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 | 643,829,431 | 10,564 | azat | >Got it, can I set distributed_replica_error_cap to 0? would that + load_balancing = random give me basically round-robin load balancing? Cause there is no round-robin strategy natively in CH.
Sorry, I misread your message (thought that you wrote `distributed_replica_error_half_life`), so yes `distributed_replica_er... | 2020-06-14T22:16:59 | 2020-06-14T22:16:59 | {} | MEMBER |
ClickHouse | ClickHouse | 643,853,586 | 11,639 | alexey-milovidov | > TODO: support downgrade scenario.
Done. | 2020-06-15T01:13:29 | 2020-06-15T01:13:29 | {} | MEMBER |
ClickHouse | ClickHouse | 643,917,162 | 11,647 | filimonov | > This does not fix #11630
But st-fixed label added and issue closed :) | 2020-06-15T05:50:30 | 2020-06-15T05:50:30 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 644,227,008 | 10,987 | Akazz | About the suggested alternative. The way I see it, it is hardly possible to just limit number of fetches, because both merges and fetches are performed by the same service task [StorageReplicatedMergeTree::queueTask()](https://github.com/ClickHouse/ClickHouse/blob/e460d7c2bd58b379cc68a2a2e144305af36f9c02/src/Storages/S... | 2020-06-15T16:07:34 | 2020-06-15T16:11:27 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 644,256,101 | 11,641 | abyss7 | Your test doesn't reflect any changes - the output is the same on master for me. | 2020-06-15T17:04:01 | 2020-06-15T17:04:01 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 644,329,332 | 10,076 | alexey-milovidov | @Siao-pin Could you please share the query example and the result of `SHOW PROCESSLIST`? | 2020-06-15T19:21:43 | 2020-06-15T19:21:43 | {} | MEMBER |
ClickHouse | ClickHouse | 644,355,994 | 5,075 | alexey-milovidov | Unassigned from @abyss7 because nothing was done. | 2020-06-15T20:16:07 | 2020-06-15T20:16:07 | {} | MEMBER |
ClickHouse | ClickHouse | 644,371,014 | 4,406 | alexey-milovidov | @renatga Ok, let's change the log level of this message to warning. | 2020-06-15T20:29:29 | 2020-06-15T20:29:29 | {} | MEMBER |
ClickHouse | ClickHouse | 644,509,850 | 11,618 | 1818hero | > What CH version do you use?
> Can you show `show create table tablename` ?
>
> It was an issue #10566
Thank you for your reply.
I use version 20.3.8.53-2
```
CREATE TABLE gec.tablename (`id` Int64, `meCode` String, `deCode` String, `dataValue` String, `eventTime` DateTime, `recordDate` Date DEFAULT toDate... | 2020-06-16T03:29:20 | 2020-06-16T03:36:51 | {} | NONE |
ClickHouse | ClickHouse | 644,613,657 | 11,516 | alesapin | Performance `bloom_filter` seems flaky. No diff on previous commit https://clickhouse-test-reports.s3.yandex.net/11516/2c9ce0f3fa4e7f02294bbb6e6021eee3633cc289/performance_comparison/report.html#fail1. | 2020-06-16T08:22:22 | 2020-06-16T08:22:22 | {} | MEMBER |
ClickHouse | ClickHouse | 644,669,954 | 11,673 | mjodeh | Thanx @arashkhavari | 2020-06-16T10:09:45 | 2020-06-16T10:09:45 | {} | NONE |
ClickHouse | ClickHouse | 643,813,416 | 11,624 | den-crane | @wallace-clickhouse
Oh, I just realized that you wanted to load data by clickhouse-server process from files at server. It actually possible using file function https://clickhouse.tech/docs/en/sql-reference/table-functions/file/
But still I would use HTTP or JDBC in case of JAVA.
HTTP https://clickhouse.tech... | 2020-06-14T19:47:36 | 2020-06-14T19:47:36 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 643,829,950 | 11,646 | alexey-milovidov | Superseded by #11647 | 2020-06-14T22:22:19 | 2020-06-14T22:22:19 | {} | MEMBER |
ClickHouse | ClickHouse | 644,322,992 | 9,002 | alexey-milovidov | It is implemented in a series of pull requests:
#10623
#11144
#11421
And available in master and in version 20.5 (testing). | 2020-06-15T19:07:58 | 2020-06-15T19:07:58 | {
"hooray": 3
} | MEMBER |
ClickHouse | ClickHouse | 644,325,709 | 3,137 | alexey-milovidov | And also we have implemented support for non-constant arguments in pointInPolygon function. | 2020-06-15T19:14:08 | 2020-06-15T19:14:08 | {} | MEMBER |
ClickHouse | ClickHouse | 644,335,063 | 8,470 | alexey-milovidov | @qoega What is the status of this task?
Did you reproduce the error on your laptop?
You are the only person from our team with Mac. | 2020-06-15T19:34:14 | 2020-06-15T19:34:14 | {} | MEMBER |
ClickHouse | ClickHouse | 644,698,033 | 11,401 | akuzm | Reran experiment no.5 with async metrics (no MADV_FREE/DONTNEED at all), here is a graph for MemoryResident.
Looks like it doesn't unmap the memory after all...
```
$ clickhouse-local --qu... | 2020-06-16T11:12:27 | 2020-06-16T11:12:27 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 643,985,025 | 11,640 | KochetovNicolai | So, the main idea is: row with `sign = -1` may *cancel* only **previously** inserted row. If there is no such row, `SELECT FINAL` just ignores it as a wrong data. | 2020-06-15T08:31:25 | 2020-06-15T08:31:25 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 644,074,298 | 11,625 | wallace-clickhouse | @den-crane Thank you for your reply!
I thought of the method you mentioned. I think it is a risk reduction measure, but in fact it is still unsafe. Can this be modified? For example, we configure a reversible ciphertext in the metrika.xml file, and when clickhouse needs to connect to other shards, it automatically dec... | 2020-06-15T11:34:10 | 2020-06-15T11:34:10 | {} | NONE |
ClickHouse | ClickHouse | 644,381,457 | 10,851 | tavplubix | Could you please add comments to key places? (such as query converting rules) | 2020-06-15T20:48:46 | 2020-06-15T20:48:46 | {} | MEMBER |
ClickHouse | ClickHouse | 424,558,423 | 3,206 | rajadayalan | Created another table access_logs_local_attach_new and used attach partition, count looks same as using attach part
```
SELECT count()
FROM test.access_logs_local_attach_new
┌────count()─┐
│ 2349986051 │
└────────────┘
```
Also noticed that using attach partitions does not attach all parts. When our script ran... | 2018-09-26T01:55:22 | 2020-06-16T01:19:23 | {} | NONE |
ClickHouse | ClickHouse | 643,809,900 | 11,662 | Avogar | I forgot to add include path in cmake, I'l add it later | 2020-06-14T19:18:12 | 2020-06-14T19:18:12 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 643,926,394 | 11,643 | ghost | @alexey-milovidov
The previous version of pipeline in `InterpreterSelectQuery` has gone.... lol
I have to modify this feature to use the new version of QueryPipeline. | 2020-06-15T06:17:20 | 2020-06-15T06:17:20 | {} | NONE |
ClickHouse | ClickHouse | 643,949,515 | 11,624 | blinkov | @wallace-clickhouse both protocols (HTTP and Native) are not secure by default but can be configured to be encrypted. | 2020-06-15T07:17:12 | 2020-06-15T07:17:12 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 644,209,519 | 11,672 | filimonov | ```
set optimize_move_to_prewhere = 0;
```
or rewriting `foo.server_date <= '2020-11-07'` to `foo.server_date <= toDate('2020-11-07')`
solves the issue. | 2020-06-15T15:36:38 | 2020-06-15T15:36:38 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 644,327,750 | 11,682 | KochetovNicolai | Please don't merge it until perftests result. | 2020-06-15T19:18:23 | 2020-06-15T19:18:23 | {} | MEMBER |
ClickHouse | ClickHouse | 644,336,973 | 3,853 | alexey-milovidov | The issue is still relevant and it can be solved pretty easily.
Just replace single-line comments with multi-line comments with the help of Lexer. | 2020-06-15T19:38:29 | 2020-06-15T19:38:29 | {} | MEMBER |
ClickHouse | ClickHouse | 644,503,210 | 11,694 | alexey-milovidov | > ClickHouse build check
> g++-9: fatal error: Killed signal terminated program cc1plus
This is the issue in notorious Sandbox infrastructure. | 2020-06-16T03:03:47 | 2020-06-16T03:03:47 | {} | MEMBER |
ClickHouse | ClickHouse | 644,623,585 | 11,516 | alesapin | The only synchronizer checks important. | 2020-06-16T08:41:10 | 2020-06-16T08:41:10 | {} | MEMBER |
ClickHouse | ClickHouse | 644,144,190 | 11,681 | alesapin | Need to cherry-pick into 20.1, 20.3 and 20.4. | 2020-06-15T13:44:41 | 2020-06-15T13:44:41 | {} | MEMBER |
ClickHouse | ClickHouse | 644,196,223 | 2,739 | filimonov | Related: https://github.com/ClickHouse/ClickHouse/issues/2739 | 2020-06-15T15:13:47 | 2020-06-15T15:13:47 | {
"confused": 1,
"laugh": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 644,352,112 | 11,686 | alexey-milovidov | A test:
```
curl http://localhost:8123/ -d "SELECT 'ab
cd' /* hello */ -- world
, 1"
``` | 2020-06-15T20:08:21 | 2020-06-15T20:08:21 | {} | MEMBER |
ClickHouse | ClickHouse | 644,448,244 | 11,300 | alexey-milovidov | > 00600_replace_running_query
Should investigate.
PS. https://github.com/ClickHouse/ClickHouse/pull/11300/commits/b8611cf46cd2d6f15f2a6e678961c06c686fa9ed
Ok. | 2020-06-15T23:48:04 | 2020-06-15T23:49:15 | {} | MEMBER |
ClickHouse | ClickHouse | 644,642,720 | 11,627 | Namnd-gem | It's working with newer version. | 2020-06-16T09:16:09 | 2020-06-16T09:16:09 | {} | NONE |
ClickHouse | ClickHouse | 643,800,727 | 11,655 | alexey-milovidov | #11663 | 2020-06-14T18:01:35 | 2020-06-14T18:01:35 | {} | MEMBER |
ClickHouse | ClickHouse | 644,326,681 | 8,165 | KochetovNicolai | Closed as duplicate was closed too. | 2020-06-15T19:16:11 | 2020-06-15T19:16:11 | {} | MEMBER |
ClickHouse | ClickHouse | 644,669,344 | 11,488 | 4ertus2 | If `xyz` is a correct table name from current database it should return columns from it. And it should return an error otherwise. | 2020-06-16T10:08:31 | 2020-06-16T10:09:06 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 643,900,356 | 11,639 | 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. | 2020-06-15T04:49:09 | 2020-06-15T04:49:09 | {} | MEMBER |
ClickHouse | ClickHouse | 644,211,755 | 11,638 | akuzm | Build failure and perftest error are CI problem. | 2020-06-15T15:40:42 | 2020-06-15T15:40:42 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 644,334,284 | 9,215 | alexey-milovidov | Unassigned because nothing was done. | 2020-06-15T19:32:27 | 2020-06-15T19:32:27 | {} | MEMBER |
ClickHouse | ClickHouse | 644,547,035 | 11,618 | 1818hero | I have solved this problem by upgrade clickhouse to version 20.4.5.36 | 2020-06-16T05:54:41 | 2020-06-16T05:54:41 | {} | NONE |
ClickHouse | ClickHouse | 643,873,079 | 11,484 | alexey-milovidov | @alesapin
> Fix for test_always_fetch_merged/test.py #11582
It did not help: https://clickhouse-test-reports.s3.yandex.net/11648/6467302ad32b4dd6205542675d54fa89944d0ff1/integration_tests_(release).html#fail1 | 2020-06-15T02:46:17 | 2020-06-15T02:46:17 | {} | MEMBER |
ClickHouse | ClickHouse | 644,120,682 | 5,681 | calebeaires | Hey team. I vote to @alex-krash suggestion! | 2020-06-15T13:01:55 | 2020-06-15T13:01:55 | {} | NONE |
ClickHouse | ClickHouse | 644,124,316 | 11,625 | wallace-clickhouse | Can we do like this?
We can preset a default password (for example, 123456) in users.xml for the "default" user, and then recommend that users change it when using clickhouse. At the same time, we use a system-generated secure random number as the key (the key shoud be stored in a certain file) to encrypt the default... | 2020-06-15T13:08:48 | 2020-06-15T13:08:48 | {} | NONE |
ClickHouse | ClickHouse | 644,289,230 | 11,641 | azat | Egh, you are right and besides the code was correct before and this patch does not changes the behavior, sorry for the noise! | 2020-06-15T18:13:00 | 2020-06-15T18:13:00 | {} | MEMBER |
ClickHouse | ClickHouse | 644,499,946 | 11,639 | alexey-milovidov | @nikitamikhaylov
I have found that the new behaviour in test `test_quorum_inserts/test.py::test_insert_quorum_with_ttl` is absolutely correct.
Old behaviour:
insert_quorum = 2
two replicas: 0 and 1
create table on replica 0 and 1
replica 0 becomes leader
stop fetches on replica 1
insert on replica 0
q... | 2020-06-16T02:52:27 | 2020-06-16T03:00:48 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 644,649,738 | 11,516 | alesapin | Sync is OK | 2020-06-16T09:28:43 | 2020-06-16T09:28:43 | {} | MEMBER |
ClickHouse | ClickHouse | 644,715,049 | 11,701 | den-crane | CH 19.16.17.1 | 2020-06-16T11:52:35 | 2020-06-16T11:52:35 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 643,869,874 | 11,347 | alexey-milovidov | @TCeason that will be awesome! | 2020-06-15T02:32:46 | 2020-06-15T02:32:46 | {} | MEMBER |
ClickHouse | ClickHouse | 643,985,874 | 11,677 | dependabot-preview[bot] | Looks like googletrans is up-to-date now, so this is no longer needed. | 2020-06-15T08:33:03 | 2020-06-15T08:33:03 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 644,323,568 | 7,937 | alexey-milovidov | It is implemented in a series of pull requests:
#10623
#11144
#11421
And available in master and in version 20.5 (testing).
Duplicate of #9002. | 2020-06-15T19:09:15 | 2020-06-15T19:09:15 | {} | MEMBER |
ClickHouse | ClickHouse | 644,665,063 | 11,488 | hexiaoting | In which situation, will
```
select xyz.*
```
sucessful?
If xyz is a database name , it also should be failed, right？
I'm new to clickhouse, and I want to try to fix this. | 2020-06-16T09:59:59 | 2020-06-16T09:59:59 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 644,096,657 | 5,319 | calebeaires | Hey folks of Clickhouse! Great job!
Is this solution implemented? If so, who to enable it? | 2020-06-15T12:16:56 | 2020-06-15T12:16:56 | {} | NONE |
ClickHouse | ClickHouse | 644,310,183 | 11,658 | blinkov | cc @zhang2014 @sundy-li | 2020-06-15T18:42:08 | 2020-06-15T18:42:08 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 644,337,729 | 3,484 | ZzEeKkAa | Worked for me as well | 2020-06-15T19:40:05 | 2020-06-15T19:40:05 | {} | NONE |
ClickHouse | ClickHouse | 644,633,418 | 11,508 | nikitamikhaylov | Your commits now live here #11703. I will probably close this PR. | 2020-06-16T08:59:07 | 2020-06-16T08:59:07 | {} | MEMBER |
ClickHouse | ClickHouse | 644,636,179 | 11,690 | alesapin | Poco uses non-blocking sockets for `::connect` and I have found the following code https://github.com/ClickHouse-Extras/poco/blob/master/Net/src/SocketImpl.cpp#L133. It uses `pollImpl` to check connection https://github.com/ClickHouse-Extras/poco/blob/master/Net/src/SocketImpl.cpp#L459-L475 only if connect return non ... | 2020-06-16T09:04:09 | 2020-06-16T09:07:47 | {} | MEMBER |
ClickHouse | ClickHouse | 644,678,719 | 11,516 | alesapin | Tidy build passed. | 2020-06-16T10:28:18 | 2020-06-16T10:28:18 | {} | MEMBER |
ClickHouse | ClickHouse | 644,695,343 | 11,615 | nikitamikhaylov | @azat Stress test found a datarace https://gist.githubusercontent.com/nikitamikhaylov/acf45aa112f3941494ca508eca35d4c8/raw/25c38d24b47801ada2c95eb09a0cd3ae93e6e7f1/datarace.md
It is inconvenient that such things are shown only in the private network when the test is timeout. I suggest you to run tests locally under th... | 2020-06-16T11:06:27 | 2020-06-16T11:08:49 | {} | MEMBER |
ClickHouse | ClickHouse | 644,702,620 | 11,681 | alesapin | Tried locally `01079` and `01076` in debug build more than 10 times in a row. | 2020-06-16T11:22:58 | 2020-06-16T11:23:27 | {} | MEMBER |
ClickHouse | ClickHouse | 644,711,313 | 4,565 | alesapin | This is very strange, but we have tests for this functionality for more than two years https://github.com/ClickHouse/ClickHouse/blame/6ccb720659b0d8be99223190b686f5444ba218ea/dbms/tests/queries/0_stateless/00030_alter_table.sql#L26-L28. Also I have added more tests https://github.com/ClickHouse/ClickHouse/pull/11706 an... | 2020-06-16T11:43:29 | 2020-06-16T11:43:29 | {} | MEMBER |
ClickHouse | ClickHouse | 643,921,242 | 11,647 | alexey-milovidov | @filimonov
Follow-up PR is also merged. | 2020-06-15T06:02:49 | 2020-06-15T06:02:49 | {} | MEMBER |
ClickHouse | ClickHouse | 643,947,340 | 11,624 | wallace-clickhouse | @den-crane
Thank you for your reply!
I use ClickHouse-Native-JDBC as JDBC Driver in my program because I think HTTP is not secure,so is there any similar interface in ClickHouse-Native-JDBC for importing data? | 2020-06-15T07:11:50 | 2020-06-15T07:12:24 | {} | NONE |
ClickHouse | ClickHouse | 644,158,916 | 11,680 | blinkov | @long2ice could you add the same link to other language versions of this page? | 2020-06-15T14:11:01 | 2020-06-15T14:11:01 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 644,448,376 | 11,300 | alexey-milovidov | > test_mysql_protocol/test.py::test_mysql_federated
Known flaky test. | 2020-06-15T23:48:33 | 2020-06-15T23:48:33 | {} | MEMBER |
ClickHouse | ClickHouse | 644,616,646 | 11,682 | KochetovNicolai | `01269_create_with_null` fails but doesn't seem to be caused by this pr changes. | 2020-06-16T08:28:06 | 2020-06-16T08:28:06 | {} | MEMBER |
ClickHouse | ClickHouse | 644,698,274 | 11,672 | vanzi | @filimonov `toDate` does fix it in this place, but it seems that this issue is more global and occurs in different places that throw random errors, using tables from the first example, this query:
```sql
SELECT toDateTime(foo.server_time, 'UTC')
FROM foo
ANY INNER JOIN bar USING (dimension_1)
WHERE toDate(foo.s... | 2020-06-16T11:12:58 | 2020-06-16T11:12:58 | {} | NONE |
ClickHouse | ClickHouse | 643,764,790 | 11,565 | filimonov | Maybe just adding smth like `priority` would be enough? I.e. do random / round-robin between nodes of same (highest) priority, if none of them is avaliable - check nodes with lower priority, etc. | 2020-06-14T13:11:36 | 2020-06-14T13:11:36 | {
"+1": 2
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 643,804,222 | 3,950 | alexey-milovidov | The task should be not hard to implement, but it is blocked by another task:
Don't ignore data from duplicate INSERTS that were sent to replicas.
So, if user send the same data to several replicas manually, this data should be accepted as if it was received by replication protocol.
This task is unassigned, but I... | 2020-06-14T18:30:11 | 2020-06-14T18:30:11 | {} | MEMBER |
ClickHouse | ClickHouse | 643,865,715 | 11,639 | alexey-milovidov | > TODO: cleanupThread.
Done. | 2020-06-15T02:13:18 | 2020-06-15T02:13:18 | {} | MEMBER |
ClickHouse | ClickHouse | 644,115,411 | 11,625 | den-crane | >>Decrypt how? Using another password stored in /etc/secret.aes ?
>when clickhouse needs to connect to other shards, it automatically decrypts
@wallace-clickhouse OK. Excellent idea. Please explain how CH automatically decrypts encrypted passwords? By what password? Hard-coded in CH ? 12345678?
| 2020-06-15T12:51:20 | 2020-06-15T12:51:40 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 644,263,973 | 11,053 | alexey-milovidov | FYI This pull request is not merged. | 2020-06-15T17:20:28 | 2020-06-15T17:20:28 | {} | MEMBER |
ClickHouse | ClickHouse | 644,431,811 | 11,484 | alexey-milovidov | Fix: https://github.com/ClickHouse/ClickHouse/pull/11691 | 2020-06-15T22:57:19 | 2020-06-15T22:57:19 | {} | MEMBER |
ClickHouse | ClickHouse | 644,333,735 | 9,226 | alexey-milovidov | I have checked the marks for contribs with our own CMake. | 2020-06-15T19:31:12 | 2020-06-15T19:31:12 | {} | MEMBER |
ClickHouse | ClickHouse | 644,715,048 | 11,642 | bookeezhou | in ARM platform , clickhouse build system can't find zstd.
solve the problem,need dowload zstd and
make && make install
then build goes ok. | 2020-06-16T11:52:35 | 2020-06-16T11:52:35 | {} | NONE |
ClickHouse | ClickHouse | 643,994,691 | 11,624 | wallace-clickhouse | @blinkov Thank you!
I have tried to merge the code of "Clear password from command line" to my version(20.3.2.1). Fortunately it works fine and shows no plaintext password while execute 'ps'.History log still records the comand line with plaintext password,so I decide to shield it by using HISTIGNORE.
@den-crane @... | 2020-06-15T08:49:45 | 2020-06-15T08:49:45 | {} | NONE |
ClickHouse | ClickHouse | 644,159,676 | 11,680 | long2ice | OK | 2020-06-15T14:12:25 | 2020-06-15T14:12:25 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 644,335,762 | 6,813 | alexey-milovidov | Reassigned to @vitlibar because @abyss7 did not respond. | 2020-06-15T19:35:41 | 2020-06-15T19:35:41 | {} | MEMBER |
ClickHouse | ClickHouse | 644,428,045 | 11,484 | alexey-milovidov | Race condition between table startup/shutdown and SYSTEM SYNC REPLICA.
The command SYSTEM SYNC REPLICA may be issued before table was started up.
https://clickhouse-test-reports.s3.yandex.net/9813/05eed28b775e0029e6ccbe4655411e34681c1b67/functional_stateless_tests_(debug)/clickhouse-server.log
https://clickhouse... | 2020-06-15T22:47:06 | 2020-06-15T22:47:06 | {} | MEMBER |
ClickHouse | ClickHouse | 644,434,409 | 11,691 | alexey-milovidov | I have added a test that is reproducing this issue easily. | 2020-06-15T23:04:28 | 2020-06-15T23:04:28 | {} | MEMBER |
ClickHouse | ClickHouse | 644,459,421 | 11,300 | alexey-milovidov | Almost Ok but a few changes remain. | 2020-06-16T00:27:33 | 2020-06-16T00:27:33 | {} | MEMBER |
ClickHouse | ClickHouse | 643,846,419 | 11,647 | alexey-milovidov | Performance test Ok. | 2020-06-15T00:33:42 | 2020-06-15T00:33:42 | {} | MEMBER |
ClickHouse | ClickHouse | 644,099,018 | 8,834 | calebeaires | Hey @alexey-milovidov, is your solution implemented? If so, who take it on the instance!? | 2020-06-15T12:22:04 | 2020-06-15T12:22:04 | {} | NONE |
ClickHouse | ClickHouse | 644,325,512 | 3,137 | alexey-milovidov | This issue is addressed by "Geo Dictionaries" feature.
It is being implemented in #9278. The status is: final code review and evaluation.
The Geo Dictionary allows you to create an index on top of the set of polygons with some attributes and quickly extract attributes from matched polygon by location. | 2020-06-15T19:13:41 | 2020-06-15T19:13:41 | {} | MEMBER |
ClickHouse | ClickHouse | 644,332,135 | 8,746 | alexey-milovidov | Did you fail? | 2020-06-15T19:27:44 | 2020-06-15T19:27:44 | {} | MEMBER |
ClickHouse | ClickHouse | 644,366,915 | 2,987 | alexey-milovidov | The server will print this error message and continue to work with IPv4. This is as expected. | 2020-06-15T20:25:59 | 2020-06-15T20:25:59 | {} | MEMBER |
ClickHouse | ClickHouse | 644,447,991 | 11,300 | alexey-milovidov | https://clickhouse-test-reports.s3.yandex.net/11300/b8611cf46cd2d6f15f2a6e678961c06c686fa9ed/unit_tests_relwithdebuginfo_gcc-9/test_run.txt.out.log
Flaky unit test. | 2020-06-15T23:47:16 | 2020-06-15T23:47:16 | {} | MEMBER |
ClickHouse | ClickHouse | 644,502,735 | 11,484 | alexey-milovidov | g++-9: fatal error: Killed signal terminated program cc1plus
https://clickhouse-builds.s3.yandex.net/11694/edff54e3abb7b7f7d1baa47b8b13d1f02be379d7/report.html#fail1 | 2020-06-16T03:02:17 | 2020-06-16T03:02:17 | {} | MEMBER |
ClickHouse | ClickHouse | 644,714,735 | 11,672 | filimonov | That is definitely a bug that should be fixed. You can use `set optimize_move_to_prewhere = 0` temporary as a workaround, but it may have negative impact on the performance. | 2020-06-16T11:51:52 | 2020-06-16T11:51:52 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 643,873,202 | 11,648 | alexey-milovidov | > test_always_fetch_merged/test.py
Reported in https://github.com/ClickHouse/ClickHouse/issues/11484 | 2020-06-15T02:46:47 | 2020-06-15T02:46:47 | {} | MEMBER |
ClickHouse | ClickHouse | 643,982,910 | 11,640 | KochetovNicolai | Actually, `FINAL` for `CollapsingMergeTree` have being working that way form the beginning.
The main idea, as I understand it, that every row with `sign = -1` should have coupling **previous** row with `sign = 1`.
Data in your example is assumed by `CollapsingMergeTree` as incorrect. However, at the moment data is me... | 2020-06-15T08:27:20 | 2020-06-15T08:27:20 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 644,014,718 | 11,636 | KochetovNicolai | Simplified test:
```
drop table if exists final_bug;
create table final_bug (x UInt64, y UInt8) engine = ReplacingMergeTree(y) order by x settings index_granularity = 8;
insert into final_bug select number % 10, 1 from numbers(1000);
insert into final_bug select number % 10, 1 from numbers(1000);
select x from fi... | 2020-06-15T09:27:22 | 2020-06-15T09:27:22 | {} | MEMBER |
ClickHouse | ClickHouse | 644,141,742 | 11,673 | mjodeh | the solution is here https://github.com/ClickHouse/ClickHouse/issues/2739 | 2020-06-15T13:40:07 | 2020-06-15T13:40:07 | {} | NONE |
ClickHouse | ClickHouse | 644,146,499 | 11,625 | den-crane | You talking about metrika.xml . Do you understand that usage of this file is not mandatory? You can put a password directly in /etc/clickhouse-server/users.d/user_name.xml ?
>the key shoud be stored in a certain file)
What is the difference where to store password in user_name.xml or `certain file` ?
You just ... | 2020-06-15T13:48:54 | 2020-06-15T13:55:18 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 644,328,232 | 11,131 | alexey-milovidov | You're right, thank you!
I will remove this line: #11685. | 2020-06-15T19:19:24 | 2020-06-15T19:19:24 | {} | MEMBER |
ClickHouse | ClickHouse | 644,564,293 | 11,505 | zhbdesign | Execute 'optimize table ods.user 'click house exception, code: 209, host: null, port: 0; Code: 209, e.displaytext() = DB:: netexception: timeout: connect timed out: 192.168.10.178:9002 (hadoop-node7:9002) (version 19.16.12.49)), 192.168.10.178 is the replica machine. At this time, the load and network are normal | 2020-06-16T06:40:28 | 2020-06-16T06:40:28 | {} | NONE |
ClickHouse | ClickHouse | 644,628,187 | 11,508 | nikitamikhaylov | Hmm, you checkout the current branch from 20.4 and try to merge into 20.4. I will fix. | 2020-06-16T08:49:47 | 2020-06-16T08:50:04 | {} | MEMBER |
ClickHouse | ClickHouse | 644,666,265 | 11,673 | arashkhavari | ```sql
SELECT
userId,
minIf(eventTime, eventName = 'event1') AS step0_time,
arrayFilter(time -> ((step0_time > 0) AND ((time - step0_time) > 0)), groupArrayIf(eventTime, eventName = 'event2'))[1] AS step1_time,
step1_time - step0_time AS timeToStep1
FROM
(
SELECT
userId,
... | 2020-06-16T10:02:14 | 2020-06-16T10:02:14 | {} | NONE |
ClickHouse | ClickHouse | 642,646,510 | 11,484 | alexey-milovidov | Fix for "00992 and 00993 failed to remove something from zookeeper": #11592 | 2020-06-11T13:25:43 | 2020-06-11T13:25:43 | {} | MEMBER |
ClickHouse | ClickHouse | 642,819,691 | 11,601 | alexey-milovidov | NULL pointer in
2020.06.11 18:12:51.946411 [ 14042 ] {} <Fatal> BaseDaemon: 5. /build/obj-x86_64-linux-gnu/../src/Storages/MergeTree/MergeTreeData.cpp:1640: DB::MergeTreeData::PartsTemporaryRename::tryRenameAll() @ 0x10f0f2bb in /usr/bin/clickhouse
\- something new... | 2020-06-11T17:16:08 | 2020-06-11T17:16:08 | {} | MEMBER |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.