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 | 539,542,986 | 7,223 | MahmoudGoda0 | Kindly find below the results: Note that before running the 2 queries i run the below query to force the merge process in the third table
optimize table3 final
Now i have 105 part active for table 3 and 75 for table 1
First Table:
```
<Debug> executeQuery: (from [::1]:47416) SELECT count(*) FROM ts1day WHERE (st... | 2019-10-08T14:34:04 | 2019-10-09T12:49:57 | {} | NONE |
ClickHouse | ClickHouse | 540,238,620 | 7,204 | alexey-milovidov | @divc99 Sure. It will be very appreciated! | 2019-10-09T22:48:27 | 2019-10-09T22:48:27 | {} | MEMBER |
ClickHouse | ClickHouse | 540,331,271 | 7,229 | Tasselmi | > @Tasselmi it sound like you are looking for that:
> #2741
> #3483
>
> BTW. @BayoNet looks like `from_env` is not documented.
actually I think `from_env` is little different from what I expect.
I want to parse `macros` from our server configuration.
for example:
```
<remote_servers>
<cluster_1st> --T... | 2019-10-10T03:41:26 | 2019-10-10T03:41:26 | {} | NONE |
ClickHouse | ClickHouse | 540,585,568 | 7,232 | ghost | @alesapin
Colleagues, when you can upload to https://packagecloud.io/Altinity/clickhouse? | 2019-10-10T13:36:05 | 2019-10-10T13:36:05 | {} | NONE |
ClickHouse | ClickHouse | 540,629,907 | 6,403 | filimonov | Should be fixed by #7175 | 2019-10-10T15:03:07 | 2019-10-10T15:03:07 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 540,131,882 | 7,248 | filimonov | > > Do I understand correctly that ClickHouse server responds that it would like to use sha256_password authentication but mysql client cannot use it?
>
> Yes.
> It should work if password is specified using double sha1.
> [Tag in config](https://github.com/ClickHouse/ClickHouse/blob/master/dbms/programs/server/us... | 2019-10-09T18:34:57 | 2019-10-09T18:34:57 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 540,450,866 | 7,232 | ghost | @alesapin we have updated clickhouse through yum. Version - 19.15.2.2-1.el7 and Repository - altinity_clickhouse. However, see this error again( | 2019-10-10T08:04:29 | 2019-10-10T08:04:42 | {} | NONE |
ClickHouse | ClickHouse | 540,662,488 | 7,263 | alexey-milovidov | As I understant, the function should return FQDN? | 2019-10-10T16:18:03 | 2019-10-10T16:18:03 | {} | MEMBER |
ClickHouse | ClickHouse | 540,667,947 | 7,263 | alexey-milovidov | How to implement:
look at the `dbms/src/Functions/hostName.cpp` and `dbms/src/Common/getFQDNOrHostName.h`. | 2019-10-10T16:31:35 | 2019-10-10T16:31:35 | {} | MEMBER |
ClickHouse | ClickHouse | 541,036,450 | 7,276 | 4ertus2 | Code is correct. It's under the ```any_join_distinct_right_table_keys``` setting too. We are going to implement ```ANY INNER JOIN``` that means 'distinct keys for both sides'. Current ```ANY INNER JOIN``` could return duplicates from left table. That leads to inconsistency of ```t1 ANY INNER JOIN t2``` and ```t2 ANY IN... | 2019-10-11T12:05:16 | 2019-10-11T12:09:45 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 541,196,368 | 7,280 | alesapin | Wow, I didn't break anything... very suspicious. | 2019-10-11T19:35:10 | 2019-10-11T19:35:10 | {} | MEMBER |
ClickHouse | ClickHouse | 539,777,141 | 6,480 | amelkikh | Hi!
Same problem on WSL Ubuntu 18 on Win10
All settings configured by default as is after setup.
```
gdb --args clickhouse-server --config-file=/etc/clickhouse-server/config.xml
GNU gdb (Ubuntu 8.1-0ubuntu3.1) 8.1.0.20180409-git
Copyright (C) 2018 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version ... | 2019-10-09T01:28:38 | 2019-10-09T01:31:06 | {} | NONE |
ClickHouse | ClickHouse | 539,779,215 | 7,237 | alexey-milovidov | From internal task tracker:
```
Нужно реализовать generic вариант функции transform.
Для этого:
- если присутствует default, то приводим типы массива dst и default к наименьшему общему типу;
- составляем отображение %%src_hash -> array_index_in_dst%%;
- формируем результат с помощью метода %%IColumn::insertFrom... | 2019-10-09T01:35:56 | 2019-10-09T01:35:56 | {} | MEMBER |
ClickHouse | ClickHouse | 540,113,369 | 7,249 | alexey-milovidov | Start of work in Q4 (November..December).
We are considering two possible ways of implementation.
One of them is very natural but is blocked by https://github.com/ClickHouse/ClickHouse/pulls?q=is%3Apr+processors+author%3AKochetovNicolai
Another way is more hacky and we will use it only if first option will fail. | 2019-10-09T17:49:48 | 2019-10-09T17:49:48 | {} | MEMBER |
ClickHouse | ClickHouse | 540,289,935 | 7,246 | jackpgao | > `(version 19.11.4)` vs `(version 19.13.3)`
>
> There was some changes in `sytem.processes` structure recently.
> Just recreate Distibuted table will set of columns supported by both versions, or update all nodes in cluster.
Recreate seems doesn't work, and the 3 nodes above have the same version. | 2019-10-10T01:51:51 | 2019-10-10T01:51:51 | {} | NONE |
ClickHouse | ClickHouse | 540,415,807 | 4,756 | hohaftf | UPD.
Clickhouse v 19.15.3.6
Query with aliases works in local mode
Query with db names throws exception
> SQL Error [288]: ClickHouse exception, code: 288, host: localhost, port: 8123; Code: 288, e.displayText() = DB::Exception: Distributed table should have an alias when distributed_product_mode set to local.... | 2019-10-10T06:53:47 | 2019-10-10T06:53:47 | {} | NONE |
ClickHouse | ClickHouse | 540,624,750 | 7,109 | abyss7 | Simple query `SELECT groupBitmapOr(31)` crashes the server. Can you please investigate and fix this? | 2019-10-10T14:51:55 | 2019-10-10T14:51:55 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 540,715,777 | 7,268 | den-crane | version 19.16.1
SELECT count()
FROM remote('{1,127}.0.0.{2,3}', system.one)
SETTINGS skip_unavailable_shards = 1
Received exception from server (version 19.16.1):
Code: 90. DB::Exception: Received from localhost:9000. DB::Exception: Empty list of columns passed.
I think there is a deep problem in distribu... | 2019-10-10T18:33:46 | 2019-10-10T18:33:46 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 540,814,671 | 7,272 | azat | Seems that #7264 had fixed it in a more easier way | 2019-10-10T21:49:49 | 2019-10-10T21:49:49 | {} | MEMBER |
ClickHouse | ClickHouse | 539,686,152 | 7,221 | alexey-milovidov | Do you want to add my test just in case? :) | 2019-10-08T20:20:28 | 2019-10-08T20:20:28 | {} | MEMBER |
ClickHouse | ClickHouse | 539,792,822 | 7,219 | pyscala | > There is documentation on the russian version of the MergeTree page here: https://clickhouse.yandex/docs/ru/operations/table_engines/mergetree/#table_engine-mergetree-multiple-volumes_configure . Unfortunately, it's not in the english version yet. There are also some examples in a recent presentation here: https://ht... | 2019-10-09T02:25:06 | 2019-10-09T02:25:06 | {} | NONE |
ClickHouse | ClickHouse | 539,843,718 | 6,168 | YiuRULE | @Akazz If you talk about #6489, if I understand well it's about moving partition from a different physical storage, while this PR is only dealing to moving datas to a different table.
The initial issue (#4729) was about moving partition to a different table (without doing specially an operation directly on the disk)... | 2019-10-09T05:48:54 | 2019-10-09T05:49:11 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 540,447,870 | 5,371 | JackyYangPassion | > * Single shard to aggregate large data may be the bottleneck, will it have a new Engine like `DistributeHDFS` storage? It will distribute the files to each shard by some scheduling algorithms.
> * Index pushed to filter the files, usually we split the HDFS file by date, `hdfs://path/table_name/dt=YYYYMMDD`. So the ... | 2019-10-10T07:57:36 | 2019-10-10T09:02:32 | {} | NONE |
ClickHouse | ClickHouse | 540,508,399 | 7,236 | azat | Sure, still take a look into this today | 2019-10-10T10:33:54 | 2019-10-10T10:33:54 | {} | MEMBER |
ClickHouse | ClickHouse | 541,062,912 | 7,238 | rusik69 | /assign | 2019-10-11T13:24:02 | 2019-10-11T13:24:02 | {} | NONE |
ClickHouse | ClickHouse | 539,716,812 | 7,221 | azat | >Do you want to add my test just in case? :)
Due to `HashTableGrower::increaseSize` makes `HashTable` grows twice power of two, before `2^23` using `2^13 - 1` does not changes anything, since the buffer grows as follow:
- 128
- 129 elements in hashtable -> 1024 # initial, but maxFill() allows only 128 items to be ... | 2019-10-08T21:43:13 | 2019-10-08T21:43:13 | {} | MEMBER |
ClickHouse | ClickHouse | 540,080,854 | 3,261 | simPod | @filimonov ugh, 1y ago, no idea... :D not even using summing merge tree now. I'd love to elaborate but that's all we have. Can close if invalid. | 2019-10-09T16:31:02 | 2019-10-09T16:31:13 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 540,496,270 | 7,236 | akuzm | Apparently some tests were broken in this PR, but we didn't notice because I broke the CI.
Now it's failing in master: https://clickhouse-test-reports.s3.yandex.net/0/81a9c81ba7fecd2747ecf1f5050f23e5d5ff771c/functional_stateless_tests_(release).html
@azat could you look into this? It's 01017_uniqCombined_memory_usa... | 2019-10-10T10:00:45 | 2019-10-10T10:00:45 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 540,617,647 | 7,247 | Akazz | Failing test `00212_shard_aggregate_function_uniq` is irrelevant - it is broken in the master | 2019-10-10T14:36:10 | 2019-10-10T14:36:10 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 540,828,385 | 7,261 | josepowera | We have seen big drop in performance after this upgrade (19.14 to 19.15).
Since data consistency is more important (and we are really happy with progress in kafka support in CH in last 2 months) we postponed this question (we didn't know if there was any bigger change in our virtual machines provider or if old ver... | 2019-10-10T22:41:19 | 2019-10-10T23:00:43 | {} | NONE |
ClickHouse | ClickHouse | 541,002,325 | 7,109 | akuzm | An AddressSanitizer failure in master:
https://clickhouse-test-reports.s3.yandex.net/7270/e25bdc8d25656486a73493ab0724342f40eeb27b/stress_test_(address)/stderr.log | 2019-10-11T10:03:49 | 2019-10-11T10:03:49 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 541,053,280 | 7,268 | nvartolomei | I assume the problem in this case is pointed out by stacktrace (19.14):
```
7. 0x5576f2e093b0 DB::getStructureOfRemoteTable(DB::Cluster const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > cons... | 2019-10-11T12:56:44 | 2019-10-11T12:56:44 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 541,099,032 | 7,216 | KochetovNicolai | This optimization is not valid for lot of cases.
```
SELECT
'IN empty set',
count()
FROM numbers(10)
WHERE (number IN
(
SELECT toUInt64(1)
WHERE 0
)) = 0
┌─'IN empty set'─┬─count()─┐
│ IN empty set │ 0 │
└────────────────┴─────────┘
```
```
SELECT
'IN empty set not in ... | 2019-10-11T14:56:43 | 2019-10-11T14:56:43 | {} | MEMBER |
ClickHouse | ClickHouse | 540,021,537 | 6,826 | Akazz | This test challenges race condition in destructor of our `RWLockImpl::LockHolder`. After #6854 the chance of race here decreased somewhat.
This test needs to be revisited to get consistent failures with the current implementation of `RWLockImpl::LockHolder`. I suggest we hold this PR until `RWLockImpl::LockHolder` ... | 2019-10-09T14:17:07 | 2019-10-09T14:19:42 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 540,662,618 | 7,263 | alexey-milovidov | PS. Separate function name is better than parameter. | 2019-10-10T16:18:17 | 2019-10-10T16:18:17 | {} | MEMBER |
ClickHouse | ClickHouse | 540,855,182 | 5,889 | alexey-milovidov | @4ertus2 I don't understand what do you mean? | 2019-10-11T00:51:36 | 2019-10-11T00:51:36 | {} | MEMBER |
ClickHouse | ClickHouse | 541,118,200 | 7,188 | thyn | didn't find any solution | 2019-10-11T15:47:03 | 2019-10-11T15:47:03 | {} | NONE |
ClickHouse | ClickHouse | 541,333,178 | 7,095 | vdimir | > So, we can also allow only single argument or zero number of arguments. In this case, the `countIf` will accept only single or two arguments.
Some tests uses `count` with 2 arguments, e.g [00425_count_nullable.sql](https://github.com/ClickHouse/ClickHouse/blob/6f758babf6912b7ecb5424b2e669b943c588317f/dbms/tests/qu... | 2019-10-12T15:07:50 | 2019-10-12T15:08:53 | {} | MEMBER |
ClickHouse | ClickHouse | 541,344,034 | 7,095 | alexey-milovidov | Yes, we can remove support for this feature (and place a note about backward compatibility in changelog). | 2019-10-12T17:13:53 | 2019-10-12T17:13:53 | {} | MEMBER |
ClickHouse | ClickHouse | 539,775,022 | 7,240 | alexey-milovidov | Need to update Docker image for tests. | 2019-10-09T01:20:31 | 2019-10-09T01:20:31 | {} | MEMBER |
ClickHouse | ClickHouse | 540,053,677 | 6,549 | olegkv | Got same error. Mutations get stuck with permission error.
All permissions look normal.
I don't use FREEZE partitions.
I don't use Docker, just Clickhouse on Azure Ubuntu VMs.
VM restart does not help.
-stop clickhouse-server,
-chmod -R 755 on data directory
-start clickhouse-server
does help, after that m... | 2019-10-09T15:28:15 | 2019-10-09T15:28:15 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 540,616,456 | 7,256 | filimonov | That should be fixed in 19.15.3 (PR #7175) | 2019-10-10T14:33:26 | 2019-10-10T14:56:01 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 541,048,980 | 7,232 | filimonov | > @alesapin
> Colleagues, when you can upload to https://packagecloud.io/Altinity/clickhouse?
Altinity build for ClickHouse RPMs v19.15.3.6-stable published | 2019-10-11T12:44:55 | 2019-10-11T12:44:55 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 541,284,192 | 7,288 | richardartoul | After looking through the code it seems like I would just need to add an optional call to `disableChecksumming` in `Connection.cpp` here:
```c++
void Connection::initBlockInput()
{
if (!block_in)
{
if (!maybe_compressed_in)
{
if (compression == Protocol::Compression::Enable... | 2019-10-12T04:38:17 | 2019-10-12T04:38:17 | {} | NONE |
ClickHouse | ClickHouse | 539,791,983 | 7,230 | Tasselmi | > Do you know of arrayEnumerate() function? It generates indices. Also note that array indices in ClickHouse are 1-based, so if you use indices later to access the same array, your approach does not work. So better to use:
>
> `SELECT arrayMap((x, y) -> (x, y), arrayEnumerate(['a', 'b', 'c']), ['a', 'b', 'c'])`
>
... | 2019-10-09T02:20:53 | 2019-10-09T02:20:53 | {} | NONE |
ClickHouse | ClickHouse | 539,878,047 | 5,438 | clemrodriguez | Thank you for the your feedback.
Code is integrated into another local branch (v19.15.7.2) as we need in our production environment.
I'll close this PR and open a new one.
| 2019-10-09T07:35:13 | 2019-10-09T07:35:13 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 540,627,764 | 7,236 | azat | Thanks, got it! | 2019-10-10T14:58:26 | 2019-10-10T14:58:26 | {} | MEMBER |
ClickHouse | ClickHouse | 540,666,964 | 7,263 | den-crane | > As I understant, the function should return FQDN?
yes (I expect the same result as `hostname -f` ) | 2019-10-10T16:29:09 | 2019-10-10T16:31:10 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 540,834,077 | 7,261 | filimonov | Most probably both options will not really help. You can try to create one more (or few more) tables with the same structure filling the same target table (in that case it will work multithreaded, with num_consumers you are still locked in one thread)
| 2019-10-10T23:05:48 | 2019-10-10T23:06:09 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 538,990,890 | 6,168 | Akazz | But the most interesting thing is: doesn't what has been done in #6489 already address the whole motivation of this PR (as described in #4729)?
UPD: Ok, it does not. Your PR solves a somewhat different problem | 2019-10-07T12:40:52 | 2019-10-09T11:07:15 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 539,991,491 | 3,261 | filimonov | What is an advantage of using such type of PK ```ORDER BY (toStartOfFiveMinute(created_at))``` instead of ```ORDER BY created_at```.
Or it's just a synthetic sample, and it was complex key in real table? | 2019-10-09T13:04:12 | 2019-10-09T13:04:12 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 540,202,871 | 7,204 | divyansshhh | Can I work on this? | 2019-10-09T21:16:34 | 2019-10-09T21:16:34 | {} | NONE |
ClickHouse | ClickHouse | 540,618,391 | 7,236 | akuzm | The list of checks for each commit in master can be seen if you press a cross or a tick to the right of the commit time at this page:
https://github.com/ClickHouse/ClickHouse/commits/master | 2019-10-10T14:37:45 | 2019-10-10T14:37:45 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 541,165,047 | 6,337 | 4ertus2 | #7282 | 2019-10-11T18:01:52 | 2019-10-11T18:01:52 | {
"+1": 2
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 541,171,142 | 7,269 | 4ertus2 | Closed in favor of #7284 | 2019-10-11T18:19:45 | 2019-10-11T18:19:45 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 541,219,954 | 7,238 | alexey-milovidov | @loqutus Ok, now it's yours! | 2019-10-11T20:52:14 | 2019-10-11T20:52:14 | {} | MEMBER |
ClickHouse | ClickHouse | 540,159,273 | 7,248 | filimonov | > >
> > Actually i was using paswordless default user. Why does it try sha256?
>
> Empty password isn't handled in a different way, since the plugin does it efficiently itself (without encryption).
I mean if sha1 is more 'portable', why don't we use that in such cases? | 2019-10-09T19:44:27 | 2019-10-09T19:44:27 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 540,162,513 | 7,251 | alexey-milovidov | This is controlled by `join_use_nulls` setting. | 2019-10-09T19:52:31 | 2019-10-09T19:52:31 | {} | MEMBER |
ClickHouse | ClickHouse | 540,280,830 | 7,248 | yurriy | It is possible, but it seems confusing as the authentication will work/not work depending only on whether the password is empty/non-empty. | 2019-10-10T01:22:51 | 2019-10-10T01:22:51 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 540,643,814 | 7,109 | akuzm | We've also observed a ThreadSanitizer failure in Stress test on current master:
https://clickhouse-test-reports.s3.yandex.net/7247/3fca2e42ca187bd4209726da87361e93882b503d/stress_test_(thread)/stderr.log
Looks more like a use-after-free. | 2019-10-10T15:34:07 | 2019-10-10T15:34:07 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 540,708,273 | 7,245 | alexey-milovidov | Could you please validate that the fresh build from master has no such issue?
https://clickhouse-builds.s3.yandex.net/7252/3fbac603288ba9962aa9fa33c6f8d0116ce963c1/report.html | 2019-10-10T18:15:39 | 2019-10-10T18:15:39 | {} | MEMBER |
ClickHouse | ClickHouse | 540,855,586 | 5,645 | alexey-milovidov | @4ertus2 Probably you left this comment by mistake? | 2019-10-11T00:53:56 | 2019-10-11T00:53:56 | {} | MEMBER |
ClickHouse | ClickHouse | 540,091,656 | 7,248 | yurriy | > Ok. And if I understand correctly, sha256 is new authentication method that is only supported in newer mysql clients?
Yes. In documentation, sha256 plugin [is defined](https://dev.mysql.com/doc/refman/5.6/en/sha256-pluggable-authentication.html) starting from version 5.6. Our tests use 5.7. | 2019-10-09T16:58:19 | 2019-10-09T16:58:19 | {
"+1": 2
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 540,147,564 | 7,248 | yurriy | > > > Do I understand correctly that ClickHouse server responds that it would like to use sha256_password authentication but mysql client cannot use it?
> >
> >
> > Yes.
> > It should work if password is specified using double sha1.
> > [Tag in config](https://github.com/ClickHouse/ClickHouse/blob/master/dbms/pr... | 2019-10-09T19:13:45 | 2019-10-09T19:13:45 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 541,298,565 | 7,284 | KochetovNicolai | Some performance tests failed.
Need to investigate. | 2019-10-12T07:59:10 | 2019-10-12T07:59:10 | {} | MEMBER |
ClickHouse | ClickHouse | 539,746,677 | 7,237 | alexey-milovidov | Minimal example:
```
SELECT CASE 1 WHEN 1 THEN 2 END
``` | 2019-10-08T23:28:55 | 2019-10-08T23:28:55 | {} | MEMBER |
ClickHouse | ClickHouse | 539,892,872 | 7,232 | ghost | @den-crane yesterday we updated clickhouse to 19.15.2.2 yesterday on Oracle Linux Server through
`yum install clickhouse-server`, however see this errors.
`2019.10.08 17:24:17.202894 [ 88 ] {d2503d49-af1a-4ae5-88c1-0939ea87a65c} <Error> executeQuery: std::exception. Code: 1001, type: std::bad_variant_access, e.what(... | 2019-10-09T08:19:13 | 2019-10-09T08:19:13 | {} | NONE |
ClickHouse | ClickHouse | 539,927,830 | 7,209 | alesapin | Docs check failed on master | 2019-10-09T09:50:59 | 2019-10-09T09:50:59 | {} | MEMBER |
ClickHouse | ClickHouse | 539,962,912 | 3,483 | filimonov | > As the bold part said. Is there any way to solve this problem? Because we always start clickhouse with `service`.
I think you need to adjust service file in that case.
https://coreos.com/os/docs/latest/using-environment-variables-in-systemd-units.html
| 2019-10-09T11:38:44 | 2019-10-09T11:38:44 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 540,462,142 | 7,253 | alesapin | All errors from master. | 2019-10-10T08:34:17 | 2019-10-10T08:34:17 | {} | MEMBER |
ClickHouse | ClickHouse | 540,520,864 | 7,232 | alesapin | You should update to 19.15.3.6. | 2019-10-10T11:10:30 | 2019-10-10T11:10:30 | {} | MEMBER |
ClickHouse | ClickHouse | 540,569,247 | 7,207 | ishirav | Thank you @den-crane, this is very helpful.
I tried checking the total size of the marks:
```
select formatReadableSize(sum(marks_bytes)) from system.parts
```
The result is 188MB, which is strange: why would the marks cache grow to 800MB if the total size of the marks is much smaller?
Regarding being not viabl... | 2019-10-10T13:07:00 | 2019-10-10T13:07:00 | {} | NONE |
ClickHouse | ClickHouse | 541,033,333 | 7,202 | qonsth | Temporary solution:
`addHours(now(), 14)` | 2019-10-11T11:55:01 | 2019-10-11T11:55:01 | {} | NONE |
ClickHouse | ClickHouse | 541,055,741 | 7,275 | filimonov | That happens because `default` profile is used as a 'base' for other profiles. And once readonly constraint was set, it can't be changed later.
Related code part:
https://github.com/ClickHouse/ClickHouse/blob/9a89ee8eabf342821fa552e69815b05ec364f9fb/dbms/src/Interpreters/SettingsConstraints.cpp#L155
Maybe we s... | 2019-10-11T13:03:43 | 2019-10-11T13:03:43 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 539,686,896 | 7,221 | azat | >Do you want to add my test just in case? :)
Sure, along with fix for the `medium_set_size_max` | 2019-10-08T20:22:03 | 2019-10-08T20:22:03 | {} | MEMBER |
ClickHouse | ClickHouse | 540,065,430 | 7,248 | alexey-milovidov | @yurriy
> ➜ mysql -h 127.0.0.1 -P 9001 --user default
ERROR 2059 (HY000): Authentication plugin 'sha256_password' cannot be loaded: /usr/lib64/mysql/plugin/sha256_password.so: cannot open shared object file: No such file or directory
Do I understand correctly that ClickHouse server responds that it would like t... | 2019-10-09T15:54:28 | 2019-10-09T15:54:28 | {} | MEMBER |
ClickHouse | ClickHouse | 540,086,289 | 3,261 | alexey-milovidov | It's valid but minor issue. | 2019-10-09T16:44:43 | 2019-10-09T16:44:43 | {} | MEMBER |
ClickHouse | ClickHouse | 540,588,481 | 7,232 | den-crane | @twinky618 you should ask Altinity ( @alex-zaitsev ) or use official RPM https://repo.yandex.ru/clickhouse/rpm/stable/x86_64/ | 2019-10-10T13:41:17 | 2019-10-10T13:44:16 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 539,966,204 | 1,469 | shcheglovnd | Another way to make cumulative sum:
```
SELECT
number_1,
arraySum(arraySlice(numbers, 1, i)) AS cum_sum
FROM
(
SELECT groupArray(number) AS numbers
FROM
(
SELECT number
FROM system.numbers
LIMIT 10
)
)
ARRAY JOIN
numbers AS number_1,
arrayEnumerat... | 2019-10-09T11:50:02 | 2019-10-09T11:50:02 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 540,005,429 | 7,184 | blinkov | @feribg this is a separate feature to implement, see #6337 | 2019-10-09T13:39:54 | 2019-10-09T13:39:54 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 540,065,848 | 7,248 | alexey-milovidov | And the workaround is to add sha1 password for ClickHouse user? | 2019-10-09T15:55:23 | 2019-10-09T15:55:23 | {} | MEMBER |
ClickHouse | ClickHouse | 540,085,273 | 7,248 | alexey-milovidov | Ok. And if I understand correctly, sha256 is new authentication method that is only supported in newer mysql clients? | 2019-10-09T16:42:06 | 2019-10-09T16:42:06 | {} | MEMBER |
ClickHouse | ClickHouse | 540,085,941 | 7,248 | alexey-milovidov | @filimonov You can think on how we can make it as obvious to the user as possible. | 2019-10-09T16:43:48 | 2019-10-09T16:43:48 | {} | MEMBER |
ClickHouse | ClickHouse | 541,284,047 | 7,057 | zhang2014 | The Stress test (thread) error was caused by #7109
Log:
```
2019.10.11 20:48:22.861372 [ 129 ] {} <Fatal> BaseDaemon: ########################################
2019.10.11 20:48:22.861840 [ 129 ] {} <Fatal> BaseDaemon: (version 19.16.1.1486) (from thread 128) Received signal Segmentation fault (11).
2019.10.11 20:... | 2019-10-12T04:35:38 | 2019-10-12T04:45:24 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 539,788,445 | 7,229 | Tasselmi | > cmdb
so how about make `layer ` `shard` `replica` all variable?
for example:
```
<macros>
<layer>${sub_cluster_name}</layer>
<shard>${shard_name}</shard>
<replica>${hostname}</replica>
</macros>
``` | 2019-10-09T02:04:30 | 2019-10-09T02:04:30 | {} | NONE |
ClickHouse | ClickHouse | 539,929,126 | 7,232 | alesapin | We will make bugfix release today. | 2019-10-09T09:54:19 | 2019-10-09T09:54:19 | {} | MEMBER |
ClickHouse | ClickHouse | 540,552,018 | 7,057 | alexey-milovidov | > I'm not sure if need the output Skipped rows: xxx because it might affect the display of the progress bar.
Yes, let's leave it in server logs and keep default progress bar compact. | 2019-10-10T12:35:51 | 2019-10-10T12:35:51 | {} | MEMBER |
ClickHouse | ClickHouse | 540,617,485 | 7,236 | akuzm | Do you mean the test log or the build log itself?
For tests, there are links to logs at the bottom of this test page:
https://clickhouse-test-reports.s3.yandex.net/0/0769f80f4ea2516cc301257af27352c27a93df6c/functional_stateless_tests_(release).html
In the list of checks, this is Details of 'Functional stateless ... | 2019-10-10T14:35:49 | 2019-10-10T14:35:49 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 540,681,252 | 7,236 | azat | > 01017_uniqCombined_memory_usage
Hm, interesting client fails with "Attempt to read after eof" error first time (after server restart) while the server indeed reports "Memory limit (for query) exceeded", and I can reproduce it with packages from [this](https://clickhouse-builds.s3.yandex.net/0/0769f80f4ea2516cc3012... | 2019-10-10T17:05:56 | 2019-10-10T17:05:56 | {} | MEMBER |
ClickHouse | ClickHouse | 539,727,855 | 7,221 | azat | >introduce another grower, that will grow by the power of 2
@alexey-milovidov something like #7236 | 2019-10-08T22:20:01 | 2019-10-08T22:20:01 | {} | MEMBER |
ClickHouse | ClickHouse | 539,965,268 | 7,246 | filimonov | `(version 19.11.4)` vs `(version 19.13.3)`
There was some changes in `sytem.processes` structure recently.
Just recreate Distibuted table will set of columns supported by both versions, or update all nodes in cluster. | 2019-10-09T11:46:45 | 2019-10-09T11:46:45 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 540,068,980 | 7,248 | yurriy | > Do I understand correctly that ClickHouse server responds that it would like to use sha256_password authentication but mysql client cannot use it?
Yes.
It should work if password is specified using double sha1.
[Tag in config](https://github.com/ClickHouse/ClickHouse/blob/master/dbms/programs/server/users.xml#L4... | 2019-10-09T16:02:34 | 2019-10-09T16:05:13 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 540,467,955 | 7,057 | zhang2014 | I'm not sure if need the output `Skipped rows: xxx` because it might affect the display of the progress bar.
after change(running progress):
```
i-tl62qd0o :) SELECT count() FROM test.single_column_bloom_filter WHERE i32 = 6789845
SELECT count()
FROM test.single_column_bloom_filter
WHERE i32 = 6789845
↑ P... | 2019-10-10T08:49:29 | 2019-10-10T08:50:29 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 527,843,402 | 6,814 | zhang2014 | TODO:
- [x] rebase master
- [ ] Support toString Function With JSONB Type
- [ ] Support JSON Functions With JSONB Type
- [ ] Support Nullable(SmallestJSON)
- [ ] Support LowCardinality(SmallestJSON)
- [ ] Support push down required column to SmallestJSON
- [ ] Support JSON Array
- [ ] Support BigJSON (use BJSON... | 2019-09-04T10:32:30 | 2019-10-10T10:42:44 | {
"+1": 3
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 540,571,512 | 7,236 | azat | >01017_uniqCombined_memory_usage
Cannot reproduce the failure of this test, is there any chance on getting logs from the build on CI? | 2019-10-10T13:11:02 | 2019-10-10T13:11:02 | {} | MEMBER |
ClickHouse | ClickHouse | 540,585,500 | 7,207 | den-crane | > The result is 188MB, which is strange: why would the marks cache grow to 800MB if the total size of the marks is much smaller?
Not sure, I suspect because it could contain cache for marks of inactive/merged_parts. And these marks will be pushed out by LRU mechanism.
>We have systems running with only 8MB
CH... | 2019-10-10T13:35:57 | 2019-10-10T13:35:57 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 540,726,852 | 7,236 | azat | @akuzm
- `00212_shard_aggregate_function_uniq` - #7270
- `01017_uniqCombined_memory_usage` - #7272 | 2019-10-10T18:56:03 | 2019-10-10T18:56:03 | {} | MEMBER |
ClickHouse | ClickHouse | 541,021,514 | 6,168 | YiuRULE | @Akazz Just for this point
`Essentially the feature of moving a partition between tables atomically is not tested - should run concurrent SELECT's + ALTER's while checking atomicity and results consistency`
Does we got some tests for replicated tables for the `REPLACE` operation ? I wasn't able to find out and I wou... | 2019-10-11T11:11:59 | 2019-10-11T11:11:59 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 541,343,669 | 7,288 | alexey-milovidov | > If you're ok with this feature and it's not too much work I could try and implement it myself also if you could point me in the right direction.
Yes, I'm totally Ok with that.
> After looking through the code it seems like I would just need to add an optional call to disableChecksumming in Connection.cpp here
... | 2019-10-12T17:09:47 | 2019-10-12T17:09:47 | {} | MEMBER |
ClickHouse | ClickHouse | 542,979,710 | 7,077 | yingfeng | This issue is bypassed by PR [7359](https://github.com/ClickHouse/ClickHouse/pull/7359) | 2019-10-17T03:16:30 | 2019-10-17T03:16:30 | {} | NONE |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.