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 | 690,369,874 | 14,698 | amosbird | It's possible to relax the condition of "end of insert data" in https://github.com/ClickHouse/ClickHouse/blob/master/programs/client/Client.cpp#L945 (for instance to check `;\n`)
but who knows what consequences it might bring | 2020-09-10T15:32:47 | 2020-09-10T15:32:47 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 690,371,267 | 14,701 | amosbird | It's replicated. Documentation will highlight operations that are not replicated, so that makes it somehow partial-intuitive. | 2020-09-10T15:34:46 | 2020-09-10T15:34:46 | {
"+1": 1
} | COLLABORATOR |
ClickHouse | ClickHouse | 690,596,368 | 14,730 | alexey-milovidov | > Is it possible to have old instances running when doing a grayscale upgrade?
No, but it is Ok.
First upgrade cluster to the minimum version that supports projections. Then wait until the chance that downgrade may be needed will be low. Then start using the new feature. The same as we had for almost all new feat... | 2020-09-10T18:17:04 | 2020-09-10T18:18:26 | {} | MEMBER |
ClickHouse | ClickHouse | 690,669,282 | 14,694 | itzikiusa | @amosbird
sounds interesting, didn't understand the concept that well though.
in general it will allow to display only specific columns with different primary key?
also, the projection should be on table level(like order by/TTL/partition)? | 2020-09-10T19:33:15 | 2020-09-10T19:33:15 | {} | NONE |
ClickHouse | ClickHouse | 690,104,289 | 13,990 | huynhphuong10284 | Just give more information: the issue is also reproduced in version 20.7.2 | 2020-09-10T09:15:33 | 2020-09-10T09:15:33 | {} | NONE |
ClickHouse | ClickHouse | 690,301,351 | 14,691 | den-crane | Yes, clickhouse-client is the same binary and it's very fat application and shares a huge piece of code with clickhouse-server.
clickhouse-client parses SQL, understands formats (including everything protobuf/parquet), parses input files (in parallel), formats output. Basically it's doing a huge part of work instead ... | 2020-09-10T13:49:09 | 2020-09-10T13:49:09 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 690,543,408 | 13,869 | alexey-milovidov | Actually we can apply this patch as is if we will copy the relevant headers and libraries to our cross-compile toolchain.
BTW, why don't continue to use procfs? I found that it is not mounted by default, but there are many tools that still require procfs, so it should likely to be mounted on the server? Maybe we can r... | 2020-09-10T17:33:26 | 2020-09-10T17:33:26 | {} | MEMBER |
ClickHouse | ClickHouse | 689,885,539 | 14,572 | liangli1213 | Thanks @zhang2014 @BohuTANG for your effort, hope you can fix it. | 2020-09-09T23:56:20 | 2020-09-09T23:56:20 | {} | NONE |
ClickHouse | ClickHouse | 689,996,247 | 14,626 | hustnn | > Some queries stopped to work, look at the test results.
> It means that some changes are incorrect.
@alexey-milovidov , I am looking at this. Seems `exists` still needed in some case. We should only avoid to use in the case the node does not exist and it will never exist because it is sequential in this [link](ht... | 2020-09-10T05:35:53 | 2020-09-10T07:13:13 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 690,177,989 | 14,683 | KochetovNicolai | fuzzer:
```
SELECT * REPLACE(i + 65537 AS i) APPLY(sum)
FROM columns_transformers
base before fuzz: SELECT * REPLACE(i + 65537 AS i) APPLY(sum) FROM columns_transformers
base after fuzz: SELECT * REPLACE(i + 1023 AS i) APPLY(sum) FROM columns_transformers
dump before fuzz:
SelectWithUnionQuery, 0x7f08a7d991d8
... | 2020-09-10T11:18:49 | 2020-09-10T11:18:49 | {} | MEMBER |
ClickHouse | ClickHouse | 690,316,196 | 14,688 | BernieJiangDL | > No.
> No transactions.
> And the source table has no relation to the Materialized view. Materialized view watches INSERTS.
When some unexpected results occur(from bussiness side, it might be some wrong/dirty records from upstream) after inserting. Rerunning the task is easiest way and then dropping a relative pa... | 2020-09-10T14:11:51 | 2020-09-10T14:11:51 | {} | NONE |
ClickHouse | ClickHouse | 690,357,884 | 14,685 | amosbird | Will it improve performance in any way?
> use transparent huge pages
IIUC, aren't they just normal huge pages instead? | 2020-09-10T15:14:31 | 2020-09-10T15:17:10 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 690,392,024 | 14,045 | alexey-milovidov | https://github.com/ClickHouse/ClickHouse/pull/14646 | 2020-09-10T15:53:50 | 2020-09-10T15:53:50 | {} | MEMBER |
ClickHouse | ClickHouse | 690,599,716 | 13,869 | olevole | > BTW, why don't continue to use procfs? I found that it is not mounted by default, but there are many tools that still require procfs, so it should likely to be mounted on the server? Maybe we can require mounting procfs as an installation step?
Actually yes - I can add an appropriate post-install message for http:... | 2020-09-10T18:22:51 | 2020-09-10T18:39:04 | {} | NONE |
ClickHouse | ClickHouse | 690,686,642 | 14,714 | alexey-milovidov | Waiting only for builds.
All builds Ok. | 2020-09-10T20:01:50 | 2020-09-10T20:01:50 | {} | MEMBER |
ClickHouse | ClickHouse | 689,723,966 | 13,875 | den-crane | What API do you use for "INSERT INTO .... FORMAT TabSeparated " ? JDBC ? | 2020-09-09T17:58:39 | 2020-09-09T17:58:54 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,942,182 | 14,521 | hexiaoting | @myrrc OK^^ | 2020-09-10T02:48:20 | 2020-09-10T02:48:20 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 690,061,368 | 14,676 | filimonov | Duplicate of #5747
https://github.com/ClickHouse/ClickHouse/blob/7edb911e9179563aa4b8b8e7127fd9b605e70a16/contrib/libhdfs3-cmake/CMakeLists.txt#L20 | 2020-09-10T07:55:31 | 2020-09-10T07:55:31 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 690,381,954 | 14,687 | Fanduzi | > > Does this mean I cannot use clickhouse-copier to migrate data between two clusters with different time zone?
>
> You can use clickhouse-copier. You need to configure the client to see results in desired timezone.
>
> example :
> **TZ=Asia/Shanghai** clickhouse-client -q 'select toDateTime(1599746203)' **--us... | 2020-09-10T15:46:29 | 2020-09-10T15:46:29 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 690,458,473 | 13,859 | olevole | There have been several reports from FreeBSD Clickhouse users about this issue. The patches fix the problem and are applied now at the FreeBSD ports level: https://svnweb.freebsd.org/ports?view=revision&revision=548207 for 20.7.2.30 ( https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=249205 ).
Hope to see it upstrea... | 2020-09-10T16:30:25 | 2020-09-10T16:30:25 | {} | NONE |
ClickHouse | ClickHouse | 690,491,127 | 8,331 | vkingnew | what's the progress of this feature?CAN be Released in this Year? | 2020-09-10T16:48:50 | 2020-09-10T16:48:50 | {
"laugh": 1
} | NONE |
ClickHouse | ClickHouse | 689,944,614 | 14,626 | alexey-milovidov | Some queries stopped to work, look at the test results.
It means that some changes are incorrect. | 2020-09-10T02:56:17 | 2020-09-10T02:56:17 | {} | MEMBER |
ClickHouse | ClickHouse | 689,949,321 | 14,649 | alexey-milovidov | Not a bug. | 2020-09-10T03:11:14 | 2020-09-10T03:11:14 | {} | MEMBER |
ClickHouse | ClickHouse | 689,953,045 | 14,632 | alexey-milovidov | Functional tests:
> 00749_inner_join_of_unnamed_subqueries
Pipeline stuck! @KochetovNicolai | 2020-09-10T03:23:02 | 2020-09-10T03:23:02 | {} | MEMBER |
ClickHouse | ClickHouse | 689,954,250 | 14,588 | alexey-milovidov | Pipeline stuck error, @KochetovNicolai | 2020-09-10T03:26:53 | 2020-09-10T03:26:53 | {} | MEMBER |
ClickHouse | ClickHouse | 690,059,619 | 14,635 | filimonov | One of the possible reason, 10.10.200.208 send data inserted in Distribured table, but the file it tries to send is damaged. | 2020-09-10T07:52:00 | 2020-09-10T07:52:00 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 690,236,155 | 14,677 | 4ertus2 | It's more then debug assert. There's `operator << (...) noexcept` that calls `shiftLeft()` that is not a noexcept func. | 2020-09-10T12:15:20 | 2020-09-10T12:16:10 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 690,294,525 | 14,572 | tavplubix | Duplicate of #14251 | 2020-09-10T13:37:39 | 2020-09-10T13:37:39 | {} | MEMBER |
ClickHouse | ClickHouse | 690,304,763 | 14,333 | nikitamikhaylov | There is no new integration tests and feature is isolated, so lets merge. | 2020-09-10T13:54:37 | 2020-09-10T13:54:37 | {} | MEMBER |
ClickHouse | ClickHouse | 690,310,035 | 14,687 | den-crane | Datetime. The point in time is saved as a Unix timestamp, **regardless of the time zone or daylight saving time**.
Additionally, the DateTime type can store time zone that is the same for the entire column, that affects how the values of the DateTime type values are **displayed in text format** and how the values spe... | 2020-09-10T14:02:00 | 2020-09-10T14:03:21 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 690,447,506 | 14,513 | amosbird | > Partition elimination does not depend on monotonicity at all - it simply evaluates the expressions for every part.
Treat is as a minmax index with granularity 1 across data parts.
I'm afraid it's not correct.
@filimonov I'll do a case by case explanation.
- case one: `select * from intDiv_test where a = 1... | 2020-09-10T16:25:30 | 2020-09-10T16:30:40 | {
"+1": 3
} | COLLABORATOR |
ClickHouse | ClickHouse | 689,741,869 | 14,623 | alexey-milovidov | Not significant for backport. | 2020-09-09T18:34:45 | 2020-09-09T18:34:45 | {} | MEMBER |
ClickHouse | ClickHouse | 689,825,074 | 14,130 | fibersel | I can fix that bug, but the only obstacle is how to write test for that case?
I've set all environment and run some integration tests from repo, but I don't know, how to create folder into running container properly.Can someone point to similar example? | 2020-09-09T21:12:57 | 2020-09-09T21:56:20 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,954,042 | 11,234 | jjtjiang | get it.thanks.
i user docker start clickhouse but havn't set host:ip ,so it Can't contact LDAP server.
having solved this problem.
| 2020-09-10T03:26:10 | 2020-09-10T03:26:10 | {} | NONE |
ClickHouse | ClickHouse | 690,212,764 | 14,683 | KochetovNicolai | Not waiting for integration and preftests. | 2020-09-10T11:46:45 | 2020-09-10T11:46:45 | {} | MEMBER |
ClickHouse | ClickHouse | 690,378,163 | 14,701 | den-crane | And it means that all replicas need a source table. And they will attach the partition without fetching. | 2020-09-10T15:42:37 | 2020-09-10T15:42:37 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,736,564 | 14,609 | alexey-milovidov | > /ClickHouse/src/Storages/MergeTree/MergeTreePartition.cpp:33:32: error: unknown warning group '-Wstringop-overflow'
Should suppress warning only for new gcc.
| 2020-09-09T18:23:56 | 2020-09-09T18:23:56 | {} | MEMBER |
ClickHouse | ClickHouse | 689,741,844 | 14,624 | alexey-milovidov | Not significant for backport. | 2020-09-09T18:34:42 | 2020-09-09T18:34:42 | {} | MEMBER |
ClickHouse | ClickHouse | 689,949,307 | 14,650 | alexey-milovidov | Not a bug. | 2020-09-10T03:11:11 | 2020-09-10T03:11:11 | {} | MEMBER |
ClickHouse | ClickHouse | 690,344,901 | 14,687 | Fanduzi | > Datetime. The point in time is saved as a Unix timestamp, **regardless of the time zone or daylight saving time**.
> Additionally, the DateTime type can store time zone that is the same for the entire column, that affects how the values of the DateTime type values are **displayed in text format** and how the values ... | 2020-09-10T14:55:11 | 2020-09-10T14:55:11 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 690,459,652 | 13,869 | olevole | There have been several reports from FreeBSD Clickhouse users about this issue. The patches fix the problem and are applied now at the FreeBSD ports level: https://svnweb.freebsd.org/ports?view=revision&revision=548207 for 20.7.2.30 ( https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=249205 ).
Hope to see it upstrea... | 2020-09-10T16:30:59 | 2020-09-10T16:30:59 | {} | NONE |
ClickHouse | ClickHouse | 689,950,945 | 14,668 | alexey-milovidov | Integration test
> test_adaptive_granularity/test.py::test_version_update_two_nodes
- flaky test @alesapin | 2020-09-10T03:16:33 | 2020-09-10T03:16:33 | {} | MEMBER |
ClickHouse | ClickHouse | 689,951,452 | 14,585 | alexey-milovidov | @KochetovNicolai @4ertus2
This pull requests most likely caused "Pipeline stuck" errors. Revert? | 2020-09-10T03:18:16 | 2020-09-10T03:18:28 | {} | MEMBER |
ClickHouse | ClickHouse | 690,129,889 | 12,513 | adrian17 | Still reproducible on 20.8.2. | 2020-09-10T10:05:59 | 2020-09-10T10:05:59 | {} | NONE |
ClickHouse | ClickHouse | 690,139,088 | 14,563 | tavplubix | > @tavplubix thank you for adding those tests. Should this PR be considered a `Bug`. I had initially created it under `Improvement` - but makes more sense to be a bug. I did update the PR description, but label needs to be changed. Do let me know if this should rather be an`Improvement`.
Ok, let's change the changel... | 2020-09-10T10:25:02 | 2020-09-10T10:25:02 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 690,144,684 | 14,626 | hustnn | @alexey-milovidov The new idea is like this https://github.com/ClickHouse/ClickHouse/pull/14693 | 2020-09-10T10:37:15 | 2020-09-10T10:37:15 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 690,305,089 | 14,524 | tavplubix | #14564 | 2020-09-10T13:55:12 | 2020-09-10T13:55:12 | {} | MEMBER |
ClickHouse | ClickHouse | 690,311,961 | 6,917 | tavplubix | Related #12135 | 2020-09-10T14:04:52 | 2020-09-10T14:04:52 | {} | MEMBER |
ClickHouse | ClickHouse | 690,321,747 | 14,696 | bohutang | Cc @alexey-milovidov CI is unhappy | 2020-09-10T14:20:41 | 2020-09-10T14:20:41 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 690,354,699 | 5,783 | avgolubev | Guys, how to avoid this error when deleting, this clickchouse has already f.. my brains? | 2020-09-10T15:09:18 | 2020-09-10T15:09:18 | {} | NONE |
ClickHouse | ClickHouse | 690,589,796 | 14,694 | amosbird | Should be covered by https://github.com/ClickHouse/ClickHouse/issues/14730 | 2020-09-10T18:07:28 | 2020-09-10T18:07:28 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 689,741,895 | 14,622 | alexey-milovidov | Not significant for backport. | 2020-09-09T18:34:49 | 2020-09-09T18:34:49 | {} | MEMBER |
ClickHouse | ClickHouse | 689,948,074 | 14,521 | hexiaoting | > Style check has found tabs in code (need to replace to spaces).
I have fixed it . | 2020-09-10T03:06:59 | 2020-09-10T03:06:59 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,948,327 | 14,582 | amosbird | Try https://clickhouse.tech/docs/en/sql-reference/statements/alter/partition/#alter_attach-partition-from instead. `MOVE PARTITION` has different semantics against your use case. | 2020-09-10T03:07:49 | 2020-09-10T03:07:49 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 689,949,705 | 14,647 | alexey-milovidov | LGTM even without a test. | 2020-09-10T03:12:29 | 2020-09-10T03:12:29 | {} | MEMBER |
ClickHouse | ClickHouse | 689,950,275 | 14,670 | alexey-milovidov | > 00749_inner_join_of_unnamed_subqueries
Pipeline stuck! @KochetovNicolai | 2020-09-10T03:14:20 | 2020-09-10T03:14:20 | {} | MEMBER |
ClickHouse | ClickHouse | 690,020,444 | 14,574 | lookingGordo | 
Yes, I did.
After changing the ULimit, Clickhouse does not restart
Thank you for your help！ | 2020-09-10T06:32:46 | 2020-09-10T06:32:46 | {} | NONE |
ClickHouse | ClickHouse | 690,054,103 | 14,585 | KochetovNicolai | @alexey-milovidov no, most likely it is #14412
Possible fix: #14683 | 2020-09-10T07:41:38 | 2020-09-10T07:41:45 | {} | MEMBER |
ClickHouse | ClickHouse | 689,854,005 | 14,130 | den-crane | @fibersel check tests/integration/test_backup_with_other_granularity/test.py
node1.exec_in_container(['bash', '-c', 'cp -r /var/lib/clickhouse/shadow/1/data/default/source_table/all_1_1_0/ /var/lib/clickhouse/data/default/dest_table/detached'])
you can execute node1.exec_in_container(['bash', '-c', 'mkdir ...... | 2020-09-09T22:23:49 | 2020-09-09T22:27:27 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,948,397 | 14,525 | alexey-milovidov | AST Fuzzer: https://github.com/ClickHouse/ClickHouse/issues/14677 | 2020-09-10T03:08:06 | 2020-09-10T03:08:06 | {} | MEMBER |
ClickHouse | ClickHouse | 690,038,724 | 14,093 | filimonov | Most probably it's related to filesystem used (Lustre) | 2020-09-10T07:11:59 | 2020-09-10T07:11:59 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 690,057,883 | 14,648 | filimonov | The workaround is trivial, and exception message is clean.
```
SELECT toYYYYMMDD(fromUnixTimestamp64Micro(toInt64(1599672041963889)))
┌─toYYYYMMDD(fromUnixTimestamp64Micro(toInt64(1599672041963889)))─┐
│ 20200909 │
└─────────────────────────────────────────... | 2020-09-10T07:48:42 | 2020-09-10T07:48:42 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 690,368,040 | 2,154 | rikonor | We also encountered this issue when trying to run a dockerized cluster. I would actually argue this issue presents itself not because of `<interserver_http_host>` not matching the output `hostname`, but rather because Clickhouse is examining the network interfaces available to the container (See https://github.com/Clic... | 2020-09-10T15:30:30 | 2020-09-10T15:30:30 | {} | NONE |
ClickHouse | ClickHouse | 690,419,859 | 14,588 | alexey-milovidov | https://github.com/ClickHouse/ClickHouse/issues/14045 | 2020-09-10T16:10:43 | 2020-09-10T16:10:43 | {} | MEMBER |
ClickHouse | ClickHouse | 690,524,572 | 14,685 | alexey-milovidov | > Will it improve performance in any way?
Yes, it is possible. But I expect only small gains.
> IIUC, aren't they just normal huge pages instead?
Normal huge pages have very poor usability (very difficult to set up and to keep configuration sane) and we cannot use them in our CI environment for performance tes... | 2020-09-10T17:08:49 | 2020-09-10T17:08:49 | {} | MEMBER |
ClickHouse | ClickHouse | 690,725,315 | 14,730 | den-crane | AFAIK such table cannot be mutated ? Alter update/delete breaks consistency.
Vertica has two types: Projections and LiveAggregatingProjections. Projections are updateable because they have the same not aggregated rows. | 2020-09-10T20:56:39 | 2020-09-10T20:56:58 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,741,924 | 14,621 | alexey-milovidov | Not significant for backport. | 2020-09-09T18:34:53 | 2020-09-09T18:34:53 | {} | MEMBER |
ClickHouse | ClickHouse | 689,949,561 | 14,647 | alexey-milovidov | Integration tests:
> test_adaptive_granularity/test.py::test_version_update_two_nodes
@alesapin - flaky test. | 2020-09-10T03:12:03 | 2020-09-10T03:12:03 | {} | MEMBER |
ClickHouse | ClickHouse | 690,013,764 | 14,251 | bohutang | We're trying to fix it. | 2020-09-10T06:17:08 | 2020-09-10T06:17:08 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 690,273,646 | 13,156 | vitlibar | > @vitlibar can you take a look please? (I'm planning to rebase this once everything will be resolved, but until then it is better to keep all the history)
sorry for the delay | 2020-09-10T13:06:02 | 2020-09-10T13:06:02 | {} | MEMBER |
ClickHouse | ClickHouse | 690,425,134 | 14,691 | kekekedeng | Get it, tks a lot. | 2020-09-10T16:14:36 | 2020-09-10T16:14:36 | {} | NONE |
ClickHouse | ClickHouse | 689,945,083 | 14,521 | alexey-milovidov | Style check has found tabs in code (need to replace to spaces). | 2020-09-10T02:57:42 | 2020-09-10T02:57:42 | {} | MEMBER |
ClickHouse | ClickHouse | 689,951,070 | 14,668 | alexey-milovidov | > 00858_issue_4756
Pipeline stuck! @KochetovNicolai | 2020-09-10T03:16:59 | 2020-09-10T03:16:59 | {} | MEMBER |
ClickHouse | ClickHouse | 689,954,164 | 14,414 | alexey-milovidov | Pipeline stuck: #14632 #14670 @KochetovNicolai
Test version update, two nodes: #14664 #14668 @alesapin | 2020-09-10T03:26:35 | 2020-09-10T03:26:35 | {} | MEMBER |
ClickHouse | ClickHouse | 689,956,353 | 14,609 | alexey-milovidov | Surprisingly, it didn't help. | 2020-09-10T03:33:49 | 2020-09-10T03:33:49 | {} | MEMBER |
ClickHouse | ClickHouse | 690,292,693 | 14,698 | den-crane | ```
clickhouse-client -m < test.sql
SELECT *
FROM test
┌─value─┐
│ 1 │
│ 2 │
└───────┘
``` | 2020-09-10T13:34:39 | 2020-09-10T13:34:39 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 690,335,480 | 14,688 | den-crane | There are many approaches.
You can design partitions in MV the same way as a source table. Drop partitions in the source table and MV.
You can insert into MV the same data with negative sign to cancel rows in MV and drop partition in the source table.
You can truncate MV and re-calculate it (manually) after inserti... | 2020-09-10T14:41:45 | 2020-09-10T14:41:45 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,734,344 | 14,611 | alexey-milovidov | OOM in build: #14607 | 2020-09-09T18:19:28 | 2020-09-09T18:19:28 | {} | MEMBER |
ClickHouse | ClickHouse | 689,949,281 | 14,651 | alexey-milovidov | Not a bug. | 2020-09-10T03:11:06 | 2020-09-10T03:11:06 | {} | MEMBER |
ClickHouse | ClickHouse | 689,949,294 | 14,652 | alexey-milovidov | Not a bug. | 2020-09-10T03:11:09 | 2020-09-10T03:11:09 | {} | MEMBER |
ClickHouse | ClickHouse | 690,295,603 | 14,688 | den-crane | No.
No transactions.
And the source table has no relation to the Materialized view. Materialized view watches INSERTS. | 2020-09-10T13:39:27 | 2020-09-10T13:39:27 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 690,330,950 | 14,698 | den-crane | @bgranvea
Yes. I know.
This behavior is from the beginning of CH. And everybody aware of this. It's not "unexpected".
Probably it's more a feature request. | 2020-09-10T14:34:44 | 2020-09-10T14:34:44 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 690,364,230 | 5,783 | den-crane | @avgolubev nobody sees/reads messages in closed issues. Open a new one. | 2020-09-10T15:24:49 | 2020-09-10T15:24:49 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,799,323 | 14,563 | bharatnc | @tavplubix thank you for adding those tests. Should this PR be considered a `Bug`. I had initially created it under `Improvement` - but makes more sense to be a bug. I did update the PR description, but label needs to be changed. Do let me know if this should rather be an`Improvement`. | 2020-09-09T20:18:57 | 2020-09-09T20:19:18 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 689,827,308 | 14,130 | fibersel | @alexey-milovidov , can you suggest where the folder for garbage files in ClickHouse located is?
I would move unexpected files and folders there | 2020-09-09T21:18:09 | 2020-09-09T22:25:00 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 690,006,873 | 14,592 | simPod | Thank you very much, appreciated | 2020-09-10T06:01:02 | 2020-09-10T06:01:02 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 690,398,796 | 13,893 | alexey-milovidov | Fixed in https://github.com/ClickHouse/ClickHouse/pull/12550
Need to add a test. | 2020-09-10T15:58:58 | 2020-09-10T15:58:58 | {} | MEMBER |
ClickHouse | ClickHouse | 690,587,051 | 14,657 | alexey-milovidov | > test_adaptive_granularity/test.py::test_version_update_two_nodes
- flaky test @alesapin | 2020-09-10T18:03:25 | 2020-09-10T18:03:25 | {} | MEMBER |
ClickHouse | ClickHouse | 689,738,028 | 14,443 | alexey-milovidov | @tavplubix Will look at the error in "Stress test (undefined) — Server failed to start". | 2020-09-09T18:26:47 | 2020-09-09T18:26:47 | {} | MEMBER |
ClickHouse | ClickHouse | 689,951,713 | 14,580 | jjtjiang | get it .thanks . | 2020-09-10T03:19:01 | 2020-09-10T03:19:01 | {} | NONE |
ClickHouse | ClickHouse | 689,952,429 | 14,664 | alexey-milovidov | Integration test
> test_adaptive_granularity/test.py::test_version_update_two_nodes
- flaky test @alesapin | 2020-09-10T03:21:10 | 2020-09-10T03:21:10 | {} | MEMBER |
ClickHouse | ClickHouse | 690,112,330 | 14,582 | dmgburg | It works just fine. Thank you! | 2020-09-10T09:31:18 | 2020-09-10T09:31:18 | {} | NONE |
ClickHouse | ClickHouse | 690,319,878 | 14,698 | bgranvea | > ```
> clickhouse-client -m < test.sql
>
> SELECT *
> FROM test
>
> ┌─value─┐
> │ 1 │
> │ 2 │
> └───────┘
> ```
my test case is too simple but my real use case has several queries.
BTW, the behaviour in this case seems also strange (no error although there are several statements and first m... | 2020-09-10T14:17:46 | 2020-09-10T14:17:46 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 690,401,733 | 14,701 | alvarotuso | @amosbird it also specifies some that are replicated but not others, that's what made me wary of making that assumption initially
Thanks for the response! | 2020-09-10T16:00:37 | 2020-09-10T16:00:37 | {} | NONE |
ClickHouse | ClickHouse | 690,407,383 | 14,701 | amosbird | > it also specifies some that are replicated but not others, that's what made me wary of making that assumption initially
Indeed. I think it's reasonable. | 2020-09-10T16:03:18 | 2020-09-10T16:03:18 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 689,878,029 | 13,875 | mjodeh | > What API do you use for "INSERT INTO .... FORMAT TabSeparated " ?
Yes
> JDBC ?
No I use Node js https://github.com/TimonKK/clickhouse | 2020-09-09T23:32:01 | 2020-09-09T23:32:01 | {} | NONE |
ClickHouse | ClickHouse | 689,890,648 | 14,437 | nikitamikhaylov | You can build the whole ClickHouse, go to your build directory, then src/processors/tests and there will be a binary.
But building the whole ClickHouse is very expensive (on CPU and memory) operation. There have to be another way to do this. I will check and answer. | 2020-09-10T00:11:50 | 2020-09-10T00:11:50 | {
"heart": 1,
"rocket": 1
} | MEMBER |
ClickHouse | ClickHouse | 690,696,027 | 14,694 | amosbird | > in general it will allow to display only specific columns with different primary key?
yes
> also, the projection should be on table level(like order by/TTL/partition)?
yes, actually at part level | 2020-09-10T20:13:54 | 2020-09-10T20:13:54 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 687,180,773 | 12,761 | alesapin | `test_host_ip_change/test.py::test_ip_change_update_dns_cache` flaky test. | 2020-09-04T14:29:31 | 2020-09-04T14:29:31 | {} | MEMBER |
ClickHouse | ClickHouse | 687,217,695 | 14,252 | nikitamikhaylov | Query found by fuzzer seems to be already fixed. Please, rebase on master.
| 2020-09-04T15:24:05 | 2020-09-04T15:24:05 | {
"+1": 1
} | MEMBER |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.