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 | 648,295,019 | 11,774 | azat | >I'm interested in this pull request. It is better to use the same allocator on all systems.
@alexey-milovidov freebsd is supported now (simple check passed), but I disabled jemalloc for darwin for now (since there are some issues - f76511668f3bda374f66d1176cc9d85d3259e117).
BTW any thoughts on adding binaries wi... | 2020-06-23T17:08:26 | 2020-06-23T17:08:26 | {} | MEMBER |
ClickHouse | ClickHouse | 648,556,974 | 8,659 | den-crane | jdbc-driver
https://github.com/ClickHouse/clickhouse-jdbc/blob/5ff82e2f902450ad5679d8a654f7557d2220c1cd/src/main/java/ru/yandex/clickhouse/settings/ClickHouseQueryParam.java#L285-L288 | 2020-06-24T03:14:23 | 2020-06-24T03:14:23 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,745,322 | 11,902 | filimonov | Inserts into original table & matview is not atomic (yet).
The issue may be due to some temporary errors / block deduplication on replicated tables / loosing the insert because MV executed in sequence an some of prev MV failed.
Check the settings
- parallel_view_processing
- deduplicate_blocks_in_dependent_mat... | 2020-06-24T10:45:53 | 2020-06-24T10:50:48 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,823,164 | 11,901 | bohutang | > Also please look at #11089
> We can easily close it with this PR :)
Done | 2020-06-24T13:33:59 | 2020-06-24T13:33:59 | {
"+1": 2
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,137,535 | 11,887 | den-crane | If you insert into Distributed table then your inserted data will wait for node-A in a Distributed table folder. Data will be unavailable for queries until it inserted into MergeTree at node-A. | 2020-06-23T13:13:06 | 2020-06-23T13:13:06 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,286,481 | 11,890 | filimonov | Duplicate of https://github.com/ClickHouse/ClickHouse/issues/5736#issuecomment-505543614 | 2020-06-23T16:52:11 | 2020-06-23T16:52:11 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,440,145 | 10,419 | filimonov | We can't help you unless you provide minimal reproducable sample (file with sample data in your case and ClickHouse version) | 2020-06-23T21:34:06 | 2020-06-23T21:34:06 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,444,333 | 10,884 | filimonov | How many brokers do you have, aren't they over loaded? Are they in the same DC?
Did you try the approach with `request.required.acks`?
Check https://stackoverflow.com/questions/54039216/how-come-kafka-fails-to-commit-offset-for-a-particular-partition and other similar questions
Give a try with clickhouse-20.5 - t... | 2020-06-23T21:44:25 | 2020-06-23T21:44:25 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,451,832 | 10,768 | filimonov | There are
- toDateTimeOrNull
- toDateTimeTimeOrZero
- parseDateTimeBestEffort
- parseDateTimeBestEffortOrNull
- parseDateTimeBestEffortOrZero
Pick the one depending or your needs. Variants with 'or' don't throw exceptions on bad strings.
| 2020-06-23T22:03:51 | 2020-06-23T22:03:51 | {
"+1": 3
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,730,304 | 11,861 | wallace-clickhouse | Yeah,thank you for your reply.
I don't want to write a plaintext password in the file because my customer doesn't allow me to write plaintext password in any file and will scan my file system to find if plaintext password exists.
"And I guess now you creates own Cluster for each query? (since you need to pass passwor... | 2020-06-24T10:12:46 | 2020-06-24T10:12:46 | {} | NONE |
ClickHouse | ClickHouse | 648,822,059 | 11,861 | den-crane | @wallace-clickhouse did you try to pass a password using from_env ?
`
<password from_env="SECRET" />`
and then set env variable SECRET for CH process ? | 2020-06-24T13:32:01 | 2020-06-24T13:32:01 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,204,787 | 10,417 | vadv | also did not find a way to access AWS S3 in the non-default region | 2020-06-23T14:36:46 | 2020-06-23T14:36:46 | {} | NONE |
ClickHouse | ClickHouse | 648,268,812 | 11,887 | den-crane | >the tables count in local database on each hosts not equal
You mean replicas?
>tables count
You mean rows count? | 2020-06-23T16:19:15 | 2020-06-23T16:19:15 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,428,504 | 11,879 | filimonov | Only constants can be substituted, not column names (by design & no plans to change that) | 2020-06-23T21:09:58 | 2020-06-23T21:09:58 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,435,823 | 11,897 | alexey-milovidov | It points to "special build check" for some reason... @alesapin | 2020-06-23T21:23:38 | 2020-06-23T21:23:38 | {} | MEMBER |
ClickHouse | ClickHouse | 648,465,384 | 11,894 | akuzm | ```
$ curl --compressed https://clickhouse-builds.s3.yandex.net/11894/f206d91072a05d74d3c236a4917b58ab1df28e62/clang-10-aarch64_relwithdebuginfo_none_bundled_unsplitted_disable_False_binary/clickhouse -o clickhouse
$ /system/bin/linker64 $(readlink -f ./clickhouse)
error: "/data/data/com.termux/files/home/clickhouse... | 2020-06-23T22:41:17 | 2020-06-23T22:41:17 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,469,017 | 9,595 | filimonov | @tavplubix какой статус ишью? | 2020-06-23T22:53:33 | 2020-06-23T22:53:33 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,589,069 | 11,428 | PhantomPhreak | This bug affect us quite hard and forcing us to set `readonly=2` when it's not really necessary :( | 2020-06-24T05:10:19 | 2020-06-24T05:10:19 | {} | NONE |
ClickHouse | ClickHouse | 648,670,362 | 10,417 | filimonov | Of course - there are `prestable` and `testing` channels on both rpm & deb repos.
Deb
https://repo.clickhouse.tech/deb/prestable/main/
https://repo.clickhouse.tech/deb/testing/main/
Rpm
https://repo.yandex.ru/clickhouse/rpm/prestable/x86_64/
https://repo.yandex.ru/clickhouse/rpm/prestable/x86_64/
| 2020-06-24T08:15:08 | 2020-06-24T08:15:08 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,677,832 | 10,417 | vadv | @filimonov
```
INSERT INTO FUNCTION s3('https://bucket.s3-ap-southeast-1.amazonaws.com/clickhouse-backup/envoy/2020/06/23/envoy.csv', '<key>', '<secret>', 'CSV', 'response_code String') VALUES
Received exception from server (version 20.5.1):
Code: 499. DB::Exception: Received from localhost:9001. DB::Exception... | 2020-06-24T08:30:07 | 2020-06-24T08:30:07 | {
"+1": 1,
"confused": 3
} | NONE |
ClickHouse | ClickHouse | 648,679,298 | 10,417 | vadv | ```
2020.06.24 08:27:36.270452 [ 911 ] {df3a58dd-970f-407a-bc37-3935dcceb3f6} <Warning> AWSClient: AWSClient: If the signature check failed. This could be because of a time skew. Attempting to adjust the signer.
2020.06.24 08:27:36.270478 [ 911 ] {df3a58dd-970f-407a-bc37-3935dcceb3f6} <Warning> AWSClient: AWSClient: ... | 2020-06-24T08:32:58 | 2020-06-24T08:32:58 | {} | NONE |
ClickHouse | ClickHouse | 648,695,324 | 10,417 | vadv | if clickhouse uses the AWS SDK, why are `AWS_ *` env variables not being processed? | 2020-06-24T09:04:15 | 2020-06-24T09:04:15 | {} | NONE |
ClickHouse | ClickHouse | 648,744,460 | 10,221 | alexey-milovidov | No.
The original issue is in:
- bogus error message;
- slightly less efficient parts removal. | 2020-06-24T10:43:56 | 2020-06-24T10:43:56 | {} | MEMBER |
ClickHouse | ClickHouse | 648,257,448 | 11,541 | CurtizJ | This bug was fixed in #10715. Will be available in `20.3.12`, `20.4.6` and `20.1.16`. | 2020-06-23T15:59:55 | 2020-06-23T15:59:55 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 648,153,972 | 11,876 | den-crane | it is impossible with AggregatingMergeTree. AggregatingMergeTree does not know that you need arrayReduce('sumMap' and just rewrites a value in _discrete column by function ANY on each insert.
You can use SummingMT and nested https://clickhouse.tech/docs/en/engines/table-engines/mergetree-family/summingmergetree/#nes... | 2020-06-23T13:38:26 | 2020-06-23T18:18:24 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,425,271 | 10,417 | filimonov | It should be fixed in #11036
Can you try 20.5-prestable? | 2020-06-23T21:07:05 | 2020-06-23T21:07:05 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,775,363 | 11,300 | alexey-milovidov | @blinkov If I understand correctly it means that the cache for stack traces was made ineffective, so we got performance degradation + memory leak. I'm glad that it is fixed and we have a test that showed the issue.
It is time for you to say respect to our testing infrastructure and to all the developers who spend th... | 2020-06-24T11:56:35 | 2020-06-24T12:06:06 | {} | MEMBER |
ClickHouse | ClickHouse | 648,823,001 | 11,906 | qoega | Not literally same problem, but similar reason as we have no context why we tried to convert string to Uint8
`Cannot convert string }Göbw2kUjaGMx꠪^bd to type UInt8`
``` sql
Code: 53, e.displayText() = DB::Exception: Cannot convert string }Göbw2kUjaGMx꠪^bd to type UInt8 (version 20.5.1.1) (from [::1]:47368) (in que... | 2020-06-24T13:33:43 | 2020-06-24T13:33:43 | {} | MEMBER |
ClickHouse | ClickHouse | 648,148,987 | 8,982 | alesapin | The bug seems to be fixed in all supported versions (20.1, 20.3, 20.4, etc). | 2020-06-23T13:30:57 | 2020-06-23T13:30:57 | {} | MEMBER |
ClickHouse | ClickHouse | 648,673,220 | 8,659 | filimonov | @kassak for me the best solution is to put the setting in lowercase if possible to avoid problems with lowercasing. If you have better ideas / options - please send proposals (better in form of pull request) into https://github.com/ClickHouse/ClickHouse-jdbc project | 2020-06-24T08:20:44 | 2020-06-24T08:20:44 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,720,835 | 11,894 | alexey-milovidov | Maybe write -pie explicitly? | 2020-06-24T09:53:51 | 2020-06-24T09:53:51 | {} | MEMBER |
ClickHouse | ClickHouse | 648,140,910 | 11,879 | den-crane | But it against the idea that parameters allows to define values only. | 2020-06-23T13:18:27 | 2020-06-23T13:18:27 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,384,186 | 11,498 | filimonov | > And let's assume that "proxy" user has only PROXY grants, and nothing more, but it does makes much sense in terms of secure, since it can "login" into any user anyway
That just to enforce identifucation, i.e. to forbid queries from proxy user if he don't send in which name he want to execute the command.
Actual... | 2020-06-23T19:54:36 | 2020-06-23T19:54:36 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,476,180 | 9,243 | filimonov | Most probably it's the same as #9850 #9873
Please update. | 2020-06-23T23:15:53 | 2020-06-23T23:15:53 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,489,823 | 5,932 | filimonov | Not enough information here to diagnose, seems related to very old version or some local issues.
Clickhouse just works on Centos / RedHat, and there is official repo with rpm packages. | 2020-06-24T00:04:49 | 2020-06-24T00:04:49 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,510,808 | 9,253 | filimonov | That sounds very similar to https://github.com/ClickHouse/ClickHouse/issues/9088
@4ertus2 can you check if your fix also resolve it? | 2020-06-24T00:31:24 | 2020-06-24T00:31:24 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,656,689 | 11,887 | mjodeh | I guess what @arashkhavari meant is that suppose that we have 3 shards and the insertion is being distributed between them then for any reason one of the shards went down. In this case the load will be distributed over 2 shards? and when the 3rd shard is back online does it go back to 3 shards? and the data which was n... | 2020-06-24T07:49:11 | 2020-06-24T07:49:11 | {} | NONE |
ClickHouse | ClickHouse | 648,755,930 | 11,902 | filimonov | Also https://github.com/ClickHouse/ClickHouse/issues/11909 | 2020-06-24T11:10:12 | 2020-06-24T11:10:12 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,808,534 | 11,901 | alexey-milovidov | Also please look at #11089
We can easily close it with this PR :) | 2020-06-24T13:07:47 | 2020-06-24T13:07:47 | {} | MEMBER |
ClickHouse | ClickHouse | 648,809,073 | 10,764 | alexey-milovidov | Fixed by @ucasFL in #11878. | 2020-06-24T13:08:43 | 2020-06-24T13:08:43 | {} | MEMBER |
ClickHouse | ClickHouse | 648,383,967 | 11,875 | alexey-milovidov | I will also add a functional test to demonstrate what code was totally and absolutely wrong. | 2020-06-23T19:54:09 | 2020-06-23T19:54:09 | {} | MEMBER |
ClickHouse | ClickHouse | 648,463,565 | 10,221 | filimonov | Isn't it a bugfix that should be backported? | 2020-06-23T22:35:21 | 2020-06-23T22:35:21 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,486,238 | 10,521 | filimonov | Confusing error message was changed #10828 and now it's warning, not an error #11687 | 2020-06-23T23:51:33 | 2020-06-23T23:51:33 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,560,950 | 11,880 | boy-xiaozhang | > 抱歉，无法根据所提供的信息为您提供帮助。检查日志，在这种情况下不可能有有效的日志（除非您配置错误）
Two versions of ClickHouse were found to be incompatible have already solved,thank you for your help | 2020-06-24T03:28:14 | 2020-06-24T03:28:14 | {} | NONE |
ClickHouse | ClickHouse | 648,617,506 | 10,417 | vadv | >Can you try 20.5-prestable?
problem with a long build:
```
23:13
[123/7955] Building CXX
...
08:55
[6346/7955] Building CXX object
```
is there any pre-build? :) | 2020-06-24T06:22:56 | 2020-06-24T06:22:56 | {} | NONE |
ClickHouse | ClickHouse | 648,624,601 | 11,901 | bohutang | CC @alexey-milovidov
Please give a 'can be tested' lable, thanks. | 2020-06-24T06:40:28 | 2020-06-24T06:40:28 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 633,008,155 | 11,003 | azat | >openshift sets a other id
How?
>here is a example of an working docker image for clickhouse:
You have tons of `ch*` (chown/chmod) commands, does not looks as a basic image
Original question that I was answering:
> is there a other solution available to setup the docker images without root ?
If you ne... | 2020-05-23T08:40:31 | 2020-06-24T08:25:13 | {} | MEMBER |
ClickHouse | ClickHouse | 648,678,470 | 10,884 | brefre | Hi,
I have 5 brokers with ~90% idle CPU (16 core systems), 80/96GiB ram available, in two racks next to each other. They do not look overloaded.
__consumer_offsets have a replication factor of 4, and min.insync.replicas is 2
I have gotten some progress by setting kafka_num_consumers = 1, that got rid of most ... | 2020-06-24T08:31:24 | 2020-06-24T08:31:24 | {} | NONE |
ClickHouse | ClickHouse | 648,824,873 | 11,901 | phuslu | sorry for jumping this thread, is possible add mysql "connection_id()" fake function in this stage?
some mysql frontend tool (e.g. mycli) need "connection_id()" during connect. | 2020-06-24T13:36:51 | 2020-06-24T13:36:51 | {} | NONE |
ClickHouse | ClickHouse | 648,306,430 | 11,498 | azat | >The use case when we need to limit the list of users in which name that proxy can talk is still not very clear for me.
For me to, but I guess that there can be some potential use cases.
For example, suppose you have "local" user (the user who executes remote query) and "proxy" user (the user which is used to log... | 2020-06-23T17:30:00 | 2020-06-23T17:30:00 | {} | MEMBER |
ClickHouse | ClickHouse | 648,448,534 | 10,824 | filimonov | A bit too long opened for crit. Any movements here? | 2020-06-23T21:55:22 | 2020-06-23T21:55:22 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,474,761 | 9,276 | filimonov | The code around mutations was significantly refactored in 20.3. Please update to 20.3 or newer | 2020-06-23T23:10:50 | 2020-06-23T23:10:50 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,631,033 | 11,897 | azat | > pushed again to trigger CI
Same problem | 2020-06-24T06:55:25 | 2020-06-24T06:55:25 | {} | MEMBER |
ClickHouse | ClickHouse | 648,746,073 | 11,804 | alexey-milovidov | It is not possible to track memory usage for subqueries. | 2020-06-24T10:47:37 | 2020-06-24T10:47:37 | {} | MEMBER |
ClickHouse | ClickHouse | 648,810,399 | 11,668 | akuzm | > Thank you @akuzm for your help with fixing the failing test.
You're welcome.
'Yandex third-party checks' are broken for a different reason, will merge now. | 2020-06-24T13:11:13 | 2020-06-24T13:11:13 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,053,419 | 11,876 | xiemeilong | I found it is still not having expected behave. | 2020-06-23T10:25:38 | 2020-06-23T10:25:38 | {} | NONE |
ClickHouse | ClickHouse | 648,308,526 | 11,875 | alexey-milovidov | No surprise: unit test is wrong, getLeastSuperType implementation is also wrong. | 2020-06-23T17:34:02 | 2020-06-23T17:34:02 | {} | MEMBER |
ClickHouse | ClickHouse | 648,465,644 | 9,978 | filimonov | That sounds weird. May be you have some issue with your schema. We need a minimal table schema reproducing the issue. | 2020-06-23T22:42:04 | 2020-06-23T22:42:04 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,533,112 | 11,888 | nicelulu | > We don't read from one fd from multiple threads.
> If we read from multiple threads, file is opened multiple times.
Thanks for your reply, i get it. | 2020-06-24T01:46:33 | 2020-06-24T01:46:33 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,046,708 | 11,815 | alesapin | > 01079_parallel_alter_detach_table_zookeeper timeout https://clickhouse-test-reports.s3.yandex.net/0/d147cd646a1e6cc41d42f92d57f016a5c49d04de/functional_stateless_tests_(thread).html#fail1
Fix in master https://github.com/ClickHouse/ClickHouse/commit/ca552d63940a3652da72c24e65e4aedd8793546a. Table was not attached ... | 2020-06-23T10:12:30 | 2020-06-23T10:12:30 | {} | MEMBER |
ClickHouse | ClickHouse | 648,060,937 | 11,745 | alesapin | ```
020-06-23 12:29:41 00953_moving_functions: [ FAIL ] 0.15 sec. - result differs with reference:
2020-06-23 12:29:41 --- /usr/share/clickhouse-test/queries/0_stateless/00953_moving_functions.reference 2020-06-23 11:24:32.000000000 +0300
2020-06-23 12:29:41 +++ /tmp/c... | 2020-06-23T10:40:11 | 2020-06-23T10:40:32 | {} | MEMBER |
ClickHouse | ClickHouse | 648,154,331 | 11,745 | alesapin | > ```
> 020-06-23 12:29:41 00953_moving_functions: [ FAIL ] 0.15 sec. - result differs with reference:
> 2020-06-23 12:29:41 --- /usr/share/clickhouse-test/queries/0_stateless/00953_moving_functions.reference 2020-06-23 11:24:32.000000000 +0300
> 2020-06-23 12:29:41 ++... | 2020-06-23T13:38:58 | 2020-06-23T13:38:58 | {} | MEMBER |
ClickHouse | ClickHouse | 648,277,691 | 11,883 | alesapin | `00953_moving_functions`flaky. Integration tests -- need to check. | 2020-06-23T16:35:47 | 2020-06-23T16:35:47 | {} | MEMBER |
ClickHouse | ClickHouse | 648,433,460 | 11,897 | azat | >ClickHouse build check — 19/20 builds are OK
Interesting, build check report contains only one row | 2020-06-23T21:18:10 | 2020-06-23T21:18:10 | {} | MEMBER |
ClickHouse | ClickHouse | 648,479,333 | 9,914 | vitlibar | This bug has been fixed in 20.4 and the fix has been backported to 20.3. | 2020-06-23T23:26:25 | 2020-06-23T23:35:43 | {} | MEMBER |
ClickHouse | ClickHouse | 648,550,283 | 11,876 | xiemeilong | Clickhouse is so powerful，Thanks. Should it be rejected when createing unaggregatable column with AggregatingMergeTree? | 2020-06-24T02:51:27 | 2020-06-24T02:51:27 | {} | NONE |
ClickHouse | ClickHouse | 648,813,206 | 11,908 | alexey-milovidov | 01133_max_result_rows
https://clickhouse-test-reports.s3.yandex.net/11908/5a7e921135d27de93cf25479fd46a0eb7527ba3b/functional_stateless_tests_(release,_polymorphic_parts_enabled).html#fail1
Something new. | 2020-06-24T13:16:14 | 2020-06-24T13:16:14 | {} | MEMBER |
ClickHouse | ClickHouse | 648,817,826 | 11,912 | alexey-milovidov | Can you please rebase on master, add a description and change title? | 2020-06-24T13:24:38 | 2020-06-24T13:25:02 | {} | MEMBER |
ClickHouse | ClickHouse | 648,496,011 | 8,353 | filimonov | > curl 'http://localhost:8123' # error if port is not provided default config could be -p 8123:8123
docke
That is standard docker behaviour. No issue here.
> Listen [::]:8123 failed:
Confusing error message was changed #10828 and now it's warning, not an error #11687
> CAP_NET_ADMIN problem
It it not a... | 2020-06-24T00:16:50 | 2020-06-24T00:16:50 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,680,506 | 11,861 | azat | > I don’t want to configure the plaintext password into the cluster configuration file metrika.xml
Why you don't want to write it in the file? Setting permissions to 600 should be secure enough (if you worry about users will not be able to get the cluster topology, then you shouldn't, since there is `system.cluster... | 2020-06-24T08:35:20 | 2020-06-24T08:35:20 | {} | MEMBER |
ClickHouse | ClickHouse | 648,806,379 | 7,154 | Hipska | The example config does not show how the `remote_url_allow_hosts` items should look like. Also nothing about this can be found in the documentation. | 2020-06-24T13:03:44 | 2020-06-24T13:03:44 | {
"+1": 5
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,152,846 | 11,887 | arashkhavari | > If you insert into Distributed table then your inserted data will wait for node-A in a Distributed table folder. Data will be unavailable for queries until it inserted into MergeTree at node-A.
My concern about local databases in shard distributed clickhouse. the tables count in local database on each hosts not eq... | 2020-06-23T13:36:48 | 2020-06-23T13:36:48 | {} | NONE |
ClickHouse | ClickHouse | 648,432,045 | 11,880 | filimonov | Sorry, there is no way to help you based on provided info. Check the logs, it is not possible to have emply logs in such situation (unless you misconfigure something) | 2020-06-23T21:14:59 | 2020-06-23T21:16:00 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,466,796 | 9,914 | filimonov | @vitlibar what is the status here? Should it be closed? | 2020-06-23T22:45:51 | 2020-06-23T22:45:51 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,467,882 | 9,914 | nezed | If this issues is fixed in v20.4 than it is missing in changelog | 2020-06-23T22:49:35 | 2020-06-23T22:49:35 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,479,880 | 8,659 | filimonov | ClickHouse dont lowercase the settings, ClickHouse does not work on Windows. So that sounds like jdbc issue. | 2020-06-23T23:28:17 | 2020-06-23T23:28:17 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,526,769 | 11,861 | wallace-clickhouse | @azat Thank you for your reply!
My scenario is that I don’t want to configure the plaintext password into the cluster configuration file metrika.xml, so I designed to call an external interface to get the password when starting the database, and then the password is kept in memory by an extern variable during the data... | 2020-06-24T01:22:11 | 2020-06-24T01:24:04 | {} | NONE |
ClickHouse | ClickHouse | 648,704,140 | 9,595 | tavplubix | Такой же, сегфолт исправился, но настройки запроса не используются табличной функцией. Это можно исправить, если выполнять табличную функцию в контексте интерпретатора и сохранять таблицу в кэш контекста запроса, но тогда запросы вроде
```
SELECT ... FROM odbc(...) SETTINGS external_table_functions_use_nulls=1
UNION... | 2020-06-24T09:21:26 | 2020-06-24T09:21:45 | {} | MEMBER |
ClickHouse | ClickHouse | 648,081,940 | 11,792 | nikitamikhaylov | This PR was needed for me only for testing. | 2020-06-23T11:20:07 | 2020-06-23T11:20:07 | {} | MEMBER |
ClickHouse | ClickHouse | 648,117,986 | 11,861 | wallace-clickhouse | How can I reference extern variables defined in Server.h and initialized in Server.cpp to Cluster.cpp.Because I am trying to achieve a feature,for which I need to initialize the extern variables in Server.cpp once the server starts up,and then I want get the variable values and reference in Cluster.cpp at anytime whi... | 2020-06-23T12:38:24 | 2020-06-23T12:38:24 | {} | NONE |
ClickHouse | ClickHouse | 648,308,741 | 11,861 | azat | >How can I reference extern variables defined in Server.h and initialized in Server.cpp to Cluster.cpp.
So if I understand you right, then you trying initialize variables that are already initialized in Server.cpp, in this case you will get multiple declarations error at the end
>I have tried to achieve it by '#i... | 2020-06-23T17:34:30 | 2020-06-23T17:34:30 | {} | MEMBER |
ClickHouse | ClickHouse | 648,454,653 | 10,594 | filimonov | Any news? | 2020-06-23T22:10:57 | 2020-06-23T22:10:57 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,675,243 | 11,003 | filimonov | @haarchri in general inside clickhouse container `clickhouse-server` process is executed as `clickhouse` user, which should be the owner of files in `/var/lib/clickhouse` (usually that folder in mounted from outside).
Since you want to make it run in OpenShift i suppose you should have some experience with OpenShif... | 2020-06-24T08:24:43 | 2020-06-24T08:32:18 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 647,474,303 | 11,745 | alesapin | - [x] Fix clang-tidy. | 2020-06-22T12:04:43 | 2020-06-23T08:12:40 | {} | MEMBER |
ClickHouse | ClickHouse | 648,098,208 | 11,605 | alesapin | I'll merge it because it's almost ready. Also, we need to add deploy scripts. | 2020-06-23T11:55:47 | 2020-06-23T11:55:47 | {} | MEMBER |
ClickHouse | ClickHouse | 648,302,546 | 11,888 | alexey-milovidov | We don't read from one fd from multiple threads.
If we read from multiple threads, file is opened multiple times. | 2020-06-23T17:22:44 | 2020-06-23T17:22:44 | {} | MEMBER |
ClickHouse | ClickHouse | 648,439,769 | 11,897 | azat | Ok, [binary](https://clickhouse-builds.s3.yandex.net/11897/ff4a44c39a3a65ae37fe44f79845ea1bed794385/clang-10-darwin_relwithdebuginfo_none_bundled_unsplitted_disable_False_binary/clickhouse) is not there (while for freebsd binary exists), so there is some failure, pushed again to trigger CI | 2020-06-23T21:33:08 | 2020-06-23T21:33:20 | {} | MEMBER |
ClickHouse | ClickHouse | 648,739,869 | 9,595 | filimonov | > сегфолт исправился
@tavplubix Может быть тогда правильнее закрыть этот ишью (если сегфолта уже нет). А открыть другой про SETTINGS? (или просто прилинковать с одному из существующих?) | 2020-06-24T10:33:31 | 2020-06-24T10:33:31 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,772,057 | 10,594 | wutchzone | There is nothing in the error log, only errors related to #10158 as I mentioned above. When we restart the ClickHouse, everything starts working again for few days, then the error appears again.
What is interesting is that it takes about 2 seconds to restart the ClickHouse, when the mentioned error is not present. ... | 2020-06-24T11:48:44 | 2020-06-24T11:48:44 | {} | NONE |
ClickHouse | ClickHouse | 648,209,072 | 11,071 | Akazz | Merged in #11874. | 2020-06-23T14:43:18 | 2020-06-23T14:45:44 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,299,882 | 11,774 | alexey-milovidov | @azat I don't know why they are built without debug information.
We can enable it any time. | 2020-06-23T17:17:45 | 2020-06-23T17:17:45 | {} | MEMBER |
ClickHouse | ClickHouse | 648,311,610 | 11,804 | akuzm | Unfortunately it turned out to be not as simple as I expected, namely because of the test case I mentioned above. If we want to support such cases, we have to slightly change memory tracking. Now we only create a per-query tracker in `ProcessList::insert`, and we would have to add a per-subquery tracker, or update the ... | 2020-06-23T17:40:05 | 2020-06-23T17:40:05 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,456,772 | 10,436 | filimonov | Any update here? | 2020-06-23T22:15:44 | 2020-06-23T22:15:44 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,156,199 | 11,890 | den-crane | CH does not support such JOIN expression.
You can only use `cartesian product` and filter in `where` section
select * from table_1, table_2
where table_1.col1 like table_2.col2 | 2020-06-23T13:41:45 | 2020-06-23T13:41:45 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,380,121 | 11,498 | filimonov | > For example let's image that you have some local table on the remote node named "salaries" and user "manager" that is allowed to query it. You can create local user "manager" and use cluster to access "salaries" on remote node (and looks pretty unsecure).
My expectations is that when cluster admin create user `man... | 2020-06-23T19:45:55 | 2020-06-23T19:46:31 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,503,545 | 11,882 | filimonov | Did you try runner script?
https://github.com/ClickHouse/ClickHouse/tree/master/tests/integration#running-with-runner-script
Also you can try adding `--privileged` flag for docker run. | 2020-06-24T00:24:03 | 2020-06-24T00:24:03 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 648,771,244 | 11,815 | alexey-milovidov | > TSan stress test sometimes killed by OOM: https://sandbox.yandex-team.ru/task/670887196/hardware
It was an issue in "Sandbox" infrastructure - looks like their servers were limited to 64 GiB memory for some period of time. | 2020-06-24T11:46:42 | 2020-06-24T11:46:42 | {} | MEMBER |
ClickHouse | ClickHouse | 648,808,320 | 11,089 | alexey-milovidov | This task is currently addressed in this PR: #11901 | 2020-06-24T13:07:24 | 2020-06-24T13:07:24 | {} | MEMBER |
ClickHouse | ClickHouse | 656,489,225 | 12,324 | filimonov | It looks like duplicate of https://github.com/ClickHouse/ClickHouse/issues/9272 | 2020-07-10T05:22:21 | 2020-07-10T05:22:21 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 656,490,397 | 12,317 | filimonov | Please create reproducable example, or provide full context (with stacktraces, log fragments etc)
Otherwise we can't help you | 2020-07-10T05:26:55 | 2020-07-10T05:26:55 | {} | CONTRIBUTOR |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.