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 | 607,803,403 | 10,015 | Slach | related to https://github.com/ClickHouse/ClickHouse/pull/7900
https://github.com/prometheus/docs/pull/1513
@vdimir @alexey-milovidov please look at this PR
@vdimir if this pull request will merge I think you can change the link in Prometheus docs | 2020-04-02T12:02:28 | 2020-04-02T12:02:28 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 608,171,724 | 10,020 | alexey-milovidov | realloc to zero size is indeed valid.
But what about the calling code?
What is the call trace? Does it look valid? | 2020-04-03T00:59:00 | 2020-04-03T00:59:27 | {} | MEMBER |
ClickHouse | ClickHouse | 608,493,469 | 10,017 | filimonov | @bashlykevich the version you use had several bugs. All the crash reports you sent may be related, and most probably the problem was fixed long ago.
Please update to one of the recent versions, for example, 19.16.16.72 or 20.1.9.54, and check if the problem get solved. If you will be able to reproduce the problem o... | 2020-04-03T15:04:18 | 2020-04-03T15:04:18 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 608,494,725 | 10,018 | filimonov | #10016 #10012 #10017 #10018 | 2020-04-03T15:06:39 | 2020-04-03T15:06:39 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 608,501,056 | 10,030 | filimonov | Exact duplicate of https://github.com/ClickHouse/ClickHouse/issues/7767#issuecomment-554306202
| 2020-04-03T15:18:35 | 2020-04-03T15:18:35 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 609,548,163 | 6,093 | bobelev | Still an issue
```plain
OPTIMIZE TABLE default.notify_rewards FINAL
Timeout exceeded while receiving data from server. Waited for 300 seconds, timeout is 300 seconds.
Cancelling query.
Received exception from server (version 20.3.5):
Code: 999. DB::Exception: Received from localhost:9000. DB::Exception: Ses... | 2020-04-06T03:35:58 | 2020-04-06T03:35:58 | {
"+1": 2
} | NONE |
ClickHouse | ClickHouse | 608,110,327 | 9,980 | filimonov | Logs would be helpful.
Did you try `Kill mutation` Command? | 2020-04-02T21:52:48 | 2020-04-02T21:52:48 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 608,121,780 | 7,141 | filimonov | https://github.com/ClickHouse/ClickHouse/pull/9923 | 2020-04-02T22:21:11 | 2020-04-02T22:21:11 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 608,170,304 | 10,025 | alexey-milovidov | There is no visible difference in performance test.
I still don't trust the new performance test for 100% (only by 99%).
@amosbird can you please also check manually with `clickhouse-benchmark`? | 2020-04-03T00:52:51 | 2020-04-03T00:52:51 | {} | MEMBER |
ClickHouse | ClickHouse | 608,253,501 | 1,726 | azat | >In Global_Master, select query on the distributed table seems to work fine.
Looks like your nested distributed table has only one shard (not sure completely about your structure though) and in this case it should work (even before #9923) | 2020-04-03T06:23:35 | 2020-04-03T06:23:35 | {} | MEMBER |
ClickHouse | ClickHouse | 608,391,399 | 9,997 | azat | Rebased again, now everything is ok (regarding conflicts and commits) | 2020-04-03T11:52:21 | 2020-04-03T11:52:21 | {} | MEMBER |
ClickHouse | ClickHouse | 608,989,719 | 10,046 | filimonov | IPv6 is disabled in docker by default (check docker docs)
Normally you see such a msg, but it should not prevent CH from starting because config for docker contains listen_try, did you change the default config? | 2020-04-04T07:29:51 | 2020-04-04T07:30:36 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 609,497,260 | 5,604 | simPod | @blinkov only as a feedback, if you wish to improve the error msg or not. Can be closed. | 2020-04-05T22:44:08 | 2020-04-05T22:44:08 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 607,742,280 | 9,993 | johnatannvmd | Quotas
```XML
<quotas>
<!-- Default quotas. -->
<default>
<duration>3600</duration>
<queries>0</queries>
<errors>0</errors>
<result_rows>0</result_rows>
<read_rows>0</read_rows>
<execution_time>0</execution_time>
</default>
</quotas... | 2020-04-02T09:51:38 | 2020-04-02T09:51:38 | {} | NONE |
ClickHouse | ClickHouse | 607,944,368 | 10,010 | johnatannvmd | You could use HaProxy or similar in front of ClickHouse servers and connect to the proxy instead of using the direct connection.
Anyway, if you use Native protocol which is TCP when the instance goes down you should manually reconnect to another or just reconnect if you use the proxy. | 2020-04-02T16:15:27 | 2020-04-02T16:15:27 | {} | NONE |
ClickHouse | ClickHouse | 608,219,785 | 9,826 | lu-yuki-matsumoto | @KochetovNicolai
I'm sorry to be very late.
In our case,
```
CREATE TABLE tableB
(`id` UInt64,
`colA` UInt32,
`colB` UInt32,
`colC` Enum8(2 vals),
`Aid` UInt32,
`colD` Enum8(8 vals),
`colE` UInt8,
`colF` UInt8,
`colG` UInt8,
`colH` DateTime,
`colDate` Date
) ENGINE = ReplacingMergeTree(... | 2020-04-03T04:17:35 | 2020-04-03T04:18:14 | {} | NONE |
ClickHouse | ClickHouse | 608,305,956 | 9,336 | bohutang | Comments:
mysql-connector-java-5.1.34:
`
/* mysql-connector-java-5.1.34 ( Revision: jess.balint@oracle.com-20141014163213-wqbwpf1ok2kvo1om ) */SHOW VARIABLES WHERE Variable_name ='language' OR Variable_name = 'net_write_timeout' OR Variable_name = 'interactive_timeout' OR Variable_name = 'wait_timeout' OR Variable... | 2020-04-03T08:34:18 | 2020-04-03T08:34:18 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 609,438,329 | 4,628 | sundarv85 | Thanks @blinkov. No questions anymore. It worked. | 2020-04-05T15:54:15 | 2020-04-05T15:54:15 | {} | NONE |
ClickHouse | ClickHouse | 609,662,823 | 1,661 | Slind14 | We still haven't found a solution to achieve this in a way that scales. | 2020-04-06T08:52:39 | 2020-04-06T08:52:39 | {
"+1": 1,
"confused": 1
} | NONE |
ClickHouse | ClickHouse | 609,674,861 | 10,052 | qoega | Did you try to test it after changes? | 2020-04-06T09:16:04 | 2020-04-06T09:16:04 | {} | MEMBER |
ClickHouse | ClickHouse | 607,626,106 | 4,022 | TomaszUrugOlszewski | No, thank you | 2020-04-02T05:16:08 | 2020-04-02T05:16:08 | {} | NONE |
ClickHouse | ClickHouse | 607,711,842 | 4,470 | dstruck | Tested with version 20.3.4.10 on Debian Buster. Fix was not needed. | 2020-04-02T08:50:58 | 2020-04-02T08:50:58 | {} | NONE |
ClickHouse | ClickHouse | 608,091,715 | 9,997 | azat | FWIW: I rebased against upstream/master, but github failed to update PR | 2020-04-02T21:12:11 | 2020-04-02T21:12:11 | {} | MEMBER |
ClickHouse | ClickHouse | 608,394,187 | 10,011 | alesapin | Can you show your table schema (show create table)? | 2020-04-03T11:59:35 | 2020-04-03T11:59:35 | {} | MEMBER |
ClickHouse | ClickHouse | 608,498,845 | 10,025 | akuzm | I don't see a significant difference on website.xml perftest, and a couple percent slowdown on term4 in string_hash_map synthetic test (both bigger and smaller keys look OK for some reason). | 2020-04-03T15:14:35 | 2020-04-03T15:14:35 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 608,509,973 | 10,031 | filimonov | > The TTL is dynamic so that it changes when I change the x, right? It won't change frequently, maybe once or twice.
If you plan to change TTL, ALTER DELETE will work better for you (with TTL you will need to process the same data 2 times - when column value changed, and then MATERIALIZE TTL)
| 2020-04-03T15:35:22 | 2020-04-03T15:35:22 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 609,515,878 | 9,836 | hustnn | @filimonov You are correct. Bigger parts are better. Just imaging some worst case. Not sure the pipelining execution model helps, so I am asking you guys. | 2020-04-06T00:55:08 | 2020-04-06T00:55:08 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 609,523,509 | 10,057 | alexey-milovidov | PS. It's unconvinient that I cannot write `LAYOUT(HASHED)` instead of `LAYOUT(HASHED())`. | 2020-04-06T01:33:29 | 2020-04-06T01:33:29 | {} | MEMBER |
ClickHouse | ClickHouse | 607,987,635 | 9,607 | CurtizJ | Closed in favor to #10026. | 2020-04-02T17:32:33 | 2020-04-02T17:32:33 | {} | MEMBER |
ClickHouse | ClickHouse | 608,121,485 | 1,726 | filimonov | https://github.com/ClickHouse/ClickHouse/pull/9923 | 2020-04-02T22:20:24 | 2020-04-02T22:20:24 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 607,692,535 | 9,997 | azat | Failures looks unrelated | 2020-04-02T08:14:39 | 2020-04-02T08:14:39 | {} | MEMBER |
ClickHouse | ClickHouse | 607,694,877 | 6,595 | gasabr | No, everything was solved, sorry, I forgot to close the issue. | 2020-04-02T08:18:59 | 2020-04-02T08:18:59 | {} | NONE |
ClickHouse | ClickHouse | 607,941,698 | 10,005 | den-crane | could not reproduce
```
SELECT toString(number) LIKE concat('%', '1', '%')
FROM numbers(10)
┌─like(toString(number), concat('%', '1', '%'))─┐
│ 0 │
│ 1 │
│ 0 │
│ ... | 2020-04-02T16:11:03 | 2020-04-02T16:11:03 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 608,500,304 | 8,814 | filimonov | No. Most probably it is #6833 (was fixed recenly). Please update to one of the latest ClickHouse versions
v19.16.16.72-stable
v20.1.9.54-stable
or
v20.3.5.21-stable (there are few unfixed regressions here yet).
| 2020-04-03T15:17:11 | 2020-04-03T15:17:11 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 608,503,797 | 9,461 | hellvesper | I have this issue too, is there some workaround? | 2020-04-03T15:23:41 | 2020-04-03T15:23:41 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 608,953,831 | 6,565 | ChannelerH | > @ChannelerH, do you have any further questions?
no,I donot | 2020-04-04T01:54:02 | 2020-04-04T01:54:02 | {} | NONE |
ClickHouse | ClickHouse | 609,497,091 | 7,749 | simPod | @blinkov I'd be interested | 2020-04-05T22:43:05 | 2020-04-05T22:43:05 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 609,520,313 | 10,056 | alexey-milovidov | The reason is obvious. It does not process the case when database name is specified as identifier: `conv_main`. | 2020-04-06T01:18:45 | 2020-04-06T01:18:45 | {} | MEMBER |
ClickHouse | ClickHouse | 609,681,687 | 10,052 | Vxider | > Did you try to test it after changes?
Sorry, I don’t have a complete test environment locally, so I can’t run all the tests (e.g. Kafka, zookeeper) on my computer. | 2020-04-06T09:29:20 | 2020-04-06T09:29:20 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 609,813,959 | 10,062 | den-crane | > Thanks for explanation! Does it mean that other view (not materialized) can be created on top of it to apply grouping?
yes.
Be aware that double groupping and reading excessive columns may slow-down your queries up to 10 times.
```
select sum(view_count), user_id
from grouped_user_video_view
group by ... | 2020-04-06T14:01:34 | 2020-04-06T14:02:12 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 607,767,111 | 9,993 | johnatannvmd | set `read_overflow_mode` to `throw` + set `result_overflow_mode` to `throw` got exceptions about read rows limit. set `max_rows_to_read` to 0 and re run query with `groupUniqArray` and got my results
```
14 rows in set. Elapsed: 460.128 sec. Processed 19.63 billion rows, 40.70 GB (42.66 million rows/s., 88.45 MB/s.... | 2020-04-02T10:41:33 | 2020-04-02T10:41:33 | {} | NONE |
ClickHouse | ClickHouse | 607,796,565 | 10,014 | abyss7 | Looks like I made a final merge in a wrong way for this file, before moving folders. | 2020-04-02T11:48:09 | 2020-04-02T11:48:09 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 608,188,461 | 10,005 | M1NGc |
also wrong | 2020-04-03T02:04:42 | 2020-04-03T02:04:42 | {} | NONE |
ClickHouse | ClickHouse | 608,407,597 | 10,020 | abyss7 | > realloc to zero size is indeed valid.
> But what about the calling code?
> What is the call trace? Does it look valid?
I found it while investigating the bug with zero-containing strings in new StringHashMap. | 2020-04-03T12:32:20 | 2020-04-03T12:32:20 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 608,493,719 | 10,018 | filimonov | @bashlykevich the version you use had several bugs. All the crash reports you sent may be related, and most probably the problem was fixed long ago.
Please update to one of the recent versions, for example, 19.16.16.72 or 20.1.9.54, and check if the problem get solved. If you will be able to reproduce the problem o... | 2020-04-03T15:04:46 | 2020-04-03T15:04:46 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 608,995,106 | 9,681 | l1t1 | i think the old pdf is better than the new pdf
both the lines and the Chinese character of new pdf are unreadable
new
old
old English
![old_en2020-04-04_163555](https://user-images.githubusercontent.com/23301704/78422579-d7708600-7692-11ea-8350-00e... | 2020-04-04T08:40:32 | 2020-04-04T08:40:32 | {} | NONE |
ClickHouse | ClickHouse | 608,998,971 | 10,049 | azat | Any thoughts on this? If this is worth it, then I will update the tests. | 2020-04-04T09:01:07 | 2020-04-04T09:01:07 | {} | MEMBER |
ClickHouse | ClickHouse | 609,856,095 | 10,003 | alesapin | 00992 and 00993 https://github.com/ClickHouse/ClickHouse/issues/10066 | 2020-04-06T15:13:44 | 2020-04-06T15:13:44 | {} | MEMBER |
ClickHouse | ClickHouse | 608,052,351 | 10,027 | hellvesper | https://github.com/ClickHouse/ClickHouse/blob/4075f26583f847df0e7f5368375ff7060b59b83d/dbms/Processors/RowsBeforeLimitCounter.h#L20
`std::memory_order::release` to `std::memory_order_release`
fixed problem | 2020-04-02T19:14:19 | 2020-04-02T19:14:19 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 608,236,429 | 10,024 | filimonov | What's about putting a line in your
.bashrc?
```
CLICKHOUSE_HISTORY_FILE=$XDG_DATA_HOME/.clickhouse-client-history
``` | 2020-04-03T05:24:29 | 2020-04-03T05:24:29 | {
"+1": 1,
"-1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 608,499,927 | 7,687 | yamrzou | Thanks @blinkov. No, this issue can be closed. | 2020-04-03T15:16:32 | 2020-04-03T15:16:32 | {} | NONE |
ClickHouse | ClickHouse | 608,516,147 | 10,025 | amosbird | Unfortunately I just verified with string_hash_map test locally and observed 10% regression on term48
```
each test is the best time of six runs
term1 : 0.788 0.772
term2 : 1.527 1.521
term4 : 3.635 3.652
term8 : 7.851 7.856
term16: 9.223 9.359
term24: 10.774 10.622
term48: 13.337 ... | 2020-04-03T15:47:22 | 2020-04-03T15:48:13 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 609,032,924 | 8,373 | blinkov | @yehiaelbehery, do you have any further questions? | 2020-04-04T14:00:47 | 2020-04-04T14:00:47 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 607,764,360 | 4,797 | ghost | @blinkov No, thanks for asking!
The idea about exposing Kafka metadata fields (https://github.com/ClickHouse/ClickHouse/issues/4797#issuecomment-476324427 and https://github.com/ClickHouse/ClickHouse/issues/4797#issuecomment-477359590) was somewhat off-topic, so in case of someone wants to continue discussing it, it... | 2020-04-02T10:35:57 | 2020-04-02T10:35:57 | {} | NONE |
ClickHouse | ClickHouse | 607,950,587 | 9,948 | alesapin | It's not ready for review. Code for polymorphic parts left. | 2020-04-02T16:26:23 | 2020-04-02T16:26:23 | {} | MEMBER |
ClickHouse | ClickHouse | 608,470,108 | 10,031 | th0th | Thanks a lot @den-crane! I didn't know that.
The TTL is dynamic so that it changes when I change the `x`, right? It won't change frequently, maybe once or twice. Also, does changing `x` bring a great cost or is it a pretty common case? | 2020-04-03T14:38:45 | 2020-04-03T14:38:45 | {} | NONE |
ClickHouse | ClickHouse | 609,015,536 | 8,076 | nvartolomei | I'll avoid rebasing this until it gets a green light, did it too many times already. | 2020-04-04T11:39:35 | 2020-04-04T11:39:35 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 608,168,915 | 10,025 | alexey-milovidov | @amosbird zero byte inside a string is perfectly legal.
We allow to store arbitrary binary data in strings and thus we don't need a separate blob type.
Any issues with zero bytes are usually attributed to ignorant practices in old-fashioned C code. | 2020-04-03T00:47:03 | 2020-04-03T00:50:31 | {} | MEMBER |
ClickHouse | ClickHouse | 608,390,293 | 10,037 | alesapin | Seems like you have updated from the extremely old version. `replica_name/metadata` node was introduced about two years ago. Now you can just copy `/table/metadata` node to `/table/replicas/replica_name/metadata` with `zkCli.sh` util. | 2020-04-03T11:49:22 | 2020-04-03T11:49:22 | {} | MEMBER |
ClickHouse | ClickHouse | 608,505,966 | 10,011 | erste | Sorry, i cant due to security issues.
All i could provide is in the first message:
```CREATE TABLE loadtest.stat_rpl (`mdate` Date) ENGINE = ReplicatedMergeTree('/clickhouse/tables/loadtest/{shard}/stat_rpl', '{replica}') PARTITION BY mdate ORDER BY mdate TTL mdate + toIntervalDay(3) SETTINGS index_granularity = 8192... | 2020-04-03T15:27:52 | 2020-04-03T15:27:52 | {
"+1": 1
} | NONE |
ClickHouse | ClickHouse | 609,742,728 | 10,011 | alesapin | Sorry, I cannot reproduce the problem, both on your version and on current master. I recommend you to update to the latest 20.3 stable version. | 2020-04-06T11:41:29 | 2020-04-06T11:41:29 | {} | MEMBER |
ClickHouse | ClickHouse | 609,772,767 | 10,061 | dependabot-preview[bot] | Looks like certifi is up-to-date now, so this is no longer needed. | 2020-04-06T12:47:24 | 2020-04-06T12:47:24 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 609,793,883 | 4,717 | blinkov | @SaltTan, do you have any further questions? | 2020-04-06T13:27:22 | 2020-04-06T13:27:22 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 609,005,686 | 4,393 | simPod | @blinkov Hi, yes. WDYT of my last comment? | 2020-04-04T10:00:11 | 2020-04-04T10:00:11 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 609,770,149 | 10,062 | den-crane | Yes. SumminngMT collapses rows during the Merge.
SummingMT and AggregatingMT expect that select query will do final aggregation using group by. Merges are eventual and may never happen.
So CH expects that all queries will do final summing using
```
select sum(view_count) , ...
from grouped_user_video_view
gro... | 2020-04-06T12:41:43 | 2020-04-06T12:42:39 | {
"eyes": 1,
"heart": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 607,953,378 | 1,784 | Boris-Barboris | We also have this problem on about 200 million rows, OPTIMIZE in any shape and form does nothing and old rows are still duplicated. SELECT FINAL does **not** help as well.
ClickHouse client version 19.17.2.4 (official build).
@blinkov
Am I right to assume that in the end there is no strong guarantee about ReplRep... | 2020-04-02T16:31:23 | 2020-04-02T16:44:46 | {} | NONE |
ClickHouse | ClickHouse | 607,990,627 | 10,025 | amosbird | I wonder, what's the case of having `\0` in a string. Is it somehow a binary instead? Perhaps we need a blob type | 2020-04-02T17:37:39 | 2020-04-02T17:37:39 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 608,313,082 | 8,110 | ghost | @blinkov
Yes, I still wonder how to do this.
Thank you. | 2020-04-03T08:45:50 | 2020-04-03T08:45:50 | {} | NONE |
ClickHouse | ClickHouse | 608,359,115 | 9,826 | KochetovNicolai | @lu-yuki-matsumoto thank you.
Is the query I used (in my previous comment) the same as the your query?
I still can't reproduce the issue on my artificial generated data.
You may help me to investigate this issue:
* Try to reduce the query. Usually, most parts of the query can be removed, until you get a small que... | 2020-04-03T10:28:58 | 2020-04-03T10:28:58 | {} | MEMBER |
ClickHouse | ClickHouse | 608,465,841 | 10,034 | den-crane | > is there any efficient way to re-shard the old data to achieve my purpose ?
No. There is no efficient way.
You need to reload all data somehow.
insert... select
clickhouse-copier | 2020-04-03T14:30:40 | 2020-04-03T14:30:40 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 608,495,112 | 10,012 | filimonov | #10016 #10012 #10017 #10018 | 2020-04-03T15:07:24 | 2020-04-03T15:07:24 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 607,699,119 | 9,993 | johnatannvmd | > 1. Try to disable use_processors
> set experimental_use_processors=0
Same result
> 2. Need 2 traces of queries with different results
First query:
```
SELECT DISTINCT
field1,
groupUniqArray(field2)
FROM default.table1
GROUP BY field1
ORDER BY field1 ASC
[clk01] 2020.04.02 08:07:34.889302... | 2020-04-02T08:27:03 | 2020-04-02T08:27:53 | {} | NONE |
ClickHouse | ClickHouse | 607,711,655 | 5,292 | igor-sh8 | Hi.
No, on the new versions of the clickhouse, we do not encounter such problems.
So I think the problem is solved. | 2020-04-02T08:50:39 | 2020-04-02T08:50:39 | {} | NONE |
ClickHouse | ClickHouse | 607,892,344 | 8,193 | geoffrank | No, thanks. I've switched to using underscores instead of dots in column names for non-nested columns. | 2020-04-02T14:48:50 | 2020-04-02T14:48:50 | {} | NONE |
ClickHouse | ClickHouse | 607,942,409 | 9,991 | den-crane | toString function is not monotonic.
See a different result
```
SELECT
11 > 9,
toString(11) > toString(9)
┌─greater(11, 9)─┬─greater(toString(11), toString(9))─┐
│ 1 │ 0 │
└────────────────┴────────────────────────────────────┘
``` | 2020-04-02T16:12:14 | 2020-04-02T16:12:14 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 608,100,919 | 9,836 | filimonov | @hustnn and what are the advantages of having 1 mln tiny parts, instead of (lets say) 1000 bigger parts? | 2020-04-02T21:30:50 | 2020-04-02T21:30:50 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 608,180,818 | 9,980 | leoones | > Logs would be helpful.
>
> Did you try `Kill mutation` Command?
I try "Kill mutation" and "system stop merges fct_apa_report_sp_productads"
But it didn't work .
System.merges table is still displayed in merge。
Merge is running continuously。
After I restart the Clickhouse service, merge is released. | 2020-04-03T01:35:53 | 2020-04-03T01:35:53 | {} | NONE |
ClickHouse | ClickHouse | 608,402,082 | 10,015 | Slach | > Don't forget to mirror changes to Russian version, please :-)
done, I hope is enough
@BayoNet please look at this PR again | 2020-04-03T12:19:04 | 2020-04-03T12:19:04 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 608,432,385 | 10,031 | th0th | Thanks for the reply @filimonov, I saw it but I don't think it would work for my case.
How can I handle different retention periods with TTL? What should my data structure be? Something like this?
```sql
CREATE TABLE example_table
(
date_time DateTime,
valid_until DateTime,
a Int TTL valid_until ... | 2020-04-03T13:26:24 | 2020-04-03T13:26:51 | {} | NONE |
ClickHouse | ClickHouse | 608,494,833 | 10,016 | filimonov | #10016 #10012 #10017 #10018 | 2020-04-03T15:06:49 | 2020-04-03T15:06:49 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 607,761,517 | 9,826 | KochetovNicolai | Things I tried:
```
create table tableA (id UInt64, col1 UInt64, colDate Date) engine = MergeTree order by colDate;
create table tableB (Aid UInt64) engine = MergeTree order by tuple();
insert into tableA select number, number % 10, '2020-01-01' from numbers(10000000);
insert into tableA select number, number % ... | 2020-04-02T10:30:18 | 2020-04-02T10:30:18 | {} | MEMBER |
ClickHouse | ClickHouse | 608,134,848 | 4,797 | filimonov | > @blinkov No, thanks for asking!
>
> The idea about exposing Kafka metadata fields ([#4797 (comment)](https://github.com/ClickHouse/ClickHouse/issues/4797#issuecomment-476324427) and [#4797 (comment)](https://github.com/ClickHouse/ClickHouse/issues/4797#issuecomment-477359590)) was somewhat off-topic, so in case of... | 2020-04-02T22:59:37 | 2020-04-02T22:59:37 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 608,185,540 | 10,005 | alexey-milovidov | @4ertus2 This is an issue with JOIN (major regression). | 2020-04-03T01:54:09 | 2020-04-03T01:54:29 | {} | MEMBER |
ClickHouse | ClickHouse | 608,494,103 | 10,016 | filimonov | @bashlykevich the version you use had several bugs. All the crash reports you sent may be related, and most probably the problem was fixed long ago.
Please update to one of the recent versions, for example, 19.16.16.72 or 20.1.9.54, and check if the problem get solved. If you will be able to reproduce the problem o... | 2020-04-03T15:05:29 | 2020-04-03T15:05:29 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 608,494,996 | 10,017 | filimonov | #10016 #10012 #10017 #10018 | 2020-04-03T15:07:08 | 2020-04-03T15:07:08 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 608,574,864 | 10,043 | Tri0L | related to #7900 | 2020-04-03T17:47:33 | 2020-04-03T17:47:33 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 609,043,283 | 8,373 | yehiaelbehery | No, Thanks for your help. | 2020-04-04T15:14:11 | 2020-04-04T15:14:11 | {} | NONE |
ClickHouse | ClickHouse | 609,797,742 | 10,062 | grachov | Thanks for explanation! Does it mean that other view (not materialized) can be created on top of it to apply grouping? | 2020-04-06T13:34:18 | 2020-04-06T13:34:18 | {} | NONE |
ClickHouse | ClickHouse | 607,753,620 | 9,335 | chhaya-sharma | @den-crane It doesn't work for me in UTC.
As the table was created without providing any zone value, so the default value of datetime would be UTC based. And as you saw in my previous comment, it returned corrupted values on retrieval. | 2020-04-02T10:14:17 | 2020-04-02T10:14:17 | {} | NONE |
ClickHouse | ClickHouse | 608,493,177 | 10,012 | filimonov | @bashlykevich the version you use had several bugs. All the crash reports you sent may be related, and most probably the problem was fixed long ago.
Please update to one of the recent versions, for example, 19.16.16.72 or 20.1.9.54, and check if the problem get solved. If you will be able to reproduce the problem o... | 2020-04-03T15:03:42 | 2020-04-03T15:03:42 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 608,179,711 | 9,980 | leoones | > At least we need to know ClickHouse version and `clickhouse-server.log` to investigate the issue.
version 19.15.3.6 | 2020-04-03T01:30:55 | 2020-04-03T01:30:55 | {} | NONE |
ClickHouse | ClickHouse | 608,183,561 | 10,005 | leeleehero | fields are string | 2020-04-03T01:46:45 | 2020-04-03T01:46:45 | {} | NONE |
ClickHouse | ClickHouse | 608,256,111 | 9,997 | azat | Okay, now it shows that there are no conflicts (it is correct), but also it shows 13 commits while it there is only 4 (other 9 commits already in upstream/master and sha1 are the same) | 2020-04-03T06:32:14 | 2020-04-03T06:32:14 | {} | MEMBER |
ClickHouse | ClickHouse | 608,325,419 | 7,572 | zhang2014 | I will quickly go through the code again(because it's pending too long, I need to make sure I'm not missing anything) | 2020-04-03T09:12:42 | 2020-04-03T09:12:42 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 608,238,135 | 10,031 | filimonov | Did you check TTL functionality?
https://clickhouse.tech/docs/en/operations/table_engines/mergetree/#table_engine-mergetree-ttl | 2020-04-03T05:31:13 | 2020-04-03T10:59:02 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 608,385,833 | 10,015 | BayoNet | Don't forget to mirror changes to Russian version, please :-) | 2020-04-03T11:37:52 | 2020-04-03T11:37:52 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 608,467,083 | 10,031 | den-crane | TTL can contain an expression
```
CREATE TABLE example_table
(
d DateTime,
a Int,
x Int
)
ENGINE = MergeTree
PARTITION BY toYYYYMM(d)
ORDER BY d
TTL d + INTERVAL if(x=1, 1, 30) day ;
```
or
```
CREATE TABLE example_table
(
d DateTime,
a Int,
x Int
)
ENGINE = MergeTree... | 2020-04-03T14:32:58 | 2020-04-03T14:33:38 | {
"+1": 2
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 608,842,610 | 10,025 | alexey-milovidov | @amosbird I see no way how to deal with it. Maybe we can find some another sort of magic. | 2020-04-03T23:49:25 | 2020-04-03T23:49:25 | {} | MEMBER |
ClickHouse | ClickHouse | 610,965,234 | 9,908 | alesapin | Do not test until merged with master. | 2020-04-08T13:38:19 | 2020-04-08T13:38:19 | {} | MEMBER |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.