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 | 662,481,188 | 11,222 | den-crane | SELECT toDateTime64('2019-05-05 20:20:12.050', 3) > today() -- 1
SELECT toDateTime('2019-05-05 20:20:12') > today() -- 0 | 2020-07-22T14:19:25 | 2020-07-22T14:19:25 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 661,820,572 | 12,542 | 4ertus2 | It's not enough to implement `cloneResized()` to get some reasonable result for join with aggregate func column. First of all it's not clear what is reasonable result for such a join. | 2020-07-21T12:17:22 | 2020-07-21T12:17:22 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 661,899,792 | 11,933 | keliss | Got it, thank you. | 2020-07-21T14:35:49 | 2020-07-21T14:35:49 | {} | NONE |
ClickHouse | ClickHouse | 662,039,104 | 12,642 | vzakaznikov | @vitlibar, could you please review this PR? This PR includes syntax tests for RBAC. Thanks! | 2020-07-21T18:42:37 | 2020-07-21T18:42:37 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 662,401,289 | 12,642 | vitlibar | > @vitlibar, could you please review this PR? This PR includes syntax tests for RBAC. Thanks!
Yes, I'll look at it carefully. | 2020-07-22T11:30:25 | 2020-07-22T11:30:25 | {} | MEMBER |
ClickHouse | ClickHouse | 662,454,452 | 12,653 | zhang2014 | test failure: 00993_system_parts_race_condition_drop_zookeeper
is not related to this patch | 2020-07-22T13:31:00 | 2020-07-22T13:31:00 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 662,471,439 | 12,667 | den-crane | >Table is in readonly mode.
ZK issue at one of shards.
check `select database,table from system.replicas where is_readonly;` at all nodes.
check /var/log/clickhouse-server/clickhouse-server.err.log /var/log/clickhouse-server/clickhouse-server.log for ZK errors / messages about table is readonly | 2020-07-22T14:02:14 | 2020-07-22T14:02:14 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 662,504,669 | 11,501 | den-crane | > It would be nice if ClickHouse would set a "to be deleted" flag on rows that are awaiting deletion, and those would be transparently skipped by all SELECT queries (time of execution for alter delete would not be instant, but rather the time needed to set the flag).
>
> Do we know if this was ever considered?
It... | 2020-07-22T15:00:44 | 2020-07-22T15:02:09 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 661,741,196 | 12,596 | sundy-li | `windowFunnel` will return approximated results when your `event_date_time` has same value for same `e_device_id` | 2020-07-21T09:20:09 | 2020-07-21T09:20:09 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 662,311,100 | 12,488 | azat | Ok, will take a look | 2020-07-22T08:10:57 | 2020-07-22T08:10:57 | {} | MEMBER |
ClickHouse | ClickHouse | 662,453,968 | 12,652 | alexey-milovidov | > Yandex third-party checks (only for Yandex employees)
Temporarily broken by Yandex employees: `sandbox/projects/clickhouse/Backport`
| 2020-07-22T13:30:08 | 2020-07-22T13:30:08 | {} | MEMBER |
ClickHouse | ClickHouse | 662,509,299 | 12,079 | den-crane | CH does have such features and it against current design -- share nothing. | 2020-07-22T15:08:41 | 2020-07-22T15:08:41 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 662,548,203 | 8,331 | Vxider | @akuzm Thanks for the review, I have added check before `Field::get` and `static_cast`. The code seems to pass the AST-based query fuzzer now. | 2020-07-22T16:18:05 | 2020-07-22T16:18:05 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 661,860,215 | 11,495 | vzakaznikov | @KochetovNicolai, would you be able to add a `pr-bugfix` label to this PR? | 2020-07-21T13:26:11 | 2020-07-21T13:26:11 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 662,720,539 | 11,883 | excitoon | @alesapin please continue reviewing it in #12670 | 2020-07-22T22:03:54 | 2020-07-22T22:03:54 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 661,726,143 | 12,571 | 4ertus2 | Thanx.
Summary:
Right table in memory: 31.05 GiB. Not very good but expected (10 mil rows * 2kb = raw data 20 GB)
Additional memory usage even with with `max_threads = 1`: more than 60 GiB. And it grows. Not expected.
Summary 2:
Probably we need lazy load for big strings (some kind of BLOB logic). | 2020-07-21T08:51:50 | 2020-07-21T09:03:29 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 661,777,354 | 12,598 | 4ertus2 | > Is it expected that the setting does not affect the queries with `length(URL)`?
It should affect this part of query `? 'LONG' : 'SHORT'` | 2020-07-21T10:34:44 | 2020-07-21T10:34:44 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 661,961,727 | 12,317 | den-crane | > > Network timeout happens before optimise finishes. Check timeout settings both on server and client
>
> Is there any other way to solve this problem?
No.
I.e. `optimize table` runs for 3600 seconds. During execution it sends nothing to a client. JDBC driver considers such calls like timed out because they ex... | 2020-07-21T16:23:02 | 2020-07-21T18:30:50 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 662,454,948 | 12,652 | alexey-milovidov | > Performance — Errors while building the report.
The script is broken by this PR but it means that the script for performance test is not wise. | 2020-07-22T13:31:55 | 2020-07-22T13:31:55 | {} | MEMBER |
ClickHouse | ClickHouse | 662,596,172 | 12,583 | azat | @den-crane thanks for the reproducer!
>http+input_format_parallel_parsing=0
Yes, the problem is in `input_format_parallel_parsing` (`=1`)
It is reproduced only via HTTP, since `clickhouse-client` is too smart and do not use `input_format_parallel_parsing` | 2020-07-22T17:51:44 | 2020-07-22T17:51:44 | {} | MEMBER |
ClickHouse | ClickHouse | 661,734,505 | 12,540 | KochetovNicolai | Here is a test:
```
SET max_bytes_before_external_sort = 33554432;
set max_block_size = 1048576;
SELECT number FROM (SELECT number FROM numbers(2097152)) ORDER BY number * 1234567890123456789 LIMIT 2097142, 10;
Code: 49. DB::Exception: Received from localhost:9000. DB::Exception: Pipeline stuck. Current state:
... | 2020-07-21T09:07:37 | 2020-07-21T11:20:17 | {} | MEMBER |
ClickHouse | ClickHouse | 661,800,939 | 12,619 | alexey-milovidov | @blinkov The text was outstandingly incorrect and ignorant. | 2020-07-21T11:31:14 | 2020-07-21T11:31:14 | {} | MEMBER |
ClickHouse | ClickHouse | 661,802,572 | 12,596 | Johnhz | > `windowFunnel` will return approximated results when your `event_date_time` has same value for same `e_device_id`
there may be `event_date_time` has same value for same `e_device_id` in business scenarios | 2020-07-21T11:35:14 | 2020-07-21T11:35:14 | {} | NONE |
ClickHouse | ClickHouse | 662,048,135 | 11,066 | alexey-milovidov | `SELECT if(number % 2 = 0, [toUUID('00000000-e1fe-11e9-bb8f-853d60c00749')], [toUUID('11111111-e1fe-11e9-bb8f-853d60c00749')]) FROM numbers(5)` | 2020-07-21T19:00:50 | 2020-07-21T19:00:50 | {} | MEMBER |
ClickHouse | ClickHouse | 662,317,951 | 12,596 | Johnhz | > You can set set `event_date_time ` to be UInt64 (microseconds or millseconds), windowFunnel support all UIntXX types.
ok,thks ,After the time type is changed to Uint64, the data is normal. Time is in milliseconds, and whether windowFunnel seconds should be multiplied by 1000. The current test result needs to be mu... | 2020-07-22T08:25:39 | 2020-07-22T08:25:39 | {} | NONE |
ClickHouse | ClickHouse | 662,453,096 | 12,635 | alexey-milovidov | > 00992_system_parts_race_condition_zookeeper
Already mentioned in CI fest. | 2020-07-22T13:28:34 | 2020-07-22T13:28:34 | {} | MEMBER |
ClickHouse | ClickHouse | 662,453,275 | 12,635 | alexey-milovidov | Performance test Ok. | 2020-07-22T13:28:51 | 2020-07-22T13:28:51 | {} | MEMBER |
ClickHouse | ClickHouse | 662,688,542 | 12,664 | CurtizJ | Seems that clone of ast is still needed. Previously there was the same problem.
https://github.com/ClickHouse/ClickHouse/pull/11529#discussion_r437420658 | 2020-07-22T20:48:34 | 2020-07-22T20:48:45 | {} | MEMBER |
ClickHouse | ClickHouse | 662,734,308 | 12,672 | azat | >Will rebase against pre #12418
This won't help I guess, since the PR is applied on top of the upstream/master, so first the original issue should be addressed | 2020-07-22T22:44:55 | 2020-07-22T22:44:55 | {} | MEMBER |
ClickHouse | ClickHouse | 661,873,320 | 12,628 | gane5hvarma | @alexey-milovidov looked into the logs. i was missing server.key file. it worked after adding these files server.crt, server.key and dhpharam.pem
| 2020-07-21T13:49:54 | 2020-07-21T13:49:54 | {} | NONE |
ClickHouse | ClickHouse | 662,314,022 | 11,326 | weiyongyuan | Afterwards,we had debug clickhouse with gdb when it happened this issue.The stacktrace shows that clickhouse has a deadlock when i turn on the system.text_log.at the 98th line of OwnSplitChannel.cpp
```
std::lock_guard<std::mutex> lock(text_log_mutex);
if (auto log = text_log.lock())
log->add(elem);
... | 2020-07-22T08:17:27 | 2020-07-22T08:17:51 | {} | NONE |
ClickHouse | ClickHouse | 662,472,842 | 12,577 | alexey-milovidov | There are different tables that can be implemented:
1. Current snapshot of the repository: information about every line of every file in master with full blame info.
Allows to answer questions like:
- list files with maximum number of authors;
- show me the oldest lines of code in the repository;
- show me the f... | 2020-07-22T14:04:36 | 2020-07-22T14:07:43 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 662,607,033 | 12,666 | alesapin | Strange build failure for freebsd:
```
In file included from ../src/Storages/MergeTree/MergeTreeMutationStatus.cpp:1:
/build/src/Storages/MergeTree/MergeTreeMutationStatus.h:17:5: error: unknown type name 'time_t'
time_t create_time = 0;
^
/build/src/Storages/MergeTree/MergeTreeMutationStatus.h:27:5: erro... | 2020-07-22T18:14:04 | 2020-07-22T18:14:04 | {} | MEMBER |
ClickHouse | ClickHouse | 661,801,851 | 12,598 | akuzm | > > Is it expected that the setting does not affect the queries with `length(URL)`?
>
> It should affect this part of query `? 'LONG' : 'SHORT'`
Looks like it doesn't -- the time is the same with and without setting. The other query speeds up three times. | 2020-07-21T11:33:23 | 2020-07-21T11:33:23 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 661,859,758 | 11,495 | vzakaznikov | This PR should be considered as a bugfix. Need to add the `bugfix` tag so that it can be automatically backported to 20.3. | 2020-07-21T13:25:16 | 2020-07-21T13:25:16 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 662,450,340 | 12,654 | den-crane | duplucate https://github.com/ClickHouse/ClickHouse/issues/12504 | 2020-07-22T13:23:25 | 2020-07-22T13:23:25 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 662,570,419 | 12,665 | den-crane | And probably it solved in 20.5.3.27 (just released)
| 2020-07-22T16:59:36 | 2020-07-22T16:59:36 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 662,679,532 | 12,664 | 4ertus2 | This query fails
```
WITH any(number * 1) AS x SELECT x FROM numbers(1);
``` | 2020-07-22T20:28:55 | 2020-07-22T20:28:55 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 661,758,251 | 12,577 | alexey-milovidov | It has very inefficient algorithm and the speed is about one month per day (for recent commits).
We will have to either leave it working for about a month or rewrite (to calculate "blame" incrementally). | 2020-07-21T09:53:39 | 2020-07-21T09:53:39 | {} | MEMBER |
ClickHouse | ClickHouse | 661,805,418 | 11,461 | alexey-milovidov | Need help from any developer with Mac OS. | 2020-07-21T11:42:20 | 2020-07-21T11:42:20 | {} | MEMBER |
ClickHouse | ClickHouse | 661,829,119 | 12,613 | KochetovNicolai | Yandex third-party checks is broken because of changes in interface. Shouldn't merge yet. | 2020-07-21T12:35:38 | 2020-07-21T12:35:38 | {} | MEMBER |
ClickHouse | ClickHouse | 661,839,934 | 12,628 | alexey-milovidov | @gane5hvarma If the server cannot start listening on port with TLS it will print explanation in it's log.
You can read it in `/var/log/clickhouse-server/clickhouse-server.log` | 2020-07-21T12:50:20 | 2020-07-21T12:50:20 | {} | MEMBER |
ClickHouse | ClickHouse | 661,882,806 | 12,586 | den-crane | duplicate https://github.com/ClickHouse/ClickHouse/issues/7677 | 2020-07-21T14:06:34 | 2020-07-21T14:06:34 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 661,893,268 | 12,623 | alesapin | @KochetovNicolai Found very suspicious line here https://github.com/ClickHouse/ClickHouse/blob/e9c47dc89c5761ff36bff58625c355d6e077037f/src/Columns/ColumnTuple.cpp#L352
The vector may be empty and we have to check it before we call `.back()`. | 2020-07-21T14:24:54 | 2020-07-21T14:25:37 | {} | MEMBER |
ClickHouse | ClickHouse | 661,902,289 | 12,636 | alexey-milovidov | @4ertus2 We can avoid this issue (and possibly many other) if we introduce another function named `castInternal` (or `_cast`, depending on your taste) that will not be affected by `cast_keep_nullable` and make all internal usages to use this function. | 2020-07-21T14:40:06 | 2020-07-21T14:40:20 | {} | MEMBER |
ClickHouse | ClickHouse | 662,358,348 | 12,621 | 4ertus2 | It's an error leaded from optimisation. `SET optimize_move_functions_out_of_any = 0` to disable. | 2020-07-22T09:51:09 | 2020-07-22T09:51:09 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 662,458,088 | 12,634 | alexey-milovidov | Performance test Ok.
> Yandex third-party checks (only for Yandex employees)
Arcadia has to be patched, @KochetovNicolai | 2020-07-22T13:37:34 | 2020-07-22T13:37:34 | {} | MEMBER |
ClickHouse | ClickHouse | 662,488,299 | 12,634 | KochetovNicolai | > Arcadia has to be patched
Done. Waiting for review from it. | 2020-07-22T14:32:03 | 2020-07-22T14:32:03 | {} | MEMBER |
ClickHouse | ClickHouse | 662,536,260 | 12,123 | akuzm | POSIX retains `/tmp` for compatibility, but encourages applications to use `TMPDIR` variable: https://pubs.opengroup.org/onlinepubs/9699919799/xrat/V4_xbd_chap10.html
We use `std::filesystem::temp_directory_path` which accounts for this and other variables: https://en.cppreference.com/w/cpp/filesystem/temp_directory... | 2020-07-22T15:56:49 | 2020-07-22T15:56:49 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 662,741,133 | 12,694 | zhlovy | select * FROM system.processes ; the result is null , so can`t kill it | 2020-07-22T23:07:13 | 2020-07-22T23:07:13 | {} | NONE |
ClickHouse | ClickHouse | 662,772,654 | 12,640 | xuyatian | @den-crane，
set distributed_aggregation_memory_efficient = 0 does not work ..All hosts are the same 1.1.54385. The clickhouse cluster was set up 2 years ago...
will upgrade to latest version.
thanks.
| 2020-07-23T01:08:47 | 2020-07-23T01:08:47 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 662,825,842 | 12,327 | ghost | @nikitamikhaylov
I improved type checking for the input of bayesAB and also replace vector with PODArray as much as possible.
May i have your review? Please let me know if there are other problems.
Cc. @alexey-milovidov | 2020-07-23T05:31:36 | 2020-07-23T05:31:36 | {} | NONE |
ClickHouse | ClickHouse | 661,762,485 | 6,852 | Jefferson111 | Yea, they should at least document it under [here](https://clickhouse.tech/docs/en/sql-reference/statements/select/where/). That current syntax reference link is as good as not existing at all... Honestly all it takes is just a few minutes for someone familiar with Clickhouse to update it with some examples and explana... | 2020-07-21T10:02:36 | 2020-07-21T10:02:36 | {} | NONE |
ClickHouse | ClickHouse | 662,307,371 | 10,019 | azat | >And the perfect solution would be to respect both PREWHERE and row-based filters at the same time
@alexey-milovidov am I understand correctly that after #12587 this can be safely done? (IOW row-based filters can be used in PREWHERE and WHERE and as a targets for optimize_move_to_prewhere optimization) | 2020-07-22T08:02:54 | 2020-07-22T08:02:54 | {} | MEMBER |
ClickHouse | ClickHouse | 662,350,298 | 12,577 | bohutang | It would be interesting if we make a git table engine, synchronized regularly :D | 2020-07-22T09:33:08 | 2020-07-22T09:33:08 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 662,422,548 | 9,756 | ShonT | Facing a similar issue
Code: 1000. DB::Exception: Received from localhost:9000. DB::Exception: mysqlxx::BadQuery
The query below was working on version 19.16.14.65 but not on 20.3.12.112 The latest Altinity LTS
```
select mysql_db.a_table.geo_region_id,
mysql_db.a_table.domain_id,
mysql_db.b_tab... | 2020-07-22T12:24:09 | 2020-07-22T12:24:58 | {} | NONE |
ClickHouse | ClickHouse | 662,429,131 | 12,666 | alesapin | Probably some tests will fail. | 2020-07-22T12:39:16 | 2020-07-22T12:39:16 | {} | MEMBER |
ClickHouse | ClickHouse | 662,530,726 | 11,121 | akuzm | Changed muzzy decay to 10s without much effect. Let's consider it done for now. | 2020-07-22T15:47:10 | 2020-07-22T15:47:10 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 662,549,296 | 12,123 | alexey-milovidov | Maybe just check for existence of tmp directory with fallback to current directory. | 2020-07-22T16:19:53 | 2020-07-22T16:19:53 | {} | MEMBER |
ClickHouse | ClickHouse | 662,716,435 | 12,672 | azat | >Functional stateless tests (address) — fail: 20
Fails in upstream too | 2020-07-22T21:52:49 | 2020-07-22T21:52:49 | {} | MEMBER |
ClickHouse | ClickHouse | 662,724,576 | 12,485 | excitoon | > Btw, all of these options are not used in the current HTTP client yet.
> I think it makes sense to add real usage of them as well.
What is current HTTP client? :wow:
https://github.com/ClickHouse/ClickHouse/blob/master/contrib/aws-s3-cmake/CMakeLists.txt#L94 | 2020-07-22T22:14:55 | 2020-07-22T22:14:55 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 661,880,931 | 11,933 | den-crane | Summary: need to increase -Djute.maxbuffer or reduce a number of active parts by any way. | 2020-07-21T14:03:15 | 2020-07-21T14:03:15 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 661,887,391 | 12,634 | alexey-milovidov | > Just added ~500 times const.
Ok. There are also a few other modifications btw. | 2020-07-21T14:14:47 | 2020-07-21T14:15:00 | {} | MEMBER |
ClickHouse | ClickHouse | 662,609,350 | 9,906 | mvcalder-xbk | My markdown in the original question was malformed, sorry. I corrected that and would like to simplify the question. I want to be able to access the row number of rows in the table.
If I start with a parquet file and create a table from it is it safe to assume the rows of the table are in the order that they are in ... | 2020-07-22T18:18:39 | 2020-07-22T18:21:31 | {} | NONE |
ClickHouse | ClickHouse | 661,794,142 | 12,611 | BayoNet | We have a topic describing the `ORDER BY` clause - https://clickhouse.tech/docs/en/sql-reference/statements/select/order-by/
I think it's better to move the *Usage* section of the setting description there under the *Optimizations* section. | 2020-07-21T11:15:21 | 2020-07-21T11:15:21 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 661,809,333 | 12,596 | Johnhz | > `windowFunnel` will return approximated results when your `event_date_time` has same value for same `e_device_id`
I check the official documents, the time type of clickhouse is accurate to the second level, there may indeed be duplication https://clickhouse.tech/docs/en/sql-reference/data-types/datetime/ | 2020-07-21T11:52:03 | 2020-07-21T11:52:03 | {} | NONE |
ClickHouse | ClickHouse | 661,826,966 | 12,596 | sundy-li | You can set set `event_date_time ` to be UInt64 (microseconds or millseconds), windowFunnel support all UIntXX types. | 2020-07-21T12:31:29 | 2020-07-21T12:31:29 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 662,245,504 | 5,214 | alexey-milovidov | Solutions discussed in #12587. | 2020-07-22T05:13:47 | 2020-07-22T05:13:47 | {} | MEMBER |
ClickHouse | ClickHouse | 662,508,098 | 11,501 | den-crane |
> This scenario is working fine. My only doubt is that I've around 20 columns in the `ORDER KEY`. I don't know if it's optimized but this is the only way of doing this `DELETE + INSERT`strategy.
20 columns in the `ORDER KEY` is OK. You can use https://clickhouse.tech/docs/en/engines/table-engines/mergetree-fam... | 2020-07-22T15:06:33 | 2020-07-22T15:06:33 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 662,682,211 | 12,640 | den-crane | Try to disable distributed_aggregation_memory_efficient
set distributed_aggregation_memory_efficient = 0
and check query with group by
Check that all host are the same 1.1.54385 ( https://github.com/ClickHouse/ClickHouse/issues/2778 )
BTW, Why do you such old version? Try at least 1.1.54394 | 2020-07-22T20:34:42 | 2020-07-22T20:53:04 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 662,784,215 | 12,656 | zhang2014 | Motivation:
in #10851 we will rewrite `ALTER TABLE table_name CHANGE old_column_name new_column_name INT NULL`(MySQL query) to `ALTER TABLE table_name MODIFY COLUMN old_column_name Nullable(Int32), RENAME COLUMN old_column_name TO new_column_name`(ClickHouse query); | 2020-07-23T02:00:09 | 2020-07-23T02:00:09 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 661,860,612 | 9,976 | vitlibar | Updated in https://github.com/ClickHouse/ClickHouse/pull/12423 | 2020-07-21T13:26:42 | 2020-07-21T13:26:56 | {} | MEMBER |
ClickHouse | ClickHouse | 661,943,052 | 12,640 | den-crane | Please provide a results for
```sql
select groupArray(shard_num), groupArray(host_name), groupArray(is_local)
from system.clusters where cluster = 'Cluster';
```
from shards 1 and 14
And
```sql
Q1:
SELECT count(),
hostName() h,
d
FROM DatabaseName.DistributedTable
WHERE dateTime >= '2... | 2020-07-21T15:50:14 | 2020-07-21T16:02:02 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 662,068,623 | 11,910 | bzon | Can we have a simple example of Redis, please? | 2020-07-21T19:43:08 | 2020-07-21T19:43:08 | {} | NONE |
ClickHouse | ClickHouse | 662,461,844 | 12,508 | alexey-milovidov | materialize_index: https://clickhouse-test-reports.s3.yandex.net/12484/767bc7a905cfce8634b5348eccafa4fb9e7210ca/functional_stateless_tests_(address).html#fail1 | 2020-07-22T13:44:50 | 2020-07-22T13:44:50 | {} | MEMBER |
ClickHouse | ClickHouse | 662,462,058 | 12,484 | alexey-milovidov | > 00943_materialize_index
flaky test, reported. | 2020-07-22T13:45:16 | 2020-07-22T13:45:16 | {} | MEMBER |
ClickHouse | ClickHouse | 661,803,198 | 12,593 | alexey-milovidov | > Yandex third-party checks (only for Yandex employees)
Deliberately broken to update simdjson. | 2020-07-21T11:36:50 | 2020-07-21T11:36:50 | {} | MEMBER |
ClickHouse | ClickHouse | 661,803,480 | 12,596 | Johnhz | > e_device_id
> `windowFunnel` will return approximated results when your `event_date_time` has same value for same `e_device_id`
In other words, an e_device_id can only have one record at the same time point? | 2020-07-21T11:37:30 | 2020-07-21T11:37:30 | {} | NONE |
ClickHouse | ClickHouse | 661,936,164 | 12,572 | qoega | If you have further questions, fell free to reopen this issue. | 2020-07-21T15:38:13 | 2020-07-21T15:38:13 | {} | MEMBER |
ClickHouse | ClickHouse | 662,457,014 | 12,652 | alexey-milovidov | > 00992_system_parts_race_condition_zookeeper
Reported in CI fest. | 2020-07-22T13:35:37 | 2020-07-22T13:35:37 | {} | MEMBER |
ClickHouse | ClickHouse | 662,501,610 | 12,184 | den-crane | Duplicates are removed from ReplacingMergeTree during parts a merge process.
Merge combines parts over partition. If you have duplicates in different partitions or at different servers they will not be removed / merged.
Try select ... from test.pool _dis **FINAL**
>And sometimes this error occurs: cannot execut... | 2020-07-22T14:55:16 | 2020-07-22T14:55:16 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 662,508,383 | 12,438 | akuzm | 
Deliberately broken for update of simdjson. | 2020-07-21T11:34:00 | 2020-07-21T11:34:00 | {} | MEMBER |
ClickHouse | ClickHouse | 661,813,909 | 12,616 | 4ertus2 | Could you describe the feature in more detail?
CH makes distributed aggregation and distributed JOIN/IN in different way. But distributed pipeline is mostly the same: you have server initiator for the query. It passed query (probably in modified form) to other nodes via Distributed table. Then nodes send their parts... | 2020-07-21T12:02:26 | 2020-07-21T12:02:26 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 661,878,640 | 12,633 | alexey-milovidov | `cast_keep_nullable` is not pushed to mutations interpreter, so we are safe unless this setting is modified globally. | 2020-07-21T13:59:12 | 2020-07-21T13:59:12 | {} | MEMBER |
ClickHouse | ClickHouse | 661,871,608 | 12,583 | den-crane | http+input_format_parallel_parsing=0
```
for i in `seq 1 50`; do curl -s "http://localhost:8123/?input_format_parallel_parsing=0&query=INSERT%20INTO%20log%20format%20TSV" --data-binary @test.json ; echo -n 'ps rss:' `ps -C clickhouse-serv -o rss=`' | '; echo "select metric,':', formatReadableSize(value) from syste... | 2020-07-21T13:46:49 | 2020-07-21T14:12:58 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 662,217,076 | 12,651 | den-crane | Use WA:
```
select key,dictGet('db.test_dict','valueCode', tuple(key)) as c1 from db.test_dict tb
group by key
```
or
```
select key,dictGetString('db.test_dict','valueCode', tuple(key)) as c1 from db.test_dict tb
group by key,c1
``` | 2020-07-22T03:14:20 | 2020-07-22T03:14:20 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 662,217,161 | 12,651 | den-crane | duplicate https://github.com/ClickHouse/ClickHouse/issues/11469 | 2020-07-22T03:14:42 | 2020-07-22T03:14:42 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 662,456,832 | 12,652 | alexey-milovidov | But it's Ok, performance test just cannot compare across revisions with different system logs structure. It should be fixed automatically after merge. @akuzm | 2020-07-22T13:35:14 | 2020-07-22T13:35:14 | {} | MEMBER |
ClickHouse | ClickHouse | 661,769,905 | 12,585 | alexey-milovidov | BTW, new division is much faster than previous but `fmod` is terrible.
`SELECT count() FROM system.numbers WHERE NOT ignore(number % 1.1)`
master (conversion + integer division): 134.35 million rows/s.
this PR (divide, multiply, truncate and subtract): 1.10 billion rows/s., fully inlined and vectorized.
this P... | 2020-07-21T10:18:10 | 2020-07-21T10:18:34 | {} | MEMBER |
ClickHouse | ClickHouse | 661,821,535 | 12,594 | 4ertus2 | Related #12527 | 2020-07-21T12:19:38 | 2020-07-21T12:19:38 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 662,111,954 | 12,648 | alexey-milovidov | This is not a bugfix because no bug was present. But it's Ok for backport, so "Bug Fix" category. | 2020-07-21T21:19:03 | 2020-07-21T21:19:03 | {} | MEMBER |
ClickHouse | ClickHouse | 662,245,580 | 1,562 | alexey-milovidov | Solutions discussed in #12587. | 2020-07-22T05:14:05 | 2020-07-22T05:14:05 | {} | MEMBER |
ClickHouse | ClickHouse | 662,444,426 | 11,526 | SaltTan | It works if absolute paths are specified:
```
clickhouse-copier --daemon --config /opt/copier/copier.xml --task-path /clickhouse/copier/task --task-file /opt/copier/task.xml
``` | 2020-07-22T13:11:50 | 2020-07-22T13:11:50 | {
"+1": 3
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 662,452,984 | 12,665 | den-crane | It's broken in 20.4 and 20.5 you can use 20.3.
it has been fixed in testing 20.6.1.4050+
Duplicate https://github.com/ClickHouse/ClickHouse/issues/12454 https://github.com/ClickHouse/ClickHouse/issues/10994
Fix https://github.com/ClickHouse/ClickHouse/pull/12116
could be fixed in 20.4.7.67 (not sure) | 2020-07-22T13:28:24 | 2020-07-22T13:32:26 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 662,749,954 | 12,700 | CurtizJ | For tuples-functions I've returned old code from version, when all explicit sets were created from AST. | 2020-07-22T23:37:12 | 2020-07-22T23:38:29 | {} | MEMBER |
ClickHouse | ClickHouse | 662,790,889 | 12,184 | gubinjie | @den-crane 20.3.9 | 2020-07-23T02:34:54 | 2020-07-23T02:34:54 | {} | NONE |
ClickHouse | ClickHouse | 672,218,016 | 12,771 | alexey-milovidov | You can see the logs of all builds here: https://clickhouse-builds.s3.yandex.net/12771/ff92812ac57d59e9c43cc90b93c0048630304ebe/clickhouse_build_check/report.html | 2020-08-11T19:28:36 | 2020-08-11T19:28:36 | {} | MEMBER |
ClickHouse | ClickHouse | 672,462,291 | 13,593 | toby080508050805 | Following is a log about the query stuck at remote server. I cannot kill or cancel remote server query. The only way to remove this query is restart server.
local server log
```
2020.08.11 14:47:58.170432 [ 367 ] {740e5ddd-7c44-4bc0-bea8-ca821b6ab41b} <Debug> executeQuery: (from [::ffff:127.0.0.1]:41978) SELECT T... | 2020-08-12T02:08:05 | 2020-08-12T02:21:50 | {} | NONE |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.