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 | 737,727,019 | 17,671 | alesapin | 
| 2020-12-03T07:44:23 | 2020-12-03T07:44:23 | {} | MEMBER |
ClickHouse | ClickHouse | 736,974,229 | 17,638 | den-crane | I don't see how it's possible
>select datetime, toStartOfMinute(datetime) as minuteTime,
>2020-11-30 08:32:01, 2020-11-30 10:32:00
for example
```sql
SELECT timezone()
┌─timezone()────┐
│ Europe/Berlin │
└───────────────┘
SELECT
now() AS x,
toStartOfMinute(x) AS y
┌───────────────────x─┬─... | 2020-12-02T03:58:00 | 2020-12-02T03:58:21 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,280,882 | 17,720 | abyss7 | It fixes the build but there is still a test that requires missing scipy Python module | 2020-12-02T14:55:23 | 2020-12-02T14:55:23 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,357,337 | 17,731 | den-crane | you can remove or clean this table `truncate table system.metric_log`
and disable it in config.xml | 2020-12-02T16:51:15 | 2020-12-02T16:52:09 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,451,265 | 17,459 | azat | >Performance — Failed to parse the report.
`countMatches.xml` looks valid (in terms of XML and SQL), @akuzm looks like some unrelated failure (should be simply restarted?), but maybe you have more details? | 2020-12-02T19:39:05 | 2020-12-02T19:39:05 | {} | MEMBER |
ClickHouse | ClickHouse | 737,793,690 | 17,755 | alexey-milovidov | Could you please make pull request to master, not to 20.12? | 2020-12-03T09:43:42 | 2020-12-03T09:43:42 | {} | MEMBER |
ClickHouse | ClickHouse | 737,850,554 | 17,659 | nautaa | @alexey-milovidov PTAL. | 2020-12-03T10:29:26 | 2020-12-03T10:29:26 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,948,826 | 10,644 | nikitamikhaylov | Sorry, I didn't know about this PR. Mann-Whitney test was added here #16883. And it is an AggregateFunction, not pure Function. | 2020-12-03T12:52:19 | 2020-12-03T12:52:19 | {} | MEMBER |
ClickHouse | ClickHouse | 737,324,875 | 17,735 | lucasguo | > 2020.12.02 21:32:38.081092 [ 61763 ] {} <Information> DatabaseOrdinary (default): Starting up tables.
> 2020.12.02 21:32:38.081206 [ 61814 ] {} <Debug> default.big_bill_net_value (ReplicatedMergeTreeRestartingThread): Activating replica.
> 2020.12.02 21:32:38.081260 [ 61815 ] {} <Debug> default.order_test (Replicat... | 2020-12-02T16:01:25 | 2020-12-02T16:01:25 | {} | NONE |
ClickHouse | ClickHouse | 737,338,577 | 17,719 | den-crane | Check this https://stackoverflow.com/a/64381858/11644308 https://stackoverflow.com/a/64390689/11644308
hash sample expr. can be the last in the tail or order by | 2020-12-02T16:22:19 | 2020-12-02T16:22:19 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,383,897 | 17,623 | alexey-milovidov | What I forgot is: pre-configured named connections with credentials in config - to avoid specifying password in command line.
Added to the list of small tasks. | 2020-12-02T17:36:20 | 2020-12-02T17:36:20 | {
"+1": 2
} | MEMBER |
ClickHouse | ClickHouse | 737,963,240 | 17,754 | filimonov | Thanks for a note :+1:
It seems no immediate actions are needed.
> Docker-produced images will continue to work in your cluster with all runtimes, as they always have.
Maybe we can also consider switching from docker to some nicer alternatives - like podman and nicer build tools like buildah, img (in CI).
... | 2020-12-03T13:02:25 | 2020-12-03T13:02:25 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 736,824,603 | 17,115 | filimonov | https://github.com/ClickHouse/ClickHouse/issues/16176 | 2020-12-01T21:14:34 | 2020-12-01T21:14:34 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 736,834,129 | 17,705 | MasterOdin | Alright, makes sense. Thanks for the feedback @den-crane, I will open a follow-up PR to adjust wording then on the system page, and have made your suggested changes here. | 2020-12-01T21:33:18 | 2020-12-01T21:33:18 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,089,002 | 17,624 | marising | > 1. llvm is an optional component, you can disable it with CMake flag
> 2. you're trying to compile an obsolete version. Use 20.8 at least.
Getting the latest image compilation 20.6 branch has the above problem.
Later, getting the image of 20.6 to compile the branch of 20.6 is no problem.
docker pull yandex... | 2020-12-02T08:56:15 | 2020-12-02T08:59:31 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,281,174 | 17,614 | TCeason | @tavplubix Need a force test label. | 2020-12-02T14:55:48 | 2020-12-02T14:55:48 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,356,248 | 17,731 | Farfaday | I could not find the string "mutation" in the logs.
But I filtered out all the
` DiskLocal: Reserving 1.00 MiB on disk default, having unreserved 16.90 TiB`
and now I can see messages of the form flowing
```
2020.12.02 16:47:41.580028 [ 40550 ] {} <Trace> SystemLog (system.metric_log): Flushing system log, 8... | 2020-12-02T16:49:27 | 2020-12-02T16:49:27 | {} | NONE |
ClickHouse | ClickHouse | 737,412,457 | 17,737 | tavplubix | Stress test (address) and Stress test (memory) - #17744 | 2020-12-02T18:27:56 | 2020-12-02T18:27:56 | {} | MEMBER |
ClickHouse | ClickHouse | 737,517,549 | 17,446 | RonanMorgan | ok, thanks, I will try tomorrow.
I just finished some tests with tshark (command colocated with my clickhouse servers), it really seems to me that all the failing requests have the following headers ...
```
Transfer-Encoding: chunked
Host: myserver.kameleoon.net:8123
Connection: Keep-Alive
User-Agent: Apache-Ht... | 2020-12-02T21:50:52 | 2020-12-02T21:54:43 | {} | NONE |
ClickHouse | ClickHouse | 737,546,824 | 17,735 | lucasguo | > You can disable system logs
>
> ```
> cat /etc/clickhouse-server/conf.d/z_log_disable.xml
> <?xml version="1.0"?>
> <yandex>
> <asynchronous_metric_log remove="1"/>
> <metric_log remove="1"/>
> <query_log remove="1"/>
> <query_thread_log remove="1"/>
> <trace_log remove="1"/>
> </y... | 2020-12-02T22:58:37 | 2020-12-02T22:58:37 | {} | NONE |
ClickHouse | ClickHouse | 737,709,418 | 17,752 | chenk818 | > Update to 20.8.7, reopen ticket if issue persist after update.
>
> See:
> #16264
thanks | 2020-12-03T07:01:48 | 2020-12-03T07:01:48 | {} | NONE |
ClickHouse | ClickHouse | 737,734,970 | 17,753 | l1t1 | the bigmap would be too big(2^128 / 8 =4.2 *10^37 | 2020-12-03T08:03:11 | 2020-12-03T08:03:11 | {} | NONE |
ClickHouse | ClickHouse | 737,743,733 | 16,621 | ilejn | Personally I have no preferences between (1) and (3) while slightly less comfortable with (2).
As far as I understand, we tend to (1), e.g. because ability to configure libhdfs3 seems really needed https://github.com/ClickHouse/ClickHouse/issues/8159
@alesapin , @filimonov further suggestions? | 2020-12-03T08:21:19 | 2020-12-03T08:21:19 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,099,368 | 17,623 | alexey-milovidov | @hexiaoting
> Will any schedule strategies for priority/resource-reserved be added to 2021 roadmap?
It is in the list of main tasks, named "Workload management".
Actually it was in roadmap for 2020 but the task was failed (not because it is difficult to implement, but because it was not even started).
This ... | 2020-12-02T09:15:17 | 2020-12-02T09:18:44 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 737,153,136 | 3,066 | Daletxt | I can use
"""INSERT INTO table_name (the_non-nullable_type_field, other_field) SELECT ifNull('contain_NULL_value_field', 'your_target_replace_value'), other_field FROM one_table;"""
to solve this problem. | 2020-12-02T10:52:48 | 2020-12-02T10:55:05 | {} | NONE |
ClickHouse | ClickHouse | 737,326,572 | 17,735 | lucasguo | Then a lot of log similar to that about adding mutation for partition. After that is
> 2020.12.02 21:32:38.133709 [ 61809 ] {} <Trace> default.quote (ReplicatedMergeTreeQueue): Marking mutation 0000000005 done because it is <= mutation_pointer (0000000005)
> 2020.12.02 21:32:38.136650 [ 61819 ] {} <Information> defa... | 2020-12-02T16:03:39 | 2020-12-02T16:03:39 | {} | NONE |
ClickHouse | ClickHouse | 737,375,454 | 17,735 | lucasguo | The result of ls
> total 16
> -rw-r-----. 1 clickhouse clickhouse 12999 Dec 2 18:23 metric_log.sql
The content in metric_log.sql
> ATTACH TABLE _ UUID 'ed5202c5-1d07-41b6-958f-836dc8c2cb32'
> (
> `event_date` Date,
> `event_time` DateTime,
> `milliseconds` UInt64,
> `ProfileEvent_Query` UInt... | 2020-12-02T17:21:33 | 2020-12-02T17:21:33 | {} | NONE |
ClickHouse | ClickHouse | 737,689,440 | 17,753 | filimonov | Afaik roaring bitmaps don't support it.
You can ask there, similar to
https://github.com/RoaringBitmap/RoaringBitmap/issues/109 | 2020-12-03T06:10:58 | 2020-12-03T06:10:58 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,714,564 | 17,379 | hoverzheng | thanks @traceon.
I used the command:
$ CC=$(brew --prefix llvm)/bin/clang CXX=$(brew --prefix llvm)/bin/clang++ cmake ..
$ ninja
and got the error:
FAILED: contrib/protobuf/protoc-3.13.0.0
: && ccache /usr/local/opt/llvm/bin/clang++ -fdiagnostics-color=always -std=c++2a -fsized-deallocation -gdwarf-arang... | 2020-12-03T07:13:47 | 2020-12-03T07:13:47 | {} | NONE |
ClickHouse | ClickHouse | 737,928,085 | 5,538 | ygj6 | ```
drop table if exists one;
CREATE TABLE one (`id` UInt32,`val` UInt32,`name` String)
ENGINE = MergeTree()
order by id;
insert into one (id, val) values (1,11),(2,12),(3,13);
drop table if exists two;
CREATE TABLE two (`id` UInt32,`word` String)
ENGINE = Join(ANY, LEFT , id);
insert into two values (1,'one... | 2020-12-03T12:37:19 | 2020-12-03T12:43:03 | {} | NONE |
ClickHouse | ClickHouse | 737,321,009 | 17,735 | lucasguo | > Need more logs. What was before "File exists: /data/clickhouse/store/c3b/c3bfe806-105d-4310-a2c5-ab0e58a02448"
Nothing more before File exists in err.log file. However, there are many log in clickhouse-server.log when startup. It's too large to post the entire content. Here I show you some lines before and after f... | 2020-12-02T15:56:09 | 2020-12-02T15:56:09 | {} | NONE |
ClickHouse | ClickHouse | 737,349,252 | 17,446 | RonanMorgan | Hello,
After more investigation I just noticed that
1) even after retries, these requests are failing again and again so it has to be reproducible
2) indeed clickhouse is receiving a "Transfer Encoding: identity" without "Length" parameters
```
2020.12.02 16:13:15.557926 [ 2440479 ] {8851d60d-95d7-4464-8131-... | 2020-12-02T16:38:05 | 2020-12-02T16:39:31 | {} | NONE |
ClickHouse | ClickHouse | 737,383,328 | 16,931 | abyss7 | Was fixed in another way | 2020-12-02T17:35:22 | 2020-12-02T17:35:22 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,387,397 | 17,739 | abyss7 | What kind of errors and in which scenarios have you encountered? | 2020-12-02T17:42:42 | 2020-12-02T17:42:42 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,404,330 | 17,735 | den-crane | >Should I delete everything under clickhouse/metadata?
No | 2020-12-02T18:12:29 | 2020-12-02T18:12:29 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,507,107 | 17,747 | tavplubix | Duplicate of #13281 | 2020-12-02T21:29:32 | 2020-12-02T21:29:32 | {} | MEMBER |
ClickHouse | ClickHouse | 737,106,430 | 17,714 | alexey-milovidov | This is not in our plans.
It can be implemented (maybe under a setting) if it will not compromise performance of data parsing in normal cases. | 2020-12-02T09:28:22 | 2020-12-02T09:28:22 | {} | MEMBER |
ClickHouse | ClickHouse | 737,646,185 | 17,752 | chenk818 | @den-crane can you help me | 2020-12-03T03:43:30 | 2020-12-03T03:43:30 | {} | NONE |
ClickHouse | ClickHouse | 737,695,483 | 17,749 | filimonov | Afaik no way for that (well - maybe export to some format and import back). | 2020-12-03T06:26:47 | 2020-12-03T06:26:47 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,719,901 | 17,614 | TCeason | @tavplubix I think the test is passed. | 2020-12-03T07:26:41 | 2020-12-03T07:26:41 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,999,670 | 17,459 | akuzm | Our CI infrastructure is just completely dead, so even the fix can't roll out... I hope we'll fix it in a couple of days. If not, will have to merge w/o checking the perf test. | 2020-12-03T13:39:21 | 2020-12-03T13:39:21 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 736,839,876 | 16,947 | azat | >In these two cases, why does the initiator care which alias the column has on a shard?
Using only positions in the block for blocks from other storages (via network) does not looks reliable to me, for example there was the following issue - https://github.com/ClickHouse/ClickHouse/pull/10621
For example in the f... | 2020-12-01T21:44:47 | 2020-12-01T21:49:49 | {} | MEMBER |
ClickHouse | ClickHouse | 736,927,711 | 17,709 | kssenii | I'll add a test later. | 2020-12-02T01:29:48 | 2020-12-02T01:29:48 | {} | MEMBER |
ClickHouse | ClickHouse | 737,131,680 | 15,179 | tavplubix | So it's still reproducible with v20.13.1.5273-testing, right? | 2020-12-02T10:13:26 | 2020-12-02T10:13:26 | {} | MEMBER |
ClickHouse | ClickHouse | 737,217,424 | 15,988 | vitlibar | I suppose it's better to use a ready out-of-the-box solution. I mean the class `ExpireLRUCache` which Poco provides. It's thread-safe, it's well-tested, and it already can do what you want. Check it out:
```
class ExternalAuthenticators
{
...
private:
using CacheOfCorrectPasswords = ExpireLRUCache<UUID ... | 2020-12-02T13:03:53 | 2020-12-02T13:03:53 | {} | MEMBER |
ClickHouse | ClickHouse | 737,328,349 | 17,735 | den-crane | check what is this file `ls -l /data/clickhouse/store/c3b/c3bfe806-105d-4310-a2c5-ab0e58a02448`
try to remove everything from `rm -rf clickhouse/data/system/*` | 2020-12-02T16:06:12 | 2020-12-02T16:06:12 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,685,788 | 17,752 | filimonov | Update to 20.8.7, reopen ticket if issue persist after update.
See:
https://github.com/ClickHouse/ClickHouse/pull/16264 | 2020-12-03T06:00:21 | 2020-12-03T06:00:21 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,474,037 | 17,681 | Akazz | Failing [**Functional stateless tests**](https://clickhouse-test-reports.s3.yandex.net/17681/62649d0dd0777f3080319830601e4abc56f4d363/functional_stateless_tests_(memory).html#fail1) in 62649d0 seem to be flapping in *01304_direct_io* (because of "... not enough space ...").
Stress tests most probably fail due to not... | 2020-12-02T20:23:17 | 2020-12-03T07:40:49 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,154,509 | 17,664 | tavplubix | SIGSEGV in Stress tests is reproducible in master: https://gist.github.com/tavplubix/6c279ad90b0554ec824671b836b43c24
`StoragePolicy::reserve(...)` [may](https://github.com/ClickHouse/ClickHouse/blob/4b65d5469f441bac67e7026ffb9a8f16337fcd65/src/Disks/StoragePolicy.cpp#L203) return `nullptr`, but we [do not](https://... | 2020-12-02T10:55:31 | 2020-12-02T10:55:31 | {} | MEMBER |
ClickHouse | ClickHouse | 737,424,974 | 14,150 | azat | >but it is not very clear for me why in some cases we can use separate threads, for example here:
This is for SELECT query, and:
- it respect max_threads setting (so if I will set max_threads=1, it will not do parallel loading, and also max_threads will be lowered to 1 in case of simple SELECT with LIMIT)
- it is ... | 2020-12-02T18:49:11 | 2020-12-02T18:58:07 | {} | MEMBER |
ClickHouse | ClickHouse | 737,681,485 | 17,731 | detailyang | same issues here (clickhouse version 20.11.4.13), after heap profiling, guess it's because wal restore many tiny parts which took more memory when deserializeBinaryBulk. See the jemalloc heap profiling in `--inuse_space` mode [https://gist.github.com/detailyang/0eba31edae8d437ee6cd51555bd5080b](https://gist.github.com/... | 2020-12-03T05:47:35 | 2020-12-03T05:55:49 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 736,867,401 | 14,160 | l1t1 | GLOBAL in | 2020-12-01T22:44:05 | 2020-12-01T22:44:05 | {} | NONE |
ClickHouse | ClickHouse | 737,738,296 | 17,385 | TimonKK | Thanks for answer, I will try again and check md5sum | 2020-12-03T08:10:16 | 2020-12-03T08:10:16 | {} | NONE |
ClickHouse | ClickHouse | 737,779,042 | 12,589 | tangxueming | My version is 20.11.4. There is 30 billions+ rows in my dataset.It spends 30+ seconds for my query with a bloom fliter skipping index (id column).
Example: select * from my_table where id = 'adasdasdqwdz';
Similarly to #7046, nothing happen for long time before appearing the progress bar.
create table query like:... | 2020-12-03T09:28:36 | 2020-12-03T09:28:36 | {} | NONE |
ClickHouse | ClickHouse | 737,852,758 | 17,659 | alexey-milovidov | Now we need to make an SQL function:
```
SELECT encodeXMLComponent('Hello, "world"!')
``` | 2020-12-03T10:31:20 | 2020-12-03T10:31:20 | {} | MEMBER |
ClickHouse | ClickHouse | 737,860,008 | 17,755 | jetgm | open a new request on master branch and close this tickets. | 2020-12-03T10:43:00 | 2020-12-03T10:43:00 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,984,292 | 13,541 | alanpaulkwan | Further, it would be great to add Parquet codec compression *levels* e.g. ZSTD(6). | 2020-12-03T13:16:41 | 2020-12-03T13:16:41 | {} | NONE |
ClickHouse | ClickHouse | 737,081,070 | 17,623 | hexiaoting | Right now in default mode, CH will use many threads(the same number of cpu cores) to execute a query.
So if we have many parallel queries , the cpu and memory resources will be in competition mode. May be some high-priority queries will be delayed and delayed.
Will any schedule strategies for priority/resource-reser... | 2020-12-02T08:41:12 | 2020-12-02T08:41:12 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,108,150 | 17,623 | hexiaoting | OK，we are looking forward to "Workload management" be implemented. | 2020-12-02T09:31:10 | 2020-12-02T09:31:10 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,130,587 | 16,724 | kitaisreal | @alexey-milovidov make test implementation of allowing different types inside in subquery. Please check 01600_select_in_different_types.sql and Set.cpp for a reference. | 2020-12-02T10:11:18 | 2020-12-02T10:11:18 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,307,790 | 17,735 | den-crane | Need more logs. What was before "File exists: /data/clickhouse/store/c3b/c3bfe806-105d-4310-a2c5-ab0e58a02448" | 2020-12-02T15:35:12 | 2020-12-02T15:35:12 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,586,931 | 17,735 | lucasguo | > it does not matter what happened after
>
> > 2020.12.03 06:53:23.987356 [ 180648 ] {} Application: Shutting down storages.
>
> Try to replace clickhouse configs with standard config files from package, leave only config.xml and user.xml from deb
I am appreciate your help.
Finally replaced the config file wi... | 2020-12-03T00:49:41 | 2020-12-03T00:49:41 | {
"+1": 1
} | NONE |
ClickHouse | ClickHouse | 736,892,017 | 16,947 | akuzm | > > In these two cases, why does the initiator care which alias the column has on a shard?
>
> Using only positions in the block for blocks from other storages (via network) does not looks reliable to me, for example there was the following issue - #10621
>
> For example in the future clickhouse can do some "opti... | 2020-12-01T23:43:33 | 2020-12-01T23:43:33 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,097,547 | 17,623 | alexey-milovidov | @filimonov
> smth about multilevel Merge-tree partitioning
Currently not in roadmap, we did not collect the use cases, we did not think about implementation details, the task is unclear. Maybe will be added later.
> dedup on MergeTree (block level)
It's in the stage of "staled review". I added it to the li... | 2020-12-02T09:11:54 | 2020-12-02T09:11:54 | {
"+1": 2
} | MEMBER |
ClickHouse | ClickHouse | 737,307,346 | 17,735 | den-crane | check `ps -ef|grep clickhouse` | 2020-12-02T15:34:36 | 2020-12-02T15:34:36 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,463,068 | 17,446 | den-crane | >User-Agent: none, Content Type: , Transfer Encoding: identity
none ? It's something bad happen in jdbc or apach-httpclient
Try to use another jvm not 1.8.0_252, something newer | 2020-12-02T20:02:04 | 2020-12-02T20:03:12 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,549,833 | 17,735 | den-crane | it does not matter what happened after
>2020.12.03 06:53:23.987356 [ 180648 ] {} Application: Shutting down storages.
Try to replace clickhouse configs with standard config files from package, leave only config.xml and user.xml from deb | 2020-12-02T23:06:54 | 2020-12-02T23:06:54 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,729,205 | 4,959 | stufently | I have same problem in latest release clickhouse-server 20.11.4.13
I try fix it run
check table events;
but after clickhouse move part to detached , this part replicate from other replica with same problem.
seems this part broken on all replicas
`2020.12.02 12:27:29.607426 [ 144664 ] {} <War... | 2020-12-03T07:49:58 | 2020-12-03T07:49:58 | {} | NONE |
ClickHouse | ClickHouse | 737,890,200 | 12,589 | Slach | @tangxueming
could you open `clickhouse-client `
and run
```
SET send_logs_level='trace';
SELECT * FROM my_table WHERE s = 'adasdasdqwdz';
```
and try to figure out with logs | 2020-12-03T11:43:22 | 2020-12-03T11:43:35 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,954,165 | 13,334 | nikitamikhaylov | @achimbab I will convert this PR to draft, because it won't be merged soon. | 2020-12-03T12:56:15 | 2020-12-03T12:56:15 | {} | MEMBER |
ClickHouse | ClickHouse | 736,914,009 | 15,179 | chang0391 | @zhang2014
Can you take care of this issue?
@tavplubix
this issue should not be closed, it‘s always exist
| 2020-12-02T00:47:21 | 2020-12-02T02:30:09 | {} | NONE |
ClickHouse | ClickHouse | 737,341,448 | 17,715 | den-crane | duplicate https://github.com/ClickHouse/ClickHouse/issues/2592 | 2020-12-02T16:26:51 | 2020-12-02T16:26:51 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,404,868 | 17,735 | den-crane | You can disable system logs
```
cat /etc/clickhouse-server/conf.d/z_log_disable.xml
<?xml version="1.0"?>
<yandex>
<asynchronous_metric_log remove="1"/>
<metric_log remove="1"/>
<query_log remove="1"/>
<query_thread_log remove="1"/>
<trace_log remove="1"/>
</yandex>
+restart
```
... | 2020-12-02T18:13:32 | 2020-12-02T18:13:32 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,511,595 | 17,738 | filimonov | > Curl is the most universal name for how to run a query.
questionable. The most natural way to run query in clickhouse is to use clickhouse-client.
Exposing a single port (HTTP only) looks weird. | 2020-12-02T21:38:00 | 2020-12-02T21:38:00 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,543,683 | 5,604 | ralphM78 | Any news about that ?
Ralph | 2020-12-02T22:50:57 | 2020-12-02T22:50:57 | {} | NONE |
ClickHouse | ClickHouse | 737,718,696 | 17,379 | traceon | @hoverzheng run this in the `ClickHouse/` repo dir:
```sh
git submodule update --init --recursive && git pull --recurse-submodule
```
...and then clean everything in build folder and run fresh `cmake` configure and build. | 2020-12-03T07:23:40 | 2020-12-03T07:24:32 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,027,782 | 16,674 | hexiaoting | The values inserted into table influence select??? a little weird.
```
┌─version()─┐
│ 20.13.1.1 │
└───────────┘
drop table p;
create table p (i int, b int) engine File(Parquet);
insert into p select number,number from numbers(5,5);
insert into p select number,number from numbers(1,5);
select * from p;
┌─... | 2020-12-02T06:40:13 | 2020-12-02T06:41:56 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,044,483 | 17,710 | yd1261137568 | SELECT * FROM id_val ANY LEFT JOIN id_val_join USING (id) SETTINGS join_use_nulls = 1;
Exception: Table test.id_val_join needs the same join_use_nulls setting as present in LEFT or FULL JOIN | 2020-12-02T07:18:54 | 2020-12-02T07:18:54 | {} | NONE |
ClickHouse | ClickHouse | 737,293,795 | 15,988 | traceon | I didn't use `ExpireLRUCache` because I didn't like the idea of decoupling something that is already perfectly mapped. There is one-to-one correspondence between the cached info and the user info. This decoupling is not necessary, and provides additional orphan cache entry cleanup problems. Also, it looks like `ExpireL... | 2020-12-02T15:14:52 | 2020-12-02T15:14:52 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,348,943 | 17,731 | den-crane | Any messages about mutations in the startup logs? | 2020-12-02T16:37:34 | 2020-12-02T16:37:34 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,517,619 | 17,738 | filimonov | Since we are in a docker conventions you can show a docker example for that, smth like:
```
echo "SELECT * FROM system.build_options" | docker run -i --rm --link some-clickhouse-server:clickhouse-server curlimages/curl 'http://clickhouse-server:8123/?query=' -s --data-binary @-
```
People using docker should kno... | 2020-12-02T21:51:04 | 2020-12-02T21:51:04 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,727,057 | 17,671 | alesapin | Not so bad | 2020-12-03T07:44:29 | 2020-12-03T07:44:29 | {} | MEMBER |
ClickHouse | ClickHouse | 737,913,527 | 17,565 | alexey-milovidov | It breaked multiline history navigation. | 2020-12-03T12:24:14 | 2020-12-03T12:24:14 | {} | MEMBER |
ClickHouse | ClickHouse | 736,940,091 | 17,682 | amosbird | It isn't related to `groupArraySample` but `distinct`. | 2020-12-02T02:08:23 | 2020-12-02T02:08:33 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 737,111,117 | 3,489 | alexey-milovidov | Yes, it's relevant. | 2020-12-02T09:36:22 | 2020-12-02T09:36:22 | {
"eyes": 3
} | MEMBER |
ClickHouse | ClickHouse | 737,258,567 | 17,346 | ralphM78 | So ```PARTITION REPLACE``` hasn't worked because of different storage policies. But ```DETACH``` and ```ATTACH``` between a ReplacingMergetree and the ReplicatedReplacingMergetree has worked. Thanks for helping.
Ralph | 2020-12-02T14:19:35 | 2020-12-02T14:19:35 | {} | NONE |
ClickHouse | ClickHouse | 737,467,694 | 17,738 | den-crane | Why only 8123 ?
Why not -p 8123:8123 -p 9000:9000 -p 9014:9014
http/tcp/mysql | 2020-12-02T20:10:45 | 2020-12-02T20:10:45 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,587,732 | 17,739 | spongedu | > What kind of errors and in which scenarios have you encountered?
@abyss7 We customized table engine of `system.part_log` by setting `<engine>` option in `config.xml`。Unfortunately we made a grammar error and before we found out what happened, the service have been running for a while. So I think it may be good if... | 2020-12-03T00:52:22 | 2020-12-03T00:52:40 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,728,301 | 17,459 | azat | >It was a global CI problem, should be fixed now, I'll add an empty commit to restart.
@akuzm does not help, by some reason
```sql
Performance — Failed to parse the report.
``` | 2020-12-03T07:47:34 | 2020-12-03T07:47:34 | {} | MEMBER |
ClickHouse | ClickHouse | 737,308,661 | 17,735 | den-crane | can you show the output of
`sudo -u clickhouse /usr/bin/clickhouse-server --config-file /etc/clickhouse-server/config.xml` | 2020-12-02T15:36:31 | 2020-12-02T15:36:31 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,316,736 | 17,735 | lucasguo | > check `ps -ef|grep clickhouse`
[lucas@server-81 ~]$ ps -ef|grep clickhouse
lucas 66550 54348 0 23:48 pts/0 00:00:00 grep --color=auto clickhouse | 2020-12-02T15:49:17 | 2020-12-02T15:49:17 | {} | NONE |
ClickHouse | ClickHouse | 737,323,453 | 17,732 | den-crane | Distributed(s4r1, metrics, dws_3, doc_id%100) -- if you insert using distributed then you can use
--distributed_group_by_no_merge arg If 1, Do not merge aggregation states from different servers for distributed query processing - in case it is for certain that there are different keys o... | 2020-12-02T15:59:32 | 2020-12-02T15:59:32 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,371,561 | 17,735 | lucasguo | After deleting everything under clickhouse/data/system/* and try to restart again, unfortunately the same error occurred. Should I delete everything under clickhouse/metadata? | 2020-12-02T17:14:39 | 2020-12-02T17:14:39 | {} | NONE |
ClickHouse | ClickHouse | 737,474,742 | 17,738 | vklimontovich | Well, the purpose of this section is to provide a very quick working example, it's unlikely that people will use it for production. Curl is the most universal name for how to run a query. And it's unlikely users will start with connecting to it through MySQL binary protocol (and for TCP protocol there's another way in ... | 2020-12-02T20:24:42 | 2020-12-02T20:24:42 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,485,898 | 17,459 | akuzm | > > Performance — Failed to parse the report.
>
> `countMatches.xml` looks valid (in terms of XML and SQL), @akuzm looks like some unrelated failure (should be simply restarted?), but maybe you have more details?
It was a global CI problem, should be fixed now, I'll add an empty commit to restart. | 2020-12-02T20:47:02 | 2020-12-02T20:47:02 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 737,537,456 | 17,731 | Farfaday | Hi,
Thanks for you reply ! I tried that (truncating the table + removing all the log stuffs from the config (query_log, metric_log, etc.)), but I am still affected by the issue described in https://github.com/ClickHouse/ClickHouse/issues/17731#issue-755226885 | 2020-12-02T22:35:45 | 2020-12-02T22:35:45 | {} | NONE |
ClickHouse | ClickHouse | 737,691,789 | 15,179 | chang0391 | yes, the issue is still exist | 2020-12-03T06:17:13 | 2020-12-03T06:17:13 | {} | NONE |
ClickHouse | ClickHouse | 725,899,954 | 16,804 | vdimir | Performance test isn't ready, before it I'd like to implement #16616 to use `SOURCE(CLICKHOUSE(...))`.
Also It may require careful testing on different platforms (especially not little-endian).
This points (performance and testing) may lead some changes in code, but I hope this changes won't be too big.
I would r... | 2020-11-12T07:41:00 | 2020-11-12T07:51:11 | {} | MEMBER |
ClickHouse | ClickHouse | 725,982,598 | 16,909 | dependabot-preview[bot] | OK, I won't notify you again about this release, but will get in touch when a new version is available. If you'd rather skip all updates until the next major or minor version, let me know by commenting `@dependabot ignore this major version` or `@dependabot ignore this minor version`.
If you change your mind, just re-... | 2020-11-12T10:15:16 | 2020-11-12T10:15:16 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 726,013,851 | 11,000 | pprogs | But it somehow works if you put full table names
this doesnt work
```sql
from customer as cust
left any join customer_details_v as det
on cust.CustomerID = det.CustomerID;
```
this works
```sql
from test.customer as cust
left any join test.customer_details_v as det
on ... | 2020-11-12T11:15:46 | 2020-11-12T11:16:31 | {} | NONE |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.