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 | 852,011,621 | 24,539 | ildus | @Avogar it's possible if we remove some features of it. The first - `sumMap` supports the input like `[keys array], [values array 1], [values array 2]`, and `[values array 1]` and `[values array 2]` can have different types which requires different nested functions too. But it's not documented, and maybe this feature c... | 2021-06-01T10:23:34 | 2021-06-01T10:23:34 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 852,012,197 | 24,603 | qingping209 | > 2021.06.01 17:16:48.860736 [ 1263773 ] {} system.metric_log (8a648132-f1ff-4e97-b40a-76b0e5b510ca) (MergerMutator): Selected 8 parts from 202106_1_22_2 to 202106_29_29_0
I believe it's the same issue. you can follow the instruction in the source issue to check some internal states of clickhouse process. | 2021-06-01T10:24:39 | 2021-06-01T10:24:39 | {} | NONE |
ClickHouse | ClickHouse | 851,470,868 | 24,763 | alexey-milovidov | @filimonov We've already done that. | 2021-05-31T12:53:00 | 2021-05-31T12:53:00 | {} | MEMBER |
ClickHouse | ClickHouse | 851,976,147 | 24,805 | tavplubix | It fails for different reason:
```
$ zgrep -Fa "CREATE TABLE test_9otca3.alter_compression_codec1" clickhouse-server1.log.gz24805
2021.05.31 19:12:05.110716 [ 89546 ] {} <Debug> DDLWorker(test_9otca3): Executing query: CREATE TABLE test_9otca3.alter_compression_codec1 UUID 'ab2c1f37-bf7b-417e-ab2c-1f37bf7b717e' (`so... | 2021-06-01T09:26:56 | 2021-06-01T09:26:56 | {} | MEMBER |
ClickHouse | ClickHouse | 851,994,341 | 23,983 | alexey-milovidov | Stale. | 2021-06-01T09:55:27 | 2021-06-01T09:55:27 | {} | MEMBER |
ClickHouse | ClickHouse | 851,352,207 | 24,774 | alesapin | ```
[12:20:25]alesap@click:~/debugtoo/debuggggeee/test_keeper_multinode_blocade_leader$ ag 'Fatal'
_instances/node3/logs/clickhouse-server.err.log
57:2021.05.31 00:51:51.824883 [ 11 ] {} <Fatal> BaseDaemon: (version 21.6.2.6998, build id: 9D64302B672743E5EA3491B26CB84DF323AF7153) (from thread 94) Terminate called fo... | 2021-05-31T09:20:57 | 2021-05-31T09:20:57 | {} | MEMBER |
ClickHouse | ClickHouse | 851,588,362 | 24,763 | filimonov | > @filimonov We've already done that.
indeed :+1:
missed that (https://github.com/ClickHouse/ClickHouse/pull/11665) | 2021-05-31T16:30:48 | 2021-05-31T16:32:18 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 851,653,931 | 23,656 | alexey-milovidov | @mergifyio update | 2021-05-31T19:54:07 | 2021-05-31T19:54:07 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 851,662,899 | 24,805 | alesapin | Suspicious:
```
2021-05-31 19:12:35 00910_zookeeper_test_alter_compression_codecs_long: [ FAIL ] 31.20 sec. - return code 159
2021-05-31 19:12:35 Received exception from server (version 21.7.1):
2021-05-31 19:12:35 Code: 159. DB::Exception: Received from localhost:9000. DB::Exception: Watching t... | 2021-05-31T20:03:38 | 2021-05-31T20:03:38 | {} | MEMBER |
ClickHouse | ClickHouse | 851,958,769 | 24,829 | alexey-milovidov | Maybe https://github.com/ClickHouse/ClickHouse/pull/24833 will be ok. | 2021-06-01T09:01:19 | 2021-06-01T09:01:19 | {} | MEMBER |
ClickHouse | ClickHouse | 851,986,820 | 24,776 | alexey-milovidov | But it's not a bugfix. | 2021-06-01T09:43:23 | 2021-06-01T09:43:23 | {} | MEMBER |
ClickHouse | ClickHouse | 851,989,651 | 24,774 | alexey-milovidov | You mean - the whole bugfix was wrong and we definitely should not backport it. | 2021-06-01T09:47:44 | 2021-06-01T09:47:44 | {} | MEMBER |
ClickHouse | ClickHouse | 851,350,787 | 24,774 | alesapin | kafka failures unrelated to changes | 2021-05-31T09:18:25 | 2021-05-31T09:18:25 | {} | MEMBER |
ClickHouse | ClickHouse | 851,471,757 | 24,784 | alexey-milovidov | The task is to make workaround in CI tests. | 2021-05-31T12:54:34 | 2021-05-31T12:54:34 | {} | MEMBER |
ClickHouse | ClickHouse | 851,573,311 | 22,217 | shacuros | @kssenii According to the changelog it should already work with the mysql table function, as the example was given in the change log:
"Added support for replicas and shards in MySQL/PostgreSQL table engine / table function. You can write SELECT * FROM mysql('host{1,2}-{1|2}', ...). Closes #20969. #22217 (Kseniia Sum... | 2021-05-31T15:55:35 | 2021-05-31T15:55:35 | {} | NONE |
ClickHouse | ClickHouse | 851,621,407 | 24,311 | nikitamikhaylov | @Mergifyio update
| 2021-05-31T18:07:25 | 2021-05-31T18:07:25 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 851,635,672 | 24,675 | alexey-milovidov | Marginal feature. | 2021-05-31T18:52:32 | 2021-05-31T18:52:32 | {} | MEMBER |
ClickHouse | ClickHouse | 851,637,762 | 24,643 | alexey-milovidov | Very minor change, not worth to backport. | 2021-05-31T18:59:02 | 2021-05-31T18:59:02 | {} | MEMBER |
ClickHouse | ClickHouse | 851,899,340 | 24,809 | alesapin | `2021.05.31 20:11:32.074243 [ 903528 ] {} <Error> test_27.dst_5 (ReplicatedMergeTreeQueue): Code: 49, e.displayText() = DB::Exception: Part 1_37_40_2 intersects next part 1_38_43_2. It is a bug., Stack trace (when copying this message, always include the lines below):` | 2021-06-01T07:37:02 | 2021-06-01T07:37:02 | {} | MEMBER |
ClickHouse | ClickHouse | 851,350,408 | 24,709 | tavplubix | Need to run integration tests, but runner is broken in 20.8 | 2021-05-31T09:17:45 | 2021-05-31T09:17:45 | {} | MEMBER |
ClickHouse | ClickHouse | 851,549,791 | 24,809 | alesapin | Fix flaky: `00652_replicated_mutations_zookeeper` and `00834_kill_mutation`. | 2021-05-31T15:08:43 | 2021-05-31T15:08:43 | {} | MEMBER |
ClickHouse | ClickHouse | 851,577,515 | 22,217 | kssenii | @shacuros Ok, its just changelog was a little incorrect: replicas were added for both table engine / table function. Shards and replicas combination only as table engine.
But I will add this support for table function also :) | 2021-05-31T16:04:36 | 2021-05-31T16:04:36 | {} | MEMBER |
ClickHouse | ClickHouse | 851,653,834 | 24,783 | den-crane | please share:
```
show create table juba_sessioncount;
show create table ...local..;
``` | 2021-05-31T19:53:48 | 2021-05-31T19:53:48 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 851,895,539 | 24,796 | filimonov | Similar concept - use PROXY protocol for SOME whitelisted ips - in Percona / MariaDB
https://mariadb.com/kb/en/server-system-variables/#proxy_protocol_networks
https://www.percona.com/doc/percona-server/8.0/flexibility/proxy_protocol_support.html
In Apache it's opposite - PROXY enabled by default, and certain IPs... | 2021-06-01T07:30:55 | 2021-06-01T07:30:55 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 851,658,207 | 6,698 | alexey-milovidov | Also need to allow specifying multiple addresses for clickhouse-client for failover. | 2021-05-31T20:01:01 | 2021-05-31T20:01:01 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 851,970,924 | 24,603 | wangzhen11aaa | How to close these stuff. CK only selects and does not do any merge.
2021.06.01 17:16:48.350993 [ 1263789 ] {} <Debug> DiskLocal: Reserving 1.00 MiB on disk `default`, having unreserved 2.97 TiB.
2021.06.01 17:16:48.446263 [ 1263776 ] {} <Debug> system.metric_log (8a648132-f1ff-4e97-b40a-76b0e5b510ca) (MergerMut... | 2021-06-01T09:18:50 | 2021-06-01T09:18:50 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 851,987,692 | 24,545 | alexey-milovidov | It is not a bugfix. Improvement in exception message. Was not required to backport it anywhere. | 2021-06-01T09:44:43 | 2021-06-01T09:45:24 | {} | MEMBER |
ClickHouse | ClickHouse | 851,993,109 | 23,983 | alexey-milovidov | ```
Key name is empty in path style S3 URI: (http://172.17.0.2/bucket/)
```
Double whitespace also has to be fixed. | 2021-06-01T09:53:23 | 2021-06-01T09:54:53 | {} | MEMBER |
ClickHouse | ClickHouse | 851,876,634 | 24,603 | qingping209 | > Hi! The logs look quite weird. Which version do you use? Also, please show the output of `select * from system.settings where changed=1` and `select * from system.merge_tree_settings where changed=1`. If you will be able to reproduce the case then gdb-traces or output of `SELECT arrayStringConcat(arrayMap(x, y -> con... | 2021-06-01T07:00:43 | 2021-06-01T10:24:52 | {} | NONE |
ClickHouse | ClickHouse | 851,638,004 | 24,686 | alexey-milovidov | No that feature in 20.8. | 2021-05-31T18:59:48 | 2021-05-31T18:59:48 | {} | MEMBER |
ClickHouse | ClickHouse | 851,641,702 | 24,539 | Avogar | @ildus How difficult is it to make Map combinator work not only with the Map type, but also with arrays as arguments (like `sumMap`)? | 2021-05-31T19:11:07 | 2021-05-31T19:11:07 | {} | MEMBER |
ClickHouse | ClickHouse | 851,646,446 | 17,623 | UnamedRus | @olafbuitelaar
https://github.com/ClickHouse/ClickHouse/pull/24539 | 2021-05-31T19:27:42 | 2021-05-31T19:27:42 | {
"heart": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 851,648,450 | 22,854 | alexey-milovidov | `toMonday` returns `Date` data type that cannot represent values before 1970-01-01. | 2021-05-31T19:34:41 | 2021-05-31T19:34:41 | {} | MEMBER |
ClickHouse | ClickHouse | 851,675,297 | 22,271 | kssenii | The library used for rabbitmq requires higher rabbitmq version. Not possible to fix. | 2021-05-31T20:40:34 | 2021-05-31T20:40:34 | {} | MEMBER |
ClickHouse | ClickHouse | 851,911,694 | 19,255 | aiwhj | > You mean empty message? I think we have test for that (which is also checked with address sanitizer).
>
> > eof() function thinks that there is a piece of data in current
>
> Why exactly? `avaliable` should be zero, so `hasPandengData` should return false.
hasPandengData returns false, but it also judges the... | 2021-06-01T07:55:23 | 2021-06-01T07:55:23 | {} | NONE |
ClickHouse | ClickHouse | 851,937,581 | 24,829 | alexey-milovidov | See
```
class StackTrace
{
public:
...
static constexpr size_t capacity =
#ifndef NDEBUG
/* The stacks are normally larger in debug version due to less inlining.
*
* NOTE: it cannot be larger then 56 right now, since otherwise it will
* not fit into minimal PIP... | 2021-06-01T08:30:59 | 2021-06-01T08:30:59 | {} | MEMBER |
ClickHouse | ClickHouse | 852,010,270 | 24,573 | alesapin | We have to reset the throttler when there were not active fetches/sends. Otherwise, it will accumulate a huge wait time and will never throttle. | 2021-06-01T10:21:19 | 2021-06-01T10:21:19 | {} | MEMBER |
ClickHouse | ClickHouse | 851,635,833 | 24,659 | alexey-milovidov | Not a bugfix. | 2021-05-31T18:52:58 | 2021-05-31T18:52:58 | {} | MEMBER |
ClickHouse | ClickHouse | 851,641,621 | 24,520 | alexey-milovidov | Experimental feature. | 2021-05-31T19:10:51 | 2021-05-31T19:10:51 | {} | MEMBER |
ClickHouse | ClickHouse | 851,670,116 | 24,603 | alesapin | Hi! The logs look quite weird. Which version do you use? Also, please show the output of `select * from system.settings where changed=1` and `select * from system.merge_tree_settings where changed=1`. If you will be able to reproduce the case then gdb-traces or output of `SELECT arrayStringConcat(arrayMap(x, y -> conca... | 2021-05-31T20:21:41 | 2021-05-31T20:21:41 | {} | MEMBER |
ClickHouse | ClickHouse | 851,740,920 | 24,783 | ljwps | > please share:
>
> ```
> show create table juba_sessioncount;
> show create table ...local..;
> ```
Thank you for your reply.
I found that if this SQL "**CREATE TABLE IF NOT EXISTS juba_sessioncount ON CLUSTER perftest_3shards_1replicas AS juba_sessioncount_replica ENGINE = Distributed(perftest_3shards_1repl... | 2021-06-01T01:20:00 | 2021-06-01T01:20:00 | {} | NONE |
ClickHouse | ClickHouse | 851,930,146 | 23,934 | hexiaoting | @kitaisreal I need your help. Can you give me some advices when you are free, Thank you. | 2021-06-01T08:20:32 | 2021-06-01T08:21:01 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 851,473,818 | 17,623 | alexey-milovidov | @AlexanderChiuluvB this task is on hold, we are waiting for some update from @bakey https://github.com/ClickHouse/ClickHouse/issues/21813#issuecomment-801214788 | 2021-05-31T12:58:01 | 2021-05-31T12:58:01 | {} | MEMBER |
ClickHouse | ClickHouse | 851,643,324 | 24,234 | alexey-milovidov | It's ok to not have this improvement in version 21.5. | 2021-05-31T19:16:34 | 2021-05-31T19:16:34 | {} | MEMBER |
ClickHouse | ClickHouse | 851,660,918 | 5,188 | alexey-milovidov | @Avogar After implementation of hedged requests we can make some twiddling around default timeouts.
E.g. make default timeout larger, like 1000 ms but start doing speculative connections to replicas after 50 ms. | 2021-05-31T20:02:31 | 2021-05-31T20:02:31 | {} | MEMBER |
ClickHouse | ClickHouse | 851,987,262 | 24,776 | alexey-milovidov | But it was merged into other releases by mistake. | 2021-06-01T09:44:04 | 2021-06-01T09:44:04 | {} | MEMBER |
ClickHouse | ClickHouse | 851,370,021 | 24,791 | kokizzu | nevermind, i wrote the answer on stackoverflow above | 2021-05-31T09:49:35 | 2021-05-31T09:49:45 | {} | NONE |
ClickHouse | ClickHouse | 851,475,114 | 24,759 | alexey-milovidov | For old versions, the only way to fix is restart.
I don't remember exact version where it was fixed. The recommended way is to simply install the latest stable version. | 2021-05-31T13:00:14 | 2021-05-31T13:00:14 | {} | MEMBER |
ClickHouse | ClickHouse | 851,530,597 | 22,217 | kssenii | @shacuros you mean trying to create table function externalDistributed? It just was implemented only as table engine. But table function is not difficult to implement also.. | 2021-05-31T14:33:07 | 2021-05-31T14:33:07 | {} | MEMBER |
ClickHouse | ClickHouse | 851,636,059 | 24,658 | alexey-milovidov | Most likely this version does not have partition pruning. | 2021-05-31T18:53:34 | 2021-05-31T18:53:34 | {} | MEMBER |
ClickHouse | ClickHouse | 851,639,907 | 24,719 | alexey-milovidov | Wrong merge. | 2021-05-31T19:05:21 | 2021-05-31T19:05:21 | {} | MEMBER |
ClickHouse | ClickHouse | 851,672,211 | 24,604 | alesapin | Just copypasted both definitions to the local files:
```
$ cat article_local.sql
ATTACH TABLE article_local
(
`I_USER_ID` UInt64,
`I_ARTICLE_ID` UInt64,
`I_MEDIA_TYPE` UInt8 DEFAULT 0,
`I_PROVINCE_ID` UInt64,
`I_DEVICE_TYPE` UInt8,
`I_CHANNEL_ID` UInt8,
`V_VERSION` String, ... | 2021-05-31T20:29:22 | 2021-05-31T20:29:22 | {} | MEMBER |
ClickHouse | ClickHouse | 851,911,679 | 24,812 | kitaisreal | Build fix https://github.com/ClickHouse/ClickHouse/pull/24821.
Integrations tests are flacky.
Functional test 00910_zookeeper_test_alter_compression_codecs_long is flacky.
SQLancer is flacky.
Arcadia is broken because of context refactoring. | 2021-06-01T07:55:21 | 2021-06-01T07:55:21 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 851,965,671 | 24,829 | benbiti | > Maybe #24833 will be ok.
+1.
I will close this。 | 2021-06-01T09:11:09 | 2021-06-01T09:11:09 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 851,993,107 | 24,523 | Algunenano | Hi, does this need any change or improvement? The CI errors seem to be completely unrelated with the changes. | 2021-06-01T09:53:23 | 2021-06-01T09:53:23 | {} | MEMBER |
ClickHouse | ClickHouse | 852,023,310 | 24,832 | akuzm | Can a user still want to limit the number of connections?
Regardless of that, silently ignoring a setting is not good UX, especially if it was set explicitly. Maybe ignoring only the default would be less bad. | 2021-06-01T10:44:33 | 2021-06-01T10:44:33 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 851,484,273 | 9,945 | KanchanaAmarathunga | [Mon May 31 14:16:35 2021] Out of memory: Kill process 9043 (clickhouse-serv) score 929 or sacrifice child
[Mon May 31 14:16:35 2021] Killed process 9043 (clickhouse-serv), UID 399, total-vm:63205848kB, anon-rss:23652388kB, file-rss:0kB, shmem-rss:0kB
I'm currently experiencing this on my single node click house. H... | 2021-05-31T13:14:49 | 2021-05-31T13:14:49 | {
"+1": 2
} | NONE |
ClickHouse | ClickHouse | 851,553,457 | 24,801 | KochetovNicolai | Flacky check - `[ FAIL ] 31.17 sec. - Test runs too long (> 30s). Make it faster.` | 2021-05-31T15:15:28 | 2021-05-31T15:15:28 | {} | MEMBER |
ClickHouse | ClickHouse | 851,635,608 | 24,677 | alexey-milovidov | Minor fix. | 2021-05-31T18:52:18 | 2021-05-31T18:52:18 | {} | MEMBER |
ClickHouse | ClickHouse | 851,637,159 | 24,627 | alexey-milovidov | Mutations in StorageMemory is a marginal feature. | 2021-05-31T18:57:04 | 2021-05-31T18:57:04 | {} | MEMBER |
ClickHouse | ClickHouse | 851,641,602 | 24,521 | alexey-milovidov | Experimental feature. | 2021-05-31T19:10:47 | 2021-05-31T19:10:47 | {} | MEMBER |
ClickHouse | ClickHouse | 851,657,805 | 24,809 | alesapin | `01414_freeze_does_not_prevent_alters` Also have a bug with non replicated merge tree mutations.... | 2021-05-31T20:00:47 | 2021-05-31T20:00:47 | {} | MEMBER |
ClickHouse | ClickHouse | 851,660,116 | 24,803 | alesapin | Database replicated -- not related to changes. | 2021-05-31T20:02:04 | 2021-05-31T20:02:04 | {} | MEMBER |
ClickHouse | ClickHouse | 851,603,121 | 17,623 | olafbuitelaar | having an arrayReduceByKeys, without the need to GROUP BY, or WITH, would be absolutely awsome, like described here; https://github.com/ClickHouse/ClickHouse/issues/11721
. Ideally the key could both be a simple value or a Tuple.
For example;
```SQL
CREATE TABLE aggSample
(
col1 Nested(
prop String,... | 2021-05-31T17:11:40 | 2021-05-31T23:31:35 | {
"+1": 2
} | NONE |
ClickHouse | ClickHouse | 851,940,664 | 24,829 | alexey-milovidov | Proper solution is to revert one of these PRs:
https://github.com/ClickHouse/ClickHouse/pull/23969
https://github.com/ClickHouse/ClickHouse/pull/21020
And not introduce more ifdefs... Ifdefs are clunky. | 2021-06-01T08:35:32 | 2021-06-01T08:35:32 | {} | MEMBER |
ClickHouse | ClickHouse | 851,950,194 | 24,404 | nvartolomei | @alexey-milovidov thoughts? Updated docs and added additional tests. | 2021-06-01T08:48:52 | 2021-06-01T08:48:52 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 851,986,661 | 24,776 | alexey-milovidov | It's very minor but still ok to merge. | 2021-06-01T09:43:08 | 2021-06-01T09:43:08 | {} | MEMBER |
ClickHouse | ClickHouse | 851,995,663 | 24,606 | filimonov | Same for UUID | 2021-06-01T09:57:37 | 2021-06-01T09:57:37 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 851,519,870 | 24,800 | alesapin | ヽ(‘ー`)ノ | 2021-05-31T14:14:57 | 2021-05-31T14:14:57 | {} | MEMBER |
ClickHouse | ClickHouse | 851,526,838 | 22,217 | shacuros | Thanks for the feature, this could simplify some parts in our system.
I tested the functionality with a sharded setup and get an error when trying it as Tablefunction. When using it as Tableengine it works as it should:
```
DESCRIBE TABLE mysql('db-shard-4:3306', 'Datamarts', 'Account_Clicks', 'user***', 'pa... | 2021-05-31T14:26:32 | 2021-05-31T14:27:15 | {} | NONE |
ClickHouse | ClickHouse | 851,532,493 | 24,777 | amosbird | > Replica r1 creates log entry PART_IS_LOST for part all_1_2_1.
How about we generating an empty `all_1_1_0` part when it's lost forever instead of canceling planned merges with `PART_IS_LOST`? | 2021-05-31T14:36:32 | 2021-05-31T14:36:32 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 851,469,842 | 24,792 | alexey-milovidov | The behaviour is correct. Division by zero produces exception for Decimal data type.
| 2021-05-31T12:51:15 | 2021-05-31T12:51:15 | {} | MEMBER |
ClickHouse | ClickHouse | 851,471,525 | 24,784 | alexey-milovidov | It is impossible. SSD cache can run only with O_DIRECT.
All other options are meaningless.
| 2021-05-31T12:54:12 | 2021-05-31T12:54:12 | {} | MEMBER |
ClickHouse | ClickHouse | 851,567,094 | 24,810 | amosbird | > Allow to combine output from projection and non aggregated main table.
I've got this feature request multiple times, however, I still cannot find a general way to solve it. One possible solution would be providing a dedicated MergeTree engine that's made for time series data, with implicit date time column. Then w... | 2021-05-31T15:41:54 | 2021-05-31T15:42:28 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 851,621,455 | 24,311 | mergify[bot] | **Command `update`: success**
> **Branch already up to date**
>
| 2021-05-31T18:07:34 | 2021-05-31T18:07:34 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 851,637,027 | 24,615 | alexey-milovidov | It's relevant for the "smarter partition pruner" that we most likely don't have in version 20.8. | 2021-05-31T18:56:40 | 2021-05-31T18:56:40 | {} | MEMBER |
ClickHouse | ClickHouse | 851,637,929 | 24,640 | alexey-milovidov | This is relevant but better avoid backporting. | 2021-05-31T18:59:31 | 2021-05-31T18:59:31 | {} | MEMBER |
ClickHouse | ClickHouse | 851,666,391 | 24,742 | alesapin | I'll close this and change the label of the source issue. | 2021-05-31T20:09:10 | 2021-05-31T20:09:10 | {} | MEMBER |
ClickHouse | ClickHouse | 851,704,063 | 24,615 | CurtizJ | Most likely in 20.8 bug can be reproduced with `ORDER BY toWeek(pk)` and `optimize_read_in_order = 1`. | 2021-05-31T22:28:35 | 2021-05-31T22:29:02 | {} | MEMBER |
ClickHouse | ClickHouse | 851,771,752 | 24,792 | mbtolou | this query with if case throw exception!
```sql
with FF as (SELECT toDecimal128((number%10)-1,5) TEST_ZERO from numbers(1000)) ,1 as FR
select toDecimal128(100* if(TEST_ZERO==0,0,ROUND(COS(ATAN(FR/TEST_ZERO)),2)),5), TEST_ZERO
from FF
with FF as (SELECT toDecimal128((number%10)-1,5) TEST_ZERO from numbers(... | 2021-06-01T03:06:28 | 2021-06-01T03:06:43 | {} | NONE |
ClickHouse | ClickHouse | 851,899,482 | 24,809 | alesapin | Ok, this fix is useful. | 2021-06-01T07:37:17 | 2021-06-01T07:37:17 | {} | MEMBER |
ClickHouse | ClickHouse | 852,023,204 | 17,623 | melin | Can directly provide functions for computing PSI (Application Station Index), WOE (Weight of Event), and IV (Information Value)? | 2021-06-01T10:44:21 | 2021-06-01T10:44:21 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 851,350,271 | 24,772 | alesapin | No related failures. | 2021-05-31T09:17:30 | 2021-05-31T09:17:30 | {} | MEMBER |
ClickHouse | ClickHouse | 851,525,932 | 24,777 | tavplubix | I realized that solution with special `PART_IS_LOST` log entry does not work if some replica was cloned during execution of such entry:
1. Let replication queue of replica `r1` contains `MERGE_PARTS` entry to merge `all_1_1_0` and `all_2_2_0` into `all_1_2_1` and part `all_1_1_0` is lost forever, so merge cannot be ... | 2021-05-31T14:24:59 | 2021-05-31T14:24:59 | {} | MEMBER |
ClickHouse | ClickHouse | 851,650,229 | 24,132 | alexey-milovidov | @mergifyio update | 2021-05-31T19:40:48 | 2021-05-31T19:40:48 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 851,953,487 | 24,829 | benbiti | decrease 56 to a proper value or just remove #if preconditions just use 32 instead?
@alexey-milovidov | 2021-06-01T08:53:39 | 2021-06-01T08:53:39 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 851,592,578 | 24,458 | sevirov | @traceon Hi. Please, review this PR. | 2021-05-31T16:42:28 | 2021-05-31T16:42:28 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 851,643,291 | 24,233 | alexey-milovidov | It's ok to not have this improvement in version 21.3. | 2021-05-31T19:16:26 | 2021-05-31T19:16:26 | {} | MEMBER |
ClickHouse | ClickHouse | 851,654,007 | 23,656 | mergify[bot] | **Command `update`: success**
> **Branch has been successfully updated**
>
| 2021-05-31T19:54:21 | 2021-05-31T19:54:21 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 851,969,908 | 24,804 | kssenii | I am not sure there is a good solution for `arr_t_int8` data type.
Data types for a table are fetched with command `SELECT attname AS name, format_type(atttypid, atttypmod) AS type, attnotnull AS not_null, attndims AS dims FROM pg_attribute WHERE attrelid = 't1'::regclass AND NOT attisdropped AND attnum > 0; `.
... | 2021-06-01T09:17:20 | 2021-06-01T09:17:20 | {} | MEMBER |
ClickHouse | ClickHouse | 852,028,792 | 24,539 | ildus | In the fuzzer I see mostly errors with invalid arguments, how they can be skipped? | 2021-06-01T10:53:37 | 2021-06-01T10:53:37 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 851,494,642 | 24,789 | alexey-milovidov | It's "performance improvement", not a "new feature". | 2021-05-31T13:32:03 | 2021-05-31T13:32:03 | {} | MEMBER |
ClickHouse | ClickHouse | 851,636,693 | 24,646 | alexey-milovidov | Metrics accounting | 2021-05-31T18:55:36 | 2021-05-31T18:55:36 | {} | MEMBER |
ClickHouse | ClickHouse | 851,650,399 | 24,132 | mergify[bot] | **Command `update`: success**
> **Branch has been successfully updated**
>
| 2021-05-31T19:41:28 | 2021-05-31T19:41:28 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 851,990,753 | 24,721 | alexey-milovidov | It was incorrect - reverted. | 2021-06-01T09:49:29 | 2021-06-01T09:49:29 | {} | MEMBER |
ClickHouse | ClickHouse | 851,999,420 | 2,161 | qingping209 | > Distibuted table over ReplicatedMergeTree tables - is how it works in ClickHouse if you need sharding + replication.
With Scheme, is the shard server aware of a dead shard and be able to find the first healthy replica to send the query ?
| 2021-06-01T10:03:31 | 2021-06-01T10:03:31 | {} | NONE |
ClickHouse | ClickHouse | 845,740,151 | 17,623 | filimonov | One of the topics which always pops up when big users try to configure RBAC - is DEFINER
https://github.com/ClickHouse/ClickHouse/issues/19228 | 2021-05-21T07:52:31 | 2021-05-21T07:52:31 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 845,807,909 | 24,349 | KochetovNicolai | Functional stateless tests (thread, ubsan)
```
2021-05-21 00:13:20 00601_kill_running_query: [ FAIL ] 16.48 sec. - result differs with reference:
2021-05-21 00:13:20 --- /usr/share/clickhouse-test/queries/0_stateless/00601_kill_running_query.reference 2021-05-20 23:04:03... | 2021-05-21T09:11:29 | 2021-05-21T09:12:33 | {} | MEMBER |
ClickHouse | ClickHouse | 845,823,994 | 24,389 | alexey-milovidov | What problem? | 2021-05-21T09:37:51 | 2021-05-21T09:37:51 | {} | MEMBER |
ClickHouse | ClickHouse | 845,663,695 | 24,370 | alexey-milovidov | @taotaizhu-pw Please answer the question:
> Make sure that git diff result is empty and you've just pulled fresh master. Try cleaning up cmake cache. Just in case, official build instructions are published here: https://clickhouse.yandex/docs/en/development/build/ | 2021-05-21T05:27:34 | 2021-05-21T05:27:34 | {} | MEMBER |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.