T18:31:15 | {} | MEMBER |
ClickHouse | ClickHouse | 695,803,719 | 15,031 | alexey-milovidov | > test_rename_column
- recently broken, maybe here: #11913 | 2020-09-20T16:02:26 | 2020-09-20T16:02:26 | {} | MEMBER |
ClickHouse | ClickHouse | 695,838,702 | 14,519 | alexey-milovidov | Fixed in #14685 | 2020-09-20T21:27:39 | 2020-09-20T21:27:39 | {} | MEMBER |
ClickHouse | ClickHouse | 695,841,115 | 11,383 | alexey-milovidov | We should not have a dedicated format to parse our text logs because the format for our text logs is not well specified an there is no guarantee about future changes.
Format `Regexp` should do the task. | 2020-09-20T21:54:06 | 2020-09-20T21:54:23 | {} | MEMBER |
ClickHouse | ClickHouse | 695,343,262 | 14,962 | alexey-milovidov | AST Fuzzer: https://github.com/ClickHouse/ClickHouse/issues/14627 | 2020-09-19T18:44:54 | 2020-09-19T18:44:54 | {} | MEMBER |
ClickHouse | ClickHouse | 695,830,284 | 14,838 | alexey-milovidov | As long as `Docs check passed` it is perfect :) | 2020-09-20T20:08:50 | 2020-09-20T20:08:50 | {} | MEMBER |
ClickHouse | ClickHouse | 695,840,308 | 14,569 | alexey-milovidov | ```
Column 26, name: dt, type: Date, parsed text: "2020-08-2"
ERROR: garbage after Date: "2230600"
ERROR: Date must be in YYYY-MM-DD format.
```
So, `2020-08-2` is not `YYYY-MM-DD`. This is most likely client-side issue. | 2020-09-20T21:45:04 | 2020-09-20T21:45:04 | {} | MEMBER |
ClickHouse | ClickHouse | 695,963,083 | 15,061 | azat | > > llvm-symbolizer does not work in the test env
>
>Fixing in #15069
@alesapin the problem there is not in version, since llvm-symbolizer 10 works fine for binaries compiled with llvm/clang 11, the problem is that this was clickhouse-local which does not read /etc/environment, see #15063 | 2020-09-21T08:00:22 | 2020-09-21T08:01:00 | {} | MEMBER |
ClickHouse | ClickHouse | 696,009,327 | 14,863 | qinglok | **As for the binary file of native mode, I would like to ask: what is the meaning of the first three bytes?**
**The original record here:**
_id name age borthday
1 linx 28 1993-07-01 00:00:00
2 jens 18 1994-07-01 00:00:00
3 rudy 13 1995-07-01 00:00:00_
**This i... | 2020-09-21T09:42:25 | 2020-09-21T09:42:25 | {} | NONE |
ClickHouse | ClickHouse | 696,038,203 | 8,089 | filimonov | Relates https://github.com/ClickHouse/ClickHouse/issues/13826 | 2020-09-21T10:47:08 | 2020-09-21T10:47:08 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 696,061,508 | 15,061 | alesapin | > > > llvm-symbolizer does not work in the test env
> >
> >
> > Fixing in #15069
>
> @alesapin the problem there is not in version, since llvm-symbolizer 10 works fine for binaries compiled with llvm/clang 11, the problem is that this was clickhouse-local which does not read /etc/environment, see #15063
Oh O... | 2020-09-21T11:42:23 | 2020-09-21T11:42:23 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 695,769,142 | 14,977 | ghost | @alexey-milovidov
Thank you for your comment.
I fixed some codes. It will work properly for the case of `indexOf(arr, x) = 0`. | 2020-09-20T10:09:51 | 2020-09-20T10:09:51 | {} | NONE |
ClickHouse | ClickHouse | 695,789,368 | 14,920 | flyfoxCI | > What version do you use?
My clickhouse version is 19.16.19.85
> Startup can take significant amount of time if you have too many tables or too many data parts inside your tables. It may happen if you partition data too much. The reason why startup takes a time will be clearly visible in server logs.
>
> In... | 2020-09-20T13:48:38 | 2020-09-20T13:48:38 | {} | NONE |
ClickHouse | ClickHouse | 695,803,535 | 14,731 | alexey-milovidov | I will take a look.
PS. All maintainers are supposedly subscribed to events on GitHub. | 2020-09-20T16:00:58 | 2020-09-20T16:00:58 | {} | MEMBER |
ClickHouse | ClickHouse | 695,803,913 | 15,028 | alexey-milovidov | > test_rename_column
- recently broken, maybe here: #11913 | 2020-09-20T16:04:08 | 2020-09-20T16:04:08 | {} | MEMBER |
ClickHouse | ClickHouse | 695,832,693 | 14,925 | alexey-milovidov | `prlimit --rss=10000000 --cpu=600` | 2020-09-20T20:32:04 | 2020-09-20T20:32:04 | {} | MEMBER |
ClickHouse | ClickHouse | 695,832,812 | 14,925 | alexey-milovidov | ```
milovidov@milovidov-desktop:~/work/ClickHouse/utils$ time prlimit --rss=10000000 --cpu=1 clickhouse-local --query "SELECT count() FROM system.numbers"
Killed
real 0m0,999s
user 0m0,869s
sys 0m0,137s
milovidov@milovidov-desktop:~/work/ClickHouse/utils$ time prlimit --rss=10000000 --cpu=1 clickhouse... | 2020-09-20T20:33:12 | 2020-09-20T20:33:12 | {} | MEMBER |
ClickHouse | ClickHouse | 695,862,963 | 14,735 | bohutang | Update:
decimal data type patch has been merged to master, you can download and experience it here:
https://clickhouse-builds.s3.yandex.net/0/3e0509af691d715747c8f6e6eb8e0f3e30942fbf/clickhouse_build_check/report.html
| 2020-09-21T01:09:00 | 2020-09-21T01:09:00 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 695,869,576 | 14,735 | DBA-Ray | Thanks a lot.
发送自 Windows 10 版邮件<https://go.microsoft.com/fwlink/?LinkId=550986>应用
发件人: BohuTANG<mailto:notifications@github.com>
发送时间: 2020年9月21日 9:09
收件人: ClickHouse/ClickHouse<mailto:ClickHouse@noreply.github.com>
抄送: Ray<mailto:rulei0501@hotmail.com>; Author<mailto:author@noreply.github.com>
主题: Re: [Clic... | 2020-09-21T01:51:57 | 2020-09-21T01:51:57 | {} | NONE |
ClickHouse | ClickHouse | 695,889,159 | 14,735 | zhang2014 | Time date type is unsupported. | 2020-09-21T03:54:32 | 2020-09-21T03:54:32 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 695,220,029 | 6,185 | alexey-milovidov | This feature still has high demand and going to be continued by @KochetovNicolai | 2020-09-19T14:28:15 | 2020-09-19T14:28:15 | {} | MEMBER |
ClickHouse | ClickHouse | 695,220,127 | 5,629 | alexey-milovidov | @4ertus2 Is it doable? | 2020-09-19T14:29:22 | 2020-09-19T14:29:22 | {
"eyes": 2
} | MEMBER |
ClickHouse | ClickHouse | 695,269,508 | 13,617 | alexey-milovidov | Proposed solution: always compare queries with disabled profiler.
Only if performance differs - run them with profiler to collect profile data. | 2020-09-19T15:32:40 | 2020-09-19T15:32:40 | {} | MEMBER |
ClickHouse | ClickHouse | 696,070,465 | 15,075 | den-crane | >Сейчас, при запросе к представлениям, созданным без явной настройки, возникает ошибка
>DB::Exception: Cannot convert NULL value to non-Nullable type
это похоже на bug , откройте новое issue с описанием и воспроизведением. | 2020-09-21T12:03:54 | 2020-09-21T12:03:54 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 695,803,811 | 15,030 | alexey-milovidov | > test_rename_column
- recently broken, maybe here: #11913 | 2020-09-20T16:03:14 | 2020-09-20T16:03:14 | {} | MEMBER |
ClickHouse | ClickHouse | 695,809,031 | 15,033 | alexey-milovidov | > test_rename_column
- recently broken, maybe here: #11913
> test_s3_with_proxy
- flaky test | 2020-09-20T16:50:46 | 2020-09-20T16:50:46 | {} | MEMBER |
ClickHouse | ClickHouse | 695,830,459 | 14,508 | alexey-milovidov | Please make some experiments and edit the text... | 2020-09-20T20:10:38 | 2020-09-20T20:10:38 | {} | MEMBER |
ClickHouse | ClickHouse | 695,834,623 | 15,060 | alexey-milovidov | We will try to lower to 5 minutes and 5 GB as soon as this PR will be merged. | 2020-09-20T20:49:19 | 2020-09-20T20:49:19 | {} | MEMBER |
ClickHouse | ClickHouse | 695,858,748 | 15,060 | alexey-milovidov | It does not pass already:
has_all.cpp - Killed
has_any.cpp - Killed
has_substr.cpp - Killed | 2020-09-21T00:39:58 | 2020-09-21T00:39:58 | {} | MEMBER |
ClickHouse | ClickHouse | 696,085,831 | 15,075 | den-crane | >если у них нет null то ничего и не сломается. их же нет :)
SQL запросы написаны таким образом что они ожидают 0 и '' , а не Null
```sql
select test.id, test.val, test2.val2, renamed
from (select 1 id, 1 val) test
left join (select 2 id, 2 val2, val2 renamed) test2 using id
where renamed = 0;
┌─id─┬─val─┬─... | 2020-09-21T12:36:33 | 2020-09-21T12:36:55 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 691,083,059 | 14,737 | ghuname | Hi @alesapin,
1. thanks for your answer. I haven't noticed that show create table shows merge_with_ttl_timeout after altering with modify setting. Thanks.
2. Regarding my test case, I am doing the following:
a) I have created a table something like:
CREATE TABLE database.table
(
...
`ipdrCreationT... | 2020-09-11T13:05:26 | 2020-09-11T13:35:08 | {} | NONE |
ClickHouse | ClickHouse | 691,156,592 | 14,697 | 4ertus2 | ya chek if broken by other test: 01473_event_time_microseconds | 2020-09-11T15:17:33 | 2020-09-11T15:17:33 | {} | CONTRIBUTOR |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.