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 | 625,246,173 | 10,721 | KochetovNicolai | Indeed. `Field` was not changed. This bug was all the time. | 2020-05-07T13:11:01 | 2020-05-07T13:11:01 | {} | MEMBER |
ClickHouse | ClickHouse | 625,667,426 | 10,725 | wenwen2019 | Is the version lower？ | 2020-05-08T06:55:48 | 2020-05-08T06:55:48 | {} | NONE |
ClickHouse | ClickHouse | 625,872,592 | 10,751 | champtar | I don't know if this apply here, but a really good read about tcp keep-alive coner cases:
https://blog.cloudflare.com/when-tcp-sockets-refuse-to-die/ | 2020-05-08T15:31:44 | 2020-05-08T15:31:44 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 625,894,517 | 10,751 | alexandr-orlov | > I don't know if this apply here, but a really good read about tcp keep-alive coner cases:
> https://blog.cloudflare.com/when-tcp-sockets-refuse-to-die/
Yep, it's excellent article! But as a summary of it, a solution against Idling ESTABLISHED connections - is just to use TCP keepalives. That's what we're doing h... | 2020-05-08T16:19:47 | 2020-05-08T16:19:47 | {} | NONE |
ClickHouse | ClickHouse | 626,068,511 | 6,432 | alexey-milovidov | @filimonov That's correct, we should make `if` with constant branches to return LowCardinality.
| 2020-05-08T23:58:39 | 2020-05-09T00:13:27 | {} | MEMBER |
ClickHouse | ClickHouse | 626,176,650 | 10,770 | azat | Fix `readability-qualified-auto`
*Previous version of the patch is [here](https://github.com/ClickHouse/ClickHouse/commit/e12acd881a78f8d30926caba017bac5e8a370355), attached for CI report* | 2020-05-09T13:30:21 | 2020-05-09T13:30:21 | {} | MEMBER |
ClickHouse | ClickHouse | 626,211,070 | 10,769 | den-crane | Please provide `show create table bi.launcher_tab_play_long2 ` output | 2020-05-09T17:40:11 | 2020-05-09T17:40:11 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 626,213,350 | 10,396 | alexey-milovidov | @abyss7
CPUID library has been broken:
https://clickhouse-test-reports.s3.yandex.net/10761/e33fb5468680f47201c0fcaa91fe5cfc3d39eaba/functional_stateless_tests_(thread)/stderr.log | 2020-05-09T17:57:54 | 2020-05-09T17:57:54 | {} | MEMBER |
ClickHouse | ClickHouse | 625,223,579 | 10,726 | 4ertus2 | Just duplicate here from mail. It's possible to extract difference betwee dates with same text representation with `dateDiff()` function
```
SELECT
toDateTime('2010-10-31 02:30:00', 'Europe/Moscow') AS a,
a + 3600 AS b,
dateDIff('hour', a, b) AS diff
┌───────────────────a─┬───────────────────b─... | 2020-05-07T12:25:53 | 2020-05-07T12:25:53 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 625,803,231 | 10,738 | VictorErmakov | Check priveleges to this folder /etc/clickhouse-server/config.xml | 2020-05-08T12:58:59 | 2020-05-08T12:58:59 | {} | NONE |
ClickHouse | ClickHouse | 625,834,378 | 10,396 | abyss7 | Self-review. Will revert if something goes wrong. | 2020-05-08T14:10:38 | 2020-05-08T14:10:38 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 625,897,496 | 10,754 | shawel | Oh I see. thanks @den-crane. It would be neat if supported. I will close this then, | 2020-05-08T16:26:31 | 2020-05-08T16:26:31 | {} | NONE |
ClickHouse | ClickHouse | 625,288,701 | 10,727 | vzakaznikov | Ok, I will add a stateless test as well. | 2020-05-07T14:27:05 | 2020-05-07T14:27:05 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 626,105,499 | 10,738 | viputh6 | Thanks for your comments... after restarting the system, it started working... issue has been resolved... | 2020-05-09T04:47:21 | 2020-05-09T04:47:21 | {} | NONE |
ClickHouse | ClickHouse | 626,108,084 | 10,680 | Crazylearner30 | Hi Any updates here ?
| 2020-05-09T05:17:36 | 2020-05-09T05:17:36 | {} | NONE |
ClickHouse | ClickHouse | 626,227,324 | 10,768 | filimonov | Most probably you have a bad string there which cant be casted to DateTime. | 2020-05-09T19:53:51 | 2020-05-09T19:53:51 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 625,818,519 | 10,750 | ObjatieGroba | TODO:
1) Run isolated process with ShellCommand
2) Find allocator for shared memory (i mean how to split shared memory for different columns)
3) Fix UDF API
4) UDF Aggregate functions
| 2020-05-08T13:34:32 | 2020-05-08T13:34:32 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 626,066,611 | 10,762 | den-crane | It's completely depends on your queries.
You can make `partition by toYYYYMMDD(timestamp) order by (host, timestamp)` or `partition by to toYYYYMM(timestamp) order by toDate(timestamp)`
Or you can you can create additional table -- `tags (tags_key)` and create `logging.logs ... partition by toYYYYMMDD(timestamp) or... | 2020-05-08T23:48:46 | 2020-05-08T23:49:31 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 626,083,866 | 10,751 | alexey-milovidov | We already have similar option: `tcp_keep_alive_timeout`. | 2020-05-09T01:31:36 | 2020-05-09T01:31:36 | {} | MEMBER |
ClickHouse | ClickHouse | 626,203,345 | 10,751 | alexey-milovidov | Ok. Let's make this option to have the same name and the same semantic as client-side option
(tcp_keep_alive_timeout, that allows to customize this timeout).
And write down in comments in config that client-side setting is also available. | 2020-05-09T16:42:10 | 2020-05-09T16:42:31 | {} | MEMBER |
ClickHouse | ClickHouse | 626,226,002 | 10,656 | filimonov | My first attempts to use the test dont hang... :/ i've played a bit with timeout, but no success yet | 2020-05-09T19:41:46 | 2020-05-09T19:41:46 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 625,892,960 | 10,751 | alexandr-orlov | > Can you add integrational test with enabled option?
I'll make an effort. I've already tested in our internal infrastructure and it works as expecting. | 2020-05-08T16:15:56 | 2020-05-08T16:15:56 | {} | NONE |
ClickHouse | ClickHouse | 626,070,651 | 6,432 | alexey-milovidov | But we must eliminate all induced issues in usability. | 2020-05-09T00:08:32 | 2020-05-09T00:08:32 | {} | MEMBER |
ClickHouse | ClickHouse | 626,120,934 | 10,693 | testlujing | the problem is happen today,messge is:
2020.05.09 12:33:27.871731 [ 594 ] {019cb4e7-89d3-4b65-b6b1-0654767dbc9b} <Error> executeQuery: Code: 173, e.displayText() = DB::ErrnoException: Allocator: Cannot malloc 128.00 KiB., errno: 12, strerror: Cannot allocate memory: (while reading column id): (while reading from par... | 2020-05-09T07:21:17 | 2020-05-09T07:21:17 | {} | NONE |
ClickHouse | ClickHouse | 626,121,684 | 6,432 | filimonov | Check that:
```
set allow_suspicious_low_cardinality_types=1;
drop table if exists test_lc_cond;
create table test_lc_cond ( a LowCardinality(UInt8), b UInt8 ) engine=Memory;
insert into test_lc_cond select rand()%2 as a, a as b from zeros(1000000000);
```
Speed of checking `LowCardinality(UInt8)` vs `UInt8`:
`... | 2020-05-09T07:28:21 | 2020-05-09T07:58:16 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 626,216,638 | 10,761 | alexey-milovidov | TSan failure is unrelated, it was accidentially introduced in #10396 | 2020-05-09T18:23:59 | 2020-05-09T18:23:59 | {} | MEMBER |
ClickHouse | ClickHouse | 625,694,710 | 10,732 | blinkov | > Should we look at some part of this PR?
discussed in chat | 2020-05-08T08:03:37 | 2020-05-08T08:03:37 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 625,896,368 | 10,656 | filimonov | Thank @azat, I have some theory about how to workaround that
Since we a test now I can try to play with that. | 2020-05-08T16:24:03 | 2020-05-08T16:24:03 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 626,129,781 | 3,157 | RalphSu | `sudo dpkg-reconfigure tzdata` works for me. Thanks | 2020-05-09T08:37:53 | 2020-05-09T08:37:53 | {} | NONE |
ClickHouse | ClickHouse | 626,192,637 | 10,766 | alexey-milovidov | Also we need to return "green" result if there are unstable queries. | 2020-05-09T15:27:56 | 2020-05-09T15:27:56 | {} | MEMBER |
ClickHouse | ClickHouse | 626,214,079 | 10,758 | bgiard | Updated the description to pass the failing description test (switched category to only Bug Fix, since we can only have one).
Is there anything else I need to do? The perf test failing doesn't seem related to that change. About the functional tests, I can't really tell why it's failing (because of the skipped tests ma... | 2020-05-09T18:03:48 | 2020-05-09T18:03:48 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 626,216,491 | 10,758 | alexey-milovidov | I have checked and validated the failures manually.
All unrelated failures are collected here: #10775 | 2020-05-09T18:22:42 | 2020-05-09T18:22:42 | {} | MEMBER |
ClickHouse | ClickHouse | 625,227,828 | 10,583 | qoega | also reproducing in 20.3.7.46 | 2020-05-07T12:34:14 | 2020-05-07T12:34:14 | {} | MEMBER |
ClickHouse | ClickHouse | 625,815,207 | 10,742 | den-crane | dup https://github.com/ClickHouse/ClickHouse/issues/6432 not a bug | 2020-05-08T13:27:02 | 2020-05-08T13:28:04 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 626,075,753 | 10,758 | alexey-milovidov | I'm Ok with backward incompatible change. | 2020-05-09T00:35:04 | 2020-05-09T00:35:04 | {} | MEMBER |
ClickHouse | ClickHouse | 626,084,863 | 10,727 | alexey-milovidov | But it still does not. | 2020-05-09T01:39:31 | 2020-05-09T01:39:31 | {} | MEMBER |
ClickHouse | ClickHouse | 626,121,735 | 10,768 | qinzl1 | 
| 2020-05-09T07:28:42 | 2020-05-09T07:28:42 | {} | NONE |
ClickHouse | ClickHouse | 626,208,927 | 10,693 | alexey-milovidov | #4086
#6732
#10488 | 2020-05-09T17:23:18 | 2020-05-09T17:23:18 | {} | MEMBER |
ClickHouse | ClickHouse | 626,210,765 | 10,768 | den-crane | **Describe the bug**
**How to reproduce**
* Which ClickHouse server version to use
* Which interface to use, if matters
* Non-default settings, if any
* `CREATE TABLE` statements for all tables involved
* Sample data for all these tables, use [clickhouse-obfuscator](https://github.com/ClickHouse/ClickHouse/bl... | 2020-05-09T17:37:24 | 2020-05-09T17:37:24 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 625,169,331 | 10,723 | filimonov | 1. Duplicate of https://github.com/ClickHouse/ClickHouse/issues/7258
Will add that soon
2. format_csv_delimiter = '\x01'
| 2020-05-07T10:27:46 | 2020-05-07T10:29:25 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 625,738,224 | 10,741 | hczhcz | See also #10267 | 2020-05-08T09:54:20 | 2020-05-08T09:54:20 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 625,838,130 | 10,460 | qoega | Have you tried it?
| 2020-05-08T14:18:44 | 2020-05-08T14:18:44 | {} | MEMBER |
ClickHouse | ClickHouse | 626,006,850 | 10,759 | filimonov | Set log_queries=1 and check the counters in system.query_log and system.query_thread_log | 2020-05-08T20:30:01 | 2020-05-08T20:30:01 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 626,070,551 | 6,432 | alexey-milovidov | Model example:
```
clickhouse-benchmark <<< "SELECT count() FROM datasets.hits_v1 WHERE NOT ignore(SHA256(toString(SHA256(rand() % 2 ? 'hello' : 'world'))))"
```
0.116 sec.
```
clickhouse-benchmark <<< "SELECT count() FROM datasets.hits_v1 WHERE NOT ignore(SHA256(toString(SHA256(toLowCardinality(rand() % 2) ?... | 2020-05-09T00:08:01 | 2020-05-09T00:08:01 | {} | MEMBER |
ClickHouse | ClickHouse | 626,076,028 | 10,758 | alexey-milovidov | avgWeighted is added recently and not widely used. It's remain compatible if weight is unsigned integer. We will warn users in changelog. | 2020-05-09T00:36:43 | 2020-05-09T00:36:43 | {} | MEMBER |
ClickHouse | ClickHouse | 626,171,048 | 10,751 | alexandr-orlov | > We already have similar option: `tcp_keep_alive_timeout`.
This option is applicable only for client side connection and enables keepalive from the client's end. But we want to avoid dead connections from the server's end. | 2020-05-09T12:47:29 | 2020-05-09T12:47:29 | {} | NONE |
ClickHouse | ClickHouse | 625,236,585 | 10,723 | mdianjun | > 2\. format_csv_delimiter = '\x01'
@filimonov do you mean this:
```
ENGINE = Kafka
SETTINGS
kafka_broker_list = 'xxx:9092',
kafka_topic_list = 'xxx',
kafka_group_name = 'xxx',
kafka_format = 'CSV',
format_csv_delimiter = '\x01',
kafka_num_consumers = 1;
```
got an error:
... | 2020-05-07T12:52:50 | 2020-05-07T12:52:50 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 625,456,520 | 10,723 | filimonov | No, right now you can only make that change globally in default profile (users.xml) | 2020-05-07T19:38:47 | 2020-05-07T19:38:47 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 625,209,738 | 10,553 | qoega | rerun tests for ✅, but previous one looks not related | 2020-05-07T11:56:32 | 2020-05-07T11:56:32 | {} | MEMBER |
ClickHouse | ClickHouse | 625,755,495 | 10,727 | vzakaznikov | @alesapin, is this exception normal?
```
2020.05.08 03:31:24.543230 [ 20 ] {} <Error> default.test_rename_with_parallel_merges: void DB::StorageReplicatedMergeTree::queueUpdatingTask(): Code: 236, e.displayText() = DB::Exception: Log pulling is cancelled, Stack trace (when copying this message, always include the lin... | 2020-05-08T10:42:18 | 2020-05-08T10:42:18 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 625,769,884 | 10,693 | testlujing | 1：ram is enough：
<max_memory_usage>100000000000</max_memory_usage>
2： the issue is not reproduces temporary
3: when the problem happen next ,i will copy log with stracktrace | 2020-05-08T11:24:00 | 2020-05-08T11:24:00 | {} | NONE |
ClickHouse | ClickHouse | 625,784,288 | 10,746 | testlujing | is my problem,sorry | 2020-05-08T12:07:46 | 2020-05-08T12:07:46 | {} | NONE |
ClickHouse | ClickHouse | 625,891,916 | 10,754 | den-crane | SimpleAggregateFunctions are not supported in SummingMergeTree https://github.com/ClickHouse/ClickHouse/issues/8052
They work only with AggregatingMergeTree | 2020-05-08T16:13:30 | 2020-05-08T16:14:25 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 626,017,295 | 10,759 | cw9 | I've found the following data from system.query_log for the replayed queries, but the hang is no longer happening, the replays are all pretty fast now. I feel the hanging is not particularly caused by this query itself, is it possible that other process in the background (merges, replications or other queries) at the t... | 2020-05-08T20:57:21 | 2020-05-08T20:57:21 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 626,122,368 | 10,768 | qinzl1 | if cast or toxxx function can't work , What does it do | 2020-05-09T07:34:06 | 2020-05-09T07:34:06 | {} | NONE |
ClickHouse | ClickHouse | 626,110,215 | 6,432 | filimonov | > It means that LowCardinality(UInt8) makes sense for computations.
Your sample shows that:
1)LowCardinality(String) makes sense for result of `if` when both branches are constant string
2) currently there is a wierd logic when type of `if` result is deducted from type of condition instead of types of branches | 2020-05-09T05:41:02 | 2020-05-09T09:42:08 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 626,202,084 | 10,760 | alexey-milovidov | The test 01247_insert_into_watch_live_view_url - reported to @vzakaznikov | 2020-05-09T16:33:24 | 2020-05-09T16:33:24 | {} | MEMBER |
ClickHouse | ClickHouse | 625,286,706 | 10,730 | qoega | I suppose that a test and docs are needed to ensure current behavior.
AFAIK files are read in alphabetical order like in `ls`. | 2020-05-07T14:23:42 | 2020-05-07T14:23:42 | {} | MEMBER |
ClickHouse | ClickHouse | 625,190,389 | 10,588 | Akazz | @excitoon
It'd be safer to have proper test when fixing such a thing. | 2020-05-07T11:13:41 | 2020-05-07T15:41:42 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 625,532,411 | 10,516 | azat | @alexey-milovidov can you take a look? | 2020-05-07T22:39:10 | 2020-05-07T22:39:10 | {} | MEMBER |
ClickHouse | ClickHouse | 625,581,768 | 8,159 | gubinjie | @yehiaelbehery
It seems that someone solved the problem using docker installation | 2020-05-08T01:39:23 | 2020-05-08T01:39:23 | {} | NONE |
ClickHouse | ClickHouse | 625,878,373 | 10,742 | filimonov | ok, let's continue in old ticket, to avoid duplicates. | 2020-05-08T15:44:24 | 2020-05-08T15:44:24 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 626,074,998 | 10,761 | alexey-milovidov | TSan failure is reported in #10277 | 2020-05-09T00:30:36 | 2020-05-09T00:30:36 | {} | MEMBER |
ClickHouse | ClickHouse | 625,234,612 | 9,959 | hhell | > getTimeZone
A bit of a hack, but it can be implemented by
`if(toTypeName(val) like 'DateTime%(%', splitByChar('\'', toTypeName(val))[2], NULL))`
Example:
```(sql)
select
toDateTime(now(), 'Europe/Moscow') as val,
if(
toTypeName(val) = 'DateTime',
timezone(),
if(
... | 2020-05-07T12:48:49 | 2020-05-07T12:48:49 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 625,384,613 | 10,732 | qoega | Should we look at some part of this PR? | 2020-05-07T17:12:52 | 2020-05-07T17:12:52 | {} | MEMBER |
ClickHouse | ClickHouse | 625,816,327 | 10,405 | 4ertus2 | Do you use `min_bytes_to_use_direct_io`? Would the error be the same it it's 0? | 2020-05-08T13:29:40 | 2020-05-08T13:29:40 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 625,878,079 | 6,432 | filimonov | @alexey-milovidov @den-crane
I don't see any logic behind current behavior, explanations above don't clarify anything.
```
SELECT toTypeName(if(toLowCardinality('a') LIKE 'a', 1, 2))
┌─toTypeName(if(like(toLowCardinality('a'), 'a'), 1, 2))─┐
│ LowCardinality(UInt8) │
└─────... | 2020-05-08T15:43:46 | 2020-05-08T16:06:10 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 626,001,424 | 10,142 | filimonov | > Same issue, but it works before last version
Which version worked? | 2020-05-08T20:20:29 | 2020-05-08T20:20:29 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 626,024,580 | 10,759 | cw9 | btw, during that period of time `17:45:59 - 17:46:09`, other `SELECT` queries are also not logging any logs like `AggregatingTransform: Aggregated.`, they all halted after 17:45:59 and resumed after 17:46:09.
During that 10s, some new `SELECT` queries came in as well but also no progress like `AggregatingTransform:... | 2020-05-08T21:17:18 | 2020-05-08T21:26:47 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 626,132,285 | 10,763 | azat | >What we should add to our CI to check your use case?
Basically I'm using unbundled + shared + split (with some quirks for my particular env), so this is should be smth like:
```sh
docker/packager/packager --unbundled --split-binary
```
P.S. That will be great! | 2020-05-09T08:57:58 | 2020-05-09T08:57:58 | {} | MEMBER |
ClickHouse | ClickHouse | 626,203,725 | 10,733 | alexey-milovidov | ```
Андрей Некрашевич, [08.05.20 11:29]
Привет! А я правильно понимаю, что мне теперь не нужно так (https://github.com/ClickHouse/ClickHouse/pull/10733/files) копипастить эти (https://clickhouse.tech/docs/ru/whats-new/extended-roadmap/#funktsii-dlia-fuzzing) функции, потому что уже есть generateRandom, и сейчас нужно... | 2020-05-09T16:44:46 | 2020-05-09T16:44:46 | {} | MEMBER |
ClickHouse | ClickHouse | 626,209,193 | 6,287 | alexey-milovidov | @CurtizJ This issue is still relevant. | 2020-05-09T17:25:21 | 2020-05-09T17:25:21 | {} | MEMBER |
ClickHouse | ClickHouse | 626,215,879 | 10,277 | alexey-milovidov | Continued in #10775 | 2020-05-09T18:18:01 | 2020-05-09T18:18:01 | {} | MEMBER |
ClickHouse | ClickHouse | 625,323,015 | 8,885 | kssenii | @filimonov , if you want, you can review this there https://github.com/ClickHouse/ClickHouse/pull/10716, but there are still a few weeks till the deadline, so it is not 100% done, but 14 rabbitmq integration tests passed OK :) | 2020-05-07T15:24:36 | 2020-05-07T15:24:36 | {} | MEMBER |
ClickHouse | ClickHouse | 626,084,182 | 10,763 | alexey-milovidov | I see that unbundled build is already Ok: https://clickhouse-builds.s3.yandex.net/10396/0f4dc8329ebd4e6f6ff097aa101cae8d6caa7950/report.html
What we should add to our CI to check your use case? | 2020-05-09T01:34:22 | 2020-05-09T01:34:22 | {} | MEMBER |
ClickHouse | ClickHouse | 626,208,726 | 10,693 | alexey-milovidov | You are not using official build.
If it is debug build, the issue may happen due to the shortage of memory maps. | 2020-05-09T17:21:52 | 2020-05-09T17:21:52 | {} | MEMBER |
ClickHouse | ClickHouse | 626,210,567 | 10,767 | den-crane | **Describe the bug**
**How to reproduce**
* Which ClickHouse server version to use
* Which interface to use, if matters
* Non-default settings, if any
* `CREATE TABLE` statements for all tables involved
* Sample data for all these tables, use [clickhouse-obfuscator](https://github.com/ClickHouse/ClickHouse/blob... | 2020-05-09T17:35:52 | 2020-05-09T17:35:52 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 626,213,740 | 10,396 | alexey-milovidov | The reason is that you *did not even look* at our patches to libcpuid. | 2020-05-09T18:01:04 | 2020-05-09T18:01:24 | {} | MEMBER |
ClickHouse | ClickHouse | 626,220,619 | 10,733 | alexey-milovidov | Let's finish this PR and proceed to other functions. | 2020-05-09T18:55:56 | 2020-05-09T18:55:56 | {} | MEMBER |
ClickHouse | ClickHouse | 626,225,183 | 10,759 | filimonov | Page cache? | 2020-05-09T19:34:43 | 2020-05-09T19:34:43 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 625,266,661 | 6,935 | YoannBuch | Hi @filimonov, what is the status of this ticket?
We changed our schema to use `Array(LowCardinality(String))` and `LowCardinality(String)` for many of our columns and now inserts are a lot slower.
(We are inserting data through a distributed table).
Are there any workarounds? | 2020-05-07T13:47:59 | 2020-05-07T13:48:30 | {} | NONE |
ClickHouse | ClickHouse | 625,702,703 | 10,679 | sundy-li | BTW What's `Yandex synchronization check`? I can't look into the Details.
I'll look into the `Integration tests`, seems the Ephemeral node `/is_acitve` not released when we kill the `node_1_2`
| 2020-05-08T08:23:57 | 2020-05-08T08:23:57 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 625,819,216 | 10,739 | den-crane | clickhouse-client --receive_timeout 900
| 2020-05-08T13:36:08 | 2020-05-08T13:36:08 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 626,035,083 | 10,753 | amosbird | closed in favor of https://github.com/ClickHouse/ClickHouse/pull/10761 | 2020-05-08T21:46:48 | 2020-05-08T21:46:48 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 626,213,440 | 10,396 | alexey-milovidov | Don't revert. I'll fix it for you :) | 2020-05-09T17:58:42 | 2020-05-09T17:58:42 | {} | MEMBER |
ClickHouse | ClickHouse | 626,218,483 | 10,767 | alexey-milovidov | @Crazylearner30 It happens because your data file or table structure is incorrect.
You are trying to insert negative number into unsigned type.
> Unsigned type must not contain '-' symbol | 2020-05-09T18:38:48 | 2020-05-09T18:38:48 | {} | MEMBER |
ClickHouse | ClickHouse | 626,220,575 | 10,733 | alexey-milovidov | Need to add performance test. | 2020-05-09T18:55:31 | 2020-05-09T18:55:31 | {} | MEMBER |
ClickHouse | ClickHouse | 625,446,539 | 8,159 | yehiaelbehery | I also tried all sorts of configurations and it gives me the same errro:
"Cannot parse URI: hdfs://nameservice1, missing port or invalid HA configuration Caused by: HdfsConfigNotFound: Config key: dfs.ha.namenodes.nameservice1 not found"
I tried to use working directory as:
- /etc/clickhouse-server
- /var/lib/... | 2020-05-07T19:16:53 | 2020-05-07T19:16:53 | {} | NONE |
ClickHouse | ClickHouse | 625,757,369 | 10,727 | vzakaznikov | Also, there is a couple of these exceptions.
```
2020.05.08 03:32:36.156681 [ 71 ] {a7595df6-dfc0-43ad-b63d-000e5cb0a3a7} <Error> executeQuery: Code: 48, e.displayText() = DB::Exception: There was an error on [node2:9000]: Cannot execute replicated DDL query on leader (version 20.5.1.3280) (from 172.19.0.1:56838) (... | 2020-05-08T10:48:00 | 2020-05-08T10:48:00 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 625,829,520 | 10,142 | dmitry3325 | Same issue, but it works before last version | 2020-05-08T13:59:27 | 2020-05-08T13:59:27 | {} | NONE |
ClickHouse | ClickHouse | 625,868,172 | 10,751 | qoega | Can you add integrational test with enabled option? | 2020-05-08T15:22:34 | 2020-05-08T15:22:34 | {} | MEMBER |
ClickHouse | ClickHouse | 625,894,088 | 10,740 | filimonov | Most probably you hit https://github.com/ClickHouse/ClickHouse/issues/7260
Recently a reproducible test case for that issue was added, so I will try to fix that.
https://github.com/ClickHouse/ClickHouse/pull/10656 | 2020-05-08T16:18:48 | 2020-05-08T16:26:08 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 626,203,011 | 10,763 | alexey-milovidov | @alesapin What do you think? | 2020-05-09T16:39:45 | 2020-05-09T16:39:45 | {} | MEMBER |
ClickHouse | ClickHouse | 626,217,710 | 9,666 | alexey-milovidov | > Do we want to have such changes in the fork or upstream first ( i.e.
https://github.com/pocoproject/poco) ?
Up to you. | 2020-05-09T18:32:26 | 2020-05-09T18:32:26 | {} | MEMBER |
ClickHouse | ClickHouse | 625,760,102 | 10,383 | KochetovNicolai | I will add more comments to `IAggregateFunction` interface to clarify your questions. | 2020-05-08T10:56:25 | 2020-05-08T10:56:25 | {} | MEMBER |
ClickHouse | ClickHouse | 625,776,489 | 10,727 | vzakaznikov | Updated test `test_rename_with_parallel_ttl_move` shows that renaming column used inside the TTL expression after the data is inserted breaks TTL move. Opened an issue https://github.com/ClickHouse/ClickHouse/issues/10747. | 2020-05-08T11:44:15 | 2020-05-08T12:54:07 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 626,016,276 | 8,159 | yehiaelbehery | @gubinjie OK, any idea about the detailed steps to make it work via docker? | 2020-05-08T20:54:32 | 2020-05-08T20:54:32 | {} | NONE |
ClickHouse | ClickHouse | 626,133,753 | 10,656 | azat | >I have some theory about how to workaround that
Great!
FWIW I was thinking about adding `RD_KAFKA_DESTROY_F_NO_CONSUMER_CLOSE` support to cppkafka and use it during shutdown, that should fix everything but will make it less graceful (I guess this is acceptable in this particular case). | 2020-05-09T09:08:06 | 2020-05-09T09:08:32 | {} | MEMBER |
ClickHouse | ClickHouse | 619,563,205 | 10,208 | tavplubix | Fixed in #7512 | 2020-04-26T14:42:22 | 2020-04-26T14:42:22 | {} | MEMBER |
ClickHouse | ClickHouse | 619,634,840 | 4,548 | alexey-milovidov | There are no updates on this feature. | 2020-04-26T22:30:02 | 2020-04-26T22:30:02 | {} | MEMBER |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.