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 | 595,356,997 | 9,374 | den-crane |
[testx.txt](https://github.com/ClickHouse/ClickHouse/files/4294283/testx.txt)
```
clickhouse-client -mn <testx.txt
---------------------------------------------------------
DIFF before merge
---------------------------------------------------------
-----------------------sum(avg over all---------------------... | 2020-03-05T17:42:07 | 2020-03-05T17:42:07 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 595,372,884 | 9,489 | alexey-milovidov | ```
CREATE TABLE test (a DEFAULT b, b DEFAULT a) ENGINE = Memory
```
It works!
... and what are the types of `a` and `b`? | 2020-03-05T18:16:41 | 2020-03-05T18:16:41 | {} | MEMBER |
ClickHouse | ClickHouse | 595,454,594 | 8,836 | alexey-milovidov | The task that has been added did not pass. | 2020-03-05T21:23:44 | 2020-03-05T21:23:44 | {} | MEMBER |
ClickHouse | ClickHouse | 595,812,917 | 8,994 | alexey-milovidov | @amosbird I have thoughts about how to make this approach more generic:
1. Implement all random generators as (ordinary) SQL functions, similar to rand, rand64, randomPrintableASCII.
2. The table function `generate` should have no clue about generating random data, instead - it should generate default values accord... | 2020-03-06T15:14:43 | 2020-03-06T15:16:20 | {} | MEMBER |
ClickHouse | ClickHouse | 595,982,463 | 9,541 | den-crane | >WHERE date = yesterday()
You MV will get a new data only if `insert into mytable_sharded` will insert `date = yesterday()`
You can create MV without this condition `WHERE date = yesterday()`
With daily partitioning ` PARTITION BY toYYYYMMDD(date)`
And remove data older than yesterday by `drop partition` or by ... | 2020-03-06T21:57:52 | 2020-03-06T22:01:14 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 596,058,880 | 9,544 | filimonov | Also - please use the latest 20.1 release (there were few bugs fixed just 2 days ago) | 2020-03-07T07:52:51 | 2020-03-07T07:52:51 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 596,075,744 | 9,556 | mbtolou | ```
SELECT
TEMP_ENRG.FA,
T2.FA
FROM
MDM.ENERGY_C_T TEMP_ENRG,
MDM.ENERGY_C_T T2
where
T2.TV = TEMP_ENRG.TV
and T2.SERVER_ID = TEMP_ENRG.SERVER_ID
and T2.DATA_ID = TEMP_ENRG.DATA_ID
AND TEMP_ENRG.SERVER_ID = 1
AND TEMP_ENRG.DATA_ID = '711688325'
```
Simple join for one table in triple... | 2020-03-07T10:58:09 | 2020-03-07T14:17:28 | {} | NONE |
ClickHouse | ClickHouse | 595,212,038 | 9,525 | alexey-milovidov | Also removed `test` prefix to use random database. | 2020-03-05T12:49:33 | 2020-03-05T12:49:33 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 595,281,911 | 4,132 | alexey-milovidov | @dmkoltsov
Please share all the build options that were used for ppc64le build. | 2020-03-05T15:14:31 | 2020-03-05T15:14:31 | {} | MEMBER |
ClickHouse | ClickHouse | 595,355,367 | 4,132 | alexey-milovidov | @dmkoltsov Thank you! Could you please also share these changes?
You can make a pull request (even if it cannot be merged to master as is, we can use it). | 2020-03-05T17:38:26 | 2020-03-05T17:38:26 | {} | MEMBER |
ClickHouse | ClickHouse | 595,430,559 | 9,529 | alexey-milovidov | ```
2596 git checkout master
2597 git checkout -b tac0x2a-gitlocalize
2598 git pull https://github.com/tac0x2a/ClickHouse.git gitlocalize-10969
2599 git pull https://github.com/tac0x2a/ClickHouse.git gitlocalize-10968
2600 cd ../..
2601 ls -l docs/ja/introduction/features_considered_disadvantages.md
... | 2020-03-05T20:27:06 | 2020-03-05T20:27:06 | {} | MEMBER |
ClickHouse | ClickHouse | 595,431,991 | 9,374 | den-crane | @alexey-milovidov so it's not a bug , but expected behavior ? Right? | 2020-03-05T20:30:52 | 2020-03-05T20:30:52 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 595,720,225 | 8,994 | qoega | @amosbird
Actually it is possible to make per-column configuration, but you probably need a specific case that you want to solve using this storage/tableFuntion and such distributions.
I tried to help with following cases:
1) I get a ClickHouse running and it has no data inside. I create a table with given nontri... | 2020-03-06T11:08:07 | 2020-03-06T11:08:07 | {} | MEMBER |
ClickHouse | ClickHouse | 595,782,824 | 9,538 | qoega | Can you show how you reproduce problem without fix? I would check it on Mac | 2020-03-06T14:07:12 | 2020-03-06T14:07:12 | {} | MEMBER |
ClickHouse | ClickHouse | 596,024,234 | 9,546 | alexey-milovidov | We have already addressed this issue, and don't allow to create executable dictionaries from SQL queries. | 2020-03-07T00:57:54 | 2020-03-07T00:59:31 | {} | MEMBER |
ClickHouse | ClickHouse | 596,029,347 | 9,271 | den-crane | `Break` interrupts execution only at the block level. This means that amount of returned rows is greater than [max_result_rows](#setting-max_result_rows), multiple of [max_block_size](settings.md#setting-max_block_size) and depends on [max_threads](settings.md#settings-max_threads). | 2020-03-07T01:34:42 | 2020-03-07T01:34:42 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 595,187,678 | 8,989 | alexey-milovidov | > @alexey-milovidov please, try to fix issue #8890 in 20.2 release. Can't work properly (expire policy doesn't work) with external dictionaries with source_type DBMS::MySQL at now.
We'll try. | 2020-03-05T11:43:10 | 2020-03-05T11:43:10 | {} | MEMBER |
ClickHouse | ClickHouse | 595,336,202 | 4,132 | dmkoltsov | 2 @alexey-milovidov
Here is the final string for build options, standing in ./Clickhouse_src_dir/build/ directory:
`# CC=cc CXX=c++ CXXFLAGS="-I/usr/local/include/" cmake -DUSE_INTERNAL_ZLIB_LIBRARY=FALSE -DUSE_INTERNAL_LIBGSASL_LIBRARY=OFF -DENABLE_HDFS=0 ..
`
But it shoud be said that many preliminary addi... | 2020-03-05T16:59:17 | 2020-03-05T17:02:53 | {} | NONE |
ClickHouse | ClickHouse | 595,423,854 | 9,515 | alexey-milovidov | `performance_introspection_and_logging` - failed as it should. | 2020-03-05T20:10:22 | 2020-03-05T20:10:22 | {} | MEMBER |
ClickHouse | ClickHouse | 595,431,447 | 9,526 | nikitamikhaylov | integration test doesn't work because i wrote wrong code | 2020-03-05T20:29:33 | 2020-03-05T20:29:33 | {} | MEMBER |
ClickHouse | ClickHouse | 595,569,542 | 9,531 | alexey-milovidov | > fatal error: error in backend: Sanitizer interface function redefined: void ()* bitcast (i32 (...)* @__msan_init to void ()*) | 2020-03-06T02:40:14 | 2020-03-06T02:40:14 | {} | MEMBER |
ClickHouse | ClickHouse | 595,584,691 | 9,519 | peaksnail | thanks, the subselects works for me
here is the bug
https://github.com/ClickHouse/ClickHouse/pull/7665
https://github.com/ClickHouse/ClickHouse/issues/5126
| 2020-03-06T03:39:30 | 2020-03-06T03:39:30 | {} | NONE |
ClickHouse | ClickHouse | 595,748,071 | 9,526 | qoega | - Probably it is better rename it to APPEND, as it is closer to what operation does
- It would be great to add idempotency as INSERT does for same data. I don't want to check manually whether command was successful if my connection closed unexpectedly. | 2020-03-06T12:31:55 | 2020-03-06T12:31:55 | {} | MEMBER |
ClickHouse | ClickHouse | 595,314,693 | 7,714 | d-ganchar | I had the same problem when some mutations were stuck and any new mutation didn't work through. So I **KILL** them but the problem remained. Query example:
```
SELECT field1,
field2,
count() counter
FROM my_table
WHERE toDate(dt) = toDate('2020-03-05')
AND field3 IN ('value1')
AN... | 2020-03-05T16:17:35 | 2020-03-05T16:17:35 | {} | NONE |
ClickHouse | ClickHouse | 595,430,843 | 9,529 | alexey-milovidov | Now I have to integrate your commit from https://github.com/ClickHouse/ClickHouse/pull/9222 to git history... | 2020-03-05T20:27:49 | 2020-03-05T20:27:49 | {} | MEMBER |
ClickHouse | ClickHouse | 595,480,907 | 9,378 | filimonov | Please check version 19.16.14.65-stable | 2020-03-05T22:33:04 | 2020-03-05T22:33:04 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 595,796,543 | 9,526 | nikitamikhaylov | https://clickhouse.tech/docs/en/query_language/alter/#alter_attach-partition-from
This feature does exactly what I need. :expressionless: Will try it on cluster copier and probably close this PR. | 2020-03-06T14:38:44 | 2020-03-06T14:44:07 | {} | MEMBER |
ClickHouse | ClickHouse | 595,911,210 | 6,478 | alexey-milovidov | Implemented and works in CI. | 2020-03-06T18:58:08 | 2020-03-06T18:58:08 | {} | MEMBER |
ClickHouse | ClickHouse | 596,023,673 | 9,508 | 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-... | 2020-03-07T00:54:07 | 2020-03-07T00:54:07 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 596,084,145 | 9,555 | alexey-milovidov | `00834_kill_mutation` - flacky test | 2020-03-07T12:40:16 | 2020-03-07T12:40:16 | {} | MEMBER |
ClickHouse | ClickHouse | 595,283,822 | 9,525 | alesapin | You have to set setting to 2, because 1 is default, so in the begging of the test:
```SET replication_alter_partitions_sync = 2;``` | 2020-03-05T15:17:57 | 2020-03-05T15:17:57 | {} | MEMBER |
ClickHouse | ClickHouse | 595,293,233 | 9,523 | alexey-milovidov | DDLWorker runs in the background. What kind of asynchronouty do you need? | 2020-03-05T15:36:12 | 2020-03-05T15:36:12 | {} | MEMBER |
ClickHouse | ClickHouse | 595,297,189 | 7,407 | filimonov | Failing test use newer feature which not exists yet in 19.16
Came from https://github.com/ClickHouse/ClickHouse/pull/9463/
```
2020-03-05 16:37:42 01090_zookeeper_mutations_and_insert_quorum: [ FAIL ] - return code 115
2020-03-05 16:37:42 Error on processing query: ALTER TABLE mutations_a... | 2020-03-05T15:43:49 | 2020-03-05T16:01:50 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 595,726,365 | 9,534 | qoega | There is a better name suggestion: `datagen` | 2020-03-06T11:25:54 | 2020-03-06T11:25:54 | {} | MEMBER |
ClickHouse | ClickHouse | 596,022,927 | 9,535 | alexey-milovidov | @blinkov Please review our `requirements.txt` file. | 2020-03-07T00:49:19 | 2020-03-07T00:49:19 | {} | MEMBER |
ClickHouse | ClickHouse | 596,055,840 | 9,544 | filimonov | It may be related to kafka_max_block_size (i recommend to change it to 512K in newer version).
Please see PR description and comments:
https://github.com/ClickHouse/ClickHouse/pull/8917#issue-369249074
What values of kafka_max_block_size, stream_poll_timeout_ms, stream_flush_interval_ms do you use?
Which forma... | 2020-03-07T07:11:25 | 2020-03-07T07:21:00 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 596,103,309 | 9,556 | den-crane | Join does not use indexes. Because currently CH supports only HashJoin. And HashJoin puts a right table into memory into HashTable with joinKey as a lookup key.
Try to swap tables
`MDM.ENERGY_C_T TEMP_ENRG, MDM.ENERGY_C_T T2` -->`MDM.ENERGY_C_T T2, MDM.ENERGY_C_T TEMP_ENRG`
Try `SET partial_merge_join = 1... | 2020-03-07T16:06:46 | 2020-03-07T16:06:46 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 595,200,362 | 9,484 | alexey-milovidov | I have added suppression for ASan and Unbundled builds. | 2020-03-05T12:17:22 | 2020-03-05T12:17:22 | {} | MEMBER |
ClickHouse | ClickHouse | 595,475,832 | 9,519 | 4ertus2 | 1. Have you tryed to use subselects for the right table? Also, there's an issue for left table's FINAL #8655
2. What's the bug with ANY INNER JOIN you are talking about? ANY INNER JOIN was changed in its semantic and gets ANY values from both tables. If you need ANY value from right table + filter with it you should ... | 2020-03-05T22:19:01 | 2020-03-05T22:19:01 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 595,901,278 | 9,539 | filimonov | I didn't check what's going exactly, if it doesn't affect server-to-server connections - it's not important. | 2020-03-06T18:32:41 | 2020-03-06T18:32:41 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 595,906,352 | 8,746 | alexey-milovidov | Failed? | 2020-03-06T18:45:57 | 2020-03-06T18:45:57 | {} | MEMBER |
ClickHouse | ClickHouse | 595,187,591 | 8,989 | alexey-milovidov | > FlatDictionary related segfault
It was induced bug, does not reproduce anymore. | 2020-03-05T11:42:54 | 2020-03-05T11:42:54 | {} | MEMBER |
ClickHouse | ClickHouse | 595,213,764 | 9,525 | alexey-milovidov | It did not help:
```
THREAD_FUZZER_CPU_TIME_PERIOD_US=1000 THREAD_FUZZER_SLEEP_PROBABILITY=0.1 THREAD_FUZZER_SLEEP_TIME_US=100000 ./clickhouse-server
```
```
00910_zookeeper_test_alter_compression_codecs: [ OK ]
1 tests passed. 0 tests skipped.
Running 1 stateless tests.
00910... | 2020-03-05T12:53:54 | 2020-03-05T12:53:54 | {} | MEMBER |
ClickHouse | ClickHouse | 595,392,756 | 9,489 | alexey-milovidov | > CREATE TABLE test (a DEFAULT b, b DEFAULT a) ENGINE = Memory
I can execute this query and it will prevent the server to start! | 2020-03-05T19:01:55 | 2020-03-05T19:01:55 | {} | MEMBER |
ClickHouse | ClickHouse | 596,097,194 | 2,829 | alexey-milovidov | This is quite easy task to do. Just add support for settings clause in configuration of dictionaries and in CREATE DICTIONARY query. | 2020-03-07T15:01:04 | 2020-03-07T15:01:04 | {} | MEMBER |
ClickHouse | ClickHouse | 595,432,102 | 9,222 | alexey-milovidov | @qoega You have failed merging this PR in 15 days. | 2020-03-05T20:31:12 | 2020-03-05T20:31:12 | {} | MEMBER |
ClickHouse | ClickHouse | 595,442,313 | 9,523 | filimonov | Check `distributed_ddl_task_timeout` setting | 2020-03-05T20:55:48 | 2020-03-05T20:55:48 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 595,554,014 | 9,360 | LiemLin |
> Try to update. There were several crashes in EngineJoin in your version.
May I ask which version to upgrade to？ | 2020-03-06T01:40:50 | 2020-03-06T01:40:50 | {} | NONE |
ClickHouse | ClickHouse | 595,840,173 | 9,338 | alexey-milovidov | `count(*)` is a synonim for `count()`
Despite the fact that the `*` is inside `count` it is not interpreted as count(all, columns).
If the user will explicitly list more than one column inside count, exception will be thrown.
This logic is implemented somewhere inside SyntaxAnalyzer.
But when the user creates m... | 2020-03-06T16:11:28 | 2020-03-06T16:12:00 | {} | MEMBER |
ClickHouse | ClickHouse | 595,271,119 | 9,142 | alexey-milovidov | > 00910_zookeeper_test_alter_compression_codecs
broken in master. | 2020-03-05T14:55:47 | 2020-03-05T14:55:47 | {} | MEMBER |
ClickHouse | ClickHouse | 595,272,504 | 9,142 | alexey-milovidov | The relevant changes are here: https://github.com/ClickHouse/ClickHouse/pull/9142/files#diff-a03f181ffa68622a640f819e4386cc3cR604 | 2020-03-05T14:58:20 | 2020-03-05T14:58:20 | {} | MEMBER |
ClickHouse | ClickHouse | 595,376,597 | 9,489 | alexey-milovidov | Will merge and then fix the case of
```
CREATE TABLE test (a DEFAULT b, b DEFAULT a) ENGINE = Memory
```
In a separate PR. | 2020-03-05T18:25:03 | 2020-03-05T18:25:03 | {} | MEMBER |
ClickHouse | ClickHouse | 595,520,260 | 9,518 | den-crane | > Is it possible that the failed mutation caused dirty data?
It could. We know that `alter delete` corrupts data in 20.1 https://github.com/ClickHouse/ClickHouse/issues/8821 . Maybe `alter update` corrupts it too.
Please provide reproducible example and more information: CH version, DDL, logs, data samlples, etc... | 2020-03-06T00:50:57 | 2020-03-06T00:50:57 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 595,904,503 | 9,538 | alexey-milovidov | Build is Ok. | 2020-03-06T18:41:10 | 2020-03-06T18:41:10 | {} | MEMBER |
ClickHouse | ClickHouse | 595,909,838 | 6,604 | alexey-milovidov | @KochetovNicolai decided to implement this feature. | 2020-03-06T18:54:49 | 2020-03-06T18:54:49 | {} | MEMBER |
ClickHouse | ClickHouse | 595,377,102 | 9,489 | alexey-milovidov | @alesapin I think that we can permit tables like this:
```
CREATE TABLE default.test (`a` String DEFAULT b, `b` String DEFAULT a) ENGINE = Memory
```
To allow a user to insert one or another column. This have its use. | 2020-03-05T18:26:11 | 2020-03-05T18:26:32 | {} | MEMBER |
ClickHouse | ClickHouse | 595,518,341 | 4,567 | den-crane | USING -- select * выбирает поля (по которым join) только из левой таблицы и поэтому в where как бы ссылаемся на левую таблицу, ровно так как это было в изначальном КХ.
ON -- select * выбирает из обеих таблиц | 2020-03-06T00:44:05 | 2020-03-06T00:46:03 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 595,556,166 | 9,478 | sundy-li | What about a triggered material view?
Just a prototype
```
create table a (did String, c1 String, c2 String, ingestTime DateTime );
create table b (did String, c3 String, c4 String, ingestTime DateTime);
create table a_b (did String, c1 String, c2 String, c3 String, c4 String);
CREATE MATERIALIZED VIEW a_b... | 2020-03-06T01:48:11 | 2020-03-06T01:49:38 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 595,565,236 | 9,523 | nicelulu | > DDLWorker runs in the background. What kind of asynchronouty do you need?
Because some time-consuming queries block the following DDL queries.
For example, when the data volume is large, execute ```ALTER TABLE table_name ON CLUSTER xxx MODIFY COLUMN column_name Int64```. Or a ```ReplicatedMergeTree``` table bec... | 2020-03-06T02:23:36 | 2020-03-06T02:23:36 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 595,799,837 | 8,944 | alexey-milovidov | Fixed long time ago. | 2020-03-06T14:46:15 | 2020-03-06T14:46:15 | {} | MEMBER |
ClickHouse | ClickHouse | 595,807,823 | 9,538 | alexey-milovidov | The fix looks Ok, but I cannot find `-U` flag in `man ld` or `man gold`.
Let's check - if the CI build (cross-compile for Darwin) will run (it is using `lld` linker), it's fine. | 2020-03-06T15:03:09 | 2020-03-06T15:03:25 | {} | MEMBER |
ClickHouse | ClickHouse | 595,845,281 | 8,594 | BayoNet | @excitoon
Hello.
Could you explain how to use it?
Update storage config first and call SYSTEM UPDATE CONFIG then?
Are there some tricky things? | 2020-03-06T16:22:24 | 2020-03-06T16:22:24 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 595,186,911 | 9,074 | alexey-milovidov | This task is almost trivial to implement. | 2020-03-05T11:41:21 | 2020-03-05T11:41:21 | {} | MEMBER |
ClickHouse | ClickHouse | 595,373,761 | 9,489 | alexey-milovidov | I will add a test with cyclic aliases and explicit types.
Then @alesapin will prohibit cyclic aliases with unknown types. | 2020-03-05T18:18:31 | 2020-03-05T18:18:31 | {} | MEMBER |
ClickHouse | ClickHouse | 595,374,616 | 9,489 | alexey-milovidov | But it does not work if I specify data types. | 2020-03-05T18:20:33 | 2020-03-05T18:20:33 | {} | MEMBER |
ClickHouse | ClickHouse | 595,435,506 | 9,374 | alexey-milovidov | I did not read all the thread but
if it works as I explained - yes, it is the expected behaviour.
It can be improved. | 2020-03-05T20:39:16 | 2020-03-05T20:39:16 | {} | MEMBER |
ClickHouse | ClickHouse | 595,571,054 | 9,360 | alexey-milovidov | @LiemLin You should upgrade to the latest stable version (20.1). | 2020-03-06T02:45:42 | 2020-03-06T02:45:42 | {} | MEMBER |
ClickHouse | ClickHouse | 595,902,359 | 9,539 | alexey-milovidov | The client cannot read logs that are sent with `send_logs_level` setting.
In distributed setup, servers will forward logs to the initiator server, then they are sent to the client and the client will show an error. | 2020-03-06T18:35:44 | 2020-03-06T18:35:44 | {} | MEMBER |
ClickHouse | ClickHouse | 595,911,518 | 6,387 | alexey-milovidov | Done. | 2020-03-06T18:58:45 | 2020-03-06T18:58:45 | {} | MEMBER |
ClickHouse | ClickHouse | 595,912,002 | 6,387 | alexey-milovidov | It works as expected and can find some unusual places where memory was allocated. Example:
```
trace: StackTrace::StackTrace()
MemoryTracker::alloc(long)
MemoryTracker::alloc(long)
operator new(unsigned long)
std::__1::deque<DB::PipelineExecutor::ExecutionState*, std::__1::allocator<DB::Pi... | 2020-03-06T18:59:49 | 2020-03-06T18:59:49 | {} | MEMBER |
ClickHouse | ClickHouse | 596,043,232 | 9,515 | alexey-milovidov | Another race condition found:
https://clickhouse-test-reports.s3.yandex.net/9515/aed0bf1654c31bfb5bf62deb48e329ed4036a2e3/functional_stateless_tests_(thread)/stderr.log | 2020-03-07T03:56:42 | 2020-03-07T03:56:42 | {} | MEMBER |
ClickHouse | ClickHouse | 595,751,396 | 9,526 | nikitamikhaylov | :sunglasses: | 2020-03-06T12:42:37 | 2020-03-06T12:42:37 | {} | MEMBER |
ClickHouse | ClickHouse | 595,784,626 | 9,538 | ghost | @qoega you can run `dbms/src/DataStreams/tests/collapsing_sorted_stream.cpp` to check | 2020-03-06T14:11:38 | 2020-03-06T14:11:38 | {} | NONE |
ClickHouse | ClickHouse | 595,905,255 | 9,531 | alexey-milovidov | Build is Ok, MSan version runs successfully! | 2020-03-06T18:43:00 | 2020-03-06T18:43:00 | {} | MEMBER |
ClickHouse | ClickHouse | 595,912,275 | 6,387 | alexey-milovidov | Also can be easily used to build memory flamegraphs. | 2020-03-06T19:00:28 | 2020-03-06T19:00:28 | {} | MEMBER |
ClickHouse | ClickHouse | 596,022,720 | 9,535 | alexey-milovidov | It looks like a garbage. Sphinx is not used at all:
```
~/work/ClickHouse/docs/tools$ grep -r -i sphinx .
./requirements.txt:Sphinx==1.6.5
./requirements.txt:sphinxcontrib-websupport==1.0.1
```
The `requirements.txt` is edited manually and sphinx was added two years ago:
```
bd98072259e docs/requirements.... | 2020-03-07T00:48:10 | 2020-03-07T00:48:10 | {} | MEMBER |
ClickHouse | ClickHouse | 596,041,720 | 9,516 | alexey-milovidov | Nice catch!
https://clickhouse-test-reports.s3.yandex.net/9516/27fba2278b3222198f6bb153475eb59a35b33855/functional_stateless_tests_(thread)/stderr.log | 2020-03-07T03:36:10 | 2020-03-07T03:36:10 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 595,204,682 | 9,373 | alexey-milovidov | I found that the inline documentation was absolutely wrong:
https://github.com/ClickHouse/ClickHouse/pull/9524
Please edit accordingly. | 2020-03-05T12:29:20 | 2020-03-05T12:29:20 | {} | MEMBER |
ClickHouse | ClickHouse | 595,403,967 | 9,374 | den-crane | hmm, it's reproducible on super simple test. Maybe I'm doing smth wrong.
```
drop table if exists byhour;
CREATE TABLE byhour
(
u String,
t DateTime,
fa AggregateFunction(avgIf,Float64,UInt8),
fm AggregateFunction(maxIf,Float64,UInt8),
fx Tuple( AggregateFuncti... | 2020-03-05T19:26:51 | 2020-03-05T19:26:51 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 595,404,139 | 9,527 | alexey-milovidov | @tavplubix Please add the same improvement to in-development Atomic database. | 2020-03-05T19:27:15 | 2020-03-05T19:27:15 | {} | MEMBER |
ClickHouse | ClickHouse | 595,609,277 | 8,994 | amosbird | A per-column configuration would be awesome. Some interesting distributions like zipf, pareto, binomial are often used in benchmarks. Perhaps we can steal from https://docs.scipy.org/doc/numpy/reference/random/generator.html | 2020-03-06T05:38:53 | 2020-03-06T05:38:53 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 595,635,669 | 9,373 | BayoNet | A documentation style is not only about presentation. Well organized docs more comfortable for uses and maintainers both.
Docs as code. Good text architecture simplifies:
- Visual search through text and its understanding.
- Updating and extending of docs by maintainers. | 2020-03-06T07:21:35 | 2020-03-06T07:21:35 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 595,845,649 | 9,539 | alexey-milovidov | I also have hit this issue and added a remark for changelog. | 2020-03-06T16:23:13 | 2020-03-06T16:23:13 | {} | MEMBER |
ClickHouse | ClickHouse | 595,846,604 | 8,594 | alexey-milovidov | @BayoNet The configuration will be updated automatically when user has changed the config file. | 2020-03-06T16:25:05 | 2020-03-06T16:25:16 | {} | MEMBER |
ClickHouse | ClickHouse | 596,045,369 | 9,323 | alexey-milovidov | The code is absolutely the same but this PR is up to 10% worse:
```
SELECT count() FROM system.numbers WHERE NOT ignore(toDateTime('2017-01-01 00:00:00') + number)
Bad: 855.21 million rows/s.
AddSecondsImpl constant_vector
14.85 │50: movdqu (%rdx,%rax,2),%xmm0
46.97 │ movdqu 0x10(%rdx,%rax,2),%xm... | 2020-03-07T04:29:40 | 2020-03-07T04:29:40 | {} | MEMBER |
ClickHouse | ClickHouse | 596,093,041 | 2,829 | gongled | Any news? I am facing the same issue with an external dictionaries in CSV format. I have already bypassed it, but it would be nice to implement setting overriding for dictionaries.
| 2020-03-07T14:17:36 | 2020-03-07T14:17:36 | {} | NONE |
ClickHouse | ClickHouse | 595,192,734 | 9,373 | BayoNet | Hello.
I don't see any defects in the content, but we are trying to implement a single style in the documentation. Applying the [template](https://github.com/ClickHouse/ClickHouse/blob/master/docs/dscr-templates/template-setting.md), you make your text look better :-) | 2020-03-05T11:56:13 | 2020-03-05T11:56:13 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 595,204,835 | 9,373 | alexey-milovidov | @BayoNet
> Applying the template, you make your text look better :-)
Ok, but I don't care. | 2020-03-05T12:29:44 | 2020-03-05T12:29:44 | {} | MEMBER |
ClickHouse | ClickHouse | 595,409,795 | 9,374 | alexey-milovidov | Tuple of AggregateFunction is not an AggregateFunction - it will not aggregate during merge but only select the first value. | 2020-03-05T19:39:26 | 2020-03-05T19:39:26 | {} | MEMBER |
ClickHouse | ClickHouse | 595,431,591 | 9,529 | alexey-milovidov | ```
git merge 1de7401
git status
git checkout --ours docs/ja/introduction/history.md
git add docs/ja/introduction/history.md
git commit
``` | 2020-03-05T20:29:56 | 2020-03-05T20:29:56 | {} | MEMBER |
ClickHouse | ClickHouse | 595,432,083 | 9,220 | alexey-milovidov | @qoega You have failed merging this PR in 15 days. | 2020-03-05T20:31:09 | 2020-03-05T20:31:09 | {} | MEMBER |
ClickHouse | ClickHouse | 595,952,248 | 9,545 | And42 | @alexey-milovidov discussion-ready pull request created 🎉🎉🎉 | 2020-03-06T20:29:08 | 2020-03-06T20:29:08 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 596,022,726 | 9,535 | 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-... | 2020-03-07T00:48:12 | 2020-03-07T00:48:12 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 596,024,493 | 9,546 | alexey-milovidov | https://github.com/ClickHouse/ClickHouse/pull/8127 | 2020-03-07T00:59:24 | 2020-03-07T00:59:24 | {} | MEMBER |
ClickHouse | ClickHouse | 595,215,054 | 9,525 | alexey-milovidov | The setting `SET replication_alter_partitions_sync = 1;` also does not help. | 2020-03-05T12:57:11 | 2020-03-05T12:57:11 | {} | MEMBER |
ClickHouse | ClickHouse | 595,367,008 | 8,994 | alexey-milovidov | Will fix some minor issues after merge. | 2020-03-05T18:04:22 | 2020-03-05T18:04:22 | {} | MEMBER |
ClickHouse | ClickHouse | 595,375,101 | 9,489 | alexey-milovidov | @filimonov Your queries don't work. I will add a test to ensure that they don't work. | 2020-03-05T18:21:41 | 2020-03-05T18:21:41 | {} | MEMBER |
ClickHouse | ClickHouse | 595,392,099 | 9,374 | den-crane | ```
insert into m values ('2019-10-08 10:10:51','1','1',0) ('2019-10-08 11:10:51','1','1',0);
insert into m values ('2019-10-08 10:10:11','1','1',1) ('2019-10-08 11:10:11','1','1',10);
insert into m values ('2019-10-08 10:10:51','1','1',0) ('2019-10-08 11:10:51','1','1',0);
insert into m values ('2019-10-08 10:10:1... | 2020-03-05T19:00:27 | 2020-03-05T19:04:54 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 595,671,866 | 2,345 | filimonov | Also https://github.com/ClickHouse/ClickHouse/pull/9261 (for those who look by issue title) | 2020-03-06T09:05:41 | 2020-03-06T09:12:23 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 578,454,063 | 7,207 | alexey-milovidov | I close this issue because it is resolved.
For the reference: it is possible to configure ClickHouse to run successfully even on machine with 4 GiB RAM. Even configurations with 1 GiB RAM will work after careful tuning. | 2020-01-25T23:45:53 | 2020-01-25T23:45:53 | {} | MEMBER |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.