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 | 844,410,293 | 24,302 | mzealey | oh of course I'm being stupid thanks :) | 2021-05-19T19:37:06 | 2021-05-19T19:37:06 | {} | NONE |
ClickHouse | ClickHouse | 844,426,332 | 23,504 | alexey-milovidov | Suppose the host is named `myhost` and it is resolved to `127.0.1.1`.
Why exactly we need to treat `myhost` as local? What is the scenario?
Why the user will write `myhost` in some configuration or queries?
If we don't treat it as local, I assume that everything will work correctly. | 2021-05-19T19:59:36 | 2021-05-19T20:00:06 | {} | MEMBER |
ClickHouse | ClickHouse | 844,502,483 | 24,311 | nikitamikhaylov | @Mergifyio update
| 2021-05-19T21:56:47 | 2021-05-19T21:56:47 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 844,651,045 | 19,255 | dhxxhcb | 20.8.12.2 also has the problem
![Uploading image.png…]()
| 2021-05-20T03:11:30 | 2021-05-20T03:11:30 | {} | NONE |
ClickHouse | ClickHouse | 844,747,199 | 24,314 | mathfool | Maybe you can wait and see if it will disappear later. It looks like a temp state when data is moving between volumn/disks(you configured multiple disk/volumn?). | 2021-05-20T06:28:47 | 2021-05-20T06:28:47 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 844,852,933 | 24,314 | onine007 | > Maybe you can wait and see if it will disappear later. It looks like a temp state when data is moving between volumn/disks(you configured multiple disk/volumn?).
yes i configured multiple disk , and it disappeared .
| 2021-05-20T08:21:51 | 2021-05-20T08:21:51 | {} | NONE |
ClickHouse | ClickHouse | 844,905,473 | 12,135 | tavplubix | @geekyouth, please, create new issue with detailed information and attach related logs | 2021-05-20T09:24:06 | 2021-05-20T09:24:06 | {} | MEMBER |
ClickHouse | ClickHouse | 844,919,547 | 24,245 | amosbird | > Can you commit anything to trigger build with newer master? Or I can use mergify to update branch
done | 2021-05-20T09:39:31 | 2021-05-20T09:39:31 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 845,212,296 | 24,340 | PavelKuzovkin | create objects in the database like this:
```
create table test_logging_rabbit
(
id UInt64,
value String,
createdAt DateTime
) engine = RabbitMQ SETTINGS
rabbitmq_host_port = 'rabbit:5672',
rabbitmq_exchange_name = 'test_exchange',
rabbitmq_exchange_type = '... | 2021-05-20T15:17:01 | 2021-05-20T15:25:17 | {} | NONE |
ClickHouse | ClickHouse | 845,368,578 | 24,337 | tavplubix | > BTW, what exception do you see?
Part ... intersects ...
It's not completely fixed, there are still some issues that may lead to this exception. | 2021-05-20T18:36:25 | 2021-05-20T18:36:25 | {} | MEMBER |
ClickHouse | ClickHouse | 844,392,718 | 24,294 | filimonov | It is used to distinguish 2 possible forms of addresses - when the bucket is a part of hostname, and when it's a part of the path.
How URLs for HuaweiCloud OBS typically look like? | 2021-05-19T19:12:14 | 2021-05-19T19:12:14 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 844,398,389 | 24,277 | filimonov | You don't need debs to create docker image.
See
https://github.com/ClickHouse/ClickHouse/blob/9982f05e4deb23d923f4f17ace5cef9d53146b67/docker/server/Dockerfile#L15-L17
You can try the image build this way here: altinity/clickhouse-server:21.6.1.6734-testing-arm
See also https://github.com/ClickHouse/ClickHo... | 2021-05-19T19:20:06 | 2021-05-19T19:20:06 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 844,421,494 | 24,262 | alexey-milovidov | https://github.com/ClickHouse/ClickHouse/pull/7898#issuecomment-844420914 | 2021-05-19T19:53:08 | 2021-05-19T19:53:08 | {} | MEMBER |
ClickHouse | ClickHouse | 844,810,803 | 23,901 | clowdrgn | Same question in #24268 , hope to fix it soon | 2021-05-20T07:36:31 | 2021-05-20T07:36:31 | {} | NONE |
ClickHouse | ClickHouse | 844,826,847 | 24,273 | kitaisreal | @azat cool will create separate PR because we have some issues with CI, do not want to retrigger that. | 2021-05-20T07:53:53 | 2021-05-20T07:53:53 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 844,884,190 | 24,319 | qoega | ASTFuzzer is quite useful:
```
SELECT * FROM remote('[::1][::1][::1][::1]', system.one) FORMAT Null
clickhouse-server: ../src/Common/DNSResolver.cpp:95: DNSResolver::IPAddresses DB::resolveIPAddressImpl(const std::string &): Assertion `host.ends_with(']')' failed.
``` | 2021-05-20T09:00:13 | 2021-05-20T09:00:13 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 844,678,129 | 24,259 | azat | Superseded by #24273 | 2021-05-20T04:30:31 | 2021-05-20T04:30:31 | {} | MEMBER |
ClickHouse | ClickHouse | 844,712,287 | 24,316 | alexey-milovidov | I will add a unit test. | 2021-05-20T05:36:01 | 2021-05-20T05:36:01 | {} | MEMBER |
ClickHouse | ClickHouse | 844,775,662 | 24,275 | alesapin | No related failures | 2021-05-20T07:09:39 | 2021-05-20T07:09:39 | {} | MEMBER |
ClickHouse | ClickHouse | 844,954,388 | 24,294 | siviae | In my case it is `https://some-bucket.obs.ru-moscow-1.hc.sbercloud.ru/some/path/file.txt` - we're using SberCloud.Advanced. It uses virtual-hosted style URLs.
I think, it would be better to specify 4 config properties for S3 engine:
- endpoint: i.e.`obs.ru-moscow-1.hc.sbercloud.ru` or `s3.us-east-2.amazonaws.com`.... | 2021-05-20T10:19:45 | 2021-05-20T10:19:45 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 845,186,632 | 24,250 | vdimir | I added a bit more tests for different data types and change return type for Float32/64 to Float64.
@ucasFL take a look if it's ok I'll merge when CI passed | 2021-05-20T14:46:05 | 2021-05-20T14:46:05 | {} | MEMBER |
ClickHouse | ClickHouse | 845,307,653 | 24,350 | alexey-milovidov | Also please remove *all* unnecessary configs. | 2021-05-20T17:16:51 | 2021-05-20T17:16:51 | {} | MEMBER |
ClickHouse | ClickHouse | 844,395,520 | 24,302 | filimonov | and where is the bug? type is not equal to 'xxx' on the rollup line, so it outputs countDistinct(a) just like you asked... | 2021-05-19T19:16:15 | 2021-05-19T19:16:15 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 844,758,670 | 24,270 | filimonov | when you use native clients (clickhouse-client, golang driver etc) the conversion of timestamps to human-readable strings happens on the client side, according to client timezone and preferences. See the setting `--use_client_time_zone` and similar.
toString happens on the server.
If you want to use some fixed r... | 2021-05-20T06:45:54 | 2021-05-20T06:45:54 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 845,350,410 | 24,189 | alesapin | tests ok | 2021-05-20T18:08:12 | 2021-05-20T18:08:12 | {} | MEMBER |
ClickHouse | ClickHouse | 844,385,831 | 24,290 | vdimir | @mergify update | 2021-05-19T19:02:32 | 2021-05-19T19:02:32 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 844,396,186 | 24,196 | alexey-milovidov | @alesapin There are no reasons. | 2021-05-19T19:17:05 | 2021-05-19T19:17:05 | {} | MEMBER |
ClickHouse | ClickHouse | 844,943,936 | 24,307 | mergify[bot] | **Command `update`: success**
> **Branch has been successfully updated**
>
_Hey, I reacted but my real name is @Mergifyio_ | 2021-05-20T10:06:34 | 2021-05-20T10:06:34 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 845,001,309 | 24,207 | akvasha | > Looks like lot of copied code there. Is it just a proof-of-concept or there are some fundamental issues which don't allow to reuse existing code (for example by inheritance)?
That's just an init commit, which is currently being actively pursued. Mostly copied code here is used to keep based HTTP handler correct wh... | 2021-05-20T11:24:10 | 2021-05-20T11:24:41 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 845,253,474 | 23,616 | alexey-milovidov | Need backport to 21.5. | 2021-05-20T16:09:44 | 2021-05-20T16:09:44 | {} | MEMBER |
ClickHouse | ClickHouse | 845,363,914 | 24,337 | alexey-milovidov | BTW, what exception do you see? | 2021-05-20T18:29:46 | 2021-05-20T18:29:46 | {} | MEMBER |
ClickHouse | ClickHouse | 845,532,616 | 24,294 | filimonov | @excitoon what do you think about it? | 2021-05-20T22:52:18 | 2021-05-20T22:52:18 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 844,385,354 | 24,290 | vdimir | Failed fast test seems flaky
```
┌─check_name───────────────────────────┬─test_name─────────────┬────check_start_time─┬─test_status─┬────pr─┬─commit_sha───────────────────────────────┐
│ Fast test │ 01852_hints_enum_name │ 2021-05-19 18:05:01 │ FAIL │ 24273 │ 8e68d007f30ac8b2fb7bce0... | 2021-05-19T19:01:53 | 2021-05-19T19:01:53 | {} | MEMBER |
ClickHouse | ClickHouse | 844,390,447 | 24,293 | filimonov | 20.10 - last good version. | 2021-05-19T19:09:09 | 2021-05-19T19:09:09 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 844,408,483 | 24,271 | alexey-milovidov | We need to cover this case in unit test (see `gtest_pod_array.cpp`) | 2021-05-19T19:34:25 | 2021-05-19T19:34:25 | {} | MEMBER |
ClickHouse | ClickHouse | 844,443,308 | 23,504 | filimonov | I said before - there are 2 main issues.
1) distributed ddl not working (it sees the task for `myclickhouse1.host.name.com` but it's resolved to 127.0.1.1 and which is 'not local')
2) prefer localhost replica for distributed tables don't work (it thinks all nodes remote)
Yes, alternative are also possible - just... | 2021-05-19T20:22:01 | 2021-05-19T20:22:01 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 844,451,211 | 24,252 | azat | >I think I have fixed your cases. Do you have more?
No, seems that it covers all the cases (at least those that I know), thanks! | 2021-05-19T20:33:25 | 2021-05-19T20:33:25 | {} | MEMBER |
ClickHouse | ClickHouse | 844,929,346 | 24,271 | kitaisreal | @fuzhe1989 actually asan can check memcpy overlap regions, will update gtest_pod_array to check that case. | 2021-05-20T09:50:24 | 2021-05-20T09:50:36 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 845,448,491 | 13,406 | filimonov | May be related #23512 | 2021-05-20T20:18:06 | 2021-05-20T20:18:06 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 845,564,970 | 19,912 | alexey-milovidov | Release 21.2 is no longer supported. | 2021-05-21T00:17:50 | 2021-05-21T00:17:50 | {} | MEMBER |
ClickHouse | ClickHouse | 844,457,204 | 24,281 | scottgurevich | After a through check, I believe the issue is caused by the `forceNullableLeft()` and `forceNullableRight()` functions. I believe these functions should return false when the join is a semi join or an asof join. | 2021-05-19T20:42:19 | 2021-05-19T20:42:19 | {} | NONE |
ClickHouse | ClickHouse | 844,492,971 | 24,102 | den-crane | Not sure what TTL do you use? TTL delete? TTL move? TTL groupby?
In case of `TTL delete` you can use ttl_only_drop_parts.
> how many aggregating tables I can hang off one 1M rows/sec MergeTable
1M rows with 40 bytes each (4 columns) <> 1M rows with 4000 bytes each (900 columns).
Sometimes 2 Materialized Views ... | 2021-05-19T21:39:15 | 2021-05-19T21:39:15 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 844,741,328 | 12,135 | geekyouth | > It's because with `Atomic` database engine `DROP TABLE` query just marks table as dropped. Table are actually removed later in background, when all queries to the table are finished. Even if no queries to the table are running, there is a delay between `DROP TABLE` and background removal, it can be configured by `<da... | 2021-05-20T06:19:51 | 2021-05-20T06:20:18 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 844,882,662 | 24,321 | qoega | can you check if 01149_zookeeper_mutation_stuck_after_replace_partition failure is related? | 2021-05-20T08:58:26 | 2021-05-20T08:58:26 | {} | MEMBER |
ClickHouse | ClickHouse | 844,901,797 | 20,172 | tekeri | Same problem.
I had issued `CREATE VIEW malformed AS SELECT NULL as bad_column;` to trigger this. The inferred type is `binary(0)` and thus zero sized.
Workaround: use `CAST(NULL as EXPECTED_TYPE) AS good_column,` instead. | 2021-05-20T09:19:34 | 2021-05-20T09:19:34 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 845,208,159 | 24,134 | vzakaznikov | @alesapin, let's reopen. On master the stress test still fails. I've updated the test to provide better output.
```
user@user-host:~/Projects/github/ClickHouse-vzakaznikov$ ./dev_tests.sh 01900_alter_race_condition
Using queries from 'tests/queries/' directory
Connecting to ClickHouse server... OK
Running 2 s... | 2021-05-20T15:11:55 | 2021-05-20T15:12:15 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 845,363,608 | 24,337 | tavplubix | I think that turning into readonly is ok, because it will "avoid possible further corruption of data in ZK" too. But AFAIK the only correct way to turn active replica into readonly is to expire session, just setting readonly flag may not work correctly. | 2021-05-20T18:29:20 | 2021-05-20T18:29:20 | {} | MEMBER |
ClickHouse | ClickHouse | 844,405,240 | 24,174 | filimonov | Does the situation normalizes after restart? How much RAM do you have at all? Also Kafka can prefetch messages - check if the detaching Kafka table(s) helps (but memory may be released with some delay). | 2021-05-19T19:29:46 | 2021-05-19T19:29:46 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 844,425,346 | 23,504 | alexey-milovidov | > made a 'canonical setup' of Ubuntu / debian clickhouse can't work properly
What is the issue? What is working in a wrong way? | 2021-05-19T19:58:17 | 2021-05-19T19:58:17 | {} | MEMBER |
ClickHouse | ClickHouse | 845,244,932 | 24,307 | alesapin | Any functional tests run is enough. | 2021-05-20T15:58:35 | 2021-05-20T15:58:48 | {} | MEMBER |
ClickHouse | ClickHouse | 845,253,184 | 23,327 | alexey-milovidov | > Stress test (thread) — Fatal message in clickhouse-server.log
https://github.com/ClickHouse/ClickHouse/pull/23616 | 2021-05-20T16:09:21 | 2021-05-20T16:09:21 | {} | MEMBER |
ClickHouse | ClickHouse | 845,540,224 | 24,360 | den-crane | >be left untouched forever on their respective (low) levels
Yes,be left untouched forever . No splits.
BTW, forced merge `optimize table ... final` can merge parts to any size (it does not limited by max_bytes_to_merge_at_max_space_in_pool).
* You can increase max_bytes_to_merge_at_max_space_in_pool to 1TB.
*... | 2021-05-20T23:09:39 | 2021-05-20T23:10:29 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 844,423,418 | 24,262 | alexey-milovidov | @UnamedRus most likely we will do nothing about this backward incompatibility.
Maybe it will be better to remove `avgWeighted` completely?
| 2021-05-19T19:55:35 | 2021-05-19T19:55:35 | {} | MEMBER |
ClickHouse | ClickHouse | 844,429,307 | 23,504 | alexey-milovidov | > on cluster DDL have issues with recognizing itself and queries hang. See #18341
We should rewrite the logic of ON CLUSTER queries to use host UUIDs or database UUIDs instead of IP addresses. | 2021-05-19T20:03:39 | 2021-05-19T20:04:11 | {} | MEMBER |
ClickHouse | ClickHouse | 844,431,576 | 24,218 | alexey-milovidov | Yes, it was changed in 21.4, see https://github.com/ClickHouse/ClickHouse/pull/20058 | 2021-05-19T20:07:12 | 2021-05-19T20:07:12 | {} | MEMBER |
ClickHouse | ClickHouse | 844,664,157 | 24,315 | alexey-milovidov | The latest *stable* version (e.g. 21.4) is recommended to use in production. In our experience the most recent stable version contains less amount of issues.
You can prefer *LTS* version (e.g. 21.3) in rare cases when you cannot install newer versions while still being able to install patches. We release patches for... | 2021-05-20T03:47:30 | 2021-05-20T03:47:30 | {} | MEMBER |
ClickHouse | ClickHouse | 844,716,691 | 24,157 | kssenii | @Mergifyio update | 2021-05-20T05:44:30 | 2021-05-20T05:44:30 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 844,776,674 | 24,059 | alesapin | All failures are not related to changes | 2021-05-20T07:10:40 | 2021-05-20T07:10:40 | {} | MEMBER |
ClickHouse | ClickHouse | 845,152,606 | 24,337 | alexey-milovidov | But this is controversial... | 2021-05-20T14:03:03 | 2021-05-20T14:03:03 | {} | MEMBER |
ClickHouse | ClickHouse | 845,574,067 | 24,350 | vzakaznikov | @MyroTk, please fix style check.
```
/ClickHouse/tests/testflows/extended_precision_data_types/extended-precision-data-type_env/docker-compose.yml
68:35 error no new line character at the end of file (new-line-at-end-of-file)
2021-05-20 17:43:31,478 Result parsed
2021-05-20 17:43:31,479 Result writte... | 2021-05-21T00:44:30 | 2021-05-21T00:44:30 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 845,579,552 | 24,350 | alexey-milovidov | I have no idea what error 70 means in Docker. Check the stderr. | 2021-05-21T01:02:23 | 2021-05-21T01:02:23 | {} | MEMBER |
ClickHouse | ClickHouse | 844,386,114 | 24,290 | mergify[bot] | **Command `update`: success**
> **Branch has been successfully updated**
>
_Hey, I reacted but my real name is @Mergifyio_ | 2021-05-19T19:02:54 | 2021-05-19T19:02:54 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 844,420,914 | 7,898 | alexey-milovidov | @apbodrov Why do you need `avgWeighted` aggregate function when you can write `sum(x * weight) / sum(weight)` instead? | 2021-05-19T19:52:21 | 2021-05-19T19:52:21 | {} | MEMBER |
ClickHouse | ClickHouse | 844,471,532 | 24,308 | azat | > kitaisreal added the force tests label 1 hour ago
@kitaisreal thanks!
By the way AFAIR after adding this label the PR should be updated (since existing CI jobs may be canceled already), will do a rebase | 2021-05-19T21:03:39 | 2021-05-19T21:03:56 | {} | MEMBER |
ClickHouse | ClickHouse | 844,496,749 | 24,237 | UnamedRus | Can be included in https://github.com/ClickHouse/ClickHouse/issues/23194 | 2021-05-19T21:46:02 | 2021-05-19T21:46:02 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 844,502,703 | 24,311 | mergify[bot] | **Command `update`: success**
> **Branch has been successfully updated**
>
| 2021-05-19T21:57:12 | 2021-05-19T21:57:12 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 844,889,507 | 24,245 | qoega | Can you commit anything to trigger build with newer master? Or I can use mergify to update branch | 2021-05-20T09:06:15 | 2021-05-20T09:06:15 | {} | MEMBER |
ClickHouse | ClickHouse | 844,717,617 | 24,157 | mergify[bot] | **Command `update`: success**
> **Branch has been successfully updated**
>
| 2021-05-20T05:45:12 | 2021-05-20T05:45:12 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 844,865,643 | 24,271 | fuzhe1989 | Hi alexey, from spec memcpy's behavior is undefined if the src and dst range are overlapping. However from [here](https://stackoverflow.com/a/25630217):
> on Linux, up until fairly recently, the implementation of memcpy() worked in a way that was similar enough to memmove() that overlapping memory wasn't an issue.
... | 2021-05-20T08:38:19 | 2021-05-20T08:38:19 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 844,943,606 | 24,307 | alesapin | @mergify update | 2021-05-20T10:06:08 | 2021-05-20T10:06:08 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 845,048,561 | 24,268 | abyss7 | > looks like duplicate with #23901
It is. | 2021-05-20T12:16:40 | 2021-05-20T12:16:40 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 845,208,459 | 24,250 | ucasfl | > I added a bit more tests for different data types and change return type for Float32/64 to Float64.
>
> @ucasFL take a look if it's ok I'll merge when CI passed
LGTM. | 2021-05-20T15:12:17 | 2021-05-20T15:12:17 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 845,557,055 | 24,364 | alexey-milovidov | Wow, we are so far ahead of Abseil hash map! | 2021-05-20T23:56:14 | 2021-05-20T23:56:14 | {} | MEMBER |
ClickHouse | ClickHouse | 844,391,414 | 18,787 | nikvas0 | > But it looks like it working correctly if we have only 2 groups:
Fixed | 2021-05-19T19:10:27 | 2021-05-19T19:10:36 | {
"heart": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 844,397,468 | 24,196 | alexey-milovidov | Performance is improved after updating re2 :tada: : https://clickhouse-test-reports.s3.yandex.net/24196/c312121ced4d3db38b205cce807193311ff887bf/performance_comparison/report.html | 2021-05-19T19:18:50 | 2021-05-19T19:18:50 | {
"hooray": 1
} | MEMBER |
ClickHouse | ClickHouse | 844,606,572 | 24,313 | alexey-milovidov | Fast test Ok. | 2021-05-20T01:14:26 | 2021-05-20T01:14:26 | {} | MEMBER |
ClickHouse | ClickHouse | 844,996,265 | 24,321 | amosbird | > can you check if 01149_zookeeper_mutation_stuck_after_replace_partition failure is related?
Cannot reproduce with local unsan build. It doesn't seem to be related though. | 2021-05-20T11:18:36 | 2021-05-20T11:18:36 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 845,531,614 | 24,102 | den-crane | >PARTITION BY toStartOfFiveMinute(t)
Oh, wowwowowow. Mom, it's too scary. | 2021-05-20T22:50:01 | 2021-05-20T22:50:01 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 845,562,110 | 23,001 | alexey-milovidov | Everything is resolved except #23508.
The fix for #23508 (which is under development in #24157) is too risky and will not be present in the first 21.5. It also affects very marginal use case. After it will be done we will backport it to 21.5 for experiment. | 2021-05-21T00:09:53 | 2021-05-21T00:09:53 | {} | MEMBER |
ClickHouse | ClickHouse | 844,846,584 | 24,086 | ghuname | It is indirectly related, and basically a long and complicated story.
We had to restore nodes, and during restoration, we found problems with huge number of parts in old clickhouse data directory.
After restoration, parts are under control, but looks like clickhouse wasn't good enough to keep them under control befor... | 2021-05-20T08:14:10 | 2021-05-20T08:14:10 | {} | NONE |
ClickHouse | ClickHouse | 844,879,587 | 24,301 | grundrauschen | Hello @den-crane,
I have a question concerning the change or `bug` to `feature`.
Is it not expected, that parameters work in every kind of queries or did I get the wrong impression from the docs?
The software I use Clickhouse with, creates queries dynamically but with certain values from the users.
To mitigate po... | 2021-05-20T08:54:42 | 2021-05-20T08:54:42 | {} | NONE |
ClickHouse | ClickHouse | 844,921,393 | 19,255 | chengbinghan | 20.8.12.2 also has the problem
| 2021-05-20T09:41:42 | 2021-05-20T09:41:42 | {} | NONE |
ClickHouse | ClickHouse | 845,040,277 | 24,149 | nshah14285 | @kssenii
OK. | 2021-05-20T12:09:04 | 2021-05-20T12:09:04 | {} | NONE |
ClickHouse | ClickHouse | 845,101,331 | 24,290 | akuzm | @mergifyio update | 2021-05-20T13:05:17 | 2021-05-20T13:05:17 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 845,564,584 | 22,272 | alexey-milovidov | @shiyhuan What exception do you get? | 2021-05-21T00:16:45 | 2021-05-21T00:16:45 | {} | MEMBER |
ClickHouse | ClickHouse | 844,392,128 | 24,277 | kolsys | I know, but I need at least one arm64 docker image with mysql support for local development. I see that build is possible, but don't understand why deb package builds without success. | 2021-05-19T19:11:25 | 2021-05-19T19:11:25 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 844,430,810 | 6,613 | alexey-milovidov | @MaxWk Thank you, this will be very appreciated! | 2021-05-19T20:05:52 | 2021-05-19T20:05:52 | {} | MEMBER |
ClickHouse | ClickHouse | 844,557,563 | 13,348 | alexey-milovidov | No new reports for almost a year. | 2021-05-19T23:11:10 | 2021-05-19T23:11:10 | {} | MEMBER |
ClickHouse | ClickHouse | 844,875,150 | 24,102 | larry-cdn77 | I guess it's TTL delete, my aggregated tables are created like so:
```
CREATE TABLE db.table (
t DateTime,
s AggregateFunction(sum, UInt64)
)
ENGINE = ReplicatedAggregatingMergeTree('/clickhouse/tables/{shard}/t', '{replica}')
PARTITION BY toStartOfFiveMinute(t)
ORDER BY (t)
TTL t + INTERVAL 24 HOUR
```... | 2021-05-20T08:49:34 | 2021-05-20T08:49:34 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 845,033,628 | 12,601 | azat | Fixed in #24252 | 2021-05-20T12:02:16 | 2021-05-20T12:02:16 | {} | MEMBER |
ClickHouse | ClickHouse | 845,569,720 | 24,350 | vzakaznikov | @MyroTk, tests failed because of
```
✘ 5m 8s [ Fail ] /clickhouse/extended precision data types
could not bring up docker-compose cluster
``` | 2021-05-21T00:32:04 | 2021-05-21T00:32:15 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 845,571,797 | 24,350 | vzakaznikov | @alexey-milovidov, we see these `Exit 70` errors randomly when we try to bring up clickhouse containers. Any ideas?
For example,
```
723ms [bash] ---------------------------------------------------------------------------------------------------------
-----------------------------
... | 2021-05-21T00:38:18 | 2021-05-21T00:38:18 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 844,610,831 | 24,174 | alexcd90 | I restart ck in the test env ,then it's memory usage decrease to 20%. | 2021-05-20T01:26:11 | 2021-05-20T01:26:11 | {} | NONE |
ClickHouse | ClickHouse | 845,101,692 | 24,290 | mergify[bot] | **Command `update`: success**
> **Branch has been successfully updated**
>
| 2021-05-20T13:05:36 | 2021-05-20T13:05:36 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 845,155,802 | 19,255 | alexey-milovidov | @chengbinghan Update to the latest stable version 21.4. | 2021-05-20T14:07:06 | 2021-05-20T14:07:14 | {} | MEMBER |
ClickHouse | ClickHouse | 845,490,802 | 23,643 | abyss7 | > Was it backported to 21.3? Why backport PR is still opened?
Because it has conflicts, which no one resolved. | 2021-05-20T21:30:25 | 2021-05-20T21:30:25 | {
"eyes": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 805,369,761 | 21,897 | alexey-milovidov | I confirm, it's reproduced very easily. | 2021-03-24T00:20:53 | 2021-03-24T00:20:53 | {} | MEMBER |
ClickHouse | ClickHouse | 805,488,086 | 21,904 | godliness | Found a reference pr: https://github.com/ClickHouse/ClickHouse/pull/21639 | 2021-03-24T04:43:49 | 2021-03-24T04:43:49 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 804,998,213 | 22,045 | OmarBazaraa | This should fix #21189 and #21041 | 2021-03-23T15:28:48 | 2021-03-24T11:25:52 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 805,980,522 | 22,086 | tavplubix | > Implementing atomicity + isolation for operation set of 'insert' and 'merge' is trivial. Many inserts and one merge can't conflict, and we avoid the conflict between merges because they are scheduled in a special way by the server itself. But we have to support 'drop' and 'alter' as well.
Yes, I wrote almost the s... | 2021-03-24T16:40:23 | 2021-03-24T17:14:28 | {} | MEMBER |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.