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 | 634,765,124 | 11,184 | alexey-milovidov | Performance tests failed. It's unclear whether it relates to query_log or not.
(I thought that query_log is already enabled in perf tests).
@akuzm ? | 2020-05-27T16:02:50 | 2020-05-27T16:04:23 | {} | MEMBER |
ClickHouse | ClickHouse | 634,804,949 | 11,169 | filimonov | Fixed in 20.4.4
https://github.com/ClickHouse/ClickHouse/pull/10561/commits/512687f1a3fd9708c44a8d5d2d11c7bf7e33a8d1 | 2020-05-27T17:02:56 | 2020-05-27T17:02:56 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 635,114,653 | 11,116 | NanoBjorn | Do not forget to fix doc for `system.storage_policies` | 2020-05-28T05:36:50 | 2020-05-28T05:36:50 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 635,142,793 | 11,237 | akuzm | Please add a brief description of what it does to https://github.com/ClickHouse/ClickHouse/blob/master/docs/en/sql-reference/functions/random-functions.md Or maybe bitmap functions? Not sure what's the best doc chapter. | 2020-05-28T06:49:16 | 2020-05-28T06:49:16 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 635,161,352 | 4,190 | akuzm | > Seems like #5388 #5387 #10915 are closed.
> But still can't add MongoDB Atlas ReplicaSet as external dictionary.
Yes, I added only the very basic URI support. The hard part was updating Poco library to a version that has the necessary MongoDB API. Now the external contributors can use this API to continue working... | 2020-05-28T07:24:52 | 2020-05-28T07:24:52 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 634,584,084 | 6,049 | alexey-milovidov | @nezed That's right and you have to resolve to ordinary logs in this case... | 2020-05-27T10:55:23 | 2020-05-27T10:55:23 | {} | MEMBER |
ClickHouse | ClickHouse | 634,588,444 | 11,213 | alexey-milovidov | Also need to add the suffix for debug build, so the library will be named `libcctzd.a`. | 2020-05-27T11:05:10 | 2020-05-27T11:05:10 | {} | MEMBER |
ClickHouse | ClickHouse | 634,612,495 | 11,220 | alexey-milovidov | @azat Yes, let's add as many metrics and counters as possible! | 2020-05-27T11:59:57 | 2020-05-27T11:59:57 | {} | MEMBER |
ClickHouse | ClickHouse | 634,714,297 | 5,747 | akonyaev90 | We finally started working on this issue.
Now we have this priority.
If anyone has ideas or advice, write here. | 2020-05-27T14:50:50 | 2020-05-27T14:50:50 | {
"+1": 3
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 635,134,650 | 11,216 | filimonov | It looks like the fix is going a bit too far.
Will get to that later, some drafts are here: https://github.com/filimonov/ClickHouse/commit/81b78c262499bae67359fdcd822a860ec110fb2a | 2020-05-28T06:29:33 | 2020-05-28T06:29:33 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 635,527,203 | 11,250 | calebeaires | input_format_tsv_empty_as_default help me with some issue. Thanks to clarify. | 2020-05-28T18:43:04 | 2020-05-28T18:43:04 | {} | NONE |
ClickHouse | ClickHouse | 634,551,521 | 9,334 | filimonov | Closing due to lack of activity, and because the thread goes away from the original topic.
If you still face the issue, please open a new one with the needed details. | 2020-05-27T09:45:54 | 2020-05-27T09:45:54 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 634,551,916 | 8,979 | ropiteaux | > > copied into detached folder and then attach accordingly
>
> How did you do that? Do you use FREEZE or stop the server? When you back up an active working copy you can get inconsistent copy.
We have a backup job in golang that does the backup to S3 every day, it does so by using FREEZE yes. (the backup job use... | 2020-05-27T09:46:46 | 2020-05-27T09:47:10 | {} | NONE |
ClickHouse | ClickHouse | 634,583,254 | 11,219 | alex-zaitsev | It seem to work on 20.4.4.18, so it is probably already fixed, but please look anyway @4ertus2 | 2020-05-27T10:53:23 | 2020-05-27T10:53:23 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 634,590,583 | 11,135 | zhuyoucui0909 | We did a simple test on latest stable version v20.4.4.18. On this version, the issue is not reproduced. We will test more on the new version.
| 2020-05-27T11:09:43 | 2020-05-27T11:09:43 | {} | NONE |
ClickHouse | ClickHouse | 635,363,925 | 8,683 | alexey-milovidov | @azat Are you interested implementing this task? | 2020-05-28T13:52:50 | 2020-05-28T13:52:50 | {} | MEMBER |
ClickHouse | ClickHouse | 635,451,021 | 8,624 | nikitamikhaylov | I think it is necessary to add performance tests to benchmark it in comparison with DirectDictionary and CacheDictionary. Or just CacheDictionary in memory with SSDCacheDictionary.
I don't mean performance tests in CI, I want something like the one given here #10622 | 2020-05-28T16:20:40 | 2020-05-28T16:20:40 | {} | MEMBER |
ClickHouse | ClickHouse | 634,482,670 | 11,213 | qoega | ```
FAILED: base/common/tests/date_lut2
: && /usr/bin/ccache /usr/bin/clang++-10 -fdiagnostics-color=always -std=c++2a -fsized-deallocation -gdwarf-aranges -pipe -msse4.1 -msse4.2 -mpopcnt -Wall -Wno-unused-command-line-argument -fdiagnostics-absolute-paths -Werror -Wextra -Wpedantic -Wno-vla-extension -Wno-zer... | 2020-05-27T07:31:09 | 2020-05-27T07:31:09 | {} | MEMBER |
ClickHouse | ClickHouse | 634,484,362 | 11,208 | qoega | ```
2020-05-27 02:26:08 00342_escape_sequences: [ FAIL ] 0.05 sec. - result differs with reference:
2020-05-27 02:26:08 --- /usr/share/clickhouse-test/queries/0_stateless/00342_escape_sequences.reference 2020-05-27 00:54:57.000000000 +0300
2020-05-27 02:26:08 +++ /tmp/... | 2020-05-27T07:34:38 | 2020-05-27T07:34:38 | {} | MEMBER |
ClickHouse | ClickHouse | 634,508,260 | 11,070 | azat | So you want to select the shard based on the parameters that you passed in `WITH` statement (so you are using `optimize_skip_unused_shards`), is my understanding correct?
>I want to shard this table with a sharding_key expression that takes myparam value as a parameter: Distributed(cluster, default, mytable, my_shar... | 2020-05-27T08:23:00 | 2020-05-27T08:23:00 | {} | MEMBER |
ClickHouse | ClickHouse | 634,551,965 | 6,049 | nezed | @alexey-milovidov I think its potentially dangerous way.
The main point of remote collected logs is to spot problems in real time with tools like alerts, while `CREATE MATERIALIZED VIEW ..` will not produce any logs in case if DBMS is failed to start or disk becomes read-only (eg. there is no space left). | 2020-05-27T09:46:54 | 2020-05-27T09:46:54 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 635,195,773 | 11,215 | toannhu96 | @4ertus2
I had tried to reproduce the bugs with much more simpler table but the test works flawlessly without the bugs that i mentioned before :(
IMHO I think the bugs come from GLOBAL ANY LEFT JOIN in my case but not come from the dependent on data.
This works well
```
CREATE DATABASE IF NOT EXISTS test;
... | 2020-05-28T08:24:19 | 2020-05-28T08:26:41 | {} | NONE |
ClickHouse | ClickHouse | 635,361,183 | 11,073 | filimonov | Continued here: #11252 | 2020-05-28T13:48:02 | 2020-05-28T13:48:02 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 635,407,168 | 11,253 | champtar | Hi @alexey-milovidov
On Centos7 the ninja version is too old (1.7, min required 1.8), and make build are working just fine (at least 20.3.10), could you leave an option to keep default behavior ? | 2020-05-28T15:03:47 | 2020-05-28T15:03:47 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 635,568,543 | 11,256 | filimonov | It looks like newer version behave a bit different
```
"2020.05.28 22:40:01.633450 [ 67 ] {} <Warning> StorageKafka (kafka): [thrd:main]: Consumer group session timed out (in join-state started) after 10498 ms without a successful response from the group coordinator (broker 1, last error was Success): revoking assign... | 2020-05-28T20:05:07 | 2020-05-28T20:05:07 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 634,603,747 | 11,213 | alesapin | Let's try combination of OBJECT library and INTERFACE linking. | 2020-05-27T11:39:45 | 2020-05-27T11:39:45 | {} | MEMBER |
ClickHouse | ClickHouse | 634,622,166 | 11,149 | azat | Just like I thought - Kafka engine is still slow | 2020-05-27T12:19:50 | 2020-05-27T12:19:50 | {} | MEMBER |
ClickHouse | ClickHouse | 634,623,486 | 11,149 | filimonov | I agree that there are several places that can be done better. And the performance can be improved.
But your test is not 100% fair. With commits, it gives [1.29M/s] for 2 consumers.
If we will add parsing/writing compressed data to the disk etc, the result we have currently is not bad.
| 2020-05-27T12:22:43 | 2020-05-27T12:27:26 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 634,765,692 | 11,184 | alexey-milovidov | Integration - tech debt, @alesapin | 2020-05-27T16:03:44 | 2020-05-27T16:03:44 | {} | MEMBER |
ClickHouse | ClickHouse | 634,835,717 | 11,184 | akuzm | > Performance tests failed. It's unclear whether it relates to query_log or not.
> (I thought that query_log is already enabled in perf tests).
>
> @akuzm ?
Not related, query log is already enabled. | 2020-05-27T17:55:14 | 2020-05-27T17:55:14 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 635,069,918 | 8,889 | jackpgao | The `MySQL` engine problem can be traced back to `19.14.11.16`.
`19.13.7.57` still works.
There are many changes related to MySQL version 19.14.
https://github.com/ClickHouse/ClickHouse/blob/master/docs/en/whats-new/changelog/2019.md
It seems only happens with more than one MySQL engine subquery. | 2020-05-28T03:13:13 | 2020-05-28T03:32:54 | {} | NONE |
ClickHouse | ClickHouse | 635,175,161 | 8,873 | yuzhichang | Similar problem. I notice that `hasAny(arr1, Array(LowCardinality(String)))` is ~70x slower than `hasAny(arr1, Array(UInt32))`.
| 2020-05-28T07:47:31 | 2020-05-28T07:47:31 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 635,333,433 | 11,250 | den-crane | I could not reproduce it.
20.5.1.3496
```
CREATE TABLE base (
type Nullable(String), qtd Nullable(Double)) ENGINE MergeTree ORDER BY tuple();
cat test.tsv
Car 12
Bike 0
Other
cat test.tsv |clickhouse-client -q "insert into base format TSV" --input_format_tsv_empty_as_default=1
┌─type──┬──qtd─┐
│ C... | 2020-05-28T12:57:35 | 2020-05-28T12:57:35 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 634,582,320 | 11,149 | filimonov | Did a performance tests:
Test https://github.com/filimonov/ch-kafka-consume-perftest
Format TSV
Topic with 4 partitions, prefilled with 33554432 records.
Consuming time with single table with kafka_num_consumers=2 measured.
| kafka_max_block_size | consume_time (s, before) | consume_speed (rec/sec, before... | 2020-05-27T10:51:02 | 2020-05-27T10:55:23 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 634,739,486 | 11,213 | alesapin | Build passed! | 2020-05-27T15:26:41 | 2020-05-27T15:26:41 | {} | MEMBER |
ClickHouse | ClickHouse | 634,764,344 | 11,184 | alexey-milovidov | Build failure:
> Build failed without build log. It's inner CI problem. Maybe something wrong with docker or host. Just wait for build restart.
Ok. | 2020-05-27T16:01:43 | 2020-05-27T16:01:43 | {} | MEMBER |
ClickHouse | ClickHouse | 635,325,394 | 11,101 | alexey-milovidov | @filimonov It will lose performance benefits.
| 2020-05-28T12:40:54 | 2020-05-28T12:40:54 | {} | MEMBER |
ClickHouse | ClickHouse | 635,318,953 | 11,248 | xjewer | ~can we do it on per-table basis? how tricky is this?~
my bad, I mixed terms of replica as a node, rather than table replica.
anyway it would only work for replicated tables and would stop replication, which is a bit different. | 2020-05-28T12:32:25 | 2020-05-28T12:54:57 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 635,693,941 | 11,262 | alexey-milovidov | @qoega Test infrastructure is completely broken: https://clickhouse-test-reports.s3.yandex.net/11262/db72f3959489ef54261071390e20703dea5caf51/functional_stateless_tests_(address).html#fail1 | 2020-05-29T00:51:15 | 2020-05-29T00:51:15 | {} | MEMBER |
ClickHouse | ClickHouse | 634,591,194 | 11,163 | azat | >Let's fix style check and merge.
Not sure that I know what is the error (and every time I would mistaken the CI will be triggered, sigh), I guess some missing whitespace, but style check is not verbose | 2020-05-27T11:11:06 | 2020-05-27T11:11:06 | {} | MEMBER |
ClickHouse | ClickHouse | 635,323,501 | 11,226 | And42 | And a few more links about perf events: https://gist.github.com/And42/558fba4592dbd27c9550502b0c20a1e2 | 2020-05-28T12:36:38 | 2020-05-28T12:36:38 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 635,413,007 | 5,957 | filimonov | It looks like a duplicate of https://github.com/ClickHouse/ClickHouse/issues/11203 | 2020-05-28T15:14:15 | 2020-05-28T15:14:15 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 635,547,651 | 11,251 | alexey-milovidov | @BayoNet Proposed wording:
It creates a table that is the same as the result of the table function. The created table will work in the same way as the corresponding table function. | 2020-05-28T19:24:16 | 2020-05-28T19:24:16 | {} | MEMBER |
ClickHouse | ClickHouse | 635,605,684 | 11,264 | Akazz | Exceptions are thrown for the purpose of getting stacktraces. | 2020-05-28T21:01:39 | 2020-05-28T21:01:39 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 635,719,383 | 11,276 | alexey-milovidov | ```
CREATE TABLE insert_small_block_performance
(
`x` String,
`a` DEFAULT SHA256(x),
`b` DEFAULT SHA256(toString(a)),
`c` DEFAULT SHA256(toString(b)),
`d` DEFAULT SHA256(toString(c)),
`e` DEFAULT SHA256(toString(d)),
`f` DEFAULT SHA256(toString(e)),
`g` DEFAULT SHA256(to... | 2020-05-29T02:25:14 | 2020-05-29T02:25:14 | {} | MEMBER |
ClickHouse | ClickHouse | 634,496,068 | 11,149 | filimonov | Linking error on clang-10 - unrelated.
```
2020-05-27 06:37:47 [112/249] Linking CXX executable src/Storages/tests/system_numbers
2020-05-27 06:37:47 FAILED: src/Storages/tests/system_numbers
2020-05-27 06:37:47 : && /usr/bin/ccache /usr/bin/clang++-10 -g -fno-omit-frame-pointer -DSANITIZER -fsanitize=thread -f... | 2020-05-27T07:59:10 | 2020-05-27T07:59:10 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 634,619,261 | 11,135 | 4ertus2 | Could you print it here? We could add it in our tests to avoid regression. | 2020-05-27T12:13:45 | 2020-05-27T12:13:45 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 634,706,684 | 11,217 | WiPiton | https://yadi.sk/d/MI7OZVmQQrrWUw - here is a link to data structures. Approx. record counts mentioned there.
| 2020-05-27T14:39:36 | 2020-05-27T14:43:35 | {} | NONE |
ClickHouse | ClickHouse | 635,171,864 | 9,108 | yuzhichang | Looks duplicated with #9601 | 2020-05-28T07:41:56 | 2020-05-28T07:41:56 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 635,227,137 | 11,028 | vitlibar | > MySQL also supports `BIT[(M)]`, `BOOL` (as a synonyms for `TINYINT`), and treats `FLOAT(p)` as either `FLOAT` or `DOUBLE` depending on `p`. Should we support those types too?
PostgreSQL supports `float(p)` as well:
```
PostgreSQL also supports the SQL-standard notations float and float(p) for specifying inexac... | 2020-05-28T09:25:29 | 2020-05-28T09:26:26 | {} | MEMBER |
ClickHouse | ClickHouse | 635,314,382 | 11,236 | alexey-milovidov | > Performance — 2 slower, 176 unstable, 5 errors
Ok. | 2020-05-28T12:28:25 | 2020-05-28T12:28:25 | {} | MEMBER |
ClickHouse | ClickHouse | 635,670,491 | 11,271 | alexey-milovidov | Proposed names:
`queryNormalize`
`queryHash` | 2020-05-28T23:30:40 | 2020-05-28T23:30:40 | {} | MEMBER |
ClickHouse | ClickHouse | 635,207,760 | 7,462 | MeteHanC | This issue persists with v20.3.4 as well | 2020-05-28T08:47:08 | 2020-05-28T08:47:08 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 634,593,156 | 11,215 | 4ertus2 | Could you simplify your query and make a sample that does not depend on your data? It looks like it's possible to remove most of columns and make a simple distributed query with several rows. Would it reproduced then or it depends on data? | 2020-05-27T11:15:37 | 2020-05-27T11:15:37 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 634,688,851 | 11,217 | den-crane | Do you have reproducible example with a fake data? | 2020-05-27T14:13:43 | 2020-05-27T14:13:43 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 634,768,682 | 11,227 | alexey-milovidov | Yes, it can be fixed - not only for JSONEachRow but also for TSV/CSV as well.
In the latter case we also have to ensure that Enum does not contain values that are ambiguous with numbers. | 2020-05-27T16:07:16 | 2020-05-27T16:07:39 | {} | MEMBER |
ClickHouse | ClickHouse | 635,125,927 | 11,101 | filimonov | @alexey-milovidov why doesn't it just bypass that? I think it should NOT do any magic with regexps if they seems complex, and contain some special regex syntax. | 2020-05-28T06:07:14 | 2020-05-28T06:07:14 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 635,194,507 | 8,873 | alex-zaitsev | Array performance is a known issue, see #6005. We plan to address it soon.
| 2020-05-28T08:21:54 | 2020-05-28T08:21:54 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 635,202,596 | 11,070 | yfinkelstein | Ok, thanks. getScalar is indeed relevant. We figured out most of the issues, so this question is answered. By the real topic we are trying to advance is this one: https://github.com/ClickHouse/ClickHouse/issues/10943. It would be great if you could review it and comment on it. | 2020-05-28T08:37:15 | 2020-05-28T08:37:15 | {} | NONE |
ClickHouse | ClickHouse | 635,546,661 | 11,251 | alexey-milovidov | It works exactly as it was designed.
The documentation has to be updated (unfortunately, it was wrong from the beginning). | 2020-05-28T19:22:04 | 2020-05-28T19:25:25 | {} | MEMBER |
ClickHouse | ClickHouse | 635,586,770 | 8,889 | alexey-milovidov | > and the version 20.4.3.16 seems works well.
So, should I close this issue as fixed? | 2020-05-28T20:22:48 | 2020-05-28T20:22:48 | {} | MEMBER |
ClickHouse | ClickHouse | 634,597,056 | 11,220 | azat | Also I was thinking about extending this metrics, by adding not only number of files but also
- number of bytes
- broken files
- and it will be nice to have disk as a label, but AFAIR current interface does not allows labels
Plus it may be useful add new metrics as counters over gauges (to allow rate calculation,... | 2020-05-27T11:24:27 | 2020-05-27T11:24:27 | {} | MEMBER |
ClickHouse | ClickHouse | 634,691,976 | 11,217 | WiPiton | Unfortunately, at the moment, no..
I can post full query and data structures involved, though, if it'll be of help | 2020-05-27T14:18:42 | 2020-05-27T14:20:02 | {} | NONE |
ClickHouse | ClickHouse | 634,834,122 | 11,186 | abyss7 | Self-reviewed | 2020-05-27T17:52:43 | 2020-05-27T17:52:43 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 634,814,866 | 4,190 | Olegt0rr | @alexey-milovidov @akuzm
Seems like #5388 #5387 #10915 are closed.
But still can't add MongoDB Atlas ReplicaSet as external dictionary.
Added [test case](https://github.com/ClickHouse/ClickHouse/pull/10915/files#diff-81025e29a8fa012a9e88720480f2a09dR181) covers:
Neither 3.6+ ReplicaSet connections ([descript... | 2020-05-27T17:19:24 | 2020-05-27T21:10:35 | {} | NONE |
ClickHouse | ClickHouse | 635,208,156 | 4,190 | Olegt0rr | @akuzm, could you help to draw up the detailed requirements for the necessary contribution?
I'll try to find someone and pay for his work ([ru link](https://freelance.habr.com/tasks/304258)) | 2020-05-28T08:47:55 | 2020-05-28T10:27:48 | {
"+1": 1
} | NONE |
ClickHouse | ClickHouse | 635,500,248 | 8,683 | azat | Yes, thanks!
I thought about implementing it via metrics, but maybe provide system table will be better.
Do you have some pros and cons for both?
I thought about provide this via metrics because basically you are interested in monitoring this, yes any system table can be exported via, say custom http handlers, b... | 2020-05-28T17:50:27 | 2020-05-28T17:50:27 | {} | MEMBER |
ClickHouse | ClickHouse | 635,500,714 | 8,683 | alexey-milovidov | But we don't have per-table metrics, only total. | 2020-05-28T17:51:27 | 2020-05-28T17:51:27 | {} | MEMBER |
ClickHouse | ClickHouse | 635,508,136 | 8,683 | azat | Ah, indeed! This definitely should be per table (and per disk) | 2020-05-28T18:05:34 | 2020-05-28T18:05:34 | {} | MEMBER |
ClickHouse | ClickHouse | 635,633,116 | 6,638 | alexey-milovidov | @azat
The issue is definitely fixed.
We want to add a test case. Do I understand correctly that:
- we should have a cluster with both localhost and remote nodes;
- we should have LowCardinality on initiator and different String and LowCardinality(String) on remote nodes;
- two level group by should be activated... | 2020-05-28T22:05:36 | 2020-05-28T22:06:59 | {} | MEMBER |
ClickHouse | ClickHouse | 634,553,237 | 11,213 | alesapin | Waiting only for build. | 2020-05-27T09:49:36 | 2020-05-27T09:49:36 | {} | MEMBER |
ClickHouse | ClickHouse | 634,587,209 | 11,163 | alexey-milovidov | I see that it works perfectly, so we can leave this test. | 2020-05-27T11:02:34 | 2020-05-27T11:02:34 | {} | MEMBER |
ClickHouse | ClickHouse | 634,589,058 | 11,213 | alexey-milovidov | PS. I don't know why we need this suffix, probably it's better to get rid of it. | 2020-05-27T11:06:33 | 2020-05-27T11:06:33 | {} | MEMBER |
ClickHouse | ClickHouse | 635,141,294 | 11,235 | akuzm | Please add a brief description of what it does to https://github.com/ClickHouse/ClickHouse/blob/master/docs/en/sql-reference/aggregate-functions/combinators.md | 2020-05-28T06:46:00 | 2020-05-28T06:46:00 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 635,276,838 | 10,351 | CurtizJ | Is it still draft? | 2020-05-28T11:15:00 | 2020-05-28T11:15:00 | {} | MEMBER |
ClickHouse | ClickHouse | 635,390,030 | 11,255 | alexey-milovidov | > somebody from CH team should check if it works!
Looks Ok, we will check on next release. | 2020-05-28T14:38:22 | 2020-05-28T14:38:22 | {} | MEMBER |
ClickHouse | ClickHouse | 634,620,807 | 11,149 | filimonov | Same setup, same data, your test:
1 kafkacat: 1.29M/s
2 kafkacat: 2.32M/s
4 kafkacat: 3.66M/s | 2020-05-27T12:17:02 | 2020-05-27T12:26:05 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 634,634,789 | 11,213 | alesapin | > Let's try combination of OBJECT library and INTERFACE linking.
It doesn't work. | 2020-05-27T12:43:57 | 2020-05-27T12:43:57 | {} | MEMBER |
ClickHouse | ClickHouse | 635,124,091 | 11,239 | filimonov | You need to escape slashes one more time.
```
SELECT extract('你好άλφα', '\\p{Han}+')
┌─extract('你好άλφα', '\\p{Han}+')─┐
│ 你好 │
└──────────────────────────────────┘
```
See https://github.com/ClickHouse/ClickHouse/issues/10922
| 2020-05-28T06:02:34 | 2020-05-28T06:02:34 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 635,302,659 | 11,226 | And42 | The point with providing separate Running and Enabled times is to show the user whether the result value is scaled and how much if it is. Without that knowledge some can consider the values are perfectly accurate in his situation while they are actually not (when scaled).
As for not recalculating as of now, this is a ... | 2020-05-28T12:17:27 | 2020-05-28T12:19:00 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 635,313,957 | 11,236 | alexey-milovidov | > Stress test (thread) — Sanitizer assert (in stderr.log)
Reported in CI fest. | 2020-05-28T12:28:03 | 2020-05-28T12:28:03 | {} | MEMBER |
ClickHouse | ClickHouse | 635,429,731 | 11,226 | And42 | Just a fix for 2 issues I've mentioned: https://github.com/ClickHouse/ClickHouse/pull/9545/commits/d795196eea6b94479bfd7d0e6a1d737b05f78c5c, https://github.com/ClickHouse/ClickHouse/pull/9545/commits/b65675ae5af520b8c089011f6d2e6c0701cec5de | 2020-05-28T15:43:59 | 2020-05-28T16:09:58 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 635,552,173 | 11,253 | alexey-milovidov | @champtar Ok.
We also have tested in CLion and it does not allow to simply switch to Ninja from PreLoad.cmake.
Close. | 2020-05-28T19:33:34 | 2020-05-28T19:33:34 | {} | MEMBER |
ClickHouse | ClickHouse | 634,442,190 | 11,149 | filimonov | Rebased on the newer master (was on the top of broken one), squashed.
Test fixed, another bug discovered #11216
(Will fix the bug in follow-up PR) | 2020-05-27T05:46:00 | 2020-05-27T10:58:48 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 634,761,169 | 11,227 | vzakaznikov | The problem is at https://github.com/ClickHouse/ClickHouse/blob/1658705f7c82839906b9d9b40afde08f2d37f428/src/DataTypes/DataTypeEnum.cpp#L195 | 2020-05-27T15:57:39 | 2020-05-27T15:57:39 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 634,782,149 | 8,889 | org0000h | we meet the same issues | 2020-05-27T16:27:52 | 2020-05-27T16:27:52 | {} | NONE |
ClickHouse | ClickHouse | 634,818,043 | 11,163 | azat | >You can edit check-style script if you like.
Thanks! Didn't know that this was in repo (and never searched for it somehow).
The style check issue had been found and fixed. | 2020-05-27T17:24:54 | 2020-05-27T17:25:23 | {} | MEMBER |
ClickHouse | ClickHouse | 635,326,005 | 11,101 | alexey-milovidov | I saw that in re2 library, developers are trying to implement something similar in more accurate way.
It was named "filtered_re2" (that was in "not recommended" stage for a long time). And I noticed that in recent versions of re2 it becomes better.
Maybe it can possibly replace OptimizedRegularExpression. | 2020-05-28T12:42:16 | 2020-05-28T12:42:54 | {} | MEMBER |
ClickHouse | ClickHouse | 635,380,079 | 11,208 | alexey-milovidov | > 00342_escape_sequences
New result is correct now.
> 00926_multimatch
The test had mistake.
> 01075_allowed_client_hosts
The test had mistake. | 2020-05-28T14:20:50 | 2020-05-28T14:20:50 | {} | MEMBER |
ClickHouse | ClickHouse | 635,628,228 | 9,158 | alexey-milovidov | I cannot reporoduce the issue on current master.
Most likely it was already fixed.
I will add your test to the test suite just in case.
PS. Thank you for the well formed test case with data! | 2020-05-28T21:53:01 | 2020-05-28T22:02:26 | {} | MEMBER |
ClickHouse | ClickHouse | 634,770,506 | 11,227 | alexey-milovidov | This is the relevant task: https://github.com/ClickHouse/ClickHouse/issues/10682 | 2020-05-27T16:08:42 | 2020-05-27T16:08:42 | {} | MEMBER |
ClickHouse | ClickHouse | 635,198,557 | 10,019 | vitlibar | Fixed the first part of that where rules configured for user A affect user B. | 2020-05-28T08:29:44 | 2020-05-28T08:29:44 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 635,307,147 | 11,248 | alexey-milovidov | This is a good idea. We can implement it with `SYSTEM STOP REPLICA` and `SYSTEM START REPLICA` - it will be consistent with `SYSTEM RESTART REPLICA`. | 2020-05-28T12:22:03 | 2020-05-28T12:22:03 | {} | MEMBER |
ClickHouse | ClickHouse | 634,593,355 | 11,163 | alexey-milovidov | You can edit `check-style` script if you like. | 2020-05-27T11:16:05 | 2020-05-27T11:16:05 | {} | MEMBER |
ClickHouse | ClickHouse | 634,608,417 | 11,149 | azat | Maybe worth checking with more partitions and compare it with old simple `kafkacat`, since it is unclear from the test what is the maximum throughput and `kafkacat` works more or less fast, even though it does not uses batching,especially if you run multiple instances of it?
P.S. , something like this may be used to... | 2020-05-27T11:50:47 | 2020-05-27T11:50:47 | {} | MEMBER |
ClickHouse | ClickHouse | 635,248,713 | 11,236 | abyss7 | @alexey-milovidov I have removed myself from assignees. | 2020-05-28T10:10:27 | 2020-05-28T10:10:46 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 635,315,223 | 11,226 | And42 | Grouping is only useful when you want to calculate some values based on the ones you've got from the hardware events. Let's say you want to get instructions per cycle. To get that you have to get a total number of instructions and a total number of cycles. If you want to get a precise value you have to be sure these ev... | 2020-05-28T12:29:08 | 2020-05-28T12:32:11 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 635,417,344 | 5,177 | filimonov | Support for Materialized columns in Kafka engine is highly questionable.
We expect the table with engine=Kafka should be a simple mapping of the data in the Kafka topic.
If you need to have some derived columns, you can add them on the Materialized view level or with an alias.
My expectation about MATERIALIZED ... | 2020-05-28T15:22:05 | 2020-05-28T15:22:05 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 643,779,210 | 11,626 | den-crane | >However why we don't use passdword-less way also for the data exchange among shards
It's unrelated to replication. You don't have to.
I replied in https://github.com/ClickHouse/ClickHouse/issues/11625 | 2020-06-14T15:06:29 | 2020-06-14T15:07:32 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 643,801,205 | 11,624 | den-crane | And why do you need cl-client if you use java? Why not HTTP or JDBC ? | 2020-06-14T18:05:47 | 2020-06-14T18:05:47 | {} | CONTRIBUTOR |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.