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 | 767,114,512 | 19,276 | alexey-milovidov | Fixed. | 2021-01-25T21:09:44 | 2021-01-25T21:09:44 | {} | MEMBER |
ClickHouse | ClickHouse | 767,116,740 | 19,546 | alexey-milovidov | Duplicate #6697. | 2021-01-25T21:14:13 | 2021-01-25T21:14:13 | {} | MEMBER |
ClickHouse | ClickHouse | 767,306,243 | 19,388 | bharatnc | Thank you for your reply @filimonov .
> test that feature (to confirm with real numbers / use cases that it has some advantages against os_thread_priority)
That would be great. I'm eager to help test your PR. Let me know how you would like me to help.
| 2021-01-26T05:31:53 | 2021-01-26T05:31:53 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 767,425,704 | 15,156 | kitaisreal | Fixed in https://github.com/ClickHouse/ClickHouse/pull/19571. | 2021-01-26T09:44:06 | 2021-01-26T09:44:06 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 767,108,193 | 19,130 | alexey-milovidov | Fixed in #19313 | 2021-01-25T20:58:30 | 2021-01-25T20:58:30 | {} | MEMBER |
ClickHouse | ClickHouse | 767,252,122 | 19,590 | alexey-milovidov | > 00158_cache_dictionary_has | FAIL | 600.29
Timeout. CC @nikitamikhaylov
| 2021-01-26T02:47:48 | 2021-01-26T02:47:48 | {} | MEMBER |
ClickHouse | ClickHouse | 766,706,639 | 19,030 | tavplubix | @sliontc you should change the password. Please, be careful when posting screenshots, logs, configs and other stuff, which may contain sensitive data. | 2021-01-25T10:12:24 | 2021-01-25T10:12:24 | {} | MEMBER |
ClickHouse | ClickHouse | 767,087,652 | 19,584 | azat | >it will be affected by "zero on empty" setting.
Indeed, nice catch, fixed | 2021-01-25T20:19:39 | 2021-01-25T20:19:49 | {} | MEMBER |
ClickHouse | ClickHouse | 767,313,109 | 19,583 | azat | >AST fuzzer (ASan) — 0x60d0002b4888 is located 56 bytes inside of 136-byte region [0x60d0002b4850,0x60d0002b48d8)
#19599 | 2021-01-26T05:51:54 | 2021-01-26T05:51:54 | {} | MEMBER |
ClickHouse | ClickHouse | 767,387,874 | 18,684 | amosbird | fuzzer error looks real
https://clickhouse-test-reports.s3.yandex.net/18684/5998383a3b16a33c95dd71b3dd30614a0fbf5741/fuzzer_ubsan/report.html#fail1
```
SELECT gcd(-9223372036854775808, 1024)
Error on processing query 'SELECT gcd(-9223372036854775808, 1024)': Code: 32, e.displayText() = DB::Exception: Attempt ... | 2021-01-26T08:34:10 | 2021-01-26T08:34:10 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 767,459,857 | 19,595 | XuankuF | @den-crane thanks!
I add an extra parameter `SET distributed_aggregation_memory_efficient=1`, and it seems to be working.
My cluster consists of three hosts with two shard each, and I found that one clickhouse-server node not increase the memory. The memory usage of the other two nodes was expected.
Ideally, all th... | 2021-01-26T10:44:59 | 2021-01-26T10:44:59 | {} | NONE |
ClickHouse | ClickHouse | 767,489,032 | 19,611 | ahutsunshine | Sure, but it doesnt solve the problem from the root. In prod, we need to a stable version rather than it we build by ourselves where there may have bugs and not be used in prod. | 2021-01-26T11:43:35 | 2021-01-26T11:44:12 | {} | NONE |
ClickHouse | ClickHouse | 767,494,085 | 19,626 | filimonov | It's the address in an executable. If you have debug symbols installed it can be mapped into a position of source code, for example like that:
```
apt-get update
apt-get install clickhouse-common-static-dbg=$(clickhouse-client --version-clean)
apt-get install binutils
addr2line -afiCe /path/to/clickhouse 0x4f2... | 2021-01-26T11:54:08 | 2021-01-26T11:54:08 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 766,854,236 | 18,740 | den-crane | I don't see this PR in https://github.com/ClickHouse/ClickHouse/blob/master/CHANGELOG.md | 2021-01-25T14:29:36 | 2021-01-25T14:29:36 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 767,051,809 | 18,549 | alexey-milovidov | Continued in #18549 | 2021-01-25T19:20:36 | 2021-01-25T19:20:36 | {} | MEMBER |
ClickHouse | ClickHouse | 767,091,778 | 19,586 | ben-efiz | @den-crane i can confirm i had to restart the pods several times while keeping the persistant volume claim (aka storage including DB). Only remark i have is that the IP address i redacted is actually from the current node the pod is running on. Just before the first appearance of the above mentioned stacktrace, i see
... | 2021-01-25T20:26:53 | 2021-01-25T20:34:21 | {} | NONE |
ClickHouse | ClickHouse | 767,125,842 | 19,585 | alexey-milovidov | There is a setting `enable_global_with_statement`. Enabling this setting (set to value 1) will turn on the desired behaviour.
Also we plan to enable it by default, see #18684. | 2021-01-25T21:30:56 | 2021-01-25T21:31:21 | {} | MEMBER |
ClickHouse | ClickHouse | 767,283,278 | 19,545 | den-crane | system.merge_tree_settings -- mergeTree tables def. settings <merge_tree> config.xml's section
https://clickhouse.tech/docs/en/operations/settings/merge-tree-settings/#merge-tree-settings
system.settings -- user's session settings <profile> users.xml's section
https://clickhouse.tech/docs/en/operations/settings/... | 2021-01-26T04:08:15 | 2021-01-26T04:08:35 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 767,477,060 | 19,611 | sundy-li | it's only in master, you can download from ci builds | 2021-01-26T11:18:30 | 2021-01-26T11:18:30 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 766,695,202 | 18,762 | sundy-li | > Looks great! But later we should not forget to add backward incompatible change to changelog if we change default behavior for this tables. Am I right that old behavior without enabled new flag will no longer br checked in tests after this PR?
No, old behavior will not be compatible after this PR, because we alrea... | 2021-01-25T09:53:45 | 2021-01-25T09:57:13 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 766,866,279 | 19,253 | KochetovNicolai | 1. Msan fuzzer (fails under asan also)
```sql
select sumIf(NULL, (number % 2) > -9223372036854775808) from numbers(1)
```
```
2021.01.25 17:29:53.575204 [ 24642 ] {} <Fatal> BaseDaemon: ########################################
2021.01.25 17:29:53.575571 [ 24642 ] {} <Fatal> BaseDaemon: (version 21.2.1.1, build id... | 2021-01-25T14:46:50 | 2021-01-25T14:46:50 | {} | MEMBER |
ClickHouse | ClickHouse | 766,939,709 | 19,349 | alexey-milovidov | > Do you mean that this time also includes the transmission time, if so, it is fine.
Yes. | 2021-01-25T16:30:37 | 2021-01-25T16:30:37 | {} | MEMBER |
ClickHouse | ClickHouse | 767,242,675 | 18,861 | hexiaoting | @nautaa
I add libstdcxx_printers to python syspath, and it can work if you use `gdb clickhouse` and `info pretty-printer`
sys.path.insert(0, '/home/hewenting/workspace/gdb')
from printers1 import register_libcxx_printer_loader
register_libcxx_printer_loader()
sys.path.insert(0, '/home/hewenting/workspace/gdb/... | 2021-01-26T02:26:51 | 2021-01-26T02:26:51 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 767,388,748 | 18,684 | azat | > fuzzer error looks real
> https://clickhouse-test-reports.s3.yandex.net/18684/5998383a3b16a33c95dd71b3dd30614a0fbf5741/fuzzer_ubsan/report.html#fail1
Should be fixed in #19532 | 2021-01-26T08:35:50 | 2021-01-26T08:35:50 | {} | MEMBER |
ClickHouse | ClickHouse | 767,432,401 | 10,411 | qoega | > Postponed until #10683
This feature is merged. We can continue there | 2021-01-26T09:56:28 | 2021-01-26T09:56:28 | {} | MEMBER |
ClickHouse | ClickHouse | 767,493,158 | 19,611 | sundy-li | @ahutsunshine Map is an experimental feature currently. | 2021-01-26T11:52:08 | 2021-01-26T11:52:08 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 766,883,876 | 19,578 | vdimir | Should be fixed with https://github.com/ClickHouse/ClickHouse/pull/19406
I'm checking that revision with fail does not include this commit... | 2021-01-25T15:11:05 | 2021-01-25T15:11:05 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 766,926,327 | 19,578 | vdimir | 290eadc1 is a failed commit and ad73dc6 is fix. Fix was merged later.
```
$ git br tmp-290eadc1 290eadc1
$ git br --contains ad73dc6
* master
$ git show --summary 290eadc1 | grep Date:
Date: Fri Jan 22 18:09:34 2021 +0300
$ git show --summary ad73dc6 | grep Date:
Date: Fri Jan 22 19:21:45 2021 +0300
``... | 2021-01-25T16:10:50 | 2021-01-25T16:10:50 | {} | MEMBER |
ClickHouse | ClickHouse | 767,405,108 | 19,573 | tavplubix | Could you please post the CREATE query as a text, not as screenshot? | 2021-01-26T09:06:57 | 2021-01-26T09:06:57 | {} | MEMBER |
ClickHouse | ClickHouse | 767,440,509 | 14,567 | qoega | Still reproducing.
``` sql
SELECT toTypeName(toIPv6('::1'))
Query id: 0eb293f1-cc3d-4643-abc1-e4924e32f45d
┌─toTypeName(toIPv6('::1'))─┐
│ IPv6 │
└───────────────────────────┘
1 rows in set. Elapsed: 0.030 sec.
```
`EXPLAIN` shows nothing interesting in this case. We can show wheth... | 2021-01-26T10:09:47 | 2021-01-26T10:09:47 | {} | MEMBER |
ClickHouse | ClickHouse | 767,451,344 | 19,511 | leshikus | Thanks! | 2021-01-26T10:29:19 | 2021-01-26T10:29:19 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 767,489,944 | 17,623 | filimonov | > > > I'm very interested in lightweight updates/deletes, and would like to contribute. Not sure how far @myrrc has gotten in planning this?
@stigsb @zhicwu see: https://github.com/ClickHouse/ClickHouse/issues/19627 | 2021-01-26T11:45:33 | 2021-01-26T11:45:33 | {
"+1": 2
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 766,773,175 | 19,557 | filimonov | Use: toDate(stamp) / toMonth(stamp) / toYear(stamp)
Compatibility layer for MySQL will be added in https://github.com/ClickHouse/ClickHouse/pull/19387
/cc @dankondr | 2021-01-25T12:10:57 | 2021-01-25T12:13:41 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 766,879,671 | 19,551 | alesapin | Test fix OK. | 2021-01-25T15:05:08 | 2021-01-25T15:05:08 | {} | MEMBER |
ClickHouse | ClickHouse | 766,967,064 | 17,623 | olafbuitelaar | when light transactions are implemented, would there also be a plan to support open XA (https://pubs.opengroup.org/onlinepubs/009680699/toc.pdf)? I saw the mysql binlog reading has some support for it; https://github.com/ClickHouse/ClickHouse/issues/4006.
I don't think full XA support is needed, but it would be great ... | 2021-01-25T17:10:37 | 2021-01-25T17:10:37 | {} | NONE |
ClickHouse | ClickHouse | 767,025,245 | 18,620 | alexey-milovidov | @ningyanhui
It's `max_concurrent_queries` in config.xml. | 2021-01-25T18:35:03 | 2021-01-25T18:35:03 | {} | MEMBER |
ClickHouse | ClickHouse | 767,279,401 | 19,545 | yangzan66 | > Moreover, the default settings specified in config.xml are applied and saved in created tables.
> Chaning defaults then will not change table settings - you have to do ALTER TABLE ... MODIFY SETTING.
thanks for your reply
how can i find one table default settings value?
I can only find the 'Show statement ... | 2021-01-26T03:55:13 | 2021-01-26T03:55:13 | {} | NONE |
ClickHouse | ClickHouse | 767,290,381 | 13,334 | ghost | @nikitamikhaylov
Hello, I will close this PR. Sorry.
I'll create another PR when I can work on it.
Thank you. | 2021-01-26T04:35:04 | 2021-01-26T04:35:04 | {} | NONE |
ClickHouse | ClickHouse | 767,288,915 | 8,986 | ghost | @alexey-milovidov
It's a good news to implement a more general caching mechanism.
How do you think about a different PR #18465? I wonder whether the PR #18465 can be applied to ClickHouse or not.
Thank you. | 2021-01-26T04:29:38 | 2021-01-26T04:53:21 | {} | NONE |
ClickHouse | ClickHouse | 767,331,228 | 19,253 | azat | @KochetovNicolai This patches introduce one UAF:
https://clickhouse-test-reports.s3.yandex.net/19584/5f494c898af82cd46d68411696fa2d2ca0ef6184/fuzzer_tsan/report.html#fail1
```sql
SELECT [NULL, '25.6', '-0.02', NULL], [NULL], 1024, [NULL, '10485.76', NULL, NULL], [NULL, '-922337203.6854775808', toNullable(NULL)... | 2021-01-26T06:32:49 | 2021-01-26T06:32:49 | {} | MEMBER |
ClickHouse | ClickHouse | 766,498,527 | 19,030 | Hedwiglzy | > > It's because `SET` query changes setting value in current session. Distributed DDL queries are executed in separate session and `SET` query does not affect it. You should enable `allow_experimental_database_materialize_mysql` setting globally in server config to make it work with `ON CLUSTER` queries. Example:
> >... | 2021-01-25T02:26:16 | 2021-01-26T08:16:20 | {} | NONE |
ClickHouse | ClickHouse | 766,885,070 | 19,545 | alexey-milovidov | It does not update without server restart as most of the configuration chenges.
Only a few config changes can update without restart. | 2021-01-25T15:12:56 | 2021-01-25T15:12:56 | {} | MEMBER |
ClickHouse | ClickHouse | 766,890,303 | 19,546 | tavplubix | ClickHouse does not support correlated subqueries | 2021-01-25T15:20:26 | 2021-01-25T15:20:26 | {} | MEMBER |
ClickHouse | ClickHouse | 767,483,565 | 19,527 | ucasfl | > @ucasFL after small discussion maybe it will be better to implement it using existing framework FunctionUnaryArithmetic?
> Please check for example implementation of abs.cpp.
> Also could you please add performance tests for this `sign` function ?
Ok, I have rewrite it with FunctionUnaryArithmetic. | 2021-01-26T11:31:35 | 2021-01-26T11:31:35 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 766,809,362 | 19,527 | kitaisreal | @ucasFL after small discussion maybe it will be better to implement it using existing framework FunctionUnaryArithmetic?
Please check for example implementation of abs.cpp.
Also could you please add performance tests for this `sign` function ? | 2021-01-25T13:17:49 | 2021-01-25T14:20:28 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 765,111,484 | 18,465 | ghost | @alexey-milovidov
I added an experimental meaning to the this feature.
I am waiting for your code-review.
Please let me know what I have to do for merging the PR.
Thank you. | 2021-01-22T04:23:13 | 2021-01-26T04:39:30 | {} | NONE |
ClickHouse | ClickHouse | 767,365,893 | 19,598 | alesapin | All builds OK, let's merge. | 2021-01-26T07:47:55 | 2021-01-26T07:47:55 | {} | MEMBER |
ClickHouse | ClickHouse | 767,384,041 | 19,443 | tavplubix | AST fuzzer (UBSan) - #19562
Testflows check - broken in master | 2021-01-26T08:27:08 | 2021-01-26T08:27:08 | {} | MEMBER |
ClickHouse | ClickHouse | 767,401,794 | 19,552 | tavplubix | `PRIMARY KEY` in MySQL is converted to `ORDER BY` key in ClickHouse. It's not possible to alter column from `ORDER BY` key. The only way to achieve it is to recreate table with new key and do `INSERT .. SELECT ...`. We could automatically do it on `ALTER_OF_COLUMN_IS_FORBIDDEN` and similar exceptions, but `MaterializeM... | 2021-01-26T09:00:52 | 2021-01-26T09:00:52 | {} | MEMBER |
ClickHouse | ClickHouse | 767,440,760 | 19,611 | ahutsunshine | For example, if the table `table_map` has the following data where the keys are `name`, `age` and `gender` and the values are `20` , `lisi` and `female`, how can I get them? And how to do count aggregation group by `name` key or `lisi` value?
```
select * from table_map
┌─a────────────────────────────────... | 2021-01-26T10:10:12 | 2021-01-26T10:10:12 | {} | NONE |
ClickHouse | ClickHouse | 766,802,863 | 19,560 | tavplubix | Please take a look into `Context::executeTableFunction(...)`:
https://github.com/ClickHouse/ClickHouse/blob/813b2bcc7334672889bda8c148c324a28a126c98/src/Interpreters/Context.cpp#L994-L1011
There is no problem, `ITableFunction::execute(...)` is actually called once due to `table_function_results` cache. | 2021-01-25T13:06:32 | 2021-01-25T13:06:32 | {} | MEMBER |
ClickHouse | ClickHouse | 766,883,488 | 19,577 | qoega | > Please help to DELETE this issue, I found I misunderstanding to output 0 rows in set.
> It works OK.
closed | 2021-01-25T15:10:33 | 2021-01-25T15:10:33 | {} | MEMBER |
ClickHouse | ClickHouse | 766,968,681 | 8,986 | alexey-milovidov | We are going to implement query cache in slightly different way, so it can be enabled for all query results, not only for subqueries. This PR will be closed in favour of future implementation. I expect that it will come in 2021 Q2/Q3. | 2021-01-25T17:13:04 | 2021-01-25T17:13:04 | {} | MEMBER |
ClickHouse | ClickHouse | 767,078,954 | 19,586 | den-crane | As I understand after POD (server with CH restarted) it gets a new hostname.
After that other CH nodes alert this message because they try to update DNS cache for already non-existent (outdated) hostname.
Everything else works as expected except this false error message. | 2021-01-25T20:03:51 | 2021-01-25T20:03:51 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 767,121,029 | 19,532 | alexey-milovidov | > There are other issues with float, seems that lcm/gcd should not be allowed for floats, objections?
No objections. | 2021-01-25T21:22:16 | 2021-01-25T21:22:16 | {} | MEMBER |
ClickHouse | ClickHouse | 767,170,251 | 18,762 | sundy-li | Failed tests are not related, ready to review. cc @alexey-milovidov @qoega | 2021-01-25T23:05:56 | 2021-01-25T23:05:56 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 767,361,920 | 11,643 | ghost | I closed this PR due to #18222. | 2021-01-26T07:38:20 | 2021-01-26T07:38:28 | {} | NONE |
ClickHouse | ClickHouse | 767,501,189 | 19,616 | filimonov | And what is the issue? ALTER UPDATE can take significant time if you have a significant amount of data. It really iterates through parts unpack them, check conditions, apply the update if there are matching rows, and pack the modified columns back.
In some cases ALTER UPDATE may be locked by replication queue / mer... | 2021-01-26T12:09:05 | 2021-01-26T12:09:05 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 766,850,194 | 17,803 | filimonov | Like the idea. Few concerns here.
1. management of embedded certificates can be more painful (imagine you need to update root certs on 100 clickhouse servers, `apt-get install` is much simpler than pushing certs somehow in).
2. from the other hand installing some clickhouse-specific ca certs globally in system can a... | 2021-01-25T14:23:39 | 2021-01-25T14:25:52 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 766,882,714 | 16,438 | filimonov | Also - respect system_metrics_events_show_zero_values (or introduce new setting). | 2021-01-25T15:09:25 | 2021-01-25T15:09:37 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 767,075,686 | 16,438 | azat | >Also - respect system_metrics_events_show_zero_values (or introduce new setting).
@filimonov There is no `system_metrics_events_show_zero_values`, but it does respect the `system_events_show_zero_values` | 2021-01-25T19:58:28 | 2021-01-25T19:58:28 | {} | MEMBER |
ClickHouse | ClickHouse | 767,110,411 | 19,329 | alexey-milovidov | Duplicate #19579. | 2021-01-25T21:02:37 | 2021-01-25T21:02:37 | {} | MEMBER |
ClickHouse | ClickHouse | 767,116,086 | 10,935 | alexey-milovidov | This is the expected behaviour. When client connects to ClickHouse and unexpectedly disconnects, the server correctly informs that the `SSL connection unexpectedly closed`. | 2021-01-25T21:12:47 | 2021-01-25T21:12:47 | {} | MEMBER |
ClickHouse | ClickHouse | 767,164,143 | 10,935 | filimonov | Maybe we just need to decrease error level to information? | 2021-01-25T22:50:42 | 2021-01-25T22:50:42 | {
"+1": 2
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 767,318,980 | 19,584 | azat | >AST fuzzer (ASan) — Received signal -3
https://clickhouse-test-reports.s3.yandex.net/19584/5f494c898af82cd46d68411696fa2d2ca0ef6184/fuzzer_asan/report.html#fail1
```
==86==WARNING: failed to fork (errno 12)
==86==WARNING: failed to fork (errno 12)
...
SUMMARY: AddressSanitizer: out-of-memory (/workspace/clic... | 2021-01-26T06:08:41 | 2021-01-26T06:08:41 | {} | MEMBER |
ClickHouse | ClickHouse | 767,107,229 | 19,411 | alexey-milovidov | Duplicate #19362. | 2021-01-25T20:56:30 | 2021-01-25T20:56:30 | {} | MEMBER |
ClickHouse | ClickHouse | 767,119,528 | 19,511 | alexey-milovidov | I will submit your patch. | 2021-01-25T21:19:29 | 2021-01-25T21:19:29 | {} | MEMBER |
ClickHouse | ClickHouse | 767,206,740 | 19,030 | sliontc | > @sliontc you should change the password. Please, be careful when posting screenshots, logs, configs and other stuff, which may contain sensitive data.
OMG, could you please delete the image? I already delete my post. | 2021-01-26T00:47:47 | 2021-01-26T00:47:47 | {} | NONE |
ClickHouse | ClickHouse | 767,288,044 | 19,595 | den-crane | max_bytes_before_external_group_by
max_bytes_before_external_sort
https://clickhouse.tech/docs/en/operations/settings/query-complexity/#settings-max_bytes_before_external_group_by
https://clickhouse.tech/docs/en/sql-reference/statements/select/order-by/#implementation-details | 2021-01-26T04:26:36 | 2021-01-26T04:26:36 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 767,452,247 | 19,611 | sundy-li | You can try #18788 #9273 | 2021-01-26T10:30:55 | 2021-01-26T10:30:55 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 767,477,922 | 19,591 | ka1bi4 | > I'm sorry - haven't noticed, it's a draft. Fill free to discard my changes.
Thanks for your comments, I'll approve it. | 2021-01-26T11:20:16 | 2021-01-26T11:21:12 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 766,888,135 | 19,545 | alexey-milovidov | Moreover, the default settings specified in config.xml are applied and saved in created tables.
Chaning defaults then will not change table settings - you have to do ALTER TABLE ... MODIFY SETTING. | 2021-01-25T15:17:15 | 2021-01-25T15:17:15 | {
"+1": 2
} | MEMBER |
ClickHouse | ClickHouse | 767,071,158 | 19,532 | azat | >Fuzzer fail is related
Not related, but yes this is just one more UB case, will address it too | 2021-01-25T19:50:11 | 2021-01-25T19:50:11 | {} | MEMBER |
ClickHouse | ClickHouse | 767,078,303 | 19,580 | alesapin | Wow! Some tests passed! | 2021-01-25T20:02:50 | 2021-01-25T20:02:50 | {} | MEMBER |
ClickHouse | ClickHouse | 767,126,291 | 18,684 | alexey-milovidov | Let's merge / rebase with master to rerun tests. | 2021-01-25T21:31:46 | 2021-01-25T21:31:46 | {} | MEMBER |
ClickHouse | ClickHouse | 767,425,969 | 19,079 | kitaisreal | Fixed in https://github.com/ClickHouse/ClickHouse/pull/19533. | 2021-01-26T09:44:36 | 2021-01-26T09:44:36 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 766,942,410 | 14,567 | tavplubix | It's because `IPv6` data type implemented as `FixedString(16)` with custom serialization ([link](https://github.com/ClickHouse/ClickHouse/blob/a6f787faeac4ecb1cb2eafa63ea38f0b3734ce0e/src/DataTypes/DataTypeCustomIPv4AndIPv6.cpp#L101)), so `equals` function compares binary representation of `toIPv6('::1')` as `FixedStri... | 2021-01-25T16:34:24 | 2021-01-25T16:34:24 | {} | MEMBER |
ClickHouse | ClickHouse | 767,118,747 | 19,567 | alexey-milovidov | Missing value is not supported in VALUES.
Both queries are invalid.
The query `insert into t values(,'b');` succeeded despite being incorrect due to implementation specific behaviour. | 2021-01-25T21:18:03 | 2021-01-25T21:18:03 | {} | MEMBER |
ClickHouse | ClickHouse | 767,350,881 | 19,596 | kitaisreal | Thanks for the information. Fixed in https://github.com/ClickHouse/ClickHouse/pull/19601. | 2021-01-26T07:12:17 | 2021-01-26T07:12:17 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 767,463,681 | 18,559 | tnaumova-ebay | Hi, we're using:
`version 20.9.3.45 (official build).`
And having similar issue:
`2021.01.26 10:48:12.315374 [ 23422 ] {} <Error> XXXX.XXXX: DB::StorageReplicatedMergeTree::queueTask()::<lambda(DB::StorageReplicatedMergeTree::LogEntryPtr&)>: Code: 49, e.displayText() = DB::Exception: Part 202101-32_338783_338792_... | 2021-01-26T10:52:32 | 2021-01-26T10:52:32 | {} | NONE |
ClickHouse | ClickHouse | 767,475,432 | 19,602 | filimonov | It's actually quite questionable sql extenstion. No determenism at all. May be you can rewrite the query using [any](https://clickhouse.tech/docs/en/sql-reference/aggregate-functions/reference/any/) ? | 2021-01-26T11:15:04 | 2021-01-26T11:16:44 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 766,773,405 | 19,387 | filimonov | Please add also:
toMonth => MONTH
toYear => YEAR
to close https://github.com/ClickHouse/ClickHouse/issues/19557 | 2021-01-25T12:11:26 | 2021-01-25T12:11:26 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 767,105,390 | 19,521 | adevyatova | > Atomic database engine is production grade and is used by default.
> Please remove the mentions that it's experimental from the docs.
Fix it. Atomic database engine will be described in another PR. | 2021-01-25T20:53:02 | 2021-01-25T20:53:02 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 767,109,690 | 19,359 | alexey-milovidov | Duplicate #19207. | 2021-01-25T21:01:16 | 2021-01-25T21:01:16 | {} | MEMBER |
ClickHouse | ClickHouse | 767,306,334 | 19,532 | azat | >AST fuzzer (UBSan) — OK
Finally (d35ff7d553cf99b5acd4c2e88f17633cad9f351c - https://clickhouse-test-reports.s3.yandex.net/19532/d35ff7d553cf99b5acd4c2e88f17633cad9f351c/fuzzer_ubsan/report.html)
*But I need to add missing `#pragma once` ...* | 2021-01-26T05:32:15 | 2021-01-26T05:34:40 | {} | MEMBER |
ClickHouse | ClickHouse | 767,463,107 | 19,611 | ahutsunshine | @sundy-li Can you share me the clickhouse version where I can use mapKeys or mapValues? I use the latest clickhouse version https://github.com/ClickHouse/ClickHouse/releases/tag/v21.1.2.15-stable to build the clickhouse server, but got `Unknown function mapKeys: While processing mapKeys(a)` when using it. | 2021-01-26T10:51:21 | 2021-01-26T10:51:21 | {} | NONE |
ClickHouse | ClickHouse | 767,485,555 | 19,523 | abyss7 | I'd suggest to make original error message more verbose - so it will mention that the parser treats `FROM` like a function. Instead of parser hack. | 2021-01-26T11:35:49 | 2021-01-26T11:35:49 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 766,691,706 | 18,762 | qoega | Looks great! But later we should not forget to add backward incompatible change to changelog if we change default behavior for this tables. Am I right that old behavior without enabled new flag will no longer br checked in tests after this PR? | 2021-01-25T09:47:47 | 2021-01-25T09:47:47 | {} | MEMBER |
ClickHouse | ClickHouse | 766,788,140 | 17,211 | vitlibar | The problem with AST fuzzer has been already fixed (https://github.com/ClickHouse/ClickHouse/issues/19287) | 2021-01-25T12:39:02 | 2021-01-25T12:39:02 | {} | MEMBER |
ClickHouse | ClickHouse | 766,854,766 | 18,740 | den-crane | ah, I see "Documentation (changelog entry is not required)"
| 2021-01-25T14:30:25 | 2021-01-25T14:30:25 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 766,873,435 | 19,578 | alexey-milovidov | @vdimir Duplicate? | 2021-01-25T14:57:04 | 2021-01-25T14:57:04 | {} | MEMBER |
ClickHouse | ClickHouse | 766,980,621 | 19,557 | sv3t0sl4v | Will try! Thanks! | 2021-01-25T17:29:11 | 2021-01-25T17:29:11 | {} | NONE |
ClickHouse | ClickHouse | 767,114,018 | 19,251 | alexey-milovidov | Fixed by @azat | 2021-01-25T21:08:52 | 2021-01-25T21:08:52 | {} | MEMBER |
ClickHouse | ClickHouse | 767,120,237 | 19,532 | azat | >../src/Functions/lcm.cpp:56:77: runtime error: nan is outside the range of representable values of type 'long'
There are other issues with float, seems that lcm/gcd should not be allowed for floats, objections? | 2021-01-25T21:20:53 | 2021-01-25T21:20:53 | {} | MEMBER |
ClickHouse | ClickHouse | 767,148,991 | 19,451 | azat | First part (do not ignore fatal write errors from destructors) - done | 2021-01-25T22:15:38 | 2021-01-25T22:15:38 | {} | MEMBER |
ClickHouse | ClickHouse | 767,471,459 | 19,591 | abyss7 | I'm sorry - haven't noticed, it's a draft. Fill free to discard my changes. | 2021-01-26T11:07:07 | 2021-01-26T11:07:07 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 767,503,031 | 19,616 | filimonov | Or you mean that ALTER command inself (which normally is async and instant) was waiting for something? Most probably some concurrent operation was holding the lock. Consider update to 20.8 or newer ClickHouse versions, most of the locks related to alters were removed in 20.6. | 2021-01-26T12:12:23 | 2021-01-26T12:13:00 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 764,664,077 | 18,672 | vdimir | > Does this make sense?
Got it, thanks.
Ok, I leave this PR as is and will try to do in such way. We will compare implementations, or if I'll face with some problems of such approach and we will discuss it deeper.
| 2021-01-21T14:04:03 | 2021-01-21T14:04:03 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 765,306,911 | 19,413 | akuzm | https://clickhouse-test-reports.s3.yandex.net/18446/82e7e7d9cbbb6541016ef64852c9858cc3c7e4c4/fuzzer_msan/report.html#fail1 | 2021-01-22T10:26:05 | 2021-01-22T10:26:05 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 765,593,536 | 19,396 | alexey-milovidov | It's 10G or 10Gi, not 10GB.
That's reasonable, because it's applicable for any numbers. | 2021-01-22T18:08:54 | 2021-01-22T18:08:54 | {} | MEMBER |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.