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 | 628,186,816 | 10,897 | silviucpp | @den-crane Testing on linux ? I'm testing on OSX . | 2020-05-13T19:05:45 | 2020-05-13T19:05:45 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 628,224,154 | 10,852 | dgrr | The tests show errors that didn't happen in my test cases | 2020-05-13T20:20:22 | 2020-05-13T20:20:22 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 628,532,987 | 10,907 | doom369 | Same here. | 2020-05-14T10:05:08 | 2020-05-14T10:05:08 | {} | NONE |
ClickHouse | ClickHouse | 627,831,731 | 10,797 | azat | >By the way, will this bug be fixed in version 20.1.6.30?in our production env,we use clickhouse 20.1.6.30.a version upgration will relate serval paties.Thank you a lot!
releases are immutable, so no 20.1.6.30 will not contain this patch, but I guess there will be 20.1.6.x (where x>30), which should contain only bug... | 2020-05-13T08:25:49 | 2020-05-13T08:25:49 | {} | MEMBER |
ClickHouse | ClickHouse | 627,868,001 | 10,797 | weiyongyuan | > > By the way, will this bug be fixed in version 20.1.6.30?in our production env,we use clickhouse 20.1.6.30.a version upgration will relate serval paties.Thank you a lot!
>
> releases are immutable, so no 20.1.6.30 will not contain this patch, but I guess there will be 20.1.6.x (where x>30), which should contain o... | 2020-05-13T09:31:15 | 2020-05-13T09:31:15 | {} | NONE |
ClickHouse | ClickHouse | 627,876,102 | 10,383 | KochetovNicolai | Right now I don't like to make a new type, because it does not actually represent new way of soring data, operations or performance improvement. The only thing we can do is to throw exception in respect to `isFinalized` flag.
Adding finalization is initially a clarification of `convertToValues` method. Like, we sepa... | 2020-05-13T09:47:14 | 2020-05-13T09:47:14 | {} | MEMBER |
ClickHouse | ClickHouse | 628,269,556 | 4,190 | alexey-milovidov | This PR is assigned to Alexander Kuzmenkov @akuzm, he decided to take it for this week at last. | 2020-05-13T22:03:47 | 2020-05-13T22:03:47 | {
"confused": 1
} | MEMBER |
ClickHouse | ClickHouse | 628,488,479 | 10,854 | qoega | Probably better to add this parameter to example? | 2020-05-14T08:43:51 | 2020-05-14T08:43:51 | {} | MEMBER |
ClickHouse | ClickHouse | 627,708,657 | 9,113 | CurtizJ | Also you have a lot of TODOs. Are all of them still relevant? | 2020-05-13T02:35:39 | 2020-05-13T02:35:39 | {} | MEMBER |
ClickHouse | ClickHouse | 627,996,199 | 10,883 | den-crane | Yes. It is safe. If you make a new cluster with the same shards/servers order it will have the same shard numbers.
You can verify it in `select * from system.clusters` | 2020-05-13T13:44:15 | 2020-05-13T13:44:38 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 628,211,469 | 10,897 | den-crane | I would check a compiler and build options https://clickhouse-builds.s3.yandex.net/0/c811e1f0d0c439b74843770082569aa7fd5f9743/build_log_673063988_1589261189.txt | 2020-05-13T19:54:25 | 2020-05-13T19:54:42 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 628,299,301 | 10,564 | azat | > can I set distributed_replica_error_cap to 0?
Nope, it will trigger division by zero error :) (will fix that)
So you need to use at least 1 second and see if this will work for your (since this is second, and can be pretty long)
> Cause there is no round-robin strategy natively in CH.
Actually it is pretty... | 2020-05-13T23:32:43 | 2020-05-13T23:32:43 | {} | MEMBER |
ClickHouse | ClickHouse | 628,340,976 | 10,679 | sundy-li | > Maybe it should be implemented as `SYSTEM` query instead of `ALTER`?
> Something like `SYSTEM DROP REPLICA 'replica_name' FROM table_name` to remove dead replica of existing table and `SYSTEM DROP REPLICA 'replica_name' FROM '/path/to/table/in/zk/'` to remove dead replica of some table, which does not exists on the ... | 2020-05-14T02:02:42 | 2020-05-14T02:02:42 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 627,884,551 | 10,882 | hczhcz | > New functions like for arrays
I like the idea!
Probably we can also add a special function `tupleJoin` which converts tuple elements to columns, function parameters, etc. | 2020-05-13T10:04:23 | 2020-05-13T10:04:23 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 627,911,033 | 10,865 | azat | Looks like to tiny for the integration test (and there was no harm, just a noise in logs) | 2020-05-13T11:04:37 | 2020-05-13T11:04:37 | {} | MEMBER |
ClickHouse | ClickHouse | 627,921,158 | 10,679 | tavplubix | Maybe it should be implemented as `SYSTEM` query instead of `ALTER`?
Something like `SYSTEM DROP REPLICA 'replica_name' FROM table_name` to remove dead replica of existing table and `SYSTEM DROP REPLICA 'replica_name' FROM '/path/to/table/in/zk/'` to remove dead replica of some table, which does not exists on the loca... | 2020-05-13T11:28:28 | 2020-05-13T11:28:28 | {
"+1": 3
} | MEMBER |
ClickHouse | ClickHouse | 628,044,350 | 10,892 | den-crane | Obviously it's not stable enough for you. You did the right thing and tested it first. Just open issues related to your problems. | 2020-05-13T14:53:08 | 2020-05-13T14:53:08 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 628,162,825 | 10,669 | alesapin | Will fix combined modify and rename in the next PR. | 2020-05-13T18:18:38 | 2020-05-13T18:18:38 | {} | MEMBER |
ClickHouse | ClickHouse | 628,197,952 | 10,849 | tavplubix | ```
00925_zookeeper_empty_replicated_merge_tree_optimize_final: [ FAIL ]
Existing table metadata in ZooKeeper differs in date index column
```
Fails because 00029_test_zookeeper_optimize_exception does not drop table `test_optimize_exception_replicated`
```
01019_alter_materialized_view_consistent: ... | 2020-05-13T19:26:54 | 2020-05-13T19:26:54 | {} | MEMBER |
ClickHouse | ClickHouse | 628,202,960 | 10,897 | den-crane | Did you build it?
Try official binary clang-10-darwin https://clickhouse-builds.s3.yandex.net/0/c811e1f0d0c439b74843770082569aa7fd5f9743/report.html | 2020-05-13T19:37:06 | 2020-05-13T19:37:06 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 628,347,794 | 8,467 | yuzhichang | @qoega Here's a scenario where several inserts have same data: The JDBC client sends insert command, but gets broken connection exception. The client cannot tell if the exception happened before the insert, so it retry insert the same data. The server gets the same data twice, so later analysis queries may get incorrec... | 2020-05-14T02:27:59 | 2020-05-14T02:27:59 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 628,455,843 | 9,577 | excitoon | @alexey-milovidov I have a problem in this issue. The hack is ugly indeed. The problem with this issue is that original configuration is stored in `Context` and doesn't update while reloading.
Storage configuration is lazy and only initializes on first usage, and it may miss configuration updates (because on first t... | 2020-05-14T07:46:18 | 2020-05-14T07:46:18 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 628,502,391 | 10,482 | browcio | > Same issue here. Fresh ubuntu 18.
Turned out it was caused by deb caching proxy. | 2020-05-14T09:06:40 | 2020-05-14T09:15:35 | {} | NONE |
ClickHouse | ClickHouse | 628,571,242 | 10,881 | nikitamikhaylov | Hello! Your `schema.xml` is not informative. You don't provide us any information, especially table schema. However I noticed that you have an error near the tag '**engine**'. I think the problem is that you didn't specify the engine of your table. That's why copier сan't parse this (empty string). Please read this ar... | 2020-05-14T11:27:57 | 2020-05-14T11:27:57 | {} | MEMBER |
ClickHouse | ClickHouse | 628,044,307 | 10,891 | xap9i | Sorry, thanks! | 2020-05-13T14:53:04 | 2020-05-13T14:53:04 | {} | NONE |
ClickHouse | ClickHouse | 628,186,442 | 10,897 | den-crane | No issue for me.
20.5.1.3327 https://repo.yandex.ru/clickhouse/deb/testing/main/
https://github.com/ClickHouse/ClickHouse/commit/c811e1f0d0c439b74843770082569aa7fd5f9743
Tested 4 times (clickhouse-client & http)
| 2020-05-13T19:05:02 | 2020-05-13T19:05:02 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 628,199,595 | 10,852 | alexey-milovidov | @dgrr Just keep it similar to other tests (most other tests are in lower_case).
I see that everything is alright. | 2020-05-13T19:30:17 | 2020-05-13T19:30:17 | {} | MEMBER |
ClickHouse | ClickHouse | 628,267,944 | 10,882 | alexey-milovidov | tupleConcat - Ok.
But I don't like the idea to allow creation of tuples of arbitrary size, because tuples are not cheap comparing to arrays (every tuple element is a column). | 2020-05-13T21:59:25 | 2020-05-13T21:59:25 | {} | MEMBER |
ClickHouse | ClickHouse | 628,482,624 | 10,797 | azat | >In my tests,when group_by_overflow_mode='break',the local node still return all the distinct keys on it.
I guess you checked on very small number of rows, and in case of `group_by_overflow_mode='break'` it will "break"/stop the aggregation from the next block
>the local node still return all the distinct keys on... | 2020-05-14T08:33:55 | 2020-05-14T08:33:55 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 627,800,765 | 10,591 | filimonov | > I don't understand why expression is stored for move ttl but not for delete ttl?
It was not used originally, and i think @excitoon kept it as is for backward compatibility. Otherwise older version will not be able to read ttl for parts created in a newer version (and not sure how it will fail, as the quite simple ... | 2020-05-13T07:25:40 | 2020-05-13T07:25:40 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 627,804,079 | 10,845 | shizhenchina | I used flink by mistake | 2020-05-13T07:32:25 | 2020-05-13T07:32:25 | {} | NONE |
ClickHouse | ClickHouse | 628,186,867 | 10,892 | den-crane | Try official build 20.5.1.3327 https://repo.yandex.ru/clickhouse/deb/testing/main/
https://github.com/ClickHouse/ClickHouse/releases | 2020-05-13T19:05:52 | 2020-05-13T19:05:52 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 628,192,114 | 10,458 | nikitamikhaylov | This test depends on config of the server. But I read again your comment and understand it. | 2020-05-13T19:15:53 | 2020-05-13T19:15:53 | {} | MEMBER |
ClickHouse | ClickHouse | 628,525,053 | 10,898 | 4ertus2 | There's addidional patch here 0ba4c373a98b9f592625fc31ce6227ddd374a92d | 2020-05-14T09:48:44 | 2020-05-14T09:48:44 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 627,701,473 | 10,797 | weiyongyuan | By the way, will this bug be fixed in version 20.1.6.30?in our production env,we use clickhouse 20.1.6.30.a version upgration will relate serval paties.Thank you a lot! | 2020-05-13T02:09:33 | 2020-05-13T02:09:33 | {} | NONE |
ClickHouse | ClickHouse | 627,832,132 | 10,862 | hhell | > With values `simple`, `iso` and `unix_timestamp`
A more general way would be to support date formatting expressions in the setting,
i.e. modifiers from https://clickhouse.tech/docs/en/single/#formatdatetime
But for that, the timestamp and tzoffset would need to be added as modifiers too,
i.e. `date_time_output_... | 2020-05-13T08:26:29 | 2020-05-13T08:30:07 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 628,018,495 | 4,061 | georgantasp | I get a similar issue after upgrading clickhouse server/client.
```
sh-4.2$ whoami
ssm-user
sh-4.2$ ls -ltra .clickhouse-client-history
-rw------- 1 ssm-user ssm-user 144689 May 13 14:14 .clickhouse-client-history
sh-4.2$ clickhouse-client
ClickHouse client version 20.3.8.53.
Connecting to localhost:9000 as u... | 2020-05-13T14:15:30 | 2020-05-13T14:15:49 | {} | NONE |
ClickHouse | ClickHouse | 628,162,529 | 10,669 | alesapin | > ALTER TABLE t RENAME COLUMN x TO old, rename COLUMN new to x
>
> also does not work
Fix for this: https://github.com/ClickHouse/ClickHouse/pull/10895 | 2020-05-13T18:18:11 | 2020-05-13T18:18:11 | {} | MEMBER |
ClickHouse | ClickHouse | 628,534,063 | 10,913 | akuzm | See https://github.com/ClickHouse/ClickHouse/issues/10914 as well | 2020-05-14T10:07:28 | 2020-05-14T10:07:28 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 628,537,958 | 10,914 | 4ertus2 | Duplicate #10907 | 2020-05-14T10:15:42 | 2020-05-14T10:15:42 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 627,666,567 | 10,797 | azat | >Is this development env? If so, maybe you can try latest testing - v20.5.1.3327-testing (there were some fixes in aggregation) ?
> Can you also provide structure of the distributed and local tables?
No need to, reproduced | 2020-05-13T00:13:10 | 2020-05-13T00:13:10 | {} | MEMBER |
ClickHouse | ClickHouse | 627,668,519 | 10,797 | azat | >@alexey-milovidov alexey-milovidov changed the title max_result_rows doesn't work on remote servers in distributed query group_by_overflow_mode does not work for distributed tables 8 hours ago
The issue is a general, not only for distributed | 2020-05-13T00:19:54 | 2020-05-13T00:19:54 | {} | MEMBER |
ClickHouse | ClickHouse | 627,864,927 | 10,564 | azat | >Sometimes when a flood of queries come in, they could be sent to one node way more than the other as the graph indicates so the one received most of the query could reach max_concurrent_queries
I guess the graph shows queries that are currently executed, in this case maybe one replica handle queries faster then the... | 2020-05-13T09:25:25 | 2020-05-13T09:25:25 | {} | MEMBER |
ClickHouse | ClickHouse | 627,908,083 | 10,524 | 4ertus2 | It's not clear in your sample which tables are distributed and which are not. It's even hard to find tables in your query :) It's better to simplify it as much as possible before makeing an issue.
Anyway. ClickHouse tryes to calculate as much as possible on remote nodes and fetch as low data as possible via network.... | 2020-05-13T10:57:01 | 2020-05-13T10:57:01 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 627,956,099 | 10,880 | yancheng44 | tested version 20.3.8.53 ， same | 2020-05-13T12:41:32 | 2020-05-13T12:41:32 | {} | NONE |
ClickHouse | ClickHouse | 628,187,704 | 10,897 | silviucpp | Can you paste the config files :) I want to try with your config files to see maybe something related from config | 2020-05-13T19:07:30 | 2020-05-13T19:07:30 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 628,193,368 | 10,897 | silviucpp | Will try to replace the config files with the default one and see. Can you please test https://github.com/ClickHouse/ClickHouse/issues/10900 as well to see if for you crashes ?
Silviu | 2020-05-13T19:18:12 | 2020-05-13T19:18:12 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 628,196,741 | 10,863 | nikitamikhaylov | I looked throw some articles. Here are the best
[High-Throughput, Thread-Safe, LRU Caching from Ebay](https://tech.ebayinc.com/engineering/high-throughput-thread-safe-lru-caching/)
[High Concurrency LRU Caching written in Go](https://www.openmymind.net/High-Concurrency-LRU-Caching/)
[Just cool idea](https://w... | 2020-05-13T19:24:38 | 2020-05-13T19:24:38 | {} | MEMBER |
ClickHouse | ClickHouse | 627,834,330 | 8,547 | 4ertus2 | ```
Since databases may be different
I cannot use concrete database in my query
```
As workaround you may try to make views in same db that look at original tables and make Distributed table over them. It's not well defined zone when you combine tables from different databases in one query. Something could go wrong... | 2020-05-13T08:30:29 | 2020-05-13T08:31:00 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 627,903,825 | 10,865 | qoega | Is there a way to add a test for this scenario(all batches pending) | 2020-05-13T10:46:53 | 2020-05-13T10:46:53 | {} | MEMBER |
ClickHouse | ClickHouse | 628,443,956 | 10,906 | kaijianding | I do understand what said here https://stackoverflow.com/questions/60142967/how-to-understand-part-and-partition-of-clickhouse/60155974#60155974
But for the partition key like `year`,`month`,`day`, the MinMax index should have only 1 value in all parts in one partition, this should be taken into account when evaluat... | 2020-05-14T07:25:33 | 2020-05-14T07:25:33 | {} | NONE |
ClickHouse | ClickHouse | 628,472,436 | 10,524 | LvKeWechat | this query only has one distributed table `event_data.event_wos_p15`
simplified SQL
`select uniq(uid,sid) as value,l.1 as from ,l.2 as to from (
select uid,sid,s_t from (
select distinct_id as uid ,groupArray((XXX,XXX,XXX,)) as cur,... array-function-transformation(cur) as tx
from event_data.event_wos_p15 wher... | 2020-05-14T08:15:28 | 2020-05-14T08:15:28 | {} | NONE |
ClickHouse | ClickHouse | 628,546,572 | 10,854 | BayoNet | Good point.
Updated. | 2020-05-14T10:34:08 | 2020-05-14T10:34:08 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 628,568,339 | 10,876 | IVitamin-C | > It's expected behavior. Only first level of `( )` (expr_qry) inside brackets is executed on shards. Everything outside brackets is executed at an initiator. So only one server does groupby in the first query and it needed more ungroupped data volume to be transferred from shards to the initiator.
I think this qu... | 2020-05-14T11:21:32 | 2020-05-14T11:21:32 | {} | NONE |
ClickHouse | ClickHouse | 627,966,826 | 10,855 | alesapin | Just comments, so I'll merge. | 2020-05-13T12:59:46 | 2020-05-13T12:59:46 | {} | MEMBER |
ClickHouse | ClickHouse | 628,178,946 | 10,892 | silviucpp | @den-crane Can you replicate the problem I'm describing here https://github.com/ClickHouse/ClickHouse/issues/10897 ? | 2020-05-13T18:51:03 | 2020-05-13T18:51:03 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 628,187,407 | 10,232 | 4ertus2 | Related #10898
I've tested current version with minimal fixes. It works. | 2020-05-13T19:06:53 | 2020-05-13T19:06:53 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 628,485,784 | 10,866 | qoega | I totally understand it. But I try to find a real use case where we need random __function__.
For example we may also need `randConstant` analogue. | 2020-05-14T08:39:09 | 2020-05-14T08:39:09 | {} | MEMBER |
ClickHouse | ClickHouse | 628,531,726 | 10,913 | akuzm | ```
/4/ :) select count(*) from hits l join hits r using EventDate where l.EventDate < today() and r.EventDate < today();
Received exception from server (version 20.5.1):
Code: 277. DB::Exception: Received from localhost:9000. DB::Exception: MinMax index by columns (EventDate) is not used and setting 'force_index_... | 2020-05-14T10:02:38 | 2020-05-14T10:02:38 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 627,677,497 | 10,814 | azat | >when use distributed_group_by_no_merge ，which part of SQL execute on MergeTree node and which part execute distributed node ？
In case of `distributed_group_by_no_merge=1` query executed only on the shards (I guess this is "MergeTree node" in your terminology) and the initiator node (distributed node in your termino... | 2020-05-13T00:50:05 | 2020-05-13T00:50:05 | {} | MEMBER |
ClickHouse | ClickHouse | 627,815,842 | 10,876 | IVitamin-C | clickHouse version:20.3.5
The cluster has 10 nodes, each node has 64 cores and 256G memory
| 2020-05-13T07:55:16 | 2020-05-13T07:55:16 | {} | NONE |
ClickHouse | ClickHouse | 627,832,972 | 10,591 | filimonov | Regarding the usability:
1) most of the user will use DELETE only, and storing it in the array / other complicated structures seem a bit overkill for that use case
2) if it will be stored in some complicated structures (array with expressions etc) it will be no clear way to get the list of parts which should be remov... | 2020-05-13T08:27:56 | 2020-05-13T08:29:36 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 627,854,312 | 10,524 | qoega | 19.15 is no longer supported.
>this error
I can't see any error/stacktrace | 2020-05-13T09:06:46 | 2020-05-13T09:06:46 | {} | MEMBER |
ClickHouse | ClickHouse | 627,986,365 | 10,879 | den-crane | Please provide xml with dictionary description ( at least lifecycle / layout , sections)
Do you observe any related messages in `/var/log/clickhouse-server//clickhouse-server.log` ?
| 2020-05-13T13:28:40 | 2020-05-13T13:28:40 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 628,042,663 | 10,891 | den-crane | It's impossible. You can use `select from` in MATERIALIZED column. You can use only `dictGet<Type>`. | 2020-05-13T14:50:54 | 2020-05-13T14:51:13 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 628,141,492 | 10,735 | slawomir-pryczek | Bump... any ideas? Maybe there is some vm which can be used for easily testing / compiling clickhouse? | 2020-05-13T17:39:03 | 2020-05-13T17:39:03 | {} | NONE |
ClickHouse | ClickHouse | 628,195,829 | 10,900 | den-crane | No issue. 20.5.1.3327
```
SELECT count(*) AS value
FROM tb2.query_filters_SUITE
WHERE ((timestamp >= toDateTime(1589396340)) AND (timestamp < toDateTime(1589396460))) AND ((length(array_float) = 2) AND (greatCircleDistance(CAST(array_float[1], 'Nullable(Float64)'), CAST(array_float[2], 'Nullable(Float64)'), 10.2,... | 2020-05-13T19:22:52 | 2020-05-13T19:24:52 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 628,206,685 | 10,897 | silviucpp | Ok thanks for links I see the prebuild works just fine. will recompile mine with all modules enabled then isolate the problem and post here the results. | 2020-05-13T19:44:37 | 2020-05-13T19:44:37 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 628,217,945 | 10,890 | KochetovNicolai | Squash it, please. | 2020-05-13T20:07:24 | 2020-05-13T20:07:24 | {} | MEMBER |
ClickHouse | ClickHouse | 627,983,833 | 10,876 | den-crane | It's expected behavior. Only first level of `( )` (expr_qry) inside brackets is executed on shards. Everything outside brackets is executed at an initiator. So only one server does groupby in the first query and it needed more ungroupped data volume to be transferred from shards to the initiator. | 2020-05-13T13:24:53 | 2020-05-13T20:57:16 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 627,663,868 | 4,562 | azat | Silence here, @alexey-milovidov can be closed I guess | 2020-05-13T00:03:56 | 2020-05-13T00:03:56 | {} | MEMBER |
ClickHouse | ClickHouse | 627,689,560 | 10,797 | weiyongyuan | > > Is this development env? If so, maybe you can try latest testing - v20.5.1.3327-testing (there were some fixes in aggregation) ?
>
> > Can you also provide structure of the distributed and local tables?
>
> No need to, reproduced
Thank you everyone so much!
>Yes,all tests are in dev envs.
>Tables are like ... | 2020-05-13T01:26:13 | 2020-05-13T02:11:01 | {} | NONE |
ClickHouse | ClickHouse | 627,830,128 | 8,547 | 4ertus2 | Probably we need rethink Distributed logic with different databases.
Related #10471 #6571 | 2020-05-13T08:22:40 | 2020-05-13T08:36:52 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 628,077,449 | 10,797 | weiyongyuan | Sorry,i have another question,in docs,Restrictions on aggregation (group_by_overflow_mode) also have the value any.
### throw – Throw an exception (default).
### break – Stop executing the query and return the partial result, as if the source data ran out.
>does break means that local node when aggregated keys equa... | 2020-05-13T15:45:11 | 2020-05-13T15:46:17 | {} | NONE |
ClickHouse | ClickHouse | 628,189,630 | 10,897 | den-crane | No configs. Only vendor's from deb. | 2020-05-13T19:11:11 | 2020-05-13T19:11:11 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 628,513,750 | 4,006 | bohutang | Update:
https://github.com/ClickHouse/ClickHouse/pull/10306
Binlog Network Stream:
- [x] Client handshake
- [x] Client ping
- [x] Client register slave
- [x] Client binlog event dump
- [x] Client receive binlog events over the network stream
Binlog Event:
- [x] FormatDescriptionEvent
- [x] RotateEvent
- ... | 2020-05-14T09:26:21 | 2020-05-14T09:26:21 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 627,712,197 | 10,845 | shizhenchina | `public class ClickHouse3Sink implements AllWindowFunction<CommonData, Object, TimeWindow> {
static {
}
@Override
public void apply(TimeWindow timeWindow, Iterable<CommonData> iterable, Collector<Object> collector) throws Exception {
Connection connection = null;
PreparedStat... | 2020-05-13T02:48:17 | 2020-05-13T02:50:50 | {} | NONE |
ClickHouse | ClickHouse | 628,089,281 | 10,564 | cw9 | Got it, can I set `distributed_replica_error_cap` to 0? would that + `load_balancing = random` give me basically round-robin load balancing? Cause there is no round-robin strategy natively in CH.
For my use case, I actually prefer returning partial data with better latency so I'm also using `skip_unavailable_shard... | 2020-05-13T16:05:10 | 2020-05-13T16:05:10 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 628,109,328 | 10,866 | axolm | > What is the use case for this function? Why `generateRandom('a FixedString(10)')` don't solve your case?
I haven't figured out all the nuances of working with generateRandom but @alexey-milovidov said:
```
Другие актуальны, но слегка меньше:
randomString
randomFixedString
randomStringUTF8
потому что похожа... | 2020-05-13T16:40:46 | 2020-05-13T16:40:46 | {} | NONE |
ClickHouse | ClickHouse | 628,187,229 | 10,897 | den-crane | Linux. I do not have OSX | 2020-05-13T19:06:32 | 2020-05-13T19:06:32 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 628,300,354 | 10,866 | alexey-milovidov | @qoega generateRandom is a table function, not an ordinary function. | 2020-05-13T23:36:10 | 2020-05-13T23:36:10 | {} | MEMBER |
ClickHouse | ClickHouse | 627,841,863 | 10,866 | qoega | What is the use case for this function? Why `generateRandom('a FixedString(10)')` don't solve your case? | 2020-05-13T08:43:57 | 2020-05-13T08:43:57 | {} | MEMBER |
ClickHouse | ClickHouse | 627,904,768 | 10,855 | alesapin | Ok, I'll just add a comment. | 2020-05-13T10:49:03 | 2020-05-13T10:49:03 | {} | MEMBER |
ClickHouse | ClickHouse | 628,204,464 | 10,897 | silviucpp | I build myself and disabled some modules :)) but same build instruction are used on past version which works fine. I'll try to use a precompiled binary to see. | 2020-05-13T19:40:09 | 2020-05-13T19:40:09 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 628,224,595 | 10,902 | alexey-milovidov | It may uncover more issues in tests. | 2020-05-13T20:21:19 | 2020-05-13T20:21:19 | {} | MEMBER |
ClickHouse | ClickHouse | 628,516,046 | 10,904 | azat | Refs: #10910 10910 | 2020-05-14T09:30:33 | 2020-05-14T09:30:33 | {} | MEMBER |
ClickHouse | ClickHouse | 628,084,654 | 10,564 | azat | >The two replicas are running in same HW and replicating the same data.
This is just a guess and I seems unlikely situation, so I think that right now you can play with `distributed_replica_error_half_life` to get replica online sooner (i.e. make replicas selectable for querying with random load_balancing). | 2020-05-13T15:57:23 | 2020-05-13T15:57:23 | {} | MEMBER |
ClickHouse | ClickHouse | 628,245,404 | 10,897 | silviucpp | Hmm that seems a build did on linux for osx.. https://clickhouse.tech/docs/en/development/build-cross-osx/
| 2020-05-13T21:05:29 | 2020-05-13T21:05:29 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 628,354,083 | 10,905 | den-crane | These parameters play role when CH is an http client not an http server.
_Added settings: http_connection_timeout, http_send_timeout, http_receive_timeout. In particular, these settings are used for downloading data parts for replication. Changing these settings allows for faster failover if the network is overloade... | 2020-05-14T02:49:58 | 2020-05-14T02:49:58 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 628,471,807 | 10,905 | gubinjie | @decaseal
Thank you
I configured these two parameters behind the jdbc connection string, for example:
jdbc: clickhouse: // xxxxxxxxx? http_send_timeout = 5000; http_receive_timeout = 5000
We found that it worked, but I have a question?
What is the difference between http_receive_timeout, http_send_timeout and soc... | 2020-05-14T08:14:11 | 2020-05-14T08:14:11 | {} | NONE |
ClickHouse | ClickHouse | 627,678,984 | 10,871 | alexey-milovidov | https://a.yandex-team.ru/review/1256186 | 2020-05-13T00:54:59 | 2020-05-13T00:54:59 | {} | MEMBER |
ClickHouse | ClickHouse | 627,841,653 | 8,547 | 4ertus2 | Your main problem is most probably cause of #6571
```
The error is DB::Exception: Table default.b1_replicated doesn't exist..
``` | 2020-05-13T08:43:36 | 2020-05-13T08:43:36 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 628,015,443 | 10,564 | cw9 | > I guess the graph shows queries that are currently executed, in this case maybe one replica handle queries faster then the other, hence the spike.
What would cause this? The two replicas are running in same HW and replicating the same data. | 2020-05-13T14:10:54 | 2020-05-13T14:10:54 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 628,121,668 | 10,763 | azat | Any news? | 2020-05-13T17:02:18 | 2020-05-13T17:02:18 | {} | MEMBER |
ClickHouse | ClickHouse | 628,293,175 | 10,902 | alexey-milovidov | Looks ok. | 2020-05-13T23:13:00 | 2020-05-13T23:13:00 | {} | MEMBER |
ClickHouse | ClickHouse | 628,301,669 | 10,055 | alexey-milovidov | @filimonov I get the same error if ClickHouse is compiled with OpenSSL 1.1.1. | 2020-05-13T23:40:30 | 2020-05-13T23:40:30 | {} | MEMBER |
ClickHouse | ClickHouse | 628,505,579 | 10,383 | KochetovNicolai | Closing it in favor of #10890 | 2020-05-14T09:12:02 | 2020-05-14T09:12:02 | {} | MEMBER |
ClickHouse | ClickHouse | 628,575,005 | 5,883 | nikitamikhaylov | Copier don't create tables on target cluster. It creates only internal service tables for its needs. That's why you have to specify table engine in "task.xml". But it seems that last versions of copier create target tables explicitly. Please try it and share your experience with us. Notice, that you don't have to upgra... | 2020-05-14T11:36:02 | 2020-05-14T11:36:02 | {} | MEMBER |
ClickHouse | ClickHouse | 617,476,284 | 10,285 | filimonov | Lack if query context when flushing system logs | 2020-04-22T00:17:41 | 2020-04-22T00:17:41 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 617,478,167 | 10,205 | alexey-milovidov | Duplicate of #10418 | 2020-04-22T00:24:53 | 2020-04-22T00:24:53 | {} | MEMBER |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.