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 | 494,049,532 | 5,212 | alexey-milovidov | Ok. | 2019-05-20T16:02:10 | 2019-05-20T16:02:10 | {} | MEMBER |
ClickHouse | ClickHouse | 494,125,938 | 5,354 | vitlibar | You're updating simdjson to the commit 262d9e, but it seems there are no such commit:
https://github.com/lemire/simdjson/commits/master | 2019-05-20T19:49:01 | 2019-05-20T19:49:01 | {} | MEMBER |
ClickHouse | ClickHouse | 494,281,011 | 5,356 | alesapin | `> Fix test that doesn't check anything`
What does it mean? This test check that clickhouse server can work in readonly mode without zookeeper. There is no need in retries, because all queries execute locally.
Changes seem unclear for me. | 2019-05-21T07:52:08 | 2019-05-21T07:54:46 | {} | MEMBER |
ClickHouse | ClickHouse | 493,500,116 | 5,287 | abraithwaite | @alexey-milovidov I'm curious what your thoughts on this are! I wouldn't mind taking a stab at fixing it but it might take me a while as I'm pretty unfamiliar with C++. :-)
Thanks! | 2019-05-17T15:40:50 | 2019-05-17T15:40:50 | {} | NONE |
ClickHouse | ClickHouse | 493,573,049 | 4,853 | alexey-milovidov | What is the size of the partition 201901 in compressed and uncompressed bytes?
(You can obtain this info from `system.parts` table.)
Note that OPTIMIZE FINAL forces full data rewrite and it can exceed the default limit for 150 GB.
High memory usage for merge is not expected, even for huge merges. The memory usag... | 2019-05-17T19:32:13 | 2019-05-17T19:32:13 | {} | MEMBER |
ClickHouse | ClickHouse | 493,664,821 | 5,116 | alexey-milovidov | `rows_in_set` is slightly misleading, because there is a notion of "set" in ClickHouse and it is completely different thing. Let it be `total_rows_to_read`. Yes, it's ok to have this backward incompatible change (need to write in changelog), because progress in HTTP headers is used not very wide. | 2019-05-18T10:03:24 | 2019-05-18T10:03:24 | {} | MEMBER |
ClickHouse | ClickHouse | 493,665,035 | 5,092 | alexey-milovidov | ? | 2019-05-18T10:06:59 | 2019-05-18T10:06:59 | {} | MEMBER |
ClickHouse | ClickHouse | 493,880,771 | 4,026 | wawanawna | > Hi @wawanawna,
> What is the workaround you found ?
We just stopped using MVs @nupursinghbaghel | 2019-05-20T08:04:20 | 2019-05-20T08:04:20 | {
"+1": 2,
"-1": 4
} | NONE |
ClickHouse | ClickHouse | 493,971,028 | 5,344 | filimonov | Please examine (or just attach) the contents of log files:
* /var/log/clickhouse-server/clickhouse-server.err.log
* /var/log/clickhouse-server/clickhouse-server.log
| 2019-05-20T12:43:22 | 2019-05-20T12:43:22 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 494,101,974 | 5,351 | Crazylearner30 | Hi sorry i was using todate instead of toDate('YYYY-MM-DD')
Its working and thanks for the quick reply :) | 2019-05-20T18:34:21 | 2019-05-20T18:34:21 | {} | NONE |
ClickHouse | ClickHouse | 494,251,441 | 5,339 | sundy-li | > There is UBSan report: https://clickhouse-test-reports.s3.yandex.net/5339/66c60c2985c9a0c677ea807f5c24ee80eaeedc46/functional_stateless_tests_(ubsan)/stderr.txt
I still can't figure out how to solve this panic error. :( | 2019-05-21T06:01:19 | 2019-05-21T06:01:19 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 493,259,566 | 3,182 | alexey-milovidov | Does not work at all. Tests were bogus and code review was pointless. | 2019-05-16T23:00:15 | 2019-05-16T23:00:15 | {} | MEMBER |
ClickHouse | ClickHouse | 493,398,912 | 5,317 | alexey-milovidov | > decreasing is very slow currently (halved every 10 minutes)
For whatever reason, I remember that my first intention was to halve every 60 seconds. It means that 1000 errors will vanish after 10 minutes. Let's change it accordingly...
> Maybe we should introduce some upper limit for error_count and never increas... | 2019-05-17T10:03:05 | 2019-05-17T10:05:05 | {} | MEMBER |
ClickHouse | ClickHouse | 493,402,477 | 2,384 | alexey-milovidov | Спасибо, как раз эта ошибка исправляется в моём pull request! | 2019-05-17T10:15:42 | 2019-05-17T10:15:42 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 493,682,563 | 5,274 | den-crane | Deduplication blocks can be implemented for all engines (including null) (or even at the INSERT level).
Block's hashes can be stored in-memory (or + in a file near the table).
Then the engine (for example Null) can check that it's a duplicate and reject the block (not propagate that block to MVs).
Then insert_de... | 2019-05-18T14:47:51 | 2019-05-18T14:47:51 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 493,702,726 | 5,328 | blinkov | cc @zhang2014 @sundy-li | 2019-05-18T19:50:59 | 2019-05-18T19:50:59 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 493,717,383 | 5,319 | alexey-milovidov | As an example, we allowed to use Nullables for minmax secondary index, but some issues start to appear #5246 | 2019-05-19T00:39:01 | 2019-05-19T00:39:01 | {} | MEMBER |
ClickHouse | ClickHouse | 493,969,248 | 5,301 | kreuzerkrieg | My 5 cents. Just finished building from clean clone. 40 minutes, on 8 cores, including all ./contrib/ stuff. AFAIR `cotire` is for precompiled header, in any case, there is another approach to speed-up the build. Why not to push all third parties to package manager like `vcpkg` or `conan` (upstream or your own fork), s... | 2019-05-20T12:38:21 | 2019-05-20T12:38:21 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 494,083,421 | 5,340 | filimonov | Looks like p.6 (llvm & symbolizer) is not needed here (clickhouse is executed in another container) | 2019-05-20T17:40:07 | 2019-05-20T17:40:07 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 494,178,293 | 5,341 | alexey-milovidov | How do you build ClickHouse? Does the issue reproduce if you use the official packages? | 2019-05-20T22:54:37 | 2019-05-20T22:54:37 | {} | MEMBER |
ClickHouse | ClickHouse | 493,378,673 | 5,316 | alexey-milovidov | > and copy the code from the driver directory to compiler_8.0.0
Verbatim copy wont work. | 2019-05-17T08:59:18 | 2019-05-17T08:59:18 | {} | MEMBER |
ClickHouse | ClickHouse | 493,386,975 | 5,315 | filimonov | Maybe I've used bad description for that issue.
I know that other threads are stopped, but excessive work done by them (while processing last blocks of data) are clearly visible in lot of cases. | 2019-05-17T09:23:59 | 2019-05-17T09:23:59 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 493,422,958 | 5,320 | sundy-li | Maybe the first argument of `sequenceMatch` should also accept other Unsigned Integer Types? | 2019-05-17T11:38:25 | 2019-05-17T11:38:25 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 493,446,503 | 4,548 | hellmany | Is there any updates on this feature ? | 2019-05-17T13:06:31 | 2019-05-17T13:06:31 | {} | NONE |
ClickHouse | ClickHouse | 493,706,496 | 4,965 | alexey-milovidov | @stavrolia ask @KochetovNicolai for details of this task. | 2019-05-18T20:53:25 | 2019-05-18T20:53:34 | {} | MEMBER |
ClickHouse | ClickHouse | 493,863,193 | 751 | izebit | @ludv1x Are you going to do something about it? It's actually a problem. At least, we have to notice about this in the documentation about AggregatingMergeTree. | 2019-05-20T07:05:16 | 2019-05-20T07:05:16 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 494,116,337 | 5,301 | kreuzerkrieg | > Yes, ClickHouse requires specific version (more exactly - fixed source code) of libraries for reproducible builds. This is needed to rebuild code with different options, different ABI, sanitizers, to easy debugging, etc.
I'm a bit confused, so, what the `unbundled` mode is good for? | 2019-05-20T19:17:36 | 2019-05-20T19:17:36 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 493,238,014 | 4,525 | proller | I want to add to many cmake/find_*.cmake something like
```
if (DEFINED ENABLE_XXX AND NOT ENABLE_XXX)
return()
endif()```
| 2019-05-16T21:27:14 | 2019-05-16T21:27:14 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 493,301,825 | 5,306 | dlesnikov | configuration files.
[configs.zip](https://github.com/yandex/ClickHouse/files/3189750/configs.zip)
| 2019-05-17T03:03:22 | 2019-05-17T03:03:22 | {} | NONE |
ClickHouse | ClickHouse | 493,637,082 | 5,320 | sundy-li | > Doesn't that change make the binary data (i.e. serialized function state) incompatible between ch versions?
It's done by c++ template code generation. So the oldest version(UInt32) still keep the same binary
function state size. | 2019-05-18T01:13:47 | 2019-05-18T01:13:47 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 493,665,090 | 4,724 | alexey-milovidov | ? | 2019-05-18T10:07:45 | 2019-05-18T10:07:45 | {} | MEMBER |
ClickHouse | ClickHouse | 493,932,428 | 5,116 | YiuRULE | One of the reason about why it's that sometime, it's seem during `INSERT` query ClickHouse ignore totally the Progress. (using this branch, with the initial configuration)
Like on the example I write in the beginning of the PR, with the argument `http_headers_progress_interval_ms=0`, the only time time that ClickHou... | 2019-05-20T10:38:13 | 2019-05-20T10:38:13 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 494,121,525 | 5,301 | alexey-milovidov | > I'm a bit confused, so, what the unbundled mode is good for?
Non production builds and builds for partially supported platforms where libraries are used from OS packages. | 2019-05-20T19:34:20 | 2019-05-20T19:35:00 | {} | MEMBER |
ClickHouse | ClickHouse | 494,223,470 | 5,244 | andyyzh | Automatic type of bitmap conversion is expensive,
Only add parameter type check.
| 2019-05-21T03:14:12 | 2019-05-21T03:14:12 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 493,401,362 | 5,307 | alexey-milovidov | Wrong title. | 2019-05-17T10:11:40 | 2019-05-17T10:11:40 | {} | MEMBER |
ClickHouse | ClickHouse | 493,422,157 | 5,320 | sundy-li | #5277 | 2019-05-17T11:34:48 | 2019-05-17T11:34:48 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 494,045,065 | 5,212 | andyyzh | Prepare to add these functions that similar to mysql, and add new mode:
> WEEK(date[,mode])
YEAR(date[,mode])
YEARWEEK(date[,mode])
I'll commit it in two weeks. | 2019-05-20T15:50:53 | 2019-05-20T15:50:53 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 494,181,863 | 5,339 | alexey-milovidov | There is UBSan report: https://clickhouse-test-reports.s3.yandex.net/5339/66c60c2985c9a0c677ea807f5c24ee80eaeedc46/functional_stateless_tests_(ubsan)/stderr.txt | 2019-05-20T23:12:35 | 2019-05-20T23:12:35 | {} | MEMBER |
ClickHouse | ClickHouse | 493,375,164 | 5,315 | alexey-milovidov | We already do "Stop other threads after LIMIT number of rows were found". | 2019-05-17T08:47:52 | 2019-05-17T08:47:52 | {} | MEMBER |
ClickHouse | ClickHouse | 493,788,205 | 5,321 | filimonov | BTW: it would be better if clickhouse-client will be able to use CLICKHOUSE_USER / CLICKHOUSE_PASSWORD directly (currently password is exposed into args). | 2019-05-19T20:18:02 | 2019-05-19T20:18:02 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 493,959,305 | 5,246 | nikvas0 | > What's wrong with `ColumnNullable::getExtremes`?
It's available only for Int* and Float* types. | 2019-05-20T12:08:39 | 2019-05-20T12:08:39 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 494,042,750 | 1,148 | fsanaulla | @alexey-milovidov it's great news, thank you! | 2019-05-20T15:44:45 | 2019-05-20T15:44:45 | {} | NONE |
ClickHouse | ClickHouse | 494,115,650 | 5,301 | alexey-milovidov | Yes, ClickHouse requires specific version (more exactly - fixed source code) of libraries for reproducible builds. This is needed to rebuild code with different options, different ABI, sanitizers, to easy debugging, etc. | 2019-05-20T19:15:24 | 2019-05-20T19:15:24 | {} | MEMBER |
ClickHouse | ClickHouse | 494,361,586 | 5,208 | ghuname | In our case server version is 18.14.15.
I misreported the issue. I appologize. Sorry.
We are not loosing kafka messages. After restarting clickhouse server, last messages are duplicated (not lost).
Looks like clickhouse has collected messages from kafka and put them into tables but haven't sent acknowledge ba... | 2019-05-21T12:01:55 | 2019-05-21T12:01:55 | {} | NONE |
ClickHouse | ClickHouse | 493,102,777 | 5,301 | alexey-milovidov | We have not tried `cotire`.
Slow build time is mostly unavoidable. Although some parts can be sped up: for example, translation units for `cache` dictionaries are using excessive amount of template instantiations.
To speed up build time on developer machine, you can:
- install ccache;
- use debug build;
- use ... | 2019-05-16T14:56:41 | 2019-05-17T08:53:10 | {} | MEMBER |
ClickHouse | ClickHouse | 493,378,280 | 5,311 | alexey-milovidov | @bgranvea | 2019-05-17T08:58:03 | 2019-05-17T08:58:03 | {} | MEMBER |
ClickHouse | ClickHouse | 493,471,328 | 1,779 | filimonov | Can't reproduce (checked on 18.4).
Most probably was fixed in the meanwhile. | 2019-05-17T14:20:47 | 2019-05-17T14:20:47 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 493,664,230 | 5,200 | alexey-milovidov | The only remaining is to fix UBSan:
https://clickhouse-test-reports.s3.yandex.net/5200/cf95327cba4d764aa3c12a1140e3db43a13c7421/functional_stateless_tests_(ubsan)/stderr.txt | 2019-05-18T09:53:37 | 2019-05-18T09:53:37 | {} | MEMBER |
ClickHouse | ClickHouse | 493,705,732 | 5,274 | filimonov | But also valid scenarios of such deduplucation are possible. Let's say you insert to A, matview copy block to B & C, but fails on D. Client repeats insert expecting indempotency and w/o deduplucation data in B and C will be duplicated.
Real atomic inserts needed (table - all MV) | 2019-05-18T20:40:32 | 2019-05-18T20:40:32 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 493,717,328 | 5,319 | alexey-milovidov | Hi! This is doable but multiple potential complications exist.
Relevant parts of code:
`KeyCondition`, `FieldVisitorAccurateLess`, `FieldVisitorAccurateEquals` | 2019-05-19T00:37:42 | 2019-05-19T00:37:42 | {} | MEMBER |
ClickHouse | ClickHouse | 494,028,525 | 1,148 | fsanaulla | Hello everyone, any update? | 2019-05-20T15:10:08 | 2019-05-20T15:10:08 | {} | NONE |
ClickHouse | ClickHouse | 494,038,562 | 1,148 | alexey-milovidov | Sorry, no update yet. I will coalesce this task with #5343 for @4ertus2 | 2019-05-20T15:34:04 | 2019-05-20T15:34:04 | {} | MEMBER |
ClickHouse | ClickHouse | 494,286,033 | 5,356 | alesapin | Discussed in chat. | 2019-05-21T08:06:44 | 2019-05-21T08:06:44 | {} | MEMBER |
ClickHouse | ClickHouse | 493,226,371 | 3,540 | alexey-milovidov | > Additionally, if you have single IN parameter, query is passed to mysql, but with wrong syntax (no parentheses for IN clause)
https://github.com/yandex/ClickHouse/pull/5313 | 2019-05-16T20:48:18 | 2019-05-16T20:48:18 | {} | MEMBER |
ClickHouse | ClickHouse | 493,249,858 | 5,301 | nvartolomei | @filimonov I didn't test it and have no real incentive for it, plus build systems is not something I want to spend time on. It is more like a suggestion for the team to look into this if they have the incentive, or maybe someone else who has time and motivation to contribute :)
@alexey-milovidov thanks for suggestio... | 2019-05-16T22:13:02 | 2019-05-16T22:13:02 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 493,375,744 | 5,315 | alexey-milovidov | If the query does not finish in a reasonable amount of time after result was sent, it's a bug. | 2019-05-17T08:49:59 | 2019-05-17T08:49:59 | {} | MEMBER |
ClickHouse | ClickHouse | 493,940,811 | 5,340 | alesapin | Build failures are OK. | 2019-05-20T11:06:12 | 2019-05-20T11:06:12 | {} | MEMBER |
ClickHouse | ClickHouse | 494,338,463 | 4,248 | abyss7 | According to preliminary data the rate of messages from Kafka is about 30K/s. It may cause the client to form a huge commit-requests' queue, where some of the requests may get timed out. And then begins something unclear. | 2019-05-21T10:40:52 | 2019-05-21T10:40:52 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 493,706,485 | 4,853 | alexey-milovidov | https://github.com/yandex/ClickHouse/issues/4965 | 2019-05-18T20:53:08 | 2019-05-18T20:53:08 | {} | MEMBER |
ClickHouse | ClickHouse | 493,820,685 | 5,328 | zhang2014 | LGTM | 2019-05-20T02:39:23 | 2019-05-20T02:39:23 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 493,991,868 | 5,301 | proller | its possible with unbundled mode: cmake -DUNBUNDLED=1
example for debian/ubuntu:
https://github.com/yandex/ClickHouse/blob/master/utils/build/build_debian_unbundled.sh | 2019-05-20T13:40:09 | 2019-05-20T13:41:08 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 494,244,558 | 5,202 | BayoNet | @vitlibar Hello. I have changed text a bit. Has It became clearer? | 2019-05-21T05:25:49 | 2019-05-21T05:27:08 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 493,389,938 | 2,384 | TimonKK | Схема таблицы:
CREATE TABLE users (_id String, email String) ENGINE = MySQL('localhost:3306', 'db', 'table', 'username', 'password')
Запрос:
SELECT _id FROM users AS u WHERE ( u.createdLang IN ('ru') AND u.isAdmin = 0 AND u.role = 'guest' )
Ошибка:
Received exception from server (version 19.5.2):
Code: 1000.... | 2019-05-17T09:34:01 | 2019-05-17T09:34:01 | {} | NONE |
ClickHouse | ClickHouse | 493,566,032 | 3,834 | alexey-milovidov | We are planning to fix this issue in nearest month.
More details: https://yandex.github.io/clickhouse-presentations/database_saturday_2018_2/octonica/meetup.pptx (slide 12). | 2019-05-17T19:08:11 | 2019-05-17T19:08:11 | {} | MEMBER |
ClickHouse | ClickHouse | 493,605,264 | 5,279 | alexey-milovidov | @svladykin Please check the failing tests. | 2019-05-17T21:30:35 | 2019-05-17T21:30:35 | {} | MEMBER |
ClickHouse | ClickHouse | 493,667,945 | 5,279 | svladykin | Thanks, will look! | 2019-05-18T10:55:05 | 2019-05-18T10:55:05 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 494,097,578 | 5,351 | filimonov | ```
insert into table2
select a,b,c,d,toDate('2019-12-09') from table1
``` | 2019-05-20T18:22:01 | 2019-05-20T18:22:01 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 494,099,256 | 5,351 | den-crane | > i tried this and its not working :(
with what error ??? | 2019-05-20T18:26:43 | 2019-05-20T18:26:43 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 493,232,438 | 3,893 | den-crane | CH JDBC driver overrides server settings. And sets own parameter socket_timeout = 30000 (30s) (by default).
But of course DBeaver allows to fix it for a connection `Under the Advanced tab, you can configure advanced connections settings, e.g., Character Coding.`
And set larger socket_timeout.
| 2019-05-16T21:07:48 | 2019-05-16T21:11:30 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 493,578,039 | 5,131 | alexey-milovidov | The reason for this bug:
Wrong analysis for WHERE expressions for index usage. This analysis is implemented in `KeyCondition.h`, `KeyCondition.cpp`. It uses calculations on ranges of `Field`s. But Fields lack information about data type. `Date` and `UInt16` (also `DateTime` and `UInt32`) are represented in the same ... | 2019-05-17T19:49:35 | 2019-05-17T19:50:51 | {} | MEMBER |
ClickHouse | ClickHouse | 493,684,261 | 4,853 | den-crane | Just a side note: LowCardinality(Nullable(Int8)) is always worse than Nullable(Int8).
In 99% cases LowCardinality makes sense only for String type.
Try to use CODEC Delta instead of LowCardinality(Float64). | 2019-05-18T15:11:50 | 2019-05-18T15:12:45 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 494,046,481 | 3,884 | 4ertus2 | #5350 | 2019-05-20T15:54:27 | 2019-05-20T15:54:27 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 494,077,042 | 4,248 | ajardan | We have the same issue on 19.6.2.11. CH starts consuming from Kafka, and dies soon.
Is there any information I can share to help fix this issue ? | 2019-05-20T17:21:22 | 2019-05-20T17:21:22 | {} | NONE |
ClickHouse | ClickHouse | 494,114,658 | 5,301 | kreuzerkrieg | > its possible with unbundled mode: cmake -DUNBUNDLED=1
> example for debian/ubuntu:
> https://github.com/yandex/ClickHouse/blob/master/utils/build/build_debian_unbundled.sh
Do you see any obvious drawbacks of using this approach? Doesnt the `ClickHouse` tightly coupled to some specific version of libraries in `co... | 2019-05-20T19:12:05 | 2019-05-20T19:12:05 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 494,329,041 | 4,853 | dysosmus | @alexey-milovidov
> What is the size of the partition 201901 in compressed and uncompressed bytes?
> (You can obtain this info from system.parts table.)
```
SELECT
partition,
name,
sum(rows),
sum(bytes_on_disk),
sum(data_compressed_bytes),
sum(data_uncompressed_bytes)
FROM system... | 2019-05-21T10:11:14 | 2019-05-21T10:11:44 | {} | NONE |
ClickHouse | ClickHouse | 493,375,508 | 5,315 | alexey-milovidov | The time to wait is to correctly finish the threads (wait for the last block to process).
In case of distributed query, we send cancellation to remote servers and wait them to stop the query processing and to response. | 2019-05-17T08:49:02 | 2019-05-17T08:49:02 | {} | MEMBER |
ClickHouse | ClickHouse | 493,606,250 | 5,069 | alexey-milovidov | Please review the checks. | 2019-05-17T21:34:37 | 2019-05-17T21:34:37 | {} | MEMBER |
ClickHouse | ClickHouse | 493,665,012 | 5,116 | alexey-milovidov | > The problem is currently the progress concerning the writing is not on real time (we get it on the end), so it's mean we send values who will not change on the progress but only in the summary, does it's ok ?
Why? | 2019-05-18T10:06:31 | 2019-05-18T10:06:31 | {} | MEMBER |
ClickHouse | ClickHouse | 493,322,389 | 5,305 | lzy305 | @proller
| 2019-05-17T05:13:37 | 2019-05-17T05:13:37 | {} | NONE |
ClickHouse | ClickHouse | 493,483,052 | 4,585 | SaltTan | I haven't seen it after #4581 was fixed | 2019-05-17T14:53:12 | 2019-05-17T14:53:12 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 493,501,544 | 5,320 | filimonov | Doesn't that change make the binary data (i.e. serialized function state) incompatible between ch versions? | 2019-05-17T15:45:08 | 2019-05-17T15:45:46 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 493,658,266 | 5,274 | nupursinghbaghel | Amazing this is. Thanks much @den-crane :) | 2019-05-18T08:10:42 | 2019-05-18T08:10:42 | {} | NONE |
ClickHouse | ClickHouse | 493,662,523 | 5,320 | alexey-milovidov | > Maybe the first argument of sequenceMatch should also accept other Unsigned Integer Types?
Yes, it will be much better for consistency. | 2019-05-18T09:28:04 | 2019-05-18T09:28:04 | {} | MEMBER |
ClickHouse | ClickHouse | 493,688,606 | 5,320 | sundy-li | > Do you want to make another PR for sequenceMatch - so I can merge this PR?
Ok, I will make another PR in a few days later. | 2019-05-18T16:15:24 | 2019-05-18T16:15:24 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 493,745,545 | 4,631 | simPod | I had to stabilize production solution so I dropped all kafka engine consumers, replaced by java consumer-inserters. Unfortunately, I can no longer help with this then 😞 | 2019-05-19T10:34:18 | 2019-05-19T10:34:18 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 494,122,880 | 5,301 | kreuzerkrieg | > I'm a bit confused, so, what the unbundled mode is good for?
>
> Non production builds and builds for partially supported platforms where libraries are used from OS packages.
oh... I see... was afraid this will be your answer. we dont see this approach as acceptable. We want the CI/production/dev env to yield e... | 2019-05-20T19:39:03 | 2019-05-20T19:39:03 | {
"+1": 2
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 494,127,200 | 5,354 | alexey-milovidov | Ok. | 2019-05-20T19:53:02 | 2019-05-20T19:53:02 | {} | MEMBER |
ClickHouse | ClickHouse | 493,274,376 | 5,314 | excitoon | I hereby agree to the terms of the CLA available at: https://yandex.ru/legal/cla/?lang=en
| 2019-05-17T00:20:24 | 2019-05-17T00:20:24 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 493,940,535 | 5,308 | abyss7 | > Isn't it better to change to zero by default?
I will try and check how it correlates with `skip_broken_messages` feature. | 2019-05-20T11:05:17 | 2019-05-20T11:05:17 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 494,097,867 | 5,351 | Crazylearner30 | i tried this and its not working :( | 2019-05-20T18:22:50 | 2019-05-20T18:22:50 | {} | NONE |
ClickHouse | ClickHouse | 493,259,677 | 3,182 | alexey-milovidov | https://github.com/yandex/ClickHouse/pull/5313 | 2019-05-16T23:00:45 | 2019-05-16T23:00:45 | {} | MEMBER |
ClickHouse | ClickHouse | 493,564,333 | 4,919 | alexey-milovidov | Fixed. | 2019-05-17T19:02:45 | 2019-05-17T19:02:45 | {} | MEMBER |
ClickHouse | ClickHouse | 493,564,846 | 4,714 | alexey-milovidov | Does it reproduce on 19.6?
We have made a bunch of fixes for table structure locks. | 2019-05-17T19:04:22 | 2019-05-17T19:04:22 | {} | MEMBER |
ClickHouse | ClickHouse | 493,604,576 | 5,308 | alexey-milovidov | Isn't it better to change to zero by default? | 2019-05-17T21:27:32 | 2019-05-17T21:27:32 | {} | MEMBER |
ClickHouse | ClickHouse | 493,663,068 | 5,246 | alexey-milovidov | What's wrong with `ColumnNullable::getExtremes`? | 2019-05-18T09:36:38 | 2019-05-18T09:36:38 | {} | MEMBER |
ClickHouse | ClickHouse | 493,678,641 | 5,320 | alexey-milovidov | Do you want to make another PR for sequenceMatch - so I can merge this PR? | 2019-05-18T13:50:40 | 2019-05-18T13:50:40 | {} | MEMBER |
ClickHouse | ClickHouse | 493,979,299 | 5,344 | MeteHanC | Okay, so I forgot to check the logs, sorry about that. There was an another app using the same port. It was about it. | 2019-05-20T13:06:34 | 2019-05-20T13:06:34 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 493,979,582 | 5,343 | 4ertus2 | It's in our plans to implement dictGet functions for Decimals. | 2019-05-20T13:07:23 | 2019-05-20T13:07:23 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 494,202,375 | 5,341 | leo29798 | I just run the clickhouse-server and execute query by clickhouse-client as release.Then I execute the query in the link and I have a bug when I use modelEvaluate function. I'm not good at English so sorry about that. | 2019-05-21T01:11:22 | 2019-05-21T01:11:44 | {} | NONE |
ClickHouse | ClickHouse | 497,542,232 | 5,472 | mcarbonneaux | It what you talk in my previous issue : https://github.com/yandex/ClickHouse/issues/4119#issuecomment-497542042? | 2019-05-31T01:28:49 | 2019-05-31T01:28:49 | {} | NONE |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.