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 | 646,790,437 | 11,805 | alexey-milovidov | Fonts comparison:
Yandex Sans (good):
Noto Sans (worse but still tolerable):
| 2020-06-19T17:56:21 | 2020-06-19T17:56:21 | {} | MEMBER |
ClickHouse | ClickHouse | 646,800,679 | 11,805 | alexey-milovidov | Table cells are not of fixed width and they can be too wide:
| 2020-06-19T18:13:22 | 2020-06-19T18:13:51 | {} | MEMBER |
ClickHouse | ClickHouse | 646,662,577 | 11,786 | KochetovNicolai | Test for secondary index is needed too. | 2020-06-19T14:21:05 | 2020-06-19T14:21:05 | {} | MEMBER |
ClickHouse | ClickHouse | 646,788,093 | 11,805 | alexey-milovidov | Too large paddings and margins or selectors. | 2020-06-19T17:45:42 | 2020-06-19T17:45:42 | {} | MEMBER |
ClickHouse | ClickHouse | 646,822,805 | 11,802 | alexey-milovidov | Failure under MemorySanitizer still reproduces. | 2020-06-19T19:12:08 | 2020-06-19T19:12:08 | {} | MEMBER |
ClickHouse | ClickHouse | 646,857,966 | 11,773 | den-crane | aggregation_memory_efficient_merge_threads is completely safe to tune.
M(SettingUInt64, aggregation_memory_efficient_merge_threads, 0, "Number of threads to use for merge intermediate aggregation results in memory efficient mode. When bigger, then more memory is consumed. 0 means - same as 'max_threads'.", 0)
... | 2020-06-19T20:57:58 | 2020-06-19T20:57:58 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 646,104,816 | 11,753 | qoega | Failed test #11765 | 2020-06-18T15:36:17 | 2020-06-18T15:36:17 | {} | MEMBER |
ClickHouse | ClickHouse | 646,219,857 | 11,763 | filimonov | Similar https://github.com/ClickHouse/ClickHouse/issues/3178 | 2020-06-18T18:00:09 | 2020-06-18T18:00:09 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 646,397,176 | 11,753 | hexiaoting | > Failed test #11765
Sorry, I don't understand.
what's the correct result for this SQL command:
```
SELECT less(( SELECT NULL, '', [] ) AS yrsihrr, 'W/u&A');
```
And what should I do for #11765 ?
```
ClickHouse client version 20.5.1.1.
Connecting to localhost:9000 as user default.
Connected to ClickHous... | 2020-06-19T02:18:18 | 2020-06-19T07:14:32 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 646,790,846 | 11,805 | alexey-milovidov | There is an animation on hover on selectors, it's unneeded. | 2020-06-19T17:51:52 | 2020-06-19T17:51:52 | {} | MEMBER |
ClickHouse | ClickHouse | 646,943,573 | 11,810 | alexey-milovidov | Maybe better to move values of constants to .cpp file? | 2020-06-20T05:21:21 | 2020-06-20T05:21:21 | {} | MEMBER |
ClickHouse | ClickHouse | 645,962,865 | 11,755 | Namnd-gem | Yes, the query is failed even in the case without GLOBAL. Change ENGINE to Memory but it still gets this error | 2020-06-18T11:43:00 | 2020-06-18T11:43:00 | {} | NONE |
ClickHouse | ClickHouse | 646,490,710 | 11,599 | filimonov | Formats which support for Kafka was fixed in that PR (were not working before)
- MsgPack
- RowBinaryWithNamesAndTypes
- JSONCompactEachRowWithNamesAndTypes
- TSVWithNamesAndTypes (there was segfault there)
- Regexp
- JSONAsString | 2020-06-19T07:42:41 | 2020-06-19T07:43:07 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 646,791,913 | 11,805 | alexey-milovidov | The titles of configurations are too long. | 2020-06-19T17:54:15 | 2020-06-19T17:54:15 | {} | MEMBER |
ClickHouse | ClickHouse | 646,854,249 | 11,791 | siradjev | Added query to reproduce.
| 2020-06-19T20:45:22 | 2020-06-19T20:45:22 | {} | NONE |
ClickHouse | ClickHouse | 646,943,894 | 11,795 | alexey-milovidov | > ClickHouse build check — 20/21 builds are OK
An error in infamous "Sandbox" infrastructure. | 2020-06-20T05:24:08 | 2020-06-20T05:24:08 | {} | MEMBER |
ClickHouse | ClickHouse | 646,119,816 | 11,755 | xmariachi | Hi @4ertus2 , @namnd-gem
The issue only happens when the left part of the JOIN is the Merge Engine table.
Left side Distributed, Right side Merge works:
```
SELECT sessionId
FROM orders_data_d_hist AS orders
GLOBAL ALL LEFT JOIN orderAttributions_d AS orderAttributions USING (frontendId, orderId)
Ok.
`... | 2020-06-18T15:56:04 | 2020-06-18T15:57:27 | {} | NONE |
ClickHouse | ClickHouse | 646,435,847 | 11,701 | filimonov | The exception just says that part /var/lib/clickhouse/data/default/alerts/202008_3738_5330_3/ has column alert_id corrupted on disk. Can you check if the exception happens with simple select when you read alert_id column in that month ? | 2020-06-19T05:00:50 | 2020-06-19T05:02:21 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 646,441,185 | 11,622 | filimonov | Maybe we should give tmp aliases for such columns? | 2020-06-19T05:22:04 | 2020-06-19T05:22:04 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 646,474,726 | 9,615 | filimonov | Duplicate of #4226 | 2020-06-19T07:04:47 | 2020-06-19T07:04:47 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 646,704,212 | 11,745 | alesapin | `01213_alter_rename_column_zookeeper` cannot reproduce locally and don't understand how it's possible. | 2020-06-19T15:42:08 | 2020-06-19T15:42:08 | {} | MEMBER |
ClickHouse | ClickHouse | 646,787,855 | 11,805 | alexey-milovidov | Huge empty space under and on the left of "Compare". | 2020-06-19T17:45:16 | 2020-06-19T17:45:16 | {} | MEMBER |
ClickHouse | ClickHouse | 646,718,629 | 11,800 | blinkov | cc @zhang2014 @sundy-li | 2020-06-19T16:08:29 | 2020-06-19T16:08:29 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 646,787,270 | 11,805 | alexey-milovidov | The list of selectors does not fit in screen (2560x1600 monitor).
| 2020-06-19T17:44:45 | 2020-06-19T18:15:58 | {} | MEMBER |
ClickHouse | ClickHouse | 646,856,062 | 6,549 | thyn | Problem still exists when using standalone tables (without replication). After starting using zookeeper all updates are performed ok. So, I can't approve that now it's working. | 2020-06-19T20:51:41 | 2020-06-19T20:51:41 | {} | NONE |
ClickHouse | ClickHouse | 645,945,041 | 11,755 | 4ertus2 | Does the query fail if the JOIN is not GLOBAL? Does it fail if you change `ENGINE = Kafka` to MergeTree or Memory? | 2020-06-18T11:01:41 | 2020-06-18T11:33:48 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 646,010,946 | 11,755 | 4ertus2 | Am I right that if you use GLOBAL JOIN not over engine Merge but directly over enigne Distributed it works? | 2020-06-18T13:17:28 | 2020-06-18T13:17:41 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 646,296,199 | 10,564 | azat | >Is there anything I can do to force an even distribution between them?
Looks this will be resolved after #11645 (+ #11669) ? | 2020-06-18T20:44:55 | 2020-06-18T20:44:55 | {} | MEMBER |
ClickHouse | ClickHouse | 646,694,376 | 11,794 | blinkov | We're discussing it in chat. | 2020-06-19T15:21:08 | 2020-06-19T15:21:08 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 646,703,754 | 11,786 | alesapin | > Test for secondary index is needed too.
It's just here https://github.com/ClickHouse/ClickHouse/pull/11786/files#diff-1b07ab8c26f5b3c84e792c167eb85d84R78 | 2020-06-19T15:41:08 | 2020-06-19T15:41:08 | {} | MEMBER |
ClickHouse | ClickHouse | 646,709,841 | 11,798 | blinkov | cc @zhang2014 @sundy-li | 2020-06-19T15:54:48 | 2020-06-19T15:54:48 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 646,712,271 | 11,797 | ithangzhou | description for this pull request need overwrite,sorry for my mistake.
------------------ 原始邮件 ------------------
发件人: "Ivan Blinkov"<notifications@github.com>;
发送时间: 2020年6月19日(星期五) 晚上11:34
收件人: "ClickHouse/ClickHouse"<ClickHouse@noreply.github.com>;
抄送: "-忘忧泉"<565258751@qq... | 2020-06-19T16:00:02 | 2020-06-19T16:00:02 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 646,788,215 | 11,805 | alexey-milovidov | The color of "hover" is the same as the color of selected items. | 2020-06-19T17:45:57 | 2020-06-19T17:45:57 | {} | MEMBER |
ClickHouse | ClickHouse | 646,846,719 | 11,669 | akuzm | > FWIW right now there `distributed_replica_error_cap` that is the high limit for the error numbers, so **how about rename these settings** to (not sure what is the policy for settings renames):
>
> * distributed_replica_error_ignore -> distributed_replica_error_limit_low
> * distributed_replica_error_cap -> dis... | 2020-06-19T20:21:18 | 2020-06-19T20:21:18 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 646,427,257 | 11,722 | filimonov | > IF (there is only one active part left in partition AND part.level>0 (has been merged at least once))
BTW: Why `AND part.level>0` (has been merged at least once)? To remove duplicates arrived inside a single select? | 2020-06-19T04:24:35 | 2020-06-19T04:25:05 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 646,501,762 | 11,779 | filimonov | If I understand correctly in your case you can just recreate the temporary table (or create new one each time) instead of truncating/dropping partition in a 'stale' temporary table. | 2020-06-19T08:07:04 | 2020-06-19T08:07:04 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 646,581,729 | 11,736 | alexey-milovidov | > Description check — Changelog entry is not found
Fixed.
| 2020-06-19T11:19:16 | 2020-06-19T11:19:16 | {} | MEMBER |
ClickHouse | ClickHouse | 646,584,738 | 11,746 | alexey-milovidov | > test_always_fetch_merged/test.py::test_replica_always_download
Flaky test. | 2020-06-19T11:26:22 | 2020-06-19T11:26:22 | {} | MEMBER |
ClickHouse | ClickHouse | 646,826,455 | 11,718 | alexey-milovidov | Hyperscan library fails in Memory Sanitizer and Undefined Behaviour Sanitizer.
We just did not have enough tests. | 2020-06-19T19:22:15 | 2020-06-19T19:22:15 | {} | MEMBER |
ClickHouse | ClickHouse | 646,831,869 | 11,745 | alesapin | Stress test (memory) — Killed by signal (in clickhouse-server.log)
FunctionExtractAllGroups in CI fest. | 2020-06-19T19:38:18 | 2020-06-19T19:38:18 | {} | MEMBER |
ClickHouse | ClickHouse | 646,918,472 | 11,572 | johnpyp | I'm having the same issue | 2020-06-20T01:45:34 | 2020-06-20T01:45:34 | {
"+1": 4
} | NONE |
ClickHouse | ClickHouse | 645,964,100 | 11,755 | 4ertus2 | Could you update the queries in issue when? We need as simple sample as possible. The harder sample is to understand and reproduce the less probablitity it would be fixed. | 2020-06-18T11:46:10 | 2020-06-18T11:46:10 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 646,385,851 | 11,773 | den-crane | >CREATE TEMPORARY TABLE ans ENGINE = Memory
Why do you expect that a query result will fit into 100GB of RAM ? Because CSV is 45GB? I believe it is very wrong expectation. I experienced 2-5 times overhead in such cases.
Can you show an output of
```
SET send_logs_level = 'debug' ;
SELECT id1, id2, id3, id4... | 2020-06-19T01:30:48 | 2020-06-19T01:32:03 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 646,406,536 | 11,564 | bobrik | Are you sure? The way I read the code, `MinMax index condition` is printed after parts are known, and most of the time is spent after that, but before `Selected 47 parts by date` message.
MinMax message is printed [here](https://github.com/ClickHouse/ClickHouse/blob/5c42408addc5d3532ac6941e69ac15db0da0e6af/src/Stora... | 2020-06-19T02:55:11 | 2020-06-19T02:55:11 | {
"+1": 3
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 644,016,737 | 11,567 | filimonov | But independantly from the issue with Settings (fixed in 20.5) there is an issue with Avro in kafka.
```
2020.06.15 10:56:03.473794 [ 20 ] {} <Error> void DB::StorageKafka::threadFunc(): std::exception. Code: 1001, type: avro::Exception, e.what() = EOF reached, Stack trace (when copying this message, always includ... | 2020-06-15T09:31:14 | 2020-06-19T05:23:52 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 646,656,458 | 11,783 | den-crane | >20, some of threads reported "failed to response"
Can you elaborate? Do you use JDBC driver? And got " :8123 failed to respond" ?
It is small issue in JDBC driver and you should just repeat your query.
>Since "Buffer" is memory-based,
>I guess its inserting performance should be higher, 10 thousands/second?
... | 2020-06-19T14:08:16 | 2020-06-19T14:08:16 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 646,917,534 | 11,800 | sundy-li | LGTM | 2020-06-20T01:39:28 | 2020-06-20T01:39:28 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 646,918,161 | 11,798 | sundy-li | LGTM | 2020-06-20T01:43:37 | 2020-06-20T01:43:37 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 646,481,105 | 11,776 | filimonov | > without having to create and maintain EC2 instances running Kafka.
Actually AWS also offers managed Kafka: https://aws.amazon.com/msk/ | 2020-06-19T07:19:53 | 2020-06-19T07:19:53 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 646,793,907 | 11,805 | alexey-milovidov | You cannot open and check index.html page from working copy of repository:
https://github.com/ClickHouse/ClickHouse/blob/master/website/benchmark/hardware/index.html
without going through an overwhelmingly heavy process of "building" the website. This process even involves some scripts in node.js. | 2020-06-19T17:58:39 | 2020-06-19T17:58:39 | {} | MEMBER |
ClickHouse | ClickHouse | 646,887,606 | 11,669 | azat | >Renaming settings is almost never good -- we strive to be backwards compatible where possible.
Will remember.
>Speaking about the names themselves, "cap" is a saturation limit for the number of errors we track for a replica. It's limited so that the error number decays in a sane time, and the replica is not cons... | 2020-06-19T22:51:25 | 2020-06-19T22:51:25 | {} | MEMBER |
ClickHouse | ClickHouse | 646,027,073 | 11,719 | filimonov | It seems you're trying to reinvent SETTINGS clause.
Try
```sql
select * from system.parts settings max_threads=1, max_block_size=100;
```
Does it satisfy your requirements?
Also most of the drivers can send the settings with query w/o making adjustments inside the query itself. | 2020-06-18T13:46:03 | 2020-06-18T13:47:49 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 646,438,124 | 11,719 | filimonov | @4ertus2 but it looks like @siradjev is happy with Settings, so let's keep it open for now and collect feedback.
If you need that feature and `settings` clause does not satisfy your requirements - please send a comment here with your usecase. | 2020-06-19T05:10:04 | 2020-06-19T05:10:32 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 646,476,790 | 11,776 | filimonov | It can be implemented but priority seems low because it's a vendor-specific message system (like Google PubSub).
Currently, support for RabbitMQ is in development, I would say the [Apache Pulsar](https://github.com/ClickHouse/ClickHouse/issues/4226) is the most probable next candidate.
Others include: NATS, NSQ,... | 2020-06-19T07:09:41 | 2020-06-19T07:35:14 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 646,549,110 | 11,291 | KochetovNicolai | 52 faster | 2020-06-19T09:58:38 | 2020-06-19T09:58:38 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 646,758,815 | 11,786 | KochetovNicolai | It's good to have any test with data :) | 2020-06-19T17:03:20 | 2020-06-19T17:03:20 | {} | MEMBER |
ClickHouse | ClickHouse | 646,792,220 | 11,805 | alexey-milovidov | The table does not fit on screen.
| 2020-06-19T17:54:57 | 2020-06-19T17:54:57 | {} | MEMBER |
ClickHouse | ClickHouse | 646,847,804 | 11,668 | tavplubix | It's not false-positive, test fails because there is no instance of `Poco::Util::Application` in `clickhouse-obfuscator`. Some unit tests probably fail for same reason.
```
$ clickhouse-obfuscator --structure "URL String, Title String, SearchPhrase String" --input-format TSV --output-format TSV --seed hello < ${CLI... | 2020-06-19T20:24:50 | 2020-06-19T20:24:50 | {
"+1": 2
} | MEMBER |
ClickHouse | ClickHouse | 646,593,060 | 11,725 | alexey-milovidov | Will do it in a different way. | 2020-06-19T11:49:15 | 2020-06-19T11:49:15 | {} | MEMBER |
ClickHouse | ClickHouse | 646,604,892 | 9,776 | tetafro | I tried to manually drop cache using `SYSTEM DROP MARK CACHE`, and it didn't affect cluster that much.
---
I will try to update clickhouse and check if the issue remains. | 2020-06-19T12:18:27 | 2020-06-19T12:18:27 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 646,792,443 | 11,805 | alexey-milovidov | Paddings and margins of table cells are too huge. | 2020-06-19T17:55:24 | 2020-06-19T17:55:24 | {} | MEMBER |
ClickHouse | ClickHouse | 646,855,613 | 11,791 | siradjev | This query tries to allocate too much and fails (but properly, no segfault):
```
select
attr.k1,
attr.dim1,
fact.k1,
fact.gk1val,
fact.gk2val,
fact.gk3val from
(
select k1, k2,
(groupArrayInsertAt(Null, 3)(tuple(toNullable(val)), toUInt32(indexOf(range(3), gk) - 1)).1 as arr)[1] gk1val,
arr[2] gk2val,arr[3... | 2020-06-19T20:50:01 | 2020-06-19T20:50:01 | {} | NONE |
ClickHouse | ClickHouse | 646,918,670 | 11,701 | nickevin | > The exception just says that part /var/lib/clickhouse/data/default/alerts/202008_3738_5330_3/ has column alert_id corrupted on disk. Can you check if the exception happens with simple select when you read alert_id column in that month ?
Thank you filimonov, select * from alerts in 202008 is ok.
It is good for m... | 2020-06-20T01:46:56 | 2020-06-20T01:46:56 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 646,944,377 | 11,718 | alexey-milovidov | ClickHouse build check:
> Build failed without build log. It's inner CI problem. Maybe something wrong with docker or host. Just wait for build restart. | 2020-06-20T05:28:41 | 2020-06-20T05:28:41 | {} | MEMBER |
ClickHouse | ClickHouse | 645,971,700 | 11,755 | Namnd-gem | I've updated queries on first comment | 2020-06-18T12:02:56 | 2020-06-18T12:02:56 | {} | NONE |
ClickHouse | ClickHouse | 645,994,395 | 11,028 | tavplubix | Is it expected, that empty arguments (`INT()`, `FLOAT()`) are also allowed?
| 2020-06-18T12:46:14 | 2020-06-18T12:46:14 | {} | MEMBER |
ClickHouse | ClickHouse | 646,265,375 | 11,719 | 4ertus2 | It's quite clear that it's posible to translate `/*+ <some> */` into `SETTINGS <some>` at query parser level. The only reason for this as I know is to pass queries through BI system that allows comments and do not allow our `SETTINGS` syntax. | 2020-06-18T19:35:09 | 2020-06-18T19:35:09 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 646,297,023 | 10,564 | cw9 | Awesome! Will try the new features out, thanks for the help! | 2020-06-18T20:46:49 | 2020-06-18T20:46:49 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 646,515,739 | 11,334 | nikitamikhaylov | This feature has unexpected results: (with big epsilon functions with `anon_` prefix must be equal to ordinary function without prefix)
```sql
jakalletti.N1 :) select anon_sum(100000000)(number) from numbers(10)
SELECT anon_sum(100000000)(number)
FROM numbers(10)
┌─anon_sum(100000000)(number)─┐
│ ... | 2020-06-19T08:37:30 | 2020-06-19T08:37:30 | {} | MEMBER |
ClickHouse | ClickHouse | 646,681,018 | 8,639 | alesapin | The only affected part is `alter` query. So only tests for alter is important. | 2020-06-19T14:53:32 | 2020-06-19T14:53:32 | {} | MEMBER |
ClickHouse | ClickHouse | 646,704,134 | 11,126 | sneakythr0ws | Hi! How did you fix it? | 2020-06-19T15:41:59 | 2020-06-19T15:41:59 | {} | NONE |
ClickHouse | ClickHouse | 646,788,858 | 11,805 | alexey-milovidov | Different height of table rows in diagram.
| 2020-06-19T17:47:20 | 2020-06-19T17:47:20 | {} | MEMBER |
ClickHouse | ClickHouse | 646,943,070 | 11,810 | alexey-milovidov | > ClickHouse build check — 21/22 builds are OK
Temporary error in "Sandbox" infrastructure. | 2020-06-20T05:16:46 | 2020-06-20T05:16:46 | {} | MEMBER |
ClickHouse | ClickHouse | 646,391,561 | 11,595 | zhuyoucui0909 | @filimonov Any update? Is there any work around? | 2020-06-19T01:55:34 | 2020-06-19T01:55:34 | {} | NONE |
ClickHouse | ClickHouse | 646,475,384 | 4,226 | filimonov | Originally posted by @cmdares at https://github.com/ClickHouse/ClickHouse/issues/9615#issue-579690521:
> Aapche Pulsar will be the mainstream message engine in the future.
> Transplant Kafka support capability to pulsar.
> Pulsar ref:
> https://dzone.com/articles/life-beyond-kafka-with-apache-pulsar
> https://dz... | 2020-06-19T07:06:19 | 2020-06-19T07:06:19 | {
"+1": 3
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 646,710,082 | 11,786 | KochetovNicolai | Funny thing, but we can't unmaterialize column back:
```
:) desc ttt
DESCRIBE TABLE ttt
┌─name─┬─type───┬─default_type─┬─default_expression─┬─comment─┬─codec_expression─┬─ttl_expression─┐
│ x │ UInt32 │ MATERIALIZED │ 3 │ │ │ │
│ y │ UInt32 │ ... | 2020-06-19T15:55:22 | 2020-06-19T15:55:22 | {} | MEMBER |
ClickHouse | ClickHouse | 646,759,929 | 11,682 | KochetovNicolai | Now tests are fine. | 2020-06-19T17:05:46 | 2020-06-19T17:05:46 | {} | MEMBER |
ClickHouse | ClickHouse | 646,788,971 | 11,805 | alexey-milovidov | Red color is extreme aggressive. | 2020-06-19T17:47:36 | 2020-06-19T17:47:36 | {} | MEMBER |
ClickHouse | ClickHouse | 646,789,401 | 11,805 | alexey-milovidov | The text on the left of the diagram is too bold. | 2020-06-19T17:48:37 | 2020-06-19T17:48:37 | {} | MEMBER |
ClickHouse | ClickHouse | 646,801,583 | 11,805 | alexey-milovidov | Different kind of machines:
cloud; baremetal servers; laptops and workstations
should be highlighted by different colors. | 2020-06-19T18:15:33 | 2020-06-19T18:15:33 | {} | MEMBER |
ClickHouse | ClickHouse | 646,836,918 | 11,785 | alexey-milovidov | Build check:
> Build failed without build log. It's inner CI problem. Maybe something wrong with docker or host. Just wait for build restart.
It's the issue in infamous Sandbox infrastructure. | 2020-06-19T19:52:00 | 2020-06-19T19:52:00 | {} | MEMBER |
ClickHouse | ClickHouse | 646,296,940 | 11,214 | tavplubix | Stress test failure looks not related: server segfaults on `SELECT max(multiply())` | 2020-06-18T20:46:37 | 2020-06-18T20:46:37 | {} | MEMBER |
ClickHouse | ClickHouse | 646,563,731 | 4,435 | alexey-milovidov | The result of `rowNumberInBlock` function depends on the order of function execution with respect to other data transformations during the query. ClickHouse query optimizer can reorder data transformations in assumption that all functions are "pure" (the function result only depends on its arguments). But this assumpti... | 2020-06-19T10:32:41 | 2020-06-19T10:33:03 | {} | MEMBER |
ClickHouse | ClickHouse | 646,638,948 | 11,794 | alexey-milovidov | @blinkov You have to explain. | 2020-06-19T13:34:52 | 2020-06-19T13:34:52 | {} | MEMBER |
ClickHouse | ClickHouse | 646,700,608 | 11,797 | blinkov | @ithangzhou why have you closed this? | 2020-06-19T15:34:12 | 2020-06-19T15:34:12 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 646,791,616 | 11,805 | alexey-milovidov | Slightly round corners of selectors look bad.
| 2020-06-19T17:53:31 | 2020-06-19T17:53:31 | {} | MEMBER |
ClickHouse | ClickHouse | 646,854,851 | 11,773 | jangorecki | @den-crane Thanks for trying to help. I cannot at the moment upgrade to 20.3.11.97.
I run query with settings you provided on my current version. Looks like `aggregation_memory_efficient_merge_threads` improves a lot. Is it safe to use it always or there are some drawbacks?
If I will reduce `max_bytes_before_external... | 2020-06-19T20:47:27 | 2020-06-19T20:47:27 | {} | NONE |
ClickHouse | ClickHouse | 646,061,989 | 11,758 | filimonov | Duplicate #10212
You can't use expressions in PARTITION clause anymore
It was intentionally removed in #8835 | 2020-06-18T14:38:29 | 2020-06-18T14:38:29 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 646,073,048 | 10,351 | CurtizJ | Also by our codestyle metod names should be in `camelCase` | 2020-06-18T14:55:00 | 2020-06-18T14:55:00 | {} | MEMBER |
ClickHouse | ClickHouse | 646,098,013 | 11,719 | siradjev | > It seems you're trying to reinvent SETTINGS clause.
True, I was not aware of this clause, as when it was discussed on previous meetup there was no mention of this functionality. Seems clickhouse went with "MSSQL-way", which is good enough as it still covers major use-cases.
> Does it satisfy your requirements?... | 2020-06-18T15:29:29 | 2020-06-18T16:30:21 | {} | NONE |
ClickHouse | ClickHouse | 646,515,665 | 11,779 | slimtom95 | Thanks for the advice. However, recreating table demands maintaining latest table definition in the etl workflow, i.e.Replicated Engine parameters, order by keys, which is not our optimal solution. (also, for security, we forbade using drop keyword | 2020-06-19T08:37:18 | 2020-06-19T08:37:18 | {
"+1": 1
} | NONE |
ClickHouse | ClickHouse | 646,516,090 | 11,722 | siradjev | > To remove duplicates arrived inside a single select?
To remove duplicates arrived within single *insert batch*.
As per my understanding data collapsing is not made during insert itself, so a single part that has just been inserted may have duplicates.
| 2020-06-19T08:38:19 | 2020-06-19T08:38:19 | {} | NONE |
ClickHouse | ClickHouse | 646,749,649 | 11,794 | alexey-milovidov | @blinkov Do you have any schedule to fix the ugly markup? | 2020-06-19T16:51:22 | 2020-06-19T16:51:22 | {} | MEMBER |
ClickHouse | ClickHouse | 646,789,595 | 11,805 | alexey-milovidov | The font is Noto Sans, must be Yandex Sans. | 2020-06-19T17:49:03 | 2020-06-19T17:49:03 | {} | MEMBER |
ClickHouse | ClickHouse | 646,801,079 | 11,805 | alexey-milovidov | All systems should be selected for comparison by default. | 2020-06-19T18:14:24 | 2020-06-19T18:14:24 | {} | MEMBER |
ClickHouse | ClickHouse | 646,943,743 | 11,810 | alexey-milovidov | ```
А ещё можно обозначить намерение явно: у нас же есть макрос ARCADIA_BUILD.
Убираешь weak и volatile и делаешь, что если ARCADIA_BUILD, то значение у переменной не выставляется в заголовочном файле.
``` | 2020-06-20T05:22:49 | 2020-06-20T05:22:49 | {} | MEMBER |
ClickHouse | ClickHouse | 646,944,503 | 11,718 | alexey-milovidov | > Functional stateful tests (debug) — Some queries hung, fail: 1
Hyperscan takes too much time to compile fuzzy regexp in debug. Will skip the test. | 2020-06-20T05:30:06 | 2020-06-20T05:30:06 | {} | MEMBER |
ClickHouse | ClickHouse | 650,176,328 | 11,418 | vitlibar | Yes, these system tables are implemented in https://github.com/ClickHouse/ClickHouse/pull/10387 and now that implementation is available in master and `20.5`. | 2020-06-26T13:22:47 | 2020-06-26T13:22:47 | {} | MEMBER |
ClickHouse | ClickHouse | 650,304,759 | 11,498 | azat | >I mean this option becomes an alternative to <user> and <password>, if you set <use_initial_user>, don't set <user> and <password> for that node
Thanks! So, that was my initial way (like #11391 but it does not have flag to switch, and user/password is still there plus it has some issues)
But there are some issue... | 2020-06-26T17:32:34 | 2020-06-26T17:37:42 | {} | MEMBER |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.