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 | 509,197,634 | 5,917 | 4ertus2 | #4670 | 2019-07-08T12:03:55 | 2019-07-08T12:03:55 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 509,234,915 | 5,888 | alexey-milovidov | > as I understand, your implementation will always use
The purpose is to make function signatures analyzable. It will allow to make better error codes. | 2019-07-08T13:55:00 | 2019-07-08T13:55:00 | {} | MEMBER |
ClickHouse | ClickHouse | 509,962,719 | 5,932 | projectsbird123 | When we started the clickhouse as a service we found this error.
ERROR:
<Error> Application: DNS error: EAI: -9
After this we have the updated the config.xml, by commenting "< listen_host>::</listen_host> >" and un-commented <listen_host>0.0.0.0</listen_host>.
Again started the clickhouse-server as a service an... | 2019-07-10T08:23:42 | 2019-07-10T09:02:30 | {} | NONE |
ClickHouse | ClickHouse | 509,207,502 | 5,918 | 4ertus2 | Hi,
For the A->A1 part it's possible to omit reading P_BRAND and LO_ORDERDATE columns in A1. ClickHouse tries not to read unused columns. I think it's the answer for the A->A1 issue.
A1 -> A2 is quite strange. But if the right table is small it could be some performance fluctuation. What is your right joined tabl... | 2019-07-08T12:36:22 | 2019-07-08T12:37:31 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 509,215,680 | 5,714 | alexey-milovidov | Fixed in master. | 2019-07-08T13:01:30 | 2019-07-08T13:01:30 | {} | MEMBER |
ClickHouse | ClickHouse | 509,303,239 | 5,287 | abraithwaite | This was fixed in a recent release :tada: | 2019-07-08T16:46:43 | 2019-07-08T16:46:43 | {} | NONE |
ClickHouse | ClickHouse | 509,625,469 | 5,898 | blinkov | cc @zhang2014 @sundy-li | 2019-07-09T12:39:28 | 2019-07-09T12:39:28 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 509,720,190 | 4,569 | alexey-milovidov | Fixed in master:
```
ClickHouse client version 19.11.0.
Connecting to localhost:9000 as user default.
Connected to ClickHouse server version 19.11.0 revision 54423.
milovidov-Pro-P30 :) SELECT a.ANIMAL a,
:-] b.ANIMAL b,
:-] c.ANIMAL c,
:-] d.ANIMAL d
:-] FROM ANIMAL a
:-] JOIN ANIM... | 2019-07-09T16:48:49 | 2019-07-09T16:48:49 | {} | MEMBER |
ClickHouse | ClickHouse | 509,756,884 | 5,541 | vzakaznikov | Hi @KochetovNicolai, thanks for the review! I have committed the changes based on your comments. | 2019-07-09T18:35:02 | 2019-07-09T18:35:02 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 509,888,350 | 4,987 | zhang2014 | Thanks | 2019-07-10T03:01:46 | 2019-07-10T03:01:46 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 510,088,474 | 5,959 | den-crane | It could if for example table codes has 10k columns. | 2019-07-10T14:36:26 | 2019-07-10T14:36:26 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 509,155,590 | 4,987 | zhang2014 | Need help:
- I compiled with the `-DCMAKE_BUILD_TYPE=Debug -DUSE_GPERF=0 -DUSE_INTERNAL_UNWIND_LIBRARY=1`, the run `SELECT * FROM build_options;`, and got:
```
2019.07.08 17:07:55.943252 [ 55 ] {} <Error> BaseDaemon: ########################################
2019.07.08 17:07:55.943558 [ 55 ] {} <Error> BaseDaemon: ... | 2019-07-08T09:40:15 | 2019-07-08T09:40:15 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 509,196,042 | 4,616 | bgranvea | I don't know if it is the same bug but I have a segfault with Kafka engine, Protobuf format and MV. It seems to happen if I send a message with a conversion error (for example a Protobuf int64 field mapped to a DateTime column and the value is exceeding UInt32 capacity). I tried to build a simpler test case but didn't ... | 2019-07-08T11:58:41 | 2019-07-08T11:58:41 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 509,538,226 | 5,929 | alesapin | Yes, it would be more correct, but this URL also used inside the release process. And also we don't have version `19.11.0` even in testing. I can make changes in the release script to fix this issues, but why somebody would like to build `clickhouse-server` image from our repo? What is the purpose? | 2019-07-09T08:09:42 | 2019-07-09T08:09:42 | {} | MEMBER |
ClickHouse | ClickHouse | 509,596,577 | 5,014 | 243f6a8885a308d313198a2e037 | @alexey-milovidov Thank you for your reaction.
I am recently wondering what would happen if `INSERT`, `CREATE TABLE` or `DROP TABLE` queries *partially* failed by `ZSESSIONEXPIRED` (In general, not only by xid overflow).
Leastways, I discovered "we can no more drop/create a table with this name" situation after p... | 2019-07-09T11:01:07 | 2019-07-09T11:01:07 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 509,689,918 | 1,394 | NanoBjorn | @hagen1778
> We are using gcsfuse only for RO purposes
How is performance of such reads? Like how many megabytes per second? Thinking of usage gcsfuse too | 2019-07-09T15:26:55 | 2019-07-09T15:27:03 | {
"eyes": 3
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 509,717,709 | 5,929 | alesapin | > And also we don't have version 19.11.0 even in testing.
We don't have exactly 19.11.0 version. We have 19.11.0.xxx versions. | 2019-07-09T16:41:37 | 2019-07-09T16:41:37 | {} | MEMBER |
ClickHouse | ClickHouse | 509,722,133 | 3,169 | alexey-milovidov | Does not reproduce on version 19.11:
```
milovidov-Pro-P30 :) CREATE TABLE xxxxxx (time Nullable(UInt64), mpid UInt8, siteid Nullable(UInt64)) ENGINE = Memory
CREATE TABLE xxxxxx
(
`time` Nullable(UInt64),
`mpid` UInt8,
`siteid` Nullable(UInt64)
)
ENGINE = Memory
Ok.
0 rows in set. Elap... | 2019-07-09T16:54:19 | 2019-07-09T16:54:19 | {} | MEMBER |
ClickHouse | ClickHouse | 509,924,997 | 1,510 | shivacr | > i encounter this error on large request (900M lines on table2) like that :
>
> ```
> INSERT INTO <table1> SELECT * FROM <table2>
> ```
HI,
Were you able to find any solution for this?
I am facing the same problem.
Inserting into new table from old one.
old one has 440472552 entries and paritioned by dat... | 2019-07-10T06:18:51 | 2019-07-10T06:18:51 | {} | NONE |
ClickHouse | ClickHouse | 508,985,411 | 5,233 | blind-oracle | Haven't seen it in recent versions. Let's close. | 2019-07-07T09:34:57 | 2019-07-07T09:34:57 | {} | NONE |
ClickHouse | ClickHouse | 509,033,823 | 5,915 | alexey-milovidov | Performance improvement:
```
x3.91 (0.317 s.) | (0.081 s.)
```
Almost four times - as expected. | 2019-07-07T21:55:39 | 2019-07-07T21:55:39 | {} | MEMBER |
ClickHouse | ClickHouse | 509,720,803 | 4,156 | alexey-milovidov | Fixed in recent versions. | 2019-07-09T16:50:30 | 2019-07-09T16:50:30 | {} | MEMBER |
ClickHouse | ClickHouse | 509,771,829 | 5,951 | alexey-milovidov | @blinkov has pointed out to the related feature: https://cloud.google.com/bigquery/docs/reference/standard-sql/query-syntax#select-except | 2019-07-09T19:17:39 | 2019-07-09T19:17:39 | {} | MEMBER |
ClickHouse | ClickHouse | 510,207,244 | 5,955 | abraithwaite | > try to add hostName() to your query to check that you see data from the same shard twice
This only returned to me the hostname that I ran the query on. Not sure how it plays into a simple select query (even against a distributed table, the docs suggest it won't be the remote table unless processing aside from fet... | 2019-07-10T20:02:40 | 2019-07-10T20:02:40 | {} | NONE |
ClickHouse | ClickHouse | 509,033,553 | 5,014 | alexey-milovidov | @243f6a8885a308d313198a2e037
The fix that @github1youlc is only for deadlock. Now when xid overflows, ZooKeeper session will forcefully expire and ClickHouse will establish a new ZooKeeper session - everything will continue to work normally just in the same way as after network errors.
You have made changes to ch... | 2019-07-07T21:51:06 | 2019-07-07T21:51:52 | {
"+1": 1,
"confused": 1,
"eyes": 2
} | MEMBER |
ClickHouse | ClickHouse | 509,227,909 | 5,869 | alexey-milovidov | At Yandex we use 3-node ZooKeeper cluster spread across 3 datacenters with maximum about 25ms RTT and 1500 km distance. | 2019-07-08T13:36:19 | 2019-07-08T13:36:19 | {
"+1": 3
} | MEMBER |
ClickHouse | ClickHouse | 509,605,440 | 5,929 | filimonov | > How to get the "latest" stable version and create more flexible config file?
Do you mean ClickHouse config? You can just pass it 'from outside' as a volume.
Just like that
```
docker run -v /path/to/my_config.xml:/etc/clickhouse-server/config.xml yandex/clickhouse-server:latest
```
If you need to adjust ... | 2019-07-09T11:32:06 | 2019-07-09T11:33:57 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 509,722,826 | 3,132 | alexey-milovidov | Could you please provide more info (e.g. example data) on how to reproduce this issue? | 2019-07-09T16:56:13 | 2019-07-09T16:56:13 | {} | MEMBER |
ClickHouse | ClickHouse | 509,728,219 | 2,494 | alexey-milovidov | @Akazz We can introduce a property of aggregate functions like `returnDefaultWhenOnlyNull`. If this property is set, we will not wrap the result type in Nullable and will return default value when nothing was aggregated. | 2019-07-09T17:12:26 | 2019-07-09T17:12:26 | {} | MEMBER |
ClickHouse | ClickHouse | 509,856,688 | 5,951 | alexey-milovidov | We came to the conclusion, that `COLUMNS('^x')` is the right solution. | 2019-07-10T00:15:45 | 2019-07-10T00:15:45 | {} | MEMBER |
ClickHouse | ClickHouse | 509,858,027 | 5,955 | den-crane | try to add hostName() to your query to check that you see data from the same shard twice | 2019-07-10T00:21:36 | 2019-07-10T00:21:36 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 508,963,591 | 5,543 | alexey-milovidov | > Date Date CODEC(Delta
The issue was fixed in #5786
(the fix is available in 19.11 testing) | 2019-07-07T01:01:46 | 2019-07-07T01:01:46 | {} | MEMBER |
ClickHouse | ClickHouse | 509,015,020 | 5,746 | perez1987 | I have the same problem.... | 2019-07-07T16:57:07 | 2019-07-07T16:57:07 | {} | NONE |
ClickHouse | ClickHouse | 509,111,152 | 5,847 | dmitryikh | > @dmitryikh
>
> Does the issue vanish if you install version 19.11?
We are not going to update the version in production to check this case. But what I plan to do is to remove `CODEC(Delta,...)` from `date` column and check if the fail will be repeating. I will report about results.
By the way, so far I have ... | 2019-07-08T07:24:57 | 2019-07-08T07:24:57 | {} | NONE |
ClickHouse | ClickHouse | 509,704,982 | 5,932 | filimonov | After an unsuccessful run attempt can you see any errors/warnings at the end of `/var/log/clickhouse-server/clickhouse-server.log`? | 2019-07-09T16:04:58 | 2019-07-09T16:04:58 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 509,724,010 | 2,908 | alexey-milovidov | Now we have TTL expressions: https://clickhouse.yandex/docs/en/operations/table_engines/mergetree/
It's not for fixed-size database setup but is suitable to store data for a limited time period. | 2019-07-09T16:59:35 | 2019-07-09T16:59:35 | {} | MEMBER |
ClickHouse | ClickHouse | 509,730,619 | 5,946 | akonyaev90 | I have no idea why my commit fail this test - '00908_long_http_insert' (((
| 2019-07-09T17:19:42 | 2019-07-09T17:19:42 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 509,805,856 | 5,947 | blind-oracle | @filimonov
It does not build on latest `gcc-9.1.0` either.
Currently I've worked this around by naming the classes in template in Boost's `quote.hpp` the same as `bind.hpp`: T1/T2/T3. Then it builds fine.
From my limited knowledge it seems that these files are generated by Boost's preprocessor somehow before Bo... | 2019-07-09T20:59:14 | 2019-07-09T20:59:14 | {} | NONE |
ClickHouse | ClickHouse | 509,827,386 | 4,987 | alexey-milovidov | @zhang2014 I've fixed this issue in https://github.com/yandex/ClickHouse/pull/5948 | 2019-07-09T22:05:23 | 2019-07-09T22:05:23 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 510,105,548 | 5,932 | champtar | @projectsbird123 is IPv6 disable on your RHEL 7.5 ? | 2019-07-10T15:16:20 | 2019-07-10T15:16:20 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 509,064,225 | 5,683 | den-crane | I believe the reason is different partitioning.
With partitions by date the same time values for different paths end up in one part (file) and very near (max_compress_block_size). With partitions by month these values much further than max_compress_block_size. | 2019-07-08T03:16:20 | 2019-07-08T03:16:20 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 509,274,073 | 5,938 | alexey-milovidov | This issue will probably get fixed:
```
SELECT
JSONLength(reinterpretAsString(rand64()) AS JSON) AS length,
JSON,
hex(JSON)
FROM system.numbers_mt
WHERE length > 0
┌─length─┬─JSON───┬─hex(reinterpretAsString(rand64()))─┐
│ 1 │ [0Lsʂ] │ 5B30004C73CA825D │
└────────┴──... | 2019-07-08T15:29:56 | 2019-07-08T15:29:56 | {} | MEMBER |
ClickHouse | ClickHouse | 509,304,793 | 5,929 | alesapin | These variables are used inside release process. We cannot simply delete them. | 2019-07-08T16:51:11 | 2019-07-08T16:51:11 | {} | MEMBER |
ClickHouse | ClickHouse | 509,762,065 | 5,951 | alexey-milovidov | This is very interesting and unusual feature!
But we need to consider all possible alternative implementation options... because syntax extension is the most intrusive implementation possible. I will ask my collegues and community members, what they think.
For example, another possible option is to use quoted col... | 2019-07-09T18:49:20 | 2019-07-09T18:49:20 | {} | MEMBER |
ClickHouse | ClickHouse | 510,283,321 | 5,959 | renkun2020 | the table only has less than 40 columns, and here is the user profile
<default>
<max_memory_usage>51539607552</max_memory_usage>
<max_memory_usage_for_all_queries>51539607552</max_memory_usage_for_all_queries>
<use_uncompressed_cache>1</use_uncompressed_cache>
... | 2019-07-11T00:52:20 | 2019-07-11T00:52:20 | {} | NONE |
ClickHouse | ClickHouse | 509,609,026 | 5,929 | filimonov | > > And also we don't have version 19.11.0 even in testing.
>
> We have 19.11 in https://repo.yandex.ru/clickhouse/deb/testing/main/
Yes. I've even installed that yesterday :) BTW: please remove those packages with prefix 2 from testing repo (OR add 2: to all of them).
```
clickhouse-server | 2:19.3.190 | http:... | 2019-07-09T11:45:02 | 2019-07-09T11:45:02 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 510,111,959 | 5,889 | blinkov | also there's a failing docs check | 2019-07-10T15:31:44 | 2019-07-10T15:31:44 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 509,006,029 | 5,014 | perez1987 | Any news? I have the same problem | 2019-07-07T14:56:41 | 2019-07-07T14:56:41 | {} | NONE |
ClickHouse | ClickHouse | 509,051,425 | 5,924 | alexey-milovidov | ```
milovidov-Latitude-E7470 :) SELECT * FROM cluster(test_unavailable_shard, system.one)
SELECT *
FROM cluster(test_unavailable_shard, system.one)
┌─dummy─┐
│ 0 │
└───────┘
↑ Progress: 0.00 rows, 0.00 B (0.00 rows/s., 0.00 B/s.) Received exception from server (version 19.11.0):
Code: 279. DB::Exception... | 2019-07-08T01:48:07 | 2019-07-08T01:49:16 | {} | MEMBER |
ClickHouse | ClickHouse | 509,134,125 | 5,363 | alex-krash | #5894 | 2019-07-08T08:38:41 | 2019-07-08T08:38:41 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 509,228,405 | 5,869 | alexey-milovidov | From experience of our clients, multi-continent ZooKeeper clusters are less usable. | 2019-07-08T13:37:37 | 2019-07-08T13:37:37 | {} | MEMBER |
ClickHouse | ClickHouse | 509,235,468 | 5,928 | alexey-milovidov | #5929 | 2019-07-08T13:56:17 | 2019-07-08T13:56:17 | {} | MEMBER |
ClickHouse | ClickHouse | 509,235,718 | 5,929 | alexey-milovidov | What is the purpose of this change? | 2019-07-08T13:56:52 | 2019-07-08T13:56:52 | {} | MEMBER |
ClickHouse | ClickHouse | 509,252,790 | 4,199 | igor-sh8 | Hi, @xmariachi.
The main problem in Too many parts error - is a lot of inserts and as a result a lot of merges.
So I came to the next schema: client inserts -> buffer table -> materializedview to table -> table.
In case I have a small cluster (4 servers) I decided to set failover schema on client side and not use ... | 2019-07-08T14:38:43 | 2019-07-08T14:38:43 | {} | NONE |
ClickHouse | ClickHouse | 509,275,086 | 5,938 | alexey-milovidov | See also https://github.com/lemire/simdjson/issues/193 | 2019-07-08T15:32:15 | 2019-07-08T15:32:15 | {} | MEMBER |
ClickHouse | ClickHouse | 509,306,288 | 1,469 | pramitchoudhary | Is there a plan on when the window function will be part of a stable release tentatively? | 2019-07-08T16:55:11 | 2019-07-08T16:55:11 | {
"+1": 5
} | NONE |
ClickHouse | ClickHouse | 509,942,252 | 5,951 | mfridental | I agree that COLUMNS is a better syntax. Also, in the second step I wanted to extend that to be able to retrieve only columns having at least one not-null and not-default value in them. Not sure whether is it possible with good performance, but what might be the syntax for that, something like `COLUMNS('.*') IF NOT EMP... | 2019-07-10T07:21:45 | 2019-07-10T07:21:45 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 508,962,761 | 5,614 | alexey-milovidov | @achimbab I've added your test to our CI. | 2019-07-07T00:40:05 | 2019-07-07T00:40:05 | {} | MEMBER |
ClickHouse | ClickHouse | 509,059,716 | 5,922 | alexey-milovidov | ```
$ docker run -it -v /home/milovidov/work/ClickHouse:/ClickHouse --rm ubuntu:disco /bin/bash
# apt update
# apt install wget tzdata
# (Choose Europe/Moscow)
# ls -l /etc/localtime
lrwxrwxrwx 1 root root 33 Jul 8 05:16 /etc/localtime -> /usr/share/zoneinfo/Europe/Moscow
# rm /etc/localtime
# ln -s ../usr/sha... | 2019-07-08T02:45:17 | 2019-07-08T02:45:17 | {} | MEMBER |
ClickHouse | ClickHouse | 509,184,832 | 5,500 | igor-sh8 | Thanks.
Your advises really helped us.
Just want to clarify a little about min_merge_bytes_to_use_direct_io and max_bytes_to_merge_at_max_space_in_pool. How to work with this settings ? What are the numbers in which cases we need to set ? Maybe you have some suggestions ? | 2019-07-08T11:18:01 | 2019-07-08T11:18:01 | {} | NONE |
ClickHouse | ClickHouse | 509,172,998 | 5,844 | nicelulu | Thank you for your reply.
Here is the flame graph.I did use many LowCardinality columns, about 50.DataTypeLowCardinality deserializeTextEscaped seems to create temp_nested_column for every element.
![imag... | 2019-07-08T10:34:20 | 2019-07-08T13:17:01 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 509,232,550 | 4,987 | alexey-milovidov | @zhang2014 It looks like build misconfiguration,
because we should not link to `gcc_s`.
You can use static linkage (the default). Probably we have errors in our CMake when shared linking is used. | 2019-07-08T13:48:44 | 2019-07-08T13:49:29 | {} | MEMBER |
ClickHouse | ClickHouse | 509,374,531 | 5,929 | alexey-milovidov | @alesapin
> ARG repository="deb http://repo.yandex.ru/clickhouse/deb/stable/ main/"
Maybe simply change repository to `testing`? Because `stable` doesn't have this version. | 2019-07-08T20:14:25 | 2019-07-08T20:14:25 | {} | MEMBER |
ClickHouse | ClickHouse | 509,437,953 | 5,500 | den-crane | >Your advises really helped us.
What did you change? And to what?
min_merge_bytes_to_use_direct_io -- this parameter controls a size of merge which should use direct disk read/write (without using Linux cache). The idea is to save more cache for selects. But it can cause more I/O because during merges (depends of... | 2019-07-09T00:12:05 | 2019-07-09T00:12:55 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 509,600,517 | 5,929 | polozhevets | > Yes, it would be more correct, but this URL also used inside the release process. And also we don't have version `19.11.0` even in testing. I can make changes in the release script to fix this issues, but why somebody would like to build `clickhouse-server` image from our repo? What is the purpose?
How to get the ... | 2019-07-09T11:15:11 | 2019-07-09T11:15:11 | {} | NONE |
ClickHouse | ClickHouse | 510,194,119 | 5,958 | alexey-milovidov | 00970_substring_arg_validation.sql - missing reference file. | 2019-07-10T19:22:09 | 2019-07-10T19:22:09 | {} | MEMBER |
ClickHouse | ClickHouse | 509,033,246 | 5,919 | alexey-milovidov | Continued in #5920 | 2019-07-07T21:45:43 | 2019-07-07T21:45:43 | {} | MEMBER |
ClickHouse | ClickHouse | 509,051,192 | 5,924 | alexey-milovidov | milovidov-Latitude-E7470 :) SELECT * FROM system.clusters
SELECT *
FROM system.clusters
```
┌─cluster──────────────────────────────────────┬─shard_num─┬─shard_weight─┬─replica_num─┬─host_name─┬─host_address─┬─port─┬─is_local─┬─user────┬─default_database─┐
│ single_remote_shard_at_port_9001 │ ... | 2019-07-08T01:46:22 | 2019-07-08T01:46:22 | {} | MEMBER |
ClickHouse | ClickHouse | 509,119,288 | 5,683 | blind-oracle | @den-crane
I don't think so.
The `Time` column is a monotonically-increasing unix timestamp which is in direct relation with `Date` column by which the partitioning is done. So if we slice `Time` value series by days or by months it should not make any difference - the values should be colocated in either case.
... | 2019-07-08T07:52:50 | 2019-07-08T07:52:50 | {} | NONE |
ClickHouse | ClickHouse | 509,264,397 | 5,935 | alexey-milovidov | Only if `rapidjson` is used (does not happen with `simdjson`) | 2019-07-08T15:05:48 | 2019-07-08T15:05:48 | {} | MEMBER |
ClickHouse | ClickHouse | 509,723,155 | 5,847 | perez1987 | We have the same problem. What data do you need? | 2019-07-09T16:57:08 | 2019-07-09T16:57:08 | {} | NONE |
ClickHouse | ClickHouse | 510,030,241 | 5,961 | decaseal | This error didn't throw in ClickHouse server version 19.4.3.11 (official build) | 2019-07-10T11:55:39 | 2019-07-10T12:07:00 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 510,052,226 | 4,470 | davtoro | any news on this? Can't use clickhouse-server or client without openssl.conf workaround on Debian 10 Buster default installation. | 2019-07-10T13:04:42 | 2019-07-10T13:04:42 | {
"+1": 4,
"laugh": 1
} | NONE |
ClickHouse | ClickHouse | 508,963,609 | 5,306 | alexey-milovidov | > range Int8 CODEC(Delta(1), ZSTD(15)),
The issue was fixed in #5786
(the fix is available in 19.11 testing) | 2019-07-07T01:02:12 | 2019-07-07T01:02:12 | {} | MEMBER |
ClickHouse | ClickHouse | 508,963,778 | 5,480 | alexey-milovidov | The fix is available in version 19.11 (testing). | 2019-07-07T01:06:26 | 2019-07-07T01:06:26 | {} | MEMBER |
ClickHouse | ClickHouse | 509,170,953 | 5,601 | kirankn | Our usecase also requires Kafka to have Avro encoded data. It would be very nice if Clickhouse could support Avro for input. | 2019-07-08T10:27:16 | 2019-07-08T10:27:16 | {} | NONE |
ClickHouse | ClickHouse | 509,728,449 | 2,633 | alexey-milovidov | Now we have `empty_as_default` but still don't have `null_as_default`. | 2019-07-09T17:13:13 | 2019-07-09T17:13:13 | {} | MEMBER |
ClickHouse | ClickHouse | 509,736,216 | 5,946 | alexey-milovidov | > I have no idea why my commit fail this test - '00908_long_http_insert' (((
It's unrelated. | 2019-07-09T17:36:07 | 2019-07-09T17:36:07 | {} | MEMBER |
ClickHouse | ClickHouse | 510,108,450 | 5,254 | Schaefer-Thomas | **I can confirm a problem with joins and alias in version 19.9.3:**
alias not used in group by, but underlying column:
```
SELECT test2.id AS c0
FROM test1
INNER JOIN test2 ON test2.test1_id = test1.id
INNER JOIN test3 ON test2.test3_id = test3.id
GROUP BY test2.id
```
Received exception from server (ve... | 2019-07-10T15:23:21 | 2019-07-10T15:23:21 | {} | NONE |
ClickHouse | ClickHouse | 508,963,693 | 5,847 | alexey-milovidov | @dmitryikh
Does the issue vanish if you install version 19.11? | 2019-07-07T01:04:19 | 2019-07-07T01:04:19 | {} | MEMBER |
ClickHouse | ClickHouse | 509,116,999 | 5,543 | blind-oracle | Thanks! The fix also went to 19.x stable release. | 2019-07-08T07:45:01 | 2019-07-08T07:45:01 | {} | NONE |
ClickHouse | ClickHouse | 509,724,635 | 2,802 | alexey-milovidov | > I want to know what function can be used to convert the date into week number and partitioned on that.
Note: Tried toRelativeWeekNum but am not able to understand how it works.
Yes, it will work. You can also use `toMonday` function. | 2019-07-09T17:01:29 | 2019-07-09T17:01:29 | {} | MEMBER |
ClickHouse | ClickHouse | 509,966,949 | 4,959 | AlexAkulov | Any progress? | 2019-07-10T08:35:22 | 2019-07-10T08:35:22 | {} | NONE |
ClickHouse | ClickHouse | 508,985,359 | 5,683 | blind-oracle | Probably this is expected. Solved by using `Delta` codec on `Time`. Let's close it for now. | 2019-07-07T09:33:58 | 2019-07-07T09:33:58 | {} | NONE |
ClickHouse | ClickHouse | 509,133,774 | 5,894 | alex-krash | I've finished | 2019-07-08T08:37:39 | 2019-07-08T08:37:39 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 509,169,069 | 3,815 | selitsky | > There's a `WITH` clause https://github.com/yandex/ClickHouse/blob/master/dbms/tests/queries/0_stateless/00490_with_select.sql , but it has limitations that prevent its usage in specifically your case.
WITH clause only work in particular select, but not inside subqueries. | 2019-07-08T10:20:49 | 2019-07-08T10:20:49 | {
"+1": 1
} | NONE |
ClickHouse | ClickHouse | 509,575,168 | 5,932 | projectsbird123 | I'm able to run the ClickHouse using the below command but not as a service.
Command used to run the clickhouse:
clickhouse-server --config-file=/etc/clickhouse-server/config.xml
| 2019-07-09T09:51:23 | 2019-07-09T09:51:23 | {} | NONE |
ClickHouse | ClickHouse | 509,605,684 | 5,929 | alexey-milovidov | > And also we don't have version 19.11.0 even in testing.
We have 19.11 in https://repo.yandex.ru/clickhouse/deb/testing/main/ | 2019-07-09T11:32:58 | 2019-07-09T11:32:58 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 508,963,548 | 5,499 | alexey-milovidov | Could you please share your table definitions? | 2019-07-07T01:00:41 | 2019-07-07T01:00:41 | {} | MEMBER |
ClickHouse | ClickHouse | 509,056,140 | 5,922 | alexey-milovidov | + https://github.com/yandex/ClickHouse/commit/4e1286e1e562d7ec3e7ddf51fb982a5b75359bd1 | 2019-07-08T02:21:55 | 2019-07-08T02:22:42 | {} | MEMBER |
ClickHouse | ClickHouse | 509,551,736 | 5,945 | akonyaev90 | ```
std::string userInfo = uri.getUserInfo();
if (userInfo.front() != ':') {
char *user = strtok(const_cast<char *> (userInfo.c_str()), ":");
if (user != NULL) {
hdfsBuilderSetUserName(builder.get(), user);
}
}
```
I think this code is good solution to this issue... | 2019-07-09T08:47:35 | 2019-07-09T08:47:35 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 509,618,327 | 5,436 | alexey-milovidov | @zhang2014
Sorry, I've missed your reply.
It's Ok, but I doubt about substitutions for database and table. It will be inconsistent with existing query parameters, let's postpone them. | 2019-07-09T12:16:29 | 2019-07-09T12:16:29 | {} | MEMBER |
ClickHouse | ClickHouse | 510,040,746 | 5,940 | SaltTan | Yesterday I was upgrading a 5-node cluster from 19.8 to 19.9.
After the first node was upgraded, distributed queries stopped working with the same message about unknown packet, and two of the four nodes with 19.8 received signal Segmentation fault | 2019-07-10T12:30:58 | 2019-07-10T12:30:58 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 508,963,288 | 5,025 | alexey-milovidov | This part of code was changed in recent versions. Does the issue reproduce with latest stable release? | 2019-07-07T00:53:49 | 2019-07-07T00:53:49 | {} | MEMBER |
ClickHouse | ClickHouse | 509,725,231 | 2,908 | pryg-skok | Very interesting, thanks. I'm not sure we will use it in production because we have strong space restrictions, but will give it a try. | 2019-07-09T17:03:11 | 2019-07-09T17:03:11 | {} | NONE |
ClickHouse | ClickHouse | 509,768,841 | 5,951 | alexey-milovidov | Another option is to add COLUMNS keyword to select columns by regexp:
```
SELECT a, COLUMNS('^x'), b FROM table
``` | 2019-07-09T19:09:32 | 2019-07-09T19:10:53 | {} | MEMBER |
ClickHouse | ClickHouse | 509,852,872 | 5,955 | den-crane | You mean the result is inconsistent because of replication lag? | 2019-07-09T23:56:44 | 2019-07-09T23:56:44 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 509,854,815 | 5,955 | abraithwaite | That was my first assumption, however that was not the case. I checked each shard and replica individually. The replicas in each shard were identical.
Should have called this out in the initial report.
I should also note that subsequent queries would sometimes return lower counts than the real totals and somet... | 2019-07-10T00:06:13 | 2019-07-10T00:06:13 | {} | NONE |
ClickHouse | ClickHouse | 509,907,559 | 5,955 | nvartolomei | > Switching the load_balancing setting to in_order (mostly) resolves the issue. in_order still breaks during restarts (doesn't seem to query the other healthy replica for the shard).
Sounds like you might have a misconfigured cluster. Tables metadata (replicated+distributed) + macros config for each node + cluster c... | 2019-07-10T04:55:28 | 2019-07-10T04:56:07 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 518,627,399 | 5,859 | 4ertus2 | > @4ertus2
> Hi, I found there was a segmentation fault coredump after this fix. It can be reproduced as follows.
Hi Lixiang Qian,
There were some crash fixes till this patch. I've added your test to build checks in #6362
We'll look if it fails.
| 2019-08-06T11:20:43 | 2019-08-06T11:20:43 | {} | CONTRIBUTOR |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.