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 | 678,355,529 | 13,930 | den-crane | > Is my understanding correct if I say that the engine "VersionedCollapsingMergeTree" does not have this limitation (order of inserts can by anything)?
correct. | 2020-08-21T15:37:27 | 2020-08-21T15:37:27 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,896,607 | 12,390 | etah000 | I want to add setting variables as you required, but for now, the clickhouse codes are still complicated for me, and I cannot find a suitable way to modify the codes. I don't want the request causes too much trouble for you and me.
I am glad to hear that you plan to merge it, but I cannot modify the code as you req... | 2020-08-24T04:27:57 | 2020-08-24T04:31:05 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 679,087,582 | 13,980 | alexey-milovidov | Looks like stress test is fixed but we have something strange with stateful tests... | 2020-08-24T12:08:25 | 2020-08-24T12:08:25 | {} | MEMBER |
ClickHouse | ClickHouse | 678,453,187 | 13,957 | den-crane | JSON is output format only. CH could not input JSON. https://clickhouse.tech/docs/en/interfaces/formats/#json | 2020-08-21T19:24:53 | 2020-08-21T19:24:53 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,444,923 | 13,958 | den-crane | Yes, it seems related to LowCardinality
20.8.1
```
SELECT
count(),
color,
toStartOfSecond(timestamp) AS second
FROM order_test AS i
GROUP BY
color,
second
ORDER BY
color ASC,
second DESC
LIMIT 500
┌─count()─┬─color─┬──────────────────second─┐
│ 250000 │ blue │ 2020-08... | 2020-08-21T19:04:44 | 2020-08-21T21:10:03 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,771,036 | 13,929 | alexey-milovidov | > 01396_inactive_replica_cleanup_nodes
Broken in #13402. | 2020-08-23T12:54:53 | 2020-08-23T12:54:53 | {} | MEMBER |
ClickHouse | ClickHouse | 678,771,293 | 13,947 | alexey-milovidov | https://github.com/ClickHouse/ClickHouse/issues/12508 | 2020-08-23T12:56:56 | 2020-08-23T12:56:56 | {} | MEMBER |
ClickHouse | ClickHouse | 678,831,628 | 8,367 | alexey-milovidov | It's https://github.com/ClickHouse/ClickHouse/issues/13348, @KochetovNicolai | 2020-08-23T22:15:50 | 2020-08-23T22:15:50 | {} | MEMBER |
ClickHouse | ClickHouse | 679,039,654 | 10,913 | 4ertus2 | Reproduced
```
drop table if exists t;
create table t (dt Date) engine = MergeTree ORDER BY dt PARTITION BY dt;
insert into t values (0);
set force_index_by_date = 1;
select count(*) from t l join t r
using dt
where l.dt < today();
drop table t;
``` | 2020-08-24T10:14:33 | 2020-08-24T10:14:33 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,344,579 | 13,928 | alexey-milovidov | It's not a bugfix. | 2020-08-21T15:14:42 | 2020-08-21T15:14:42 | {} | MEMBER |
ClickHouse | ClickHouse | 678,358,649 | 13,402 | alexey-milovidov | Let's check if the test `01396_inactive_replica_cleanup_nodes` will fix if we revert.
If not, we can apply this PR again... | 2020-08-21T15:44:07 | 2020-08-21T15:44:07 | {} | MEMBER |
ClickHouse | ClickHouse | 678,953,001 | 12,031 | vitlibar | > does this PR fix this kind of errors?
>
> > GRANT ALL ON `xxx`.* TO `yyyy`
> > Clickhouse error when executing command: Code: 446, Message: default: Introspection functions are disabled
>
> (with Clickhouse 20.6.3.28)
I'm fixing that error in https://github.com/ClickHouse/ClickHouse/pull/13987 | 2020-08-24T07:17:10 | 2020-08-24T07:17:10 | {} | MEMBER |
ClickHouse | ClickHouse | 678,983,281 | 13,956 | excitoon | ```
01396_inactive_replica_cleanup_nodes | OK | 517.80
-- | -- | --
``` | 2020-08-24T08:22:19 | 2020-08-24T08:22:19 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 679,008,715 | 8,661 | filimonov | #13953 | 2020-08-24T09:12:25 | 2020-08-24T09:12:25 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 679,052,856 | 4,432 | MrSsunlight | I encountered the same problem in version 20.3.11.97 | 2020-08-24T10:43:16 | 2020-08-24T10:43:16 | {} | NONE |
ClickHouse | ClickHouse | 678,351,959 | 13,937 | bohutang | Releated to https://github.com/ClickHouse/ClickHouse/issues/13953 | 2020-08-21T15:29:44 | 2020-08-21T15:29:44 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,476,343 | 12,748 | den-crane | fixed https://github.com/ClickHouse/ClickHouse/pull/13887
```
20.6.4.44
┌─attr_name_lang_ru───────────────────┬─hex(MD5(attr_name_lang_ru))──────┬─hex(object_id)───────────┬─version─┬─change_id─┬─sign─┬─_part─────────────────────┐
....
│ Хабаровский край │ AF942E528062C1FF9AB72D85F78F0AE5 │ 5F... | 2020-08-21T20:22:04 | 2020-08-21T20:22:04 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,481,737 | 13,936 | den-crane | duplicate https://github.com/ClickHouse/ClickHouse/issues/12602
**You need to upgrade to 20.5.5.74** | 2020-08-21T20:35:44 | 2020-08-21T20:35:44 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,609,815 | 13,942 | excitoon | Merged to #13956 | 2020-08-22T07:47:50 | 2020-08-22T07:47:50 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 679,077,155 | 13,862 | 4ertus2 | > flamegraph from debug build
Debug flame graph shows that we are waiting in `PipelineExecutor::executeStepImpl()`
I've checked the query with 20.3 and `experimental_use_processors = 0`. Result is mostly the same: CH utializes just around ~1.5 CPU (140-160% of cores in `top` result). So most probably it's not a r... | 2020-08-24T11:43:29 | 2020-08-24T13:04:18 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,342,869 | 13,861 | filimonov | ERROR: type should be string, got "\r\nhttps://github.com/ClickHouse/ClickHouse/compare/v20.4.6.53-stable...v20.4.7.67-stable\r\n\r\n@alesapin do you know what could be the reason? " | 2020-08-21T15:11:16 | 2020-08-21T15:11:16 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,676,094 | 10,163 | ramazanpolat | > The format is intended only for data import (input format).
> It can only parse a table with single field of type String.
>
> It will parse a sequence of JSON objects separated by newlines, spaces and/or comma.
> Each JSON object is parsed as a whole to string, without parsing internals.
> JSON object may resid... | 2020-08-22T18:34:49 | 2020-08-22T18:34:49 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,821,053 | 13,617 | alexey-milovidov | 1. Is the result correct?
2. What is the geomean difference across all tests / all queries? | 2020-08-23T20:28:08 | 2020-08-23T20:28:08 | {} | MEMBER |
ClickHouse | ClickHouse | 678,884,516 | 13,878 | hexiaoting | @ramazanpolat
```
create table json1(
root String,
id String
materialized JSONExtractString(root, '_id'),
source String
materialized JSONExtractString(root, 'source') ,
sentences Array(String)
alias JSONExtractArrayRaw(root, 'sentences'),
sentences_content Array(String)
materialized arrayMap(s ... | 2020-08-24T03:28:25 | 2020-08-24T03:28:25 | {
"+1": 4
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,325,334 | 13,937 | den-crane | It needs to include an SSH client into CH. What next? FTP, SAMBA, and so on?
And it's easy to automate with bash. | 2020-08-21T14:39:06 | 2020-08-21T15:31:39 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,721,898 | 13,965 | melin | > Can you show your table DDL?
>
> try --input_format_defaults_for_omitted_fields arg For input data calculate default expressions for omitted fields (it works for JSONEachRow, CSV and TSV formats).
>
> input_format_defaults_for_omitted_fields=1
```sql
─────────────────────────────────────────────────┐
│ CRE... | 2020-08-23T02:50:50 | 2020-08-23T05:43:34 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,823,465 | 13,956 | excitoon | Rebased onto revert of revert. | 2020-08-23T20:49:40 | 2020-08-23T20:49:40 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,835,345 | 13,984 | alexey-milovidov | Too massive. | 2020-08-23T22:54:28 | 2020-08-23T22:54:28 | {} | MEMBER |
ClickHouse | ClickHouse | 679,041,335 | 13,994 | filimonov | /cc @KochetovNicolai | 2020-08-24T10:18:22 | 2020-08-24T10:18:22 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 679,086,993 | 9,770 | alexey-milovidov | I put this comment to indicate that this feature is nice to have... and easy to implement. | 2020-08-24T12:07:08 | 2020-08-24T12:07:22 | {} | MEMBER |
ClickHouse | ClickHouse | 678,451,652 | 13,957 | den-crane | @divyankgarg Trust me. This is JSONEachRow:
```
{"device":"device1","topic":"topic1","fields.tandTimeOffset":["fields1.1","fields1.2"],"keys._playbook_name":["keys1.1","keys1.2"]}
```
This is not JSONEachRow:
```
{
"device": "10.84.130.44",
"topic": "external",
"fields": {
"high-thresho... | 2020-08-21T19:21:03 | 2020-08-21T19:22:19 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,468,867 | 13,955 | azat | Distributed queries uses `default` user (if it was not changed in the configuration) on the remote shards, and your current user for the local shard, and I guess that you have some settings with restrictions applied for the default user?
In this case you can either drop this limitations for `default` user or change th... | 2020-08-21T20:03:30 | 2020-08-21T20:04:08 | {} | MEMBER |
ClickHouse | ClickHouse | 678,578,225 | 13,875 | mjodeh | `2020.08.21 00:39:55.722883 [ 101651 ] {} <Error> HTTPHandler: Code: 241, e.displayText() = DB::Exception: Memory limit (total) exceeded: would use 26.54 GiB (attempt to allocate chunk of 4502760 bytes), maximum: 26.54 GiB, Stack trace (when copying this message, always include the lines below):
2020.08.21 00:39:55.72... | 2020-08-22T01:55:08 | 2020-08-22T01:55:08 | {} | NONE |
ClickHouse | ClickHouse | 678,674,262 | 13,564 | ramazanpolat | > Can I ask why do you need Nullable there? You expect that some messages will not have json object?
@filimonov
Also in IBM Change Data Capture (which is another data capture software just like Debezium), delete rows are sent with NULL messages. Because there is no update or inserted data, message becames NULL.
... | 2020-08-22T18:17:43 | 2020-08-22T18:17:43 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,783,718 | 13,965 | den-crane | You have 2 issues with this file.
1. space `10,683, ,1958,914406053380946067`
2. some values are not Int64
`x33` Int64,
`x34` Int64,
`x35` Int64,
`x36` Int64,
`x37` Int64,
let's replace space with 0, and x33-x37 define as float
```
CREATE TABLE datainsight.import_test_csv
(
`x`... | 2020-08-23T14:52:25 | 2020-08-23T14:53:32 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 679,134,304 | 13,991 | bohutang | You're missing a port after host in the first parameter.
Usage:
`CREATE DATABASE clickhouse_test ENGINE = MaterializeMySQL('mysqlhost:port', 'clickhouse_test', 'root', 'password')`
But the error message as above needs to be fixed too. | 2020-08-24T13:43:18 | 2020-08-24T13:43:18 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,351,175 | 12,550 | myrrc | > Resolves #10880 ?
Yes, the perf tests showed the speedup with right-hand constant arguments. | 2020-08-21T15:28:07 | 2020-08-21T15:28:07 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,420,173 | 13,957 | den-crane | You JSON is not in the JSONEachRow format.
JSONEachRow is a subset of JSON.
And fields and keys are not Nested.
```
CREATE TABLE readings_hb_nest
(
device String,
topic String,
fields Nested (tandTimeOffset String),
keys Nested (_playbook_name String)
)
ENGINE = MergeTree
ORDER BY devi... | 2020-08-21T18:08:24 | 2020-08-21T18:28:59 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,454,134 | 13,688 | it1804 | In addition, SourceRedis RedisHash is not tested at all, because a layer named complex_key_hashed_two_keys does not exist, although this test is only activated under this condition. | 2020-08-21T19:27:12 | 2020-08-21T19:37:36 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,674,850 | 13,967 | ramazanpolat | @amosbird thank you, that worked. | 2020-08-22T18:22:54 | 2020-08-22T18:22:54 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,683,419 | 13,878 | ramazanpolat | @hexiaoting How did you make it work?
Can you post your DDL's here?
| 2020-08-22T19:37:09 | 2020-08-22T19:37:09 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,827,356 | 8,367 | 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.
Infrastructure failure. | 2020-08-23T21:29:19 | 2020-08-23T21:29:19 | {} | MEMBER |
ClickHouse | ClickHouse | 678,827,798 | 8,367 | alexey-milovidov | Stress test: #13980 | 2020-08-23T21:33:36 | 2020-08-23T21:33:36 | {} | MEMBER |
ClickHouse | ClickHouse | 678,836,233 | 13,956 | excitoon | > This bug is connected to default policy issues after previous changes which were hidden because `getVolumeByName()` method was not called (or rarily) anywhere until that commit.
I was wrong, previous code was mostly right (except ignoring empty pointer which led to NULL pointer dereference), but thing is name of t... | 2020-08-23T23:03:38 | 2020-08-23T23:25:27 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 679,114,194 | 14,004 | den-crane | I would expect both SHOW & DESCRIBE to work for TEMPORARY if no database specified.
SHOW CREATE TABLE t -- TEMPORARY
SHOW CREATE TABLE default.t -- STRIPELOG | 2020-08-24T13:05:46 | 2020-08-24T13:05:46 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,529,198 | 13,954 | alexey-milovidov | Yes, it helps. | 2020-08-21T21:42:41 | 2020-08-21T21:42:41 | {} | MEMBER |
ClickHouse | ClickHouse | 678,529,401 | 13,954 | alexey-milovidov | Alternative: https://github.com/ClickHouse/ClickHouse/pull/13956 | 2020-08-21T21:43:23 | 2020-08-21T21:43:23 | {} | MEMBER |
ClickHouse | ClickHouse | 678,588,790 | 13,961 | den-crane | probably this is it https://github.com/ClickHouse/ClickHouse/issues/5237 | 2020-08-22T03:46:59 | 2020-08-22T03:46:59 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,670,307 | 13,862 | UnamedRus | > Try reorder the JOINs. Since the second one only generates ~200 million records, it's much faster.
On example it's almost 4 times faster and utilize all cpu, but on real data it's actually slower:
Probably need better example.
On real query:
t1, t2 join: 187.636 sec 73.26 million rows/s ~17 utilized... | 2020-08-22T17:43:24 | 2020-08-22T17:50:13 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,821,335 | 12,438 | alexey-milovidov | Ok, thank you! | 2020-08-23T20:29:28 | 2020-08-23T20:29:28 | {} | MEMBER |
ClickHouse | ClickHouse | 678,835,354 | 13,982 | alexey-milovidov | Too massive. | 2020-08-23T22:54:33 | 2020-08-23T22:54:33 | {} | MEMBER |
ClickHouse | ClickHouse | 678,916,308 | 12,171 | taiyang-li | Done. | 2020-08-24T05:45:16 | 2020-08-24T05:45:16 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 679,037,237 | 12,158 | tavplubix | Fixed in #13756 | 2020-08-24T10:09:19 | 2020-08-24T10:09:19 | {} | MEMBER |
ClickHouse | ClickHouse | 679,044,706 | 13,963 | filimonov | 1) Does it reproduces always on that query?
2) what is a result if `CHECK TABLE reporting_minute`
3) Can you try to create a data sample showing the problem?
| 2020-08-24T10:26:15 | 2020-08-24T10:26:15 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 679,057,100 | 13,980 | alexey-milovidov | The first reason really exists:
```
+ stop
+ timeout 120 service clickhouse-server stop
Stop clickhouse-server service: DONE
++ pidof clickhouse-server
+ kill -9 354
Killed clickhouse-server
+ echo 'Killed clickhouse-server'
``` | 2020-08-24T10:53:49 | 2020-08-24T10:54:10 | {} | MEMBER |
ClickHouse | ClickHouse | 678,448,831 | 13,957 | divyankgarg | I believe it is JSONEachRow because when I am trying to put JSON format, I am not getting any output from table.
Also, I believe you are giving list under fields and key and I my case it is dictionary and nested does not work for dictionary.
I want to insert the data having format:
insert into readings_hb_nest val... | 2020-08-21T19:14:12 | 2020-08-21T19:14:12 | {} | NONE |
ClickHouse | ClickHouse | 678,453,360 | 13,957 | divyankgarg | Actually I streched and showed you. Below is the actual format of incoming message from Kafka. That is JsonEachRow format.
{"device-group":"qfxgroup1","device":"10.84.130.44","topic":"external","rule":"check-in-traffic_copy_6sjv66","fields":{"high-threshold":800000000,"in-octets-stats-value":0,"low-threshold":50000000... | 2020-08-21T19:25:17 | 2020-08-21T19:25:17 | {} | NONE |
ClickHouse | ClickHouse | 678,694,069 | 13,957 | divyankgarg | Hi den,
For the message below can you tell me what data type should I provide for KEYS_INSTANCE_ID and DEVICE_GROUP to create table using MergerTree and Kafka engine using Materialized view. I tried string and Nullable(String) respectively but didn't worked for me for creating the table.
##Message
{"DEVICE_GROUP":nu... | 2020-08-22T21:01:17 | 2020-08-22T21:01:17 | {} | NONE |
ClickHouse | ClickHouse | 679,009,525 | 4,022 | filimonov | #13953 | 2020-08-24T09:13:53 | 2020-08-24T09:13:53 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 679,026,717 | 12,508 | qoega | test_version_update_after_mutation
https://clickhouse-test-reports.s3.yandex.net/0/55ac192417ee0d2b83e7f28c5765ae939ad2983f/integration_tests_(thread).html | 2020-08-24T09:46:39 | 2020-08-24T09:46:39 | {} | MEMBER |
ClickHouse | ClickHouse | 678,431,691 | 13,930 | heyciao | Fantastic - I can then use that one.
Thanks a lot! :) | 2020-08-21T18:35:25 | 2020-08-21T18:35:25 | {} | NONE |
ClickHouse | ClickHouse | 678,529,296 | 13,958 | den-crane |
minimal example
```
$ cat order_test1.sql
drop table if exists order_test1;
create table order_test1
(
timestamp DateTime64(3),
color LowCardinality(String)
) engine = MergeTree() ORDER BY tuple();
insert into order_test1 values ('2020-08-21 18:46:08.000','red')('2020-08-21 18:46:08.000','gr... | 2020-08-21T21:43:01 | 2020-08-21T21:43:01 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,761,200 | 13,954 | excitoon | @alexey-milovidov Let's revert it for a while. | 2020-08-23T11:14:54 | 2020-08-23T11:14:54 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,761,376 | 13,956 | excitoon | This bug is connected to default policy issues after previous changes which were hidden because `getVolumeByName()` method was not called (or rarily) anywhere until that commit. | 2020-08-23T11:16:55 | 2020-08-23T11:17:15 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,346,850 | 12,550 | filimonov | Resolves #10880 ? | 2020-08-21T15:19:14 | 2020-08-21T15:19:14 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,529,857 | 13,960 | alexey-milovidov | #13691 #12562 #13369 #13584 #12964. | 2020-08-21T21:45:00 | 2020-08-21T21:45:00 | {} | MEMBER |
ClickHouse | ClickHouse | 678,643,202 | 10,685 | amosbird | Sounds reasonable. Perhaps we can have a special operator for this, like `column ~ 'long_string'` and some additional definitions added when creating merge tree tables. | 2020-08-22T13:45:37 | 2020-08-22T13:45:37 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 678,670,919 | 13,862 | amosbird | > Probably need better example.
It means optimizing the JOIN order is non-trivial :). | 2020-08-22T17:48:51 | 2020-08-22T17:48:51 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 678,672,251 | 10,607 | ramazanpolat | ```
CREATE TABLE doc(s String) Engine=Memory;
INSERT INTO doc FORMAT JSONAsString '{"a":1}';
```
Gives this:
```
Code: 117, e.displayText() = DB::Exception: JSON object must begin with '{'.: (at row 1)
, Stack trace (when copying this message, always include the lines below):
0. Poco::Exception::Exception(std... | 2020-08-22T18:00:22 | 2020-08-22T18:00:22 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,676,558 | 13,878 | ramazanpolat | Bookmarking this for later review. | 2020-08-22T18:39:28 | 2020-08-22T18:39:28 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,706,096 | 13,965 | den-crane | Can you show your table DDL?
try --input_format_defaults_for_omitted_fields arg For input data calculate default expressions for omitted fields (it works for JSONEachRow, CSV and TSV formats).
input_format_defaults_for_omitted_fields=1 | 2020-08-22T23:02:17 | 2020-08-22T23:02:46 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,835,351 | 13,983 | alexey-milovidov | Too massive. | 2020-08-23T22:54:31 | 2020-08-23T22:54:31 | {} | MEMBER |
ClickHouse | ClickHouse | 678,981,941 | 13,861 | alesapin | Possibly broken here https://github.com/ClickHouse/ClickHouse/pull/12165/files | 2020-08-24T08:19:25 | 2020-08-24T08:19:25 | {} | MEMBER |
ClickHouse | ClickHouse | 679,097,721 | 14,002 | rubentorresbonet | As extra info: I got the source from the master repository today. | 2020-08-24T12:30:30 | 2020-08-24T12:30:30 | {} | NONE |
ClickHouse | ClickHouse | 678,351,548 | 13,402 | alexey-milovidov | @excitoon @Akazz The test https://clickhouse-test-reports.s3.yandex.net/0/55ac192417ee0d2b83e7f28c5765ae939ad2983f/functional_stateless_tests_(debug).html#fail1 started to fail in debug build.
Is there a chance that it's related to your changes? | 2020-08-21T15:28:57 | 2020-08-21T15:28:57 | {} | MEMBER |
ClickHouse | ClickHouse | 678,456,617 | 6,093 | den-crane | @xmariachi
`optimize table` does not provide any output or any progress to a client during execution.
Client waits for any byte from a server for 300s ( receive_timeout ) but the server sends nothing. After 300s the client tries to cancel the query but it's impossible for `optimize` then the client waits for a ... | 2020-08-21T19:33:28 | 2020-08-21T19:33:28 | {
"+1": 14
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,722,010 | 13,956 | excitoon | Somehow, I introduced critical bug with 5c230f65157ac3dd643d6aa94967ef673201ede2. | 2020-08-23T02:52:57 | 2020-08-23T02:52:57 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 679,094,983 | 14,002 | filimonov | https://github.com/ClickHouse/ClickHouse/pull/10425 | 2020-08-24T12:24:30 | 2020-08-24T12:24:30 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,348,028 | 13,937 | bohutang | > It needs to include an SSH client into CH. What next? FTP, SAMBA, and so on?
> And it easy to automate with bash.
here just a thought for others dumping.
| 2020-08-21T15:21:40 | 2020-08-21T15:21:40 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,447,995 | 6,093 | xmariachi | Any cause that may be leading/impacting to this? It is happening to me but it is quite predictable actually. So it makes me think there may be something wrong there.
Cannot find anything related to it on logs.
It's a very small partition (hundreds of entries).
```
docker run -it yandex/clickhouse-client --host m... | 2020-08-21T19:12:12 | 2020-08-21T19:12:12 | {} | NONE |
ClickHouse | ClickHouse | 678,461,186 | 13,957 | den-crane | Still it's not valid JSONEachRow.
Check these examples:
```
CREATE TABLE js
(
device String,
fields Nested (tandTimeOffset String)
)
ENGINE = Memory
insert into js format JSONEachRow {"device":"device1","fields.tandTimeOffset":["fields1.1","fields1.2"]}
set input_format_import_nested_json=1;
... | 2020-08-21T19:45:01 | 2020-08-21T19:45:26 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,581,728 | 13,875 | den-crane | try to upgrade 20.3.17.173 this error "DB::Exception: Memory limit (total) exceeded" should be fixed there. | 2020-08-22T02:28:55 | 2020-08-22T02:28:55 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,582,236 | 13,875 | mjodeh | I will test it and get back to report the results | 2020-08-22T02:34:10 | 2020-08-22T02:34:10 | {} | NONE |
ClickHouse | ClickHouse | 678,652,151 | 10,685 | zhang2014 | `CityHash64` function is not injective, how does it work in a sparse index in this case?
BTW: Maybe bloom filter skip index is better?
| 2020-08-22T15:09:37 | 2020-08-22T15:09:37 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,665,886 | 13,862 | amosbird | Current implementation of multiple JOINs is memory-bounded. You'll have each intermediate block materialized to evaluate the next JOIN. In your example, the first INNER JOIN gives 3 billion records, and it'll be used to build blocks in the second JOIN, which slows down the pipeline. `IColumn::replicate` becomes the hot... | 2020-08-22T17:05:15 | 2020-08-22T17:05:15 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 678,827,615 | 8,367 | alexey-milovidov | > AST fuzzer — Received signal 6
@akuzm There is no stack trace from clickhouse-server. The problem is undiagnosable.
@nikitamikhaylov The error is in cache dictionary - you should take a look https://github.com/ClickHouse/ClickHouse/issues/13981 | 2020-08-23T21:31:46 | 2020-08-23T21:33:08 | {} | MEMBER |
ClickHouse | ClickHouse | 678,827,985 | 8,367 | alexey-milovidov | Performance test: it's unclear for me if error is real. Need to investigate. | 2020-08-23T21:35:24 | 2020-08-23T21:35:24 | {} | MEMBER |
ClickHouse | ClickHouse | 678,990,691 | 13,978 | filimonov | > "uncommitted" (or: "offline") directory
Maybe that approach can also be (re)used for https://github.com/ClickHouse/ClickHouse/issues/13574
> a. mark moving parts on an old node as OLD, during SELECTs return to a root executor all OLD part ids that were processed during query
> b. mark moving parts on a new no... | 2020-08-24T08:37:21 | 2020-08-24T09:03:23 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 679,073,815 | 13,978 | alesapin |
> We can allocate local block numbers, write parts to separate "uncommitted" (or: "offline") directory - these parts will be available for SELECTs and merges and can be committed as new parts later when ZooKeeper will become available.
If I understand correctly, "uncommitted" parts can be merged only with other "u... | 2020-08-24T11:34:48 | 2020-08-24T11:34:48 | {} | MEMBER |
ClickHouse | ClickHouse | 678,343,817 | 13,869 | alexey-milovidov | @urgordeadbeef it's only allowed to use libraries from `contrib`, no system/package provided libraries are possible.
This is needed because FreeBSD build is cross-compiled. | 2020-08-21T15:13:07 | 2020-08-21T15:13:07 | {} | MEMBER |
ClickHouse | ClickHouse | 678,350,217 | 13,402 | akuzm | > This PR definitely could not have affected the failed performance test - [constant_column_search](https://clickhouse-test-reports.s3.yandex.net/13402/16d709964bef99d85dd5eb2c8cdc46669917a0a7/performance_comparison/report.html#fail1) as well as `array_element`
What about `read_in_order_many_parts`? The query `OPTIM... | 2020-08-21T15:26:01 | 2020-08-21T15:26:01 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,339,099 | 13,864 | akuzm | > @filimonov @CurtizJ The difference is that `IDataType::enumerateStreams` called more times.
Ohhhh it's this bad callback again. Do you remember that it is also responsible for like 20% of all memory allocations for an average query? Because the lamda passed to it has to be allocated on heap. I started to rewrite i... | 2020-08-21T15:03:35 | 2020-08-21T16:05:14 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,375,868 | 13,953 | bohutang | Consistent files copy backup not only ClickHouse need other databases have this need as well, we need to do:
'make consistent copies of a directory of files, even while the respective files are being edited, removed, added, and/or renamed.'
Just like [TokuBackup](https://github.com/percona/Percona-TokuBackup) did, ... | 2020-08-21T16:21:08 | 2020-08-21T16:21:08 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 678,669,954 | 13,967 | amosbird | Please try with the `-n` flag without `-m`. | 2020-08-22T17:40:10 | 2020-08-22T17:40:10 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 678,831,711 | 13,348 | alexey-milovidov | It has encountered second time. | 2020-08-23T22:16:31 | 2020-08-23T22:16:31 | {} | MEMBER |
ClickHouse | ClickHouse | 678,835,343 | 13,985 | alexey-milovidov | Too massive. | 2020-08-23T22:54:26 | 2020-08-23T22:54:26 | {} | MEMBER |
ClickHouse | ClickHouse | 678,986,878 | 12,508 | alesapin | > @alesapin https://clickhouse-test-reports.s3.yandex.net/0/d2be6c036b3b2f4bec8de985a9c620f9ae2cd28d/functional_stateless_tests_(release)/test_run.txt.out.log
>
> Is it possible that mutations finish on replica in different order?
> Maybe it is old behaviour that was already fixed?
Seems like it's very a old tes... | 2020-08-24T08:29:51 | 2020-08-24T08:29:51 | {} | MEMBER |
ClickHouse | ClickHouse | 679,128,933 | 13,991 | liangli1213 | @zhang2014 Would you please tell me the correct way to use MaterializeMySQL? | 2020-08-24T13:33:38 | 2020-08-24T13:33:38 | {} | NONE |
ClickHouse | ClickHouse | 683,161,083 | 14,201 | filimonov | I was thinking about just put the errors (together with unparsed messages) inside a rows generated by Kafka engine.
And later on MV level you can decide what to do with them (drop/store in table/push somewhere etc)
Logfile - not sure if good (they are not accessible if you have only clickhouse shell).
For pro... | 2020-08-28T21:48:11 | 2020-08-28T21:48:11 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 683,177,806 | 14,201 | filimonov | > Users would want to know about this.
They will know with exception message, why suppress exceptions and write logs silently? | 2020-08-28T22:46:11 | 2020-08-28T22:46:11 | {} | CONTRIBUTOR |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.