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 | 740,826,013 | 17,906 | Steuf | Ho nice ! Thanks.
So in this case I can start the project with Ordinary and migrate later when backup will support it.
Thank you it's clear. | 2020-12-08T18:33:47 | 2020-12-08T18:34:05 | {} | NONE |
ClickHouse | ClickHouse | 740,979,202 | 17,707 | azat | Looks like this breaks perf tests (due to port conflict for `tcp_with_proxy_port`), will fix it in #17907 (and you don't see this in the report since clickhouse-server from upstream does not interpret this new setting) | 2020-12-08T20:26:37 | 2020-12-08T20:28:22 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 741,525,434 | 17,858 | amosbird | depends on https://github.com/RoaringBitmap/CRoaring/pull/277 | 2020-12-09T04:51:55 | 2020-12-09T04:51:55 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 742,290,472 | 9,431 | FavonianKong | Hi，Is there any new progress on this proposal? Does the clickhouse support only SSL currently? | 2020-12-10T07:11:22 | 2020-12-10T07:11:22 | {} | NONE |
ClickHouse | ClickHouse | 740,803,770 | 17,748 | azat | >It doesn't install the test config, maybe it should do it.
But is sources available in that container?
If not, is there internet in that container? | 2020-12-08T18:00:24 | 2020-12-08T18:00:24 | {} | MEMBER |
ClickHouse | ClickHouse | 741,720,631 | 17,918 | alesapin | Fast test OK | 2020-12-09T11:46:52 | 2020-12-09T11:46:52 | {} | MEMBER |
ClickHouse | ClickHouse | 741,765,784 | 17,932 | den-crane | You can use arrayMap and lamdba.
For example Prime numbers:
```sql
SELECT arrayFilter( (z,x) -> not arrayExists(i -> x%i=0 and x<>i, r) , (arrayMap(z->z+2, range(20))) as r,r) prime
┌─prime─────────────────┐
│ [2,3,5,7,11,13,17,19] │
└───────────────────────┘
``` | 2020-12-09T13:19:42 | 2020-12-09T13:19:42 | {
"hooray": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 742,133,397 | 17,852 | l1t1 | what is the option of make | 2020-12-09T23:41:38 | 2020-12-09T23:41:38 | {} | NONE |
ClickHouse | ClickHouse | 742,140,317 | 17,748 | azat | @akuzm Thanks! But looks like they haven't continued? some CI issues? | 2020-12-10T00:00:02 | 2020-12-10T00:00:15 | {} | MEMBER |
ClickHouse | ClickHouse | 742,206,788 | 17,633 | hexiaoting | @Akazz I cannot get the error info for the failed check. | 2020-12-10T03:08:47 | 2020-12-10T03:08:47 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 742,207,353 | 17,953 | l1t1 | this fun has been realized.
https://docs.oracle.com/en/database/oracle/oracle-database/21/nfcon/bitwise-aggregate-functions-274057636.html | 2020-12-10T03:10:40 | 2020-12-10T03:10:40 | {} | NONE |
ClickHouse | ClickHouse | 742,229,582 | 17,953 | l1t1 | window functions
https://docs.oracle.com/en/database/oracle/oracle-database/21/nfcon/practice-using-enhanced-analytic-functions-282450990.html | 2020-12-10T04:21:55 | 2020-12-10T04:21:55 | {} | NONE |
ClickHouse | ClickHouse | 742,488,318 | 17,959 | fastio | > Now you need to list all settings which should be readonly.
>
> We want to limit all, but allow changing some of them.
What do you mean is that updating system.settings table by `alter table update ` ? | 2020-12-10T12:22:33 | 2020-12-10T12:22:33 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 742,101,767 | 17,748 | azat | Can someone put force-tests label please?
(test failure reported to #17895) | 2020-12-09T22:24:38 | 2020-12-09T22:24:38 | {} | MEMBER |
ClickHouse | ClickHouse | 741,700,089 | 17,364 | filimonov | > @CurtizJ :
> Well, you can't do much there. If `WHERE` selects a small range of rows, it will be slower with `optimize_read_in_order` than without, because the reading itself is slower. If only somehow tricky to rewrite the pipeline or disable the optimization. | 2020-12-09T11:04:18 | 2020-12-09T11:04:33 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 741,703,475 | 17,918 | KochetovNicolai | It actually does not loose any data. Wrong information is in marks. It is likely to be fixed by merge. | 2020-12-09T11:10:58 | 2020-12-09T11:10:58 | {} | MEMBER |
ClickHouse | ClickHouse | 741,760,860 | 17,932 | l1t1 | code from https://sqlite.org/lang_with.html#recursivecte | 2020-12-09T13:11:58 | 2020-12-09T13:11:58 | {} | NONE |
ClickHouse | ClickHouse | 741,770,545 | 16,954 | KochetovNicolai | Almost done. let's add couple of checks and merge. | 2020-12-09T13:28:19 | 2020-12-09T13:28:19 | {} | MEMBER |
ClickHouse | ClickHouse | 741,799,829 | 17,916 | alexey-milovidov | Theoretically we can make GROUP BY without aggregate functions to produce records as soon as they are found and to work faster with LIMIT. Also we can make DISTINCT to work as fast as GROUP BY.
Now the difference is just an implementation detail. | 2020-12-09T14:18:49 | 2020-12-09T14:18:49 | {} | MEMBER |
ClickHouse | ClickHouse | 741,145,385 | 17,844 | yuxin-jdwt | @den-crane，emmm。。。，when use the clickhouse-client command
what i print will print twice on the console,like this
i find Command Line Options --echo but how can disable it?
$ clickhouse-client
:)ss
3Q | 2020-12-08T22:33:24 | 2020-12-08T22:33:24 | {} | NONE |
ClickHouse | ClickHouse | 741,724,037 | 17,379 | hoverzheng | Thanks @traceon!
I solved it !
| 2020-12-09T11:54:31 | 2020-12-09T11:54:31 | {} | NONE |
ClickHouse | ClickHouse | 742,457,785 | 17,960 | astudnev | Without depth in order by it works just fine:
```
SELECT
depth,
SUM(coinpath.txs_ethereum_outbound_4c4dce9598b8e2addcc99.amount) / dictGet('currency', 'divider', toUInt64(1)) AS `_aq.amount`
FROM coinpath.txs_ethereum_outbound_4c4dce9598b8e2addcc99
WHERE (coinpath.txs_ethereum_outbound_4c4dce9598b8e2ad... | 2020-12-10T11:18:56 | 2020-12-10T11:18:56 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 742,487,137 | 17,916 | alexey-milovidov | @poppinlr No, both are parallel.
But GROUP BY does it better. | 2020-12-10T12:20:06 | 2020-12-10T12:20:11 | {} | MEMBER |
ClickHouse | ClickHouse | 741,708,680 | 17,659 | nautaa | I can pass the test locally, can you help me? @alexey-milovidov | 2020-12-09T11:21:29 | 2020-12-09T11:21:29 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 742,360,007 | 16,621 | ilejn | @alesapin It looks like stress test issues are not relevant, they are replication related.
Stress test was passed before last push.
Should I rebase again or what?
| 2020-12-10T08:16:46 | 2020-12-10T08:16:46 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 742,456,917 | 17,960 | qoega | And what if there is no more `depth` in `ORDER BY`? | 2020-12-10T11:17:09 | 2020-12-10T11:17:09 | {} | MEMBER |
ClickHouse | ClickHouse | 740,775,031 | 17,866 | alexey-milovidov | @dkxiaohei They were unrelated internal checks. Thank you! | 2020-12-08T17:17:03 | 2020-12-08T17:17:03 | {} | MEMBER |
ClickHouse | ClickHouse | 741,489,036 | 1,178 | ChesterHsieh | Thx @daledude. **Final** is necessary for this case. | 2020-12-09T02:41:56 | 2020-12-09T02:41:56 | {} | NONE |
ClickHouse | ClickHouse | 741,693,142 | 8,445 | filimonov | Should be fixed in #14211 (clickhouse versions 20.10 and newer). | 2020-12-09T10:51:34 | 2020-12-09T10:51:34 | {
"rocket": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 741,779,460 | 9,260 | alesapin | @alexey-milovidov's case is not related to this issue. The bug was fixed with revert of #17918. | 2020-12-09T13:43:31 | 2020-12-09T13:43:31 | {} | MEMBER |
ClickHouse | ClickHouse | 741,935,778 | 17,939 | den-crane | Probably your task can be solved without neighbor. What is your real task? Calculate what exactly? in sliding window?
groupArrayMovingSum groupArrayMovingAvg ? | 2020-12-09T17:41:34 | 2020-12-09T17:41:34 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 742,211,824 | 17,952 | allen12921 | try lastest verison first | 2020-12-10T03:23:30 | 2020-12-10T03:23:30 | {} | NONE |
ClickHouse | ClickHouse | 741,244,834 | 17,184 | ygrek | currently failing fast tests are
01200_mutations_memory_consumption (which also fails on master) with
```
01200_mutations_memory_consumption: [ FAIL ] - return code 60
[devbox1] 2020.12.08 23:40:04.368173 [ 117234 ] {8c8e0ab5-dc07-49bf-9f44-095afdc0bfc0} <Error> executeQuery: Co... | 2020-12-08T23:41:49 | 2020-12-08T23:41:49 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 741,596,501 | 17,866 | dkxiaohei | > @dkxiaohei They were unrelated internal checks. Thank you!
Thank you very much! This is my first PR ever. It is indeed encouraging. | 2020-12-09T07:45:11 | 2020-12-09T07:45:11 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 741,614,369 | 17,858 | andyyzh | @amosbird Great work, thank you! | 2020-12-09T08:22:47 | 2020-12-09T08:22:47 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 741,803,803 | 17,917 | amosbird | https://github.com/ClickHouse/ClickHouse/pull/15481
but it was closed because of no use cases. | 2020-12-09T14:25:13 | 2020-12-09T14:25:13 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 742,149,883 | 16,856 | excitoon | Checking that in 30b0c38 | 2020-12-10T00:26:18 | 2020-12-10T00:26:18 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 742,168,139 | 17,917 | alexey-milovidov | Let's continue. We can test for clang version and use the new option if it is clang 12. | 2020-12-10T01:14:49 | 2020-12-10T01:14:49 | {} | MEMBER |
ClickHouse | ClickHouse | 742,329,650 | 17,939 | TomerRZN | I have a table of events, containing the fields `[id, timestamp]`
I'd like to query a specific `id` and get the window of X rows before and Y rows after.
If this it the table:
```
┌───────────────────────────────────id─┬───────────────timestamp─┐
│ 190b52c7-55ee-52ba-b3d2-064247bb9d92 │ 2020-08-24 22:38:17.493 │... | 2020-12-10T07:48:51 | 2020-12-10T07:48:51 | {} | NONE |
ClickHouse | ClickHouse | 742,331,019 | 17,939 | TomerRZN | > SELECT rand() ?
>
> ```
> ┌────number─┬─window──────────────────────────────────────────────────────────────────────────────────────────────────────────┐
> │ 2170931 │ [0,0,0,2170931,2170931,2170931,2170931,2170931,17492443,17492443,17492443] │
> │ 17492443 │ [2170931,217... | 2020-12-10T07:49:56 | 2020-12-10T07:49:56 | {} | NONE |
ClickHouse | ClickHouse | 740,775,299 | 17,849 | den-crane | @rbeeli No. I am not a member of CH developing team. I am just a ch community member the same as you.
Yes, it would be nice to see PR with a documentation. | 2020-12-08T17:17:25 | 2020-12-08T17:17:25 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 740,821,279 | 17,906 | Steuf | Ok I see, thanks. It's more flexible for some actions on cluster.
Last question, it's possible to convert a Ordinary database to Atomic ? If not it will ne possible later ? | 2020-12-08T18:26:26 | 2020-12-08T18:26:26 | {} | NONE |
ClickHouse | ClickHouse | 741,937,441 | 17,937 | den-crane | duplicate of https://github.com/ClickHouse/ClickHouse/issues/7666 ? | 2020-12-09T17:42:39 | 2020-12-09T17:42:39 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 742,325,372 | 16,877 | alesapin |
Integration test failure is not related to changes. Tried tests multiple times and every time everything is OK.
I'm going to merge this PR without additional refactoring and comments/documentation. I think this code will change significantly after adding the RAFT implementation. | 2020-12-10T07:45:22 | 2020-12-10T07:45:22 | {} | MEMBER |
ClickHouse | ClickHouse | 740,801,507 | 17,748 | akuzm | > No it's not, missing TLD list in perf tests:
It's here:
https://github.com/ClickHouse/ClickHouse/blob/master/docker/test/performance-comparison/compare.sh#L36
It doesn't install the test config, maybe it should do it. | 2020-12-08T17:57:07 | 2020-12-08T17:57:07 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 741,618,341 | 17,623 | filimonov | Update Arrow, add support for lists in Parquet. https://github.com/ClickHouse/ClickHouse/issues/7704#issuecomment-741616424
Attempt to create zero-copy ORC / Parquet (for non seekable sources) - https://github.com/ClickHouse/ClickHouse/issues/10461#issuecomment-623006030 https://github.com/ClickHouse/ClickHouse/iss... | 2020-12-09T08:30:47 | 2020-12-09T09:01:51 | {
"+1": 6
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 741,757,319 | 14,042 | sliontc | > @sliontc service start clickhouse-server
>
> Version 18.14 is out of support.
Could you help in remote control?
I'v tried service clickhouse-server start, but still can't connect. It makes me crazy . | 2020-12-09T13:04:51 | 2020-12-09T13:04:51 | {} | NONE |
ClickHouse | ClickHouse | 741,846,304 | 9,260 | alesapin | We discussed in tg and it appeared that the problematic table was created on 20.12 version. Downgrade to 20.11 didn't help, so @alex-zaitsev reported that the error reproduces on 20.11 too. To fix the error just `optimize final` table on any version except 20.12.2. | 2020-12-09T15:29:58 | 2020-12-09T15:29:58 | {} | MEMBER |
ClickHouse | ClickHouse | 741,847,886 | 9,452 | vincentail | i got the same question, it same like unmatch fiedtype(Uint) with the value type(TRUE),you must translate boolean type to UInt8 | 2020-12-09T15:32:07 | 2020-12-09T15:32:07 | {} | NONE |
ClickHouse | ClickHouse | 742,174,547 | 16,856 | excitoon | Same test failed in #17934 -- it flaps | 2020-12-10T01:33:29 | 2020-12-10T07:54:02 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 742,566,779 | 17,957 | alesapin | This exception says that the replica has unprocessed alters. If the error doesn't disappear, then provide server logs table schema and query. | 2020-12-10T14:47:45 | 2020-12-10T14:47:45 | {} | MEMBER |
ClickHouse | ClickHouse | 742,567,523 | 17,962 | den-crane | And free space? `df -h /var/lib` ? | 2020-12-10T14:49:01 | 2020-12-10T14:49:01 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 742,574,396 | 17,960 | den-crane | What if ORDER BY identity( depth ) ASC, `_aq.amount` DESC
What if ORDER BY cast( depth as String ) ASC, `_aq.amount` DESC | 2020-12-10T15:00:30 | 2020-12-10T15:00:30 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 740,969,638 | 16,721 | nikitamikhaylov | > @nikitamikhaylov you forget about `max_memory_usage_for_user`, or that was the intention?
Totally forgot, will fix. | 2020-12-08T20:18:40 | 2020-12-08T20:18:40 | {} | MEMBER |
ClickHouse | ClickHouse | 741,750,395 | 14,042 | sliontc | @den-crane Hi, I got same error, how to fix it? My CK suddenly crashed after a SQL query with the error "AST is too big...". Then I couldn't connect the CK. It always output
| 2020-12-09T12:50:58 | 2020-12-09T12:50:58 | {} | NONE |
ClickHouse | ClickHouse | 741,769,310 | 2,797 | lgogolin | Any update on that ? | 2020-12-09T13:26:10 | 2020-12-09T13:26:10 | {} | NONE |
ClickHouse | ClickHouse | 741,956,003 | 17,364 | UnamedRus | But if we don't have LIMIT in the query, do we actually gain any benefit from optimize_read_in_order then? | 2020-12-09T18:14:20 | 2020-12-09T18:17:17 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 742,283,946 | 17,916 | poppinlr | is it because distinct is a serial execution and group by is a parallel execution?
if so, can distinct be parallel possible? | 2020-12-10T06:56:28 | 2020-12-10T06:56:28 | {} | NONE |
ClickHouse | ClickHouse | 740,808,624 | 17,748 | azat | Ok, I see it `COPY * /`, so I guess I can put a symlink to this file into the `docker/test/performance-comparison`
https://github.com/ClickHouse/ClickHouse/blob/f6f336ac7839ed52a147eb0bcae8b88f0548082a/docker/test/performance-comparison/Dockerfile#L40 | 2020-12-08T18:07:29 | 2020-12-08T18:07:29 | {} | MEMBER |
ClickHouse | ClickHouse | 741,566,795 | 17,850 | JKeyZ | thank u for help | 2020-12-09T06:37:53 | 2020-12-09T06:37:53 | {} | NONE |
ClickHouse | ClickHouse | 741,586,216 | 17,890 | zhang2014 | We will synchronize the full data when the database is first create or not found binlog file in MySQL after restart. | 2020-12-09T07:21:21 | 2020-12-09T07:21:21 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 741,692,849 | 6,104 | filimonov | Fixed in #14211 (clickhouse versions 20.10 and newer). | 2020-12-09T10:50:58 | 2020-12-09T10:50:58 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 741,750,952 | 14,042 | sliontc | > Thanks for reply. I've solved it.
Hi, How did you solve it? | 2020-12-09T12:52:13 | 2020-12-09T12:52:13 | {} | NONE |
ClickHouse | ClickHouse | 741,767,844 | 17,916 | den-crane | Distinct and Groupby use different algorithms.
Distinct produces the first row faster. For example compare the same queries with LIMIT 1. | 2020-12-09T13:23:29 | 2020-12-09T13:23:29 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 742,086,716 | 17,916 | l1t1 | if there is no limit and no aggregate functions, can distinct query change to choose the faster algorithm? | 2020-12-09T21:54:06 | 2020-12-09T21:55:19 | {} | NONE |
ClickHouse | ClickHouse | 742,375,518 | 16,621 | alesapin | > @alesapin It looks like stress test issues are not relevant, they are replication related.
> Stress test was passed before last push.
> Should I rebase again or what?
No, probably unrelated failure | 2020-12-10T08:48:42 | 2020-12-10T08:48:42 | {} | MEMBER |
ClickHouse | ClickHouse | 742,453,188 | 17,960 | astudnev | Yes, it is the same with HAVING depth=1 and without currency_id:
```
SELECT
depth,
SUM(coinpath.txs_ethereum_outbound_4c4dce9598b8e2addcc99.amount) / dictGet('currency', 'divider', toUInt64(1)) AS `_aq.amount`
FROM coinpath.txs_ethereum_outbound_4c4dce9598b8e2addcc99
WHERE (coinpath.txs_ethereum_outbou... | 2020-12-10T11:09:30 | 2020-12-10T11:09:30 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 742,473,935 | 17,932 | l1t1 | thanks | 2020-12-10T11:52:21 | 2020-12-10T11:52:21 | {} | NONE |
ClickHouse | ClickHouse | 742,486,118 | 17,959 | alexey-milovidov | I also need this feature :) | 2020-12-10T12:18:02 | 2020-12-10T12:18:02 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 741,613,507 | 16,158 | bgranvea | nice! the first limitations that I notice:
- cannot select a default database (USE xxx => no session error, if I change URL to http://localhost:8123?database=xxx, I get an error)
- only one statement.... | 2020-12-09T08:21:04 | 2020-12-09T08:21:04 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 741,924,356 | 17,939 | den-crane | SELECT rand() ?
```
┌────number─┬─window──────────────────────────────────────────────────────────────────────────────────────────────────────────┐
│ 2170931 │ [0,0,0,2170931,2170931,2170931,2170931,2170931,17492443,17492443,17492443] │
│ 17492443 │ [2170931,2170931,2170931... | 2020-12-09T17:29:31 | 2020-12-09T17:29:31 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 742,101,261 | 17,895 | azat | @Enmk test fails - https://clickhouse-test-reports.s3.yandex.net/17748/713211943d61a7f0475f9893a826effdc7ebcd51/fast_test.html#fail1
And this looks like a bug:
```sql
SELECT
dt64,
d,
dt64 < d
FROM
(
SELECT
number - 1 AS n,
toDateTime64(toStartOfInterval(now(), toIntervalS... | 2020-12-09T22:23:29 | 2020-12-09T22:29:45 | {} | MEMBER |
ClickHouse | ClickHouse | 742,130,843 | 17,934 | excitoon | lgtm | 2020-12-09T23:34:20 | 2020-12-09T23:34:20 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 742,451,560 | 17,960 | qoega | Does it still reproduce if you filter rows by `HAVING depth=1`? And you can remove `currency_id` column - it should be irrelevant. | 2020-12-10T11:06:13 | 2020-12-10T11:06:13 | {} | MEMBER |
ClickHouse | ClickHouse | 741,747,340 | 5,225 | sliontc | Got same issue, how to solve it ? @lzy305 | 2020-12-09T12:44:22 | 2020-12-09T12:44:22 | {} | NONE |
ClickHouse | ClickHouse | 725,445,521 | 16,877 | alesapin | TODO:
- [x] Correct settings processing from configs (timeouts and so on)
- [x] Fix exception handling (do not use LOGICAL_ERROR in all places)
- [ ] Make TestKeeperStorage a backend for TestKeeper (remove a bunch of duplicated code)
- [x] Check XID and ZXID for watches, now it works well with XID = ZXID =-1. -- ... | 2020-11-11T14:12:48 | 2020-12-09T15:36:30 | {} | MEMBER |
ClickHouse | ClickHouse | 741,867,353 | 16,935 | Avogar | Also I think it would be better to write about the changes in PR description. | 2020-12-09T15:58:33 | 2020-12-09T15:58:33 | {} | MEMBER |
ClickHouse | ClickHouse | 741,944,737 | 17,937 | UnamedRus | I would say it's related, right, but it doesn't has GROUP BY in it.
Issue would reproduce only if we have now64() in the query, if we remove it from the query, it would produce the right results.
| 2020-12-09T17:54:14 | 2020-12-09T17:54:14 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 741,947,035 | 17,875 | UnamedRus | > localhost :) select groupUniqArray( ['world','hello','hello']);
That expected result, it would choose unique arrays being passed to it.
Right way of doing that query is:
```
select groupUniqArrayArray( ['world','hello','hello']);
SELECT groupUniqArrayArray(['world', 'hello', 'hello'])
┌─groupUniqArrayAr... | 2020-12-09T17:58:17 | 2020-12-09T17:58:17 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 742,045,606 | 15,323 | alexey-milovidov | Fixed: https://news.ycombinator.com/item?id=25364000 | 2020-12-09T20:59:06 | 2020-12-09T20:59:06 | {} | MEMBER |
ClickHouse | ClickHouse | 742,068,169 | 17,750 | nikitamikhaylov | Merged in other PR. | 2020-12-09T21:20:45 | 2020-12-09T21:20:45 | {
"+1": 2
} | MEMBER |
ClickHouse | ClickHouse | 742,138,211 | 17,852 | alexey-milovidov | @l1t1 It is done after build with a separate command (strip or objcopy).
If you want to get the same result during build, you can compile without debug info (Release instead of RelWithDebInfo mode in CMake). | 2020-12-09T23:53:31 | 2020-12-09T23:53:31 | {} | MEMBER |
ClickHouse | ClickHouse | 742,168,940 | 17,916 | alexey-milovidov | @l1t1 Yes, faster algorithm is possible and easy to implement. | 2020-12-10T01:17:04 | 2020-12-10T01:17:04 | {} | MEMBER |
ClickHouse | ClickHouse | 740,821,307 | 17,906 | den-crane | AFAIK you already can move tables to atomic database using rename table. But later will be implemented a special method to convert database;
```sql
create database atom Engine=Atomic;
create database ord Engine=Ordinary;
create table ord.A (A Int64) Engine=MergeTree order by A;
insert into ord.A select number fr... | 2020-12-08T18:26:29 | 2020-12-08T18:26:29 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 741,779,689 | 9,260 | alesapin | > I've got the same problem during mutation in 20.11.5.18. There were no skipping indices involved.
Need a data sample or at least logs. | 2020-12-09T13:43:56 | 2020-12-09T13:43:56 | {} | MEMBER |
ClickHouse | ClickHouse | 742,003,297 | 17,919 | alesapin | test_log_family_s3/test.py::test_log_family_s3 flaky | 2020-12-09T19:40:02 | 2020-12-09T19:40:02 | {} | MEMBER |
ClickHouse | ClickHouse | 742,536,856 | 17,959 | filimonov | > > Now you need to list all settings which should be readonly.
> > We want to limit all, but allow changing some of them.
>
> What do you mean is that updating system.settings table by `alter table update ` ?
https://clickhouse.tech/docs/en/sql-reference/statements/set/ | 2020-12-10T13:57:46 | 2020-12-10T13:57:46 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 742,579,759 | 5,125 | xicmiah | Would it be possible to have an engine working like strengthened `ReplacingMergeTree`, returning one row per primary key for queries, but not necessarily the last one?
This would be roughly equivalent to running
```sql
select ... -- actual query
from (
select * from table
limit 1 by (primary key)
) table_ded... | 2020-12-10T15:09:02 | 2020-12-10T15:09:02 | {} | NONE |
ClickHouse | ClickHouse | 740,765,014 | 15,889 | nvartolomei | ```
WITH concat(addressToLine(arrayJoin(trace) AS addr), '#') AS symbol
SELECT count() > 7
FROM system.trace_log AS t
WHERE (query_id =
(
SELECT
[NULL, NULL, NULL, NULL, 0.00009999999747378752, NULL, NULL, NULL, NULL, NULL],
query_id
FROM system.query_log
WHERE (query LIKE '%test ... | 2020-12-08T17:03:03 | 2020-12-08T17:03:03 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 740,815,755 | 17,906 | den-crane | In short:
Atomic:
non-blocking drop table / rename table
tables delete (&detach) async (wait for selects finish but invisible for new selects)
atomic drop table (all files / folders removed)
atomic table swap (table swap by "EXCHANGE TABLES t1 AND t2;")
rename dictionary / rename database
unique automatic UUI... | 2020-12-08T18:18:13 | 2020-12-08T18:18:13 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 741,377,288 | 17,623 | chang0391 | @alexey-milovidov
Suggest this function which is when a new node is added, all distributed tables can be rebalanced through one command,this will make the resource utilization of each node very balanced and user operation will be more convenient | 2020-12-09T01:14:14 | 2020-12-09T01:18:02 | {
"+1": 1
} | NONE |
ClickHouse | ClickHouse | 741,424,435 | 17,858 | andyyzh | @amosbird Take time to have a look, on version 20.8 is work. | 2020-12-09T01:47:50 | 2020-12-09T01:47:50 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 741,518,561 | 17,887 | sundy-li | The failed test is irrelevant. | 2020-12-09T04:27:43 | 2020-12-09T04:27:43 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 741,520,281 | 17,748 | azat | >Performance — Failed to parse the report.
Requires #17907 | 2020-12-09T04:34:03 | 2020-12-09T04:34:03 | {} | MEMBER |
ClickHouse | ClickHouse | 741,754,821 | 14,042 | den-crane | @sliontc service start clickhouse-server
Version 18.14 is out of support. | 2020-12-09T13:00:11 | 2020-12-09T13:00:11 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 742,412,038 | 16,621 | alesapin | ```
[12:53:06]alesap@click:~/debugtoo/somech$ grep -a 'Fatal' clickhouse-server.log
2020.12.07 15:15:13.263443 [ 244 ] {} <Fatal> BaseDaemon: (version 20.13.1.5351, build id: 3A7AA5193610E1F8) (from thread 158211) Terminate called without an active exception
2020.12.07 15:15:13.283369 [ 447877 ] {} <Fatal> BaseDaem... | 2020-12-10T09:53:52 | 2020-12-10T09:53:52 | {} | MEMBER |
ClickHouse | ClickHouse | 742,456,439 | 17,960 | nikitamikhaylov | Which type do depth and _aq.amount have? Could you please execute EXPLAIN SYNTAX "your query" in 20.8? | 2020-12-10T11:16:13 | 2020-12-10T11:16:27 | {} | MEMBER |
ClickHouse | ClickHouse | 747,658,924 | 18,137 | mrAndersen | UP:
Problem FOUND! If i set max_threads=1 (from default 8) everything is working! All data are in place!
Also there are some strange messages in logs:
```
<Debug> MergeTreeReadPool: Slow read, event №1: read 1048576 bytes in 1.104 sec
<Debug> MergeTreeReadPool: Will lower number of threads to 7
```
I assum... | 2020-12-17T19:43:23 | 2020-12-17T19:43:23 | {} | NONE |
ClickHouse | ClickHouse | 747,939,781 | 18,137 | KochetovNicolai | Repro:
```
create table t (x UInt64, s String) engine = MergeTree order by x;
INSERT INTO t SELECT
number,
if(number < (8129 * 1024), arrayStringConcat(arrayMap(x -> toString(x), range(number % 128)), ' '), '')
FROM numbers((8129 * 1024) * 3);
select count(), sum(length(s)) from t settings max_threads ... | 2020-12-18T08:17:27 | 2020-12-18T08:17:27 | {} | MEMBER |
ClickHouse | ClickHouse | 747,579,384 | 18,192 | alexey-milovidov | ```
Warning: Unexpected input(s) 'image-reference', 'dockerfile-path', valid inputs are ['image', 'path', 'debug', 'fail-build', 'grype-version', 'acs-report-enable', 'severity-cutoff']
Run anchore/scan-action@main
with:
image-reference: localbuild/testimage:latest
dockerfile-path: docker/server/Dockerfi... | 2020-12-17T17:19:23 | 2020-12-17T17:19:23 | {} | MEMBER |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.