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 | 704,202,960 | 15,394 | 4ertus2 | This UB is not a trouble fo Decimals theirselfs cause they use not full integer ranges.
It's also possible to replace code
```
T result = decimal.value;
if constexpr (!keep_sign)
if (result < T(0))
result = -result;
return result % scale_multiplier;
```
with
```
T resu... | 2020-10-06T11:22:02 | 2020-10-06T11:22:34 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 704,177,114 | 15,639 | alesapin | But why we need to store the file with rows mapping on tmp disk? | 2020-10-06T10:25:15 | 2020-10-06T11:30:29 | {} | MEMBER |
ClickHouse | ClickHouse | 703,749,928 | 15,596 | OmarBazaraa | > `hostname -f` -- can be overridden in config.xml using `interserver_http`
It worked.
Thank you so much, Denny.
But do you have an explanation for [this](https://github.com/ClickHouse/ClickHouse/issues/15596#issuecomment-703591540)? | 2020-10-05T16:41:07 | 2020-10-05T16:41:07 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 704,030,689 | 15,642 | alexey-milovidov | So, it's for the list of indices to force... | 2020-10-06T05:09:56 | 2020-10-06T05:09:56 | {} | MEMBER |
ClickHouse | ClickHouse | 704,309,099 | 15,454 | KochetovNicolai | small test
```
create table dict_source_int_kv (key UInt64, value UInt64) engine = MergeTree order by key;
CREATE DICTIONARY
(
`key` UInt64,
`value` UInt64
)
PRIMARY KEY key
SOURCE(CLICKHOUSE(HOST '127.0.0.1' PORT 9000 USER 'default' PASSWORD '' DB 'default' TABLE 'dict_source_int_kv'))
LIFETIME(MI... | 2020-10-06T14:30:33 | 2020-10-06T15:31:11 | {} | MEMBER |
ClickHouse | ClickHouse | 704,488,062 | 15,596 | den-crane | > > --connect_timeout_with_failover_ms arg Connection timeout for selecting first healthy replica.
> > For example you have 3 replicas. R1, R2, R3.
> > R1 is dead right now. And connect_timeout_with_failover_ms = 1000. On distributed query CH will try to connect to R1. Will wait for 1000ms and then re-try to connect ... | 2020-10-06T18:59:53 | 2020-10-06T19:00:03 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 703,549,363 | 15,550 | alesapin | Tests failures not related to changes. | 2020-10-05T10:38:40 | 2020-10-05T10:38:40 | {} | MEMBER |
ClickHouse | ClickHouse | 704,058,240 | 15,643 | azat | >Integration tests (asan) — fail: 10, passed: 904, error: 0
Test failures looks unrelated
<details>
Test name | Test status | Test time, sec.
-- | -- | --
test_ttl_move/test.py::test_alter_with_merge_work[mt_test_alter_with_merge_do_not_work-MergeTree()-0] | FAIL | 0.0
-- | -- | --
test_ttl_move/test.py::t... | 2020-10-06T06:25:34 | 2020-10-06T06:25:34 | {} | MEMBER |
ClickHouse | ClickHouse | 704,247,416 | 12,136 | AnaUvarova | @vivarum Hi! Could you please share some cases of usage for this function? | 2020-10-06T12:52:12 | 2020-10-06T12:52:12 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 703,591,540 | 15,596 | OmarBazaraa | And btw, when I added this setting and restarted the servers, the servers that the `INSERT`s end up into were not able to start because of the following error:
```
<Error> Application: DB::Exception: Part `202009_0_0_0` was found on disk `default` which is not defined in the storage policy: Cannot attach table `myd... | 2020-10-05T12:12:55 | 2020-10-05T12:14:56 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 703,640,635 | 15,607 | filimonov | BTW that again leads us back to discussion of the transitivity of user permissions. Should the derived in-database objects (views / materialized view / dictionaries) use/require the permission of the current user both to object and underlying object, or it's the object definer need to inheritence his rights to the unde... | 2020-10-05T13:42:12 | 2020-10-05T13:42:12 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 703,844,819 | 15,552 | filimonov | > > flooded by issues like 'my clickhouse stop being available sometimes randomly'
>
> We already have seen another kind of issues, like: "All tables on my ClickHouse instance have disappear and there is a table with a request to send one BTC".
Will will have both :) Black hat guys will remove data and blackmail ... | 2020-10-05T19:36:40 | 2020-10-05T19:36:40 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 704,094,781 | 15,643 | alexey-milovidov | They are too many, we have to figure out the reason before merge :( | 2020-10-06T07:47:07 | 2020-10-06T07:47:07 | {} | MEMBER |
ClickHouse | ClickHouse | 704,354,909 | 15,670 | den-crane | grep 1601305200_29_29_0 /var/log/clickhouse/clickhouse-server.log
at all replicas
It could happend probably in ZK restart at step 1 | 2020-10-06T15:33:19 | 2020-10-06T15:35:25 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 703,611,321 | 15,552 | alexey-milovidov | > flooded by issues like 'my clickhouse stop being available sometimes randomly'
We already have seen another kind of issues, like: "All tables on my ClickHouse instance have disappear and there is a table with a request to send one BTC". | 2020-10-05T12:52:11 | 2020-10-05T12:52:11 | {} | MEMBER |
ClickHouse | ClickHouse | 703,663,360 | 15,538 | 4ertus2 | An issue for fuzzer error #15623 | 2020-10-05T14:19:15 | 2020-10-05T14:19:15 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 703,664,445 | 12,136 | AnaUvarova | Hi, please help me understand what is the difference between %y, %Y and %g, %G? | 2020-10-05T14:21:01 | 2020-10-05T14:21:01 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 704,030,369 | 15,642 | alexey-milovidov | If at least one index is required then better to name it `force_data_skipping_index`? | 2020-10-06T05:08:49 | 2020-10-06T05:08:49 | {} | MEMBER |
ClickHouse | ClickHouse | 704,073,942 | 15,636 | alesapin | Integration OK. | 2020-10-06T07:01:02 | 2020-10-06T07:01:02 | {} | MEMBER |
ClickHouse | ClickHouse | 704,119,431 | 15,596 | OmarBazaraa | > ```
> <connect_timeout_with_failover_ms>1000</connect_timeout_with_failover_ms>
> ```
Can you please further explain the effects of increasing this setting?
Will the insertion latency increase because of this 2-region setup? or the replication occur asynchronously?
| 2020-10-06T08:35:26 | 2020-10-06T08:35:26 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 704,218,789 | 15,451 | KochetovNicolai | ```
set with_global = 1;
with 10 as x select x + y from (select 2 as y) union all select x + 1 from (select 1 as x);
WITH 10 AS x
SELECT x + y
FROM
(
SELECT 2 AS y
)
UNION ALL
SELECT x + 1
FROM
(
SELECT 1 AS x
)
Received exception from server (version 20.10.1):
Code: 179. DB::Exception: ... | 2020-10-06T11:56:44 | 2020-10-06T11:56:44 | {} | MEMBER |
ClickHouse | ClickHouse | 704,220,790 | 15,598 | OmarBazaraa | > check a table system.replicas is_readonly column
Thanks Denny.
I found the readonly table when I checked `system.replicas`, but this wasn't reflected in `ReadonlyReplica` metric in `system.metrics` table. | 2020-10-06T12:00:47 | 2020-10-06T12:00:47 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 704,235,271 | 15,411 | 4ertus2 | It looks kile literal `1` clashes with `regInfo.1`. We need better column naming for litaral expressions (some DB internals). | 2020-10-06T12:29:19 | 2020-10-06T12:30:22 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 704,345,685 | 15,683 | KochetovNicolai | Cannot reproduce it locally. | 2020-10-06T15:20:28 | 2020-10-06T15:20:28 | {} | MEMBER |
ClickHouse | ClickHouse | 704,348,892 | 14,414 | KochetovNicolai | Crash in `quantilesDeterministic` #15683 | 2020-10-06T15:24:44 | 2020-10-06T15:24:44 | {} | MEMBER |
ClickHouse | ClickHouse | 704,451,334 | 15,596 | den-crane | --connect_timeout_with_failover_ms arg Connection timeout for selecting first healthy replica.
For example you have 3 replicas. R1, R2, R3.
R1 is dead right now. And connect_timeout_with_failover_ms = 1000. On distributed query CH will try to connect to R1. Will wait for 1000ms and then re-try to connect to R2 or R... | 2020-10-06T18:04:21 | 2020-10-06T18:06:37 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 703,544,773 | 15,463 | wutchzone | Thank you, now it is working. I totally missed the `-MergeState modifier`. | 2020-10-05T10:29:08 | 2020-10-05T10:29:08 | {} | NONE |
ClickHouse | ClickHouse | 703,625,257 | 13,597 | develar | No need to build from sources if you just want to get native executable for macOS. See https://github.com/ClickHouse/ClickHouse/issues/235#issuecomment-548781417 how to get precompiled binary for macOS. | 2020-10-05T13:17:01 | 2020-10-05T13:17:01 | {} | NONE |
ClickHouse | ClickHouse | 703,728,954 | 15,633 | alexey-milovidov | Fast test Ok. | 2020-10-05T16:04:27 | 2020-10-05T16:04:27 | {} | MEMBER |
ClickHouse | ClickHouse | 704,032,275 | 15,642 | alexey-milovidov | I've fixed "Arcadia". | 2020-10-06T05:15:29 | 2020-10-06T05:15:29 | {} | MEMBER |
ClickHouse | ClickHouse | 704,052,638 | 15,578 | KochetovNicolai | > I thought it was already working this way.
So did I. | 2020-10-06T06:12:43 | 2020-10-06T06:12:43 | {} | MEMBER |
ClickHouse | ClickHouse | 704,116,628 | 15,596 | OmarBazaraa | Here are my configurations:
```
<yandex>
<path>/mnt/ssd1/clickhouse/</path>
<tmp_path>/mnt/ssd1/clickhouse/tmp/</tmp_path>
<user_files_path>/mnt/ssd1/clickhouse/user_files/</user_files_path>
<access_control_path>/mnt/ssd1/clickhouse/access/</access_control_path>
<format_schema_path>/mnt/ssd... | 2020-10-06T08:30:07 | 2020-10-06T08:30:07 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 704,158,159 | 15,644 | rentocika | 1. In first disable selinux in `/etc/sysconfig/selinux` and reboot:
`SELINUX=disabled`
2. Before installing ClickHouse install chkconfig:
`dnf install chkconfig -y`
3. Install ClickHouse.
4. Enjoy =) | 2020-10-06T09:48:12 | 2020-10-06T11:04:47 | {
"+1": 1
} | NONE |
ClickHouse | ClickHouse | 704,435,132 | 15,670 | den-crane | >No lines found except for the error log on one of the replicas.
check previous logs ....log.4.gz
zgrep 1601305200_29_29_0 /var/log/clickhouse-server/clickhouse-server.log*.gz
>Could it be caused by restarting ZK cluster nodes simultaneously instead of one-by-one
Yes it could. Always restart ZK cluster... | 2020-10-06T17:35:11 | 2020-10-06T17:35:49 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 704,476,638 | 15,596 | OmarBazaraa | > --connect_timeout_with_failover_ms arg Connection timeout for selecting first healthy replica.
>
> For example you have 3 replicas. R1, R2, R3.
> R1 is dead right now. And connect_timeout_with_failover_ms = 1000. On distributed query CH will try to connect to R1. Will wait for 1000ms and then re-try to connect to... | 2020-10-06T18:42:42 | 2020-10-06T18:43:03 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 703,545,164 | 15,561 | alesapin | Failures not related to changes. | 2020-10-05T10:29:58 | 2020-10-05T10:29:58 | {} | MEMBER |
ClickHouse | ClickHouse | 703,550,935 | 15,525 | alesapin | Don't needed | 2020-10-05T10:42:02 | 2020-10-05T10:42:02 | {} | MEMBER |
ClickHouse | ClickHouse | 703,620,176 | 15,236 | alesapin | tests ok again | 2020-10-05T13:08:23 | 2020-10-05T13:08:23 | {} | MEMBER |
ClickHouse | ClickHouse | 703,692,938 | 15,608 | alexey-milovidov | > test_ttl_move
Flaky test. | 2020-10-05T15:05:44 | 2020-10-05T15:05:44 | {} | MEMBER |
ClickHouse | ClickHouse | 703,764,779 | 15,552 | UnamedRus | From chproxy [github](https://github.com/Vertamedia/chproxy)
```
Be careful when configuring limits, allowed networks, passwords etc. By default chproxy tries detecting the most obvious configuration errors such as allowed_networks: ["0.0.0.0/0"] or sending passwords via unencrypted HTTP.
Special option hack_me_pl... | 2020-10-05T17:07:32 | 2020-10-05T17:07:32 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 704,108,953 | 15,538 | 4ertus2 | There's `helpers.uexpect.ExpectTimeoutError: Timeout 20.000s` error in integration test TestLiveViewOverDistributedSuite.
There's no IN or JOIN in test queries, so PR should not affect the test.
None of perf test errors could be caused by PR. | 2020-10-06T08:15:04 | 2020-10-06T08:17:35 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 704,220,277 | 15,451 | KochetovNicolai | ```
set with_global = 1;
explain syntax with 1 as x select 1 + y from (select x + 2 as y) union all select x + y from (select 1 as y);
EXPLAIN SYNTAX
WITH 1 AS x
SELECT 1 + y
FROM
(
SELECT x + 2 AS y
)
UNION ALL
SELECT x + y
FROM
(
SELECT 1 AS y
)
Received exception from server (version ... | 2020-10-06T11:59:47 | 2020-10-06T11:59:47 | {} | MEMBER |
ClickHouse | ClickHouse | 704,260,953 | 15,656 | alesapin | Flaky check passed 6 times. Probably error fixed. | 2020-10-06T13:15:25 | 2020-10-06T13:15:25 | {} | MEMBER |
ClickHouse | ClickHouse | 704,449,498 | 15,596 | den-crane | > `<path>/mnt/ssd1/clickhouse/</path>`
>...
> `<disk_ssd_1> <path>/mnt/ssd1/clickhouse/</path> </disk_ssd_1>`
AFAIK it's a misconfiguration. This disk should named default in storage_configuration not disk_ssd_1 or you should create another folder `/mnt/ssd1/clickhousexxxx` for ` <path>`
expected:
```
... | 2020-10-06T18:01:14 | 2020-10-06T18:10:35 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 703,592,528 | 15,572 | igor9292 | > Да, похоже на проблемы с парсингом.
>
> Ищите записи в логах с severity Error примерно в то же время что и "Re-joining claimed consumer after failure" (возможно чуть позднее, емнип в них не будет упоминаться Kafka).
Нет ничего похожего, дебаг надо включить?
вот при включенном дебаге, что есть в логах
Oct... | 2020-10-05T12:14:54 | 2020-10-05T12:14:54 | {} | NONE |
ClickHouse | ClickHouse | 703,596,759 | 15,552 | filimonov | `SYSTEM STOP GLOBAL ACCESS`
I bet in that scenario valid users may also access ClickHouse via the internet using default passwordless user :hankey:
So after 'white-hat hacker' will run that command, the valid user most probably will lose the connection to clickhouse...
And the 'fast fix' the user will apply wi... | 2020-10-05T12:23:19 | 2020-10-05T12:25:02 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 704,043,415 | 15,552 | alexey-milovidov | > Will will have both :)
Yes, I'm also thinking this way. But there is one argument remaining...
These kind of ransom attacks are rare but they are destructive.
Security scans can operate on better scale. It's easy to scan all IPv4 addresses for specific port in minutes.
Maybe it's better to emit a predefined... | 2020-10-06T05:50:09 | 2020-10-06T05:50:09 | {} | MEMBER |
ClickHouse | ClickHouse | 704,076,501 | 15,642 | azat | >If at least one index is required then better to name it force_data_skipping_index?
Nope, all of them should be in use (and the test shows this, also added a few examples to the documentation to underline this there too). | 2020-10-06T07:06:34 | 2020-10-06T07:06:34 | {} | MEMBER |
ClickHouse | ClickHouse | 704,093,546 | 13,832 | xDam13N | 20.9.2.20 Clickhouse version has the same bug. Is there any workaround for this problem? | 2020-10-06T07:44:26 | 2020-10-06T07:44:43 | {} | NONE |
ClickHouse | ClickHouse | 703,545,581 | 15,559 | alesapin | Failures not related to changes. | 2020-10-05T10:30:54 | 2020-10-05T10:30:54 | {} | MEMBER |
ClickHouse | ClickHouse | 703,636,996 | 15,607 | filimonov | It will allow to create the dictionary, but once you will try to access that - you will get an exception:
```
DROP DATABASE IF EXISTS issue15607;
CREATE DATABASE issue15607;
CREATE TABLE issue15607.dict_source engine=Log as select 'a' as code_affilie, toUInt32(1) as partner_id;
CREATE DICTIONARY issue15607.dict
(... | 2020-10-05T13:36:23 | 2020-10-05T13:37:00 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 703,780,575 | 15,596 | den-crane | > > `hostname -f` -- can be overridden in config.xml using `interserver_http`
>
> It worked.
> Thank you so much, Denny.
> But do you have an explanation for [this](https://github.com/ClickHouse/ClickHouse/issues/15596#issuecomment-703591540)?
```
Part `202009_0_0_0` was found on disk `default` which is not ... | 2020-10-05T17:36:38 | 2020-10-05T17:36:38 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 703,902,555 | 15,638 | filimonov | Today i was playing a bit with arm build ( 20.10.1.4844 from [CI](https://clickhouse.tech/docs/en/getting-started/install/#from-binaries-non-linux)) and also saw those exceptions. | 2020-10-05T21:36:25 | 2020-10-05T21:36:25 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 704,116,532 | 15,638 | yard | I am a bit concerned since can’t this lead to more severe issues like exceptions not handling properly (since stack unwinding is involved there as well)? | 2020-10-06T08:29:54 | 2020-10-06T08:29:54 | {} | NONE |
ClickHouse | ClickHouse | 704,387,335 | 11,594 | crakjie | What if the column is already Nullable ? | 2020-10-06T16:10:48 | 2020-10-06T16:10:48 | {} | NONE |
ClickHouse | ClickHouse | 704,459,673 | 15,670 | keliss | I still have no idea why ZK restart caused CH to suddenly retry downloading that part given it is pretty old and it was supposed to not exist. | 2020-10-06T18:18:01 | 2020-10-06T18:56:40 | {} | NONE |
ClickHouse | ClickHouse | 703,560,746 | 15,590 | abyss7 | Do I understand correctly that this PR was intentionally merged in a broken state? | 2020-10-05T11:04:09 | 2020-10-05T11:04:09 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 703,586,070 | 15,596 | OmarBazaraa | This solves connecting to another node using the `remote` function.
But the replication problem still exists, when I insert in a region, the data is not replicated in the other region. | 2020-10-05T12:01:31 | 2020-10-05T12:01:31 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 703,592,173 | 15,595 | alesapin | Build OK, Compatibility OK, ThinLTO ok. | 2020-10-05T12:14:10 | 2020-10-05T12:14:10 | {} | MEMBER |
ClickHouse | ClickHouse | 704,179,917 | 14,916 | 4ertus2 | Works in master
```
SELECT
n1.number AS number1,
n1.number
FROM numbers(1) AS n1
INNER JOIN numbers(1) AS n2 ON n1.number = n2.number
INNER JOIN numbers(1) AS n3 ON n1.number = n3.number
┌─number1─┬─n1.number─┐
│ 0 │ 0 │
└─────────┴───────────┘
```
Try to set `multiple_joins_rewriter... | 2020-10-06T10:31:15 | 2020-10-06T10:31:15 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 704,216,588 | 15,638 | alexey-milovidov | What profiler is doing is "asynchronous unwinding" - it's more tricky than unwinding on exception.
Nevertheless, if libunwind prints these messages - it is able to detect that something is wrong with unwind tables.
If it will happen during exception handling, std::terminate will be called and the server will be ter... | 2020-10-06T11:52:10 | 2020-10-06T11:52:10 | {} | MEMBER |
ClickHouse | ClickHouse | 704,400,777 | 15,670 | keliss | No lines found except for the error log on one of the replicas. Got it, thank you. Maybe you can give advice on how to avoid this in ZK? I don't have much experience with it. Could it be caused by restarting ZK cluster nodes simultaneously instead of one-by-one? | 2020-10-06T16:33:59 | 2020-10-06T16:33:59 | {} | NONE |
ClickHouse | ClickHouse | 704,447,229 | 15,670 | den-crane | >2020.09.28 18:59:52.864637 [ 267261 ] {} <Debug> Table B: Removing part from filesystem 1601305200_29_29_0
Check server's `uptime`. This probably is a result of a spontaneous hardware reboot. | 2020-10-06T17:57:25 | 2020-10-06T17:57:37 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 703,545,943 | 15,611 | filimonov | It's just a notice saying your kernel doesn't have a newer API, and jemalloc used the fallback setting to determine the number of CPUs. Besides that, everything should work normally.
You can just ignore it, or use a newer version of OS/kernel. Centos 6 is now in maintenance mode and its hard EOL is November 30th, 2... | 2020-10-05T10:31:33 | 2020-10-05T10:31:33 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 703,620,802 | 15,596 | den-crane | >Host not found: ch-01-08-2.qlycpey12qyudls00h5anwsfzh.cx.internal.cloudapp.net
DNS problem?
On start a replica announces itself in ZK. One of your replicas announced as `hostname -f` as ch-01-08-2.qlycpey12qyudls00h5anwsfzh.
`hostname -f` -- can be overridden in config.xml using `interserver_http` | 2020-10-05T13:09:34 | 2020-10-05T13:10:34 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 704,095,962 | 15,642 | azat | >Yandex synchronization check (only for Yandex employees)
Hm it fails, although I removed structured binding (9d9877a204aa5c3a4e73462e074db0b9b5455ed3), @alexey-milovidov can you please take a look? | 2020-10-06T07:49:45 | 2020-10-06T07:50:32 | {} | MEMBER |
ClickHouse | ClickHouse | 704,346,496 | 15,645 | KochetovNicolai | fuzzer is reported here: #15683 | 2020-10-06T15:21:31 | 2020-10-06T15:21:31 | {} | MEMBER |
ClickHouse | ClickHouse | 703,544,992 | 15,569 | alesapin | Failures not related to changes. | 2020-10-05T10:29:34 | 2020-10-05T10:29:34 | {} | MEMBER |
ClickHouse | ClickHouse | 703,558,737 | 15,591 | ghost | EATE TABLE [IF NOT EXISTS] [db.]table_name [ON CLUSTER cluster]
(
name1 [type1] [DEFAULT|MATERIALIZED|ALIAS expr1] [TTL expr1],
name2 [type2] [DEFAULT|MATERIALIZED|ALIAS expr2] [TTL expr2],
...
INDEX index_name1 expr1 TYPE type1(...) GRANULARITY value1,
INDEX index_name2 expr2 TYPE type2(...) GRANULARITY value2
... | 2020-10-05T11:00:03 | 2020-10-05T11:00:03 | {} | NONE |
ClickHouse | ClickHouse | 703,569,592 | 11,298 | abyss7 | Status for now (0_stateless): `923 failed, 982 passed in 440.90s (0:07:20)` | 2020-10-05T11:23:51 | 2020-10-05T11:23:51 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 703,618,483 | 15,596 | OmarBazaraa | > **select * from system.replication_queue** on all replicas
I found this:
```
SELECT *
FROM remote('ch-01-08-3.my-domain.com', system.replication_queue)
FORMAT Vertical
Row 1:
──────
database: my_db
table: my_table_local
replica_name: 3
position: ... | 2020-10-05T13:05:25 | 2020-10-05T13:09:01 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 703,805,337 | 15,638 | yard | Update: seems to be related to query profiler - disabling it via setting `query_profiler_real_time_period_ns` / `query_profiler_cpu_time_period_ns` to 0 makes the messages go away. | 2020-10-05T18:21:35 | 2020-10-05T18:21:35 | {} | NONE |
ClickHouse | ClickHouse | 704,138,509 | 15,617 | alesapin | test failures not related to changes. | 2020-10-06T09:11:24 | 2020-10-06T09:11:24 | {} | MEMBER |
ClickHouse | ClickHouse | 703,605,230 | 15,612 | alesapin | Docker push OK. | 2020-10-05T12:40:46 | 2020-10-05T12:40:46 | {} | MEMBER |
ClickHouse | ClickHouse | 703,684,221 | 15,608 | danlark1 | Does not seem related to the changes at all
```
> 'Client failed! Return code: {}, stderr: {}'.format(self.process.returncode, stderr))
E helpers.client.QueryRuntimeException: Client failed! Return code: 209, stderr: Code: 209. DB::NetException: Timeout: connect timed out: 172.19.0.6:9000 (... | 2020-10-05T14:52:31 | 2020-10-05T14:52:31 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 703,739,077 | 12,136 | alexey-milovidov | https://en.wikipedia.org/wiki/ISO_week_date#Relation_with_the_Gregorian_calendar | 2020-10-05T16:21:54 | 2020-10-05T16:21:54 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 703,769,877 | 15,552 | alexey-milovidov | It's typical to make ClickHouse listen `::` or `0.0.0.0` inside company network that's not accessible outside. | 2020-10-05T17:17:01 | 2020-10-05T17:17:01 | {} | MEMBER |
ClickHouse | ClickHouse | 704,138,312 | 15,616 | alesapin | Test failures not related to changes. | 2020-10-06T09:11:03 | 2020-10-06T09:11:03 | {} | MEMBER |
ClickHouse | ClickHouse | 704,474,974 | 15,596 | OmarBazaraa | > or you can omit this default at all because it's already exists from <path section with name default and can point it as `<disk>default</disk>`
It worked. Thanks a lot, Denny. Although, the displayed error message seems strange. xD | 2020-10-06T18:40:25 | 2020-10-06T18:40:25 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 703,589,781 | 15,538 | 4ertus2 | `DB::Exception: Database test doesn't exist.` in python live view tests. | 2020-10-05T12:09:13 | 2020-10-05T12:09:13 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 703,649,077 | 15,611 | thangntt | > It's just a notice saying your kernel doesn't have a newer API, and jemalloc used the fallback setting to determine the number of CPUs. Besides that, everything should work normally.
>
> You can just ignore it, or use a newer version of OS/kernel. Centos 6 is now in maintenance mode and its hard EOL is November 30... | 2020-10-05T13:56:11 | 2020-10-05T13:56:11 | {} | NONE |
ClickHouse | ClickHouse | 704,038,039 | 15,552 | alexey-milovidov | > At clickhouse start, clickhouse tries to reach itself by interfaces via some preset proxy from the internet
Already been considered, but it looks like a worse solution to enable by default: it's suspicious for database server to ping some host on the internet. | 2020-10-06T05:33:42 | 2020-10-06T05:33:42 | {} | MEMBER |
ClickHouse | ClickHouse | 704,069,634 | 15,642 | azat | >I've fixed "Arcadia".
Will try a better fix (w/o preprocessor) | 2020-10-06T06:51:14 | 2020-10-06T06:51:14 | {} | MEMBER |
ClickHouse | ClickHouse | 704,208,397 | 15,639 | GrigoryPervakov | If you use S3 disk as first storage, this file uploaded to cloud and then deleted after merge. | 2020-10-06T11:33:43 | 2020-10-06T11:33:43 | {} | MEMBER |
ClickHouse | ClickHouse | 704,441,050 | 15,670 | keliss | Yep, this way I was able to find some more information about this part. I guess, the most interesting lines are
```
/var/log/clickhouse-server/clickhouse-server.log.0.gz:2020.10.06 11:48:19.644585 [ 363667 ] {} <Error> Table B: Removing locally missing part from ZooKeeper and queueing a fetch: 1601305200_29_29_0
/va... | 2020-10-06T17:46:20 | 2020-10-06T17:46:20 | {} | NONE |
ClickHouse | ClickHouse | 704,230,161 | 15,638 | yard | Thank you for the explanation! Any ideas on what we can try on our end (e.g. specific versions of libunwind, build flags, extra logging etc)? Happy to experiment and help with finding the root cause since we are really want to run our CH instance on ARM in production environment – apart from the above hiccups it seems ... | 2020-10-06T12:19:25 | 2020-10-06T12:19:25 | {} | NONE |
ClickHouse | ClickHouse | 704,295,399 | 15,667 | alesapin | Waiting for integration tests | 2020-10-06T14:10:04 | 2020-10-06T14:10:04 | {} | MEMBER |
ClickHouse | ClickHouse | 704,457,013 | 15,670 | keliss | 209 days, so it is not an option. Though it actually looks like a shutdown of the CH server itself:
```
2020.09.28 18:59:52.655514 [ 267261 ] {} <Information> Application: Closed connections.
2020.09.28 18:59:52.725762 [ 267261 ] {} <Information> Application: Shutting down storages.
...
2020.09.28 18:59:52.828610 ... | 2020-10-06T18:13:52 | 2020-10-06T18:13:52 | {} | NONE |
ClickHouse | ClickHouse | 703,544,743 | 15,566 | alesapin | Stress reported, other failures not related. | 2020-10-05T10:29:04 | 2020-10-05T10:29:04 | {} | MEMBER |
ClickHouse | ClickHouse | 703,618,054 | 15,543 | CurtizJ | ```The test 'read_in_order_many_parts' is too slow to run as a whole. Investigate whether the create and fill queries can be sped up``` looks suspicios, but it's already broken in master. Actually this test wasn't slowed down.
| 2020-10-05T13:04:35 | 2020-10-05T13:04:35 | {} | MEMBER |
ClickHouse | ClickHouse | 703,640,791 | 15,596 | den-crane | `<remote_servers>` -- cluster description is not used for Replication (Replicated engines). | 2020-10-05T13:42:27 | 2020-10-05T13:42:27 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 703,864,180 | 15,552 | UnamedRus | >It's typical to make ClickHouse listen :: or 0.0.0.0 inside company network that's not accessible outside.
At clickhouse start, clickhouse tries to reach itself by interfaces via some preset proxy from the internet, it wouldn't work if clickhouse set behind open to everyone load balancer. And that ofc can be disabl... | 2020-10-05T20:15:17 | 2020-10-05T20:15:17 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 704,102,805 | 15,642 | alexey-milovidov | > Hm it fails, although I removed structured binding
It's about `contains` method.
Don't worry and just go ahead with the ifdef. | 2020-10-06T08:03:31 | 2020-10-06T08:03:31 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 704,272,190 | 15,605 | dependabot-preview[bot] | OK, I won't notify you again about this release, but will get in touch when a new version is available. If you'd rather skip all updates until the next major or minor version, let me know by commenting `@dependabot ignore this major version` or `@dependabot ignore this minor version`.
If you change your mind, just re-... | 2020-10-06T13:33:40 | 2020-10-06T13:33:40 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 704,272,223 | 15,606 | dependabot-preview[bot] | OK, I won't notify you again about this release, but will get in touch when a new version is available. If you'd rather skip all updates until the next major or minor version, let me know by commenting `@dependabot ignore this major version` or `@dependabot ignore this minor version`.
If you change your mind, just re-... | 2020-10-06T13:33:44 | 2020-10-06T13:33:44 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 704,384,365 | 10,682 | vivarum | https://github.com/ClickHouse/ClickHouse/pull/15685 | 2020-10-06T16:06:00 | 2020-10-06T16:06:00 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 704,399,974 | 15,684 | alexey-milovidov | It's not an overflow. These metrics are negative by legitimate reason - they count memory usage by the difference of allocations and frees in the context of `BackgroundProcessingPool`. But it's often the case when memory is allocated on INSERT but freed during merge in `BackgroundProcessingPool`, so the metric counts m... | 2020-10-06T16:32:34 | 2020-10-06T16:32:34 | {} | MEMBER |
ClickHouse | ClickHouse | 694,629,000 | 14,953 | zhang2014 | Why not backport? | 2020-09-18T03:36:44 | 2020-09-18T03:36:44 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 694,494,015 | 13,393 | vitlibar | https://clickhouse-test-reports.s3.yandex.net/14880/d99a011d07bee19e20d2626de02b0a4bd5d6a3fc/fuzzer/report.html#fail1 | 2020-09-17T20:53:09 | 2020-09-17T20:53:09 | {} | MEMBER |
ClickHouse | ClickHouse | 694,622,421 | 14,786 | bohutang | Hello,
Please check your version by 'select version()'.
I have checked the build from master and can't reproduce the issue yet follow your steps.
1. source code:
```
$ git log -2
commit 76a3cc2dae22ea9ebc9b82776ef07ea95a996e0c (HEAD -> master, origin/master, origin/HEAD)
Merge: 6a1e4ccdf9 22bd22702e
Author:... | 2020-09-18T03:10:54 | 2020-09-18T03:10:54 | {} | CONTRIBUTOR |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.