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 | 408,367,024 | 1,386 | blinkov | Looks like it was fixed in #1933 | 2018-07-27T09:35:43 | 2018-07-27T09:35:59 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 408,680,145 | 1,125 | inkrement | @blinkov I am afraid that I can not reproduce the data or the exact setting :/ ... Thus, I'll close the issue for now and reopen it if I run into it again. | 2018-07-29T14:07:04 | 2018-07-29T14:07:04 | {} | NONE |
ClickHouse | ClickHouse | 408,841,202 | 2,620 | wawanawna | No answer @alexey-milovidov @ztlpn ? | 2018-07-30T12:07:10 | 2018-07-30T12:07:10 | {} | NONE |
ClickHouse | ClickHouse | 408,840,577 | 2,748 | alesapin | Please, provide command output:
`dpkg -l | grep ssl`
It seems like something went wrong with ssl version. | 2018-07-30T12:05:06 | 2018-07-30T14:05:58 | {} | MEMBER |
ClickHouse | ClickHouse | 408,926,050 | 2,754 | alexey-milovidov | Yes, aggregate function states in Arena are placed without alignment.
It is Ok most of the time, but there are some considerations:
1. Unaligned memory access is legal for x86 (and AArch 64) in almost all cases. For modern x86_64 CPU there is almost no performance penalty of unaligned access. And more compact memor... | 2018-07-30T16:30:15 | 2018-07-30T16:31:52 | {} | MEMBER |
ClickHouse | ClickHouse | 409,261,466 | 2,770 | alexey-milovidov | > Which formats we would like to support
A subset of `strftime`, that can be easily implemented by calling methods of `DateLUTImpl` and probably formatting that components in different ways.
Locale dependent formatting shouldn't be supported for now.
> What is correct approach to work with DateLUTImpl (should ... | 2018-07-31T15:25:08 | 2018-07-31T15:28:03 | {} | MEMBER |
ClickHouse | ClickHouse | 408,359,915 | 2,735 | lamberken | :+1: | 2018-07-27T09:07:35 | 2018-07-27T09:07:35 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 408,483,662 | 2,733 | alexey-milovidov | clickhouse-copier:
- coordination by ZooKeeper is by design;
- configuration complexity is unavoidable.
For progress, we look at system.parts on destination servers and at task nodes in ZK. | 2018-07-27T17:18:23 | 2018-07-27T17:18:23 | {} | MEMBER |
ClickHouse | ClickHouse | 408,502,476 | 2,738 | ztlpn | Let us count.
For Hashed dictionaries each attribute is stored as a hash table. Because we have a complex key dictionary, the key for these hash tables will have type `StringRef`. The keys themselves are stored in the arena pool. The hash table value for the `b` attribute will be `Int64`, and for the `s` attribute i... | 2018-07-27T18:30:10 | 2018-07-27T18:30:33 | {
"+1": 2
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 408,912,632 | 2,750 | alexey-milovidov | Thank you! | 2018-07-30T15:51:13 | 2018-07-30T15:51:13 | {} | MEMBER |
ClickHouse | ClickHouse | 409,275,734 | 2,769 | alexey-milovidov | #2571 | 2018-07-31T16:04:54 | 2018-07-31T16:04:54 | {} | MEMBER |
ClickHouse | ClickHouse | 409,503,365 | 2,768 | Tongmyong | @alesapin I've found that the problem may come from replication_queue, which tries again and again when the record in it fails. How do I deal with this problem?
----------------------------------------------------------------------------------------------------------
SELECT *
FROM replication_queue
FORMAT Vertic... | 2018-08-01T08:57:44 | 2018-08-01T08:57:44 | {} | NONE |
ClickHouse | ClickHouse | 408,321,917 | 2,521 | fizerkhan | @alexey-milovidov Awesome. That makes sense.
Let's say If don't use the `-If` combinator, then how the behaviour will be
```
histogram(10)(duration)
```
Here I can understand that 10 ranges will come, but what is the ceiling limit? | 2018-07-27T06:12:45 | 2018-07-27T06:12:55 | {} | NONE |
ClickHouse | ClickHouse | 408,373,332 | 1,347 | blinkov | @569934390 have you tried it? | 2018-07-27T10:02:04 | 2018-07-27T10:02:04 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 408,482,745 | 2,743 | alexey-milovidov | ClickHouse works on ZFS. We use it for backup servers.
> Is there any performance difference over ext4 ???
We have no comparison of performance, and we don't run queries on ZFS cluster (it is used for passive replicas). | 2018-07-27T17:14:56 | 2018-07-27T17:14:56 | {} | MEMBER |
ClickHouse | ClickHouse | 408,517,521 | 2,738 | den-crane | hm, the same dictionary but with hashed layout eats 1.37 GiB.
I would prefer more compact but a slower hashed dictionary with collisions and configurable number of buckets which can save memory, because in real production I have a dictionary using 14 GiB RAM with less than 1 GiB table. And all attributes compacted t... | 2018-07-27T19:31:11 | 2018-07-27T19:34:50 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 409,188,296 | 2,766 | alesapin | Which batch size do you mean? ClickHouse can't do anything, If clients make `INSERT` queries with one row. Otherwise if `INSERT` queries generate too small blocks, squashing of small blocks will be used by default (except case when you have distributed table with `insert_distributed_sync` setting). | 2018-07-31T11:30:50 | 2018-07-31T11:30:50 | {} | MEMBER |
ClickHouse | ClickHouse | 409,661,164 | 2,675 | Amper | @arizz96
Hi, I'm facing the same problem.
I'm going to add to Kafka storage engine soon a parameter that allows to specify the topic for messages that could not be processed. | 2018-08-01T17:45:22 | 2018-08-01T17:45:22 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 408,358,195 | 1,483 | blinkov | Basically it's done via buffering the data on the side of application that does inserts and then pushing many rows into ClickHouse together using `INSERT INTO ... FORMAT` functionality https://clickhouse.yandex/docs/en/query_language/insert_into/
How exactly it looks like heavily depends on the source of data and wh... | 2018-07-27T09:01:03 | 2018-07-27T09:01:03 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 408,360,245 | 1,461 | blinkov | The source code have changed much since this discussion so it does not make much sense to return to it. Please, open a new issue if you have any similar troubles on recent code. | 2018-07-27T09:08:54 | 2018-07-27T09:08:54 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 408,374,930 | 1,127 | blinkov | ClickHouse documentation resides near code right in this repository and anyone is welcome to contribute to it: https://github.com/yandex/ClickHouse/tree/master/docs
It's then published on official website https://clickhouse.yandex/docs/en/ and there is "pencil" icon that links back to edit on GitHub page.
There a... | 2018-07-27T10:09:14 | 2018-07-27T10:09:14 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 408,390,085 | 1,003 | blinkov | A lot of time have passed since this discussion. @sunsingerus please open a new issue if you have any trouble building recent ClickHouse code. | 2018-07-27T11:20:17 | 2018-07-27T11:20:17 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 408,391,941 | 2,737 | alexey-milovidov | Why we need MurmurHash2?
MurmurHash2 is an outdated hash function, superseded by MurmurHash3 and subsequently by CityHash, xxHash, HighwayHash. Usually there is no reason to use MurmurHash.
It is needed for the cases when you already have MurmurHash in some applications and you want to reproduce it in ClickHouse ... | 2018-07-27T11:29:47 | 2018-07-27T11:29:47 | {} | MEMBER |
ClickHouse | ClickHouse | 408,435,794 | 2,716 | alexey-milovidov | :+1: | 2018-07-27T14:29:57 | 2018-07-27T14:29:57 | {} | MEMBER |
ClickHouse | ClickHouse | 408,462,263 | 2,733 | blinkov | Strategically it'd be better to fix the weak spots of current `clickhouse-copier` implementation instead of making yet another alternative to choose from. | 2018-07-27T15:57:03 | 2018-07-27T15:57:03 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 408,568,798 | 859 | Oberonhive | Found a similar problem. Query below produces a error `Unknown identifier: marks_bytes, projection layer`.
```SQL
SELECT
table IN (
SELECT table
FROM system.columns
GROUP BY table
ORDER BY AVG(CASE WHEN marks_bytes>0 THEN 1 ELSE 0 END)
) AS t
FROM system.columns
GROUP BY t
```
Subquer... | 2018-07-28T00:26:39 | 2018-07-28T00:26:39 | {} | NONE |
ClickHouse | ClickHouse | 408,634,198 | 1,980 | vpanfilov | ```sql
SELECT groupArray(arrayJoin([1, NULL, 2]))
┌─groupArray(arrayJoin([1, NULL, 2]))─┐
│ [1,2] │
└─────────────────────────────────────┘
```
I found this behavior when I got unexpected `DB::Exception: Sizes of ARRAY-JOIN-ed arrays do not match` in one of my queries. Why it is ... | 2018-07-28T20:52:41 | 2018-07-28T20:52:41 | {} | NONE |
ClickHouse | ClickHouse | 408,999,104 | 1,616 | derekperkins | @alesapin What version of ClickHouse was that added to? | 2018-07-30T20:28:38 | 2018-07-30T20:28:38 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 409,141,913 | 1,531 | blinkov | @badvir do you have any further questions? | 2018-07-31T08:36:52 | 2018-07-31T08:36:52 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 409,142,750 | 1,534 | blinkov | @djeshkov was everything else working fine besides this messages in log? | 2018-07-31T08:39:57 | 2018-07-31T08:39:57 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 409,651,114 | 2,778 | alesapin | The problem with this versions is well known, but there is no solution :( You have to update all nodes of your cluster simultaneously. In newer version this problem is solved and shouldn't reproduce. | 2018-08-01T17:14:38 | 2018-08-01T17:14:38 | {
"confused": 2
} | MEMBER |
ClickHouse | ClickHouse | 408,443,609 | 2,689 | alexey-milovidov | Thank you! KILL on cluster is indeed useful.
By the way, if you have a distributed query (that generate many queries on cluster nodes), and if you KILL just the initiating query, all generated queries will also be killed.
About OPTIMIZE. In many cases excessive OPTIMIZE (especially with FINAL) may be harmful, bec... | 2018-07-27T14:55:06 | 2018-07-27T14:55:06 | {} | MEMBER |
ClickHouse | ClickHouse | 409,168,642 | 2,398 | LiuYangkuan | Yes, #2701 is an extendable solution! | 2018-07-31T10:09:44 | 2018-07-31T10:09:44 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 408,299,988 | 2,734 | zzl-wy | @thanhthai3010 I have query some history data,so i am very sure that the data have fully inserted.The setting as following:
I have just try the newest version,and the quesion still exsist.
| 2018-07-27T03:21:54 | 2018-08-01T03:03:39 | {} | NONE |
ClickHouse | ClickHouse | 408,401,907 | 2,728 | xmariachi | > will hardly be applicable "as is" to most companies":
Definitely! Not "as is", but as a starting point to get the concepts and understand the _dos_ and _don'ts_.
Integrations may be well be totally separated - it's just additions I thought of that are pretty common on incoming data scenario, esp. Kafka). | 2018-07-27T12:15:31 | 2018-07-27T12:15:55 | {} | NONE |
ClickHouse | ClickHouse | 408,614,151 | 2,746 | alexey-milovidov | TLDR:
Versions 1 and 18 are totally compatible. | 2018-07-28T15:09:49 | 2018-07-28T15:09:49 | {} | MEMBER |
ClickHouse | ClickHouse | 408,775,839 | 1,661 | blinkov | @Slind14 And why it's just 48 cores? Which hardware setup do you have? | 2018-07-30T07:43:40 | 2018-07-30T07:43:40 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 409,161,463 | 2,747 | vpanfilov | There is also an intresting key-value approch for VERY high number of diffent events and parameters, which change very often.
It is described in [this presentation](https://github.com/yandex/clickhouse-presentations/blob/master/meetup15/lightning/badoo.pdf) (only in Russian). You can see table schemas at the last sl... | 2018-07-31T09:43:54 | 2018-07-31T09:43:54 | {} | NONE |
ClickHouse | ClickHouse | 382,685,255 | 1,469 | zhangguiyu | My experience is xxxState (sumState, countState) will only work across 1-dimension, if you group using more than 1 columns, then the runningAccumulate will get messed up (not in sequence). | 2018-04-19T10:20:29 | 2018-07-31T11:14:14 | {} | NONE |
ClickHouse | ClickHouse | 409,598,212 | 2,690 | tuedelue | There is an issue with your rights management. The docker container uses its internal clickhouse user and group (id 105 and id 106).
You need to make sure that the id-corresponding user/group has rw access to the folder on the host (they are most probably *not* the same).
A workaround is to create a your own docker... | 2018-08-01T14:38:39 | 2018-08-01T14:38:39 | {} | NONE |
ClickHouse | ClickHouse | 408,628,020 | 563 | hatarist | Nah, tested it with the server v1.1.54390;
```
curl 'http://127.0.0.1:8123/?query=SELECT+1&database=non_existent_db' -H 'Accept-Encoding: gzip' --output - | gunzip
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Lef... | 2018-07-28T18:53:08 | 2018-07-28T18:53:08 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 408,649,753 | 2,733 | zhang2014 | > Table function remote is just a temporary Distributed table (a StorageDistributed object). How are you going to send data into it? The only possibility is to do the same as INSERT INTO TABLE FUNCTION remote(...).
Yes, but it's transactional.
## About transaction
> Support `ALTER TABLE table_name BEGIN TRANS... | 2018-07-29T03:43:32 | 2018-07-29T03:43:32 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 408,357,235 | 2,283 | vpanfilov | It is the same behaviour if ALIAS columns are present in MATERIALIZED or other ALIAS columns. Very annoying bug - tables are created succesfully, but all inserts fails. | 2018-07-27T08:57:16 | 2018-07-27T08:57:16 | {} | NONE |
ClickHouse | ClickHouse | 408,377,951 | 2,710 | Joozty | @ztlpn Thanks it helped. | 2018-07-27T10:22:24 | 2018-07-27T10:22:24 | {} | NONE |
ClickHouse | ClickHouse | 409,002,074 | 1,616 | alexey-milovidov | It's not released yet. I expect it to be in 18.5. | 2018-07-30T20:38:40 | 2018-07-30T20:38:40 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 409,021,267 | 2,290 | derekperkins | This should be solved when #1616 makes it into a public release. | 2018-07-30T21:42:11 | 2018-07-30T21:42:11 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 409,140,685 | 1,023 | blinkov | @hagen1778 is this issue still relevant? | 2018-07-31T08:32:36 | 2018-07-31T08:32:36 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 409,151,705 | 1,518 | blinkov | @adhamelia have you managed to use the approach described by @alexey-milovidov ? | 2018-07-31T09:10:33 | 2018-07-31T09:10:33 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 409,449,816 | 973 | alniks | No, but somehow problem resolved. | 2018-08-01T04:59:54 | 2018-08-01T04:59:54 | {} | NONE |
ClickHouse | ClickHouse | 409,515,506 | 2,766 | alesapin | Although ClickHouse has the setting `min_insert_rows` it doesn't work like you describe.
It seems very strange to prevent client from small inserts inside DB. I never heard about such functionality in DBs. Maybe it's better to reject client's small insert requests inside your application server? Or prevent client to... | 2018-08-01T09:37:24 | 2018-08-01T09:37:24 | {} | MEMBER |
ClickHouse | ClickHouse | 408,362,569 | 1,342 | blinkov | Parquet is in development at #2427 | 2018-07-27T09:18:12 | 2018-07-27T09:18:12 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 408,389,300 | 2,735 | alexey-milovidov | Could you please also add a test?
It's rather simple. You can use `remote('127.0.0.{1,2}', ...` to perform distributed query (and force serialization and deserialization of aggregate states). It would be perfect if the test will reproduce the bug in older version.
The test should have word `shard` in its name. (I... | 2018-07-27T11:16:17 | 2018-07-27T11:16:17 | {} | MEMBER |
ClickHouse | ClickHouse | 408,420,450 | 1,661 | blinkov | Which solution you ended up with at the moment? | 2018-07-27T13:36:23 | 2018-07-27T13:36:23 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 408,865,215 | 2,755 | wawanawna | UPDATE: after few tries was able to reproduce problems with merges even without MV.
After stop 1 replica (node21 with leaders) on node11 (replica) we started seen
```
2018.07.30 13:23:39.715741 [ 35 ] <Error> executeQuery: Code: 252, e.displayText() = DB::Exception: Too many parts (300). Merges are processing signi... | 2018-07-30T13:35:37 | 2018-07-30T13:35:37 | {} | NONE |
ClickHouse | ClickHouse | 409,133,906 | 2,502 | marcelbirkner | Hi @jonatasfreitasv,
I am getting the same error like you did. What did you do to get clickhouse-copier to work?
What do you mean with "copying jobs xml to Zookeeper".
I would appreciate any help.
Thank you,
Marcel | 2018-07-31T08:07:52 | 2018-07-31T08:07:52 | {} | NONE |
ClickHouse | ClickHouse | 409,418,665 | 2,766 | lamberken | @alesapin, thank for your explain.
What's I think is parameter likes `min_insert_rows`, for example, if the server use `ReplicatedMergeTree` engine and set `min_insert_rows = 1000`.
when client inserts just one row each time, it'll get error message. if so, this mechanism can prevent create many znodes.
our rea... | 2018-08-01T01:26:50 | 2018-08-01T09:02:04 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 408,434,341 | 2,731 | alexey-milovidov | You can also increase `max_ast_elements` without any fear. | 2018-07-27T14:25:00 | 2018-07-27T14:25:00 | {} | MEMBER |
ClickHouse | ClickHouse | 409,090,425 | 2,763 | Tigro | Did you add this key http://repo.yandex.ru/clickhouse/CLICKHOUSE-KEY.GPG to apt keyring? (apt-key add CLICKHOUSE-KEY.GPG). | 2018-07-31T04:25:48 | 2018-07-31T04:25:48 | {
"+1": 5
} | NONE |
ClickHouse | ClickHouse | 409,451,111 | 972 | alniks | <internal_replication>true</internal_replication> helped us | 2018-08-01T05:09:13 | 2018-08-01T05:09:13 | {} | NONE |
ClickHouse | ClickHouse | 409,515,809 | 2,747 | alesapin | If there would be any additional questions, please reopen issue. | 2018-08-01T09:38:36 | 2018-08-01T09:38:36 | {} | MEMBER |
ClickHouse | ClickHouse | 409,651,654 | 2,778 | alexey-milovidov | Newer versions behave correctly even if you have different order or types of columns on remote nodes. | 2018-08-01T17:16:14 | 2018-08-01T17:16:14 | {} | MEMBER |
ClickHouse | ClickHouse | 408,377,040 | 1,343 | blinkov | @arctica do you have any further questions? | 2018-07-27T10:18:28 | 2018-07-27T10:18:28 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 409,106,165 | 2,754 | chenxing-xc | Thanks for your response. I will try to align aggregation state in general, and see what happens | 2018-07-31T06:09:38 | 2018-07-31T06:09:38 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 409,274,926 | 2,728 | xmariachi | Indeed, and I have met some of these myself. We are using it and so far it's OK, apart from JSON parsing limitations.
However, the way I see it, I think the main benefit is in shedding light on CH for the incoming data scenario, especially using the Collapsing or other *MergeTree engines "in action" for event handli... | 2018-07-31T16:02:32 | 2018-07-31T16:02:32 | {} | NONE |
ClickHouse | ClickHouse | 409,599,720 | 2,778 | kvap | Nice debugging there @dqminh. Look like it mixes up the order of columns in one of the blocks. | 2018-08-01T14:42:57 | 2018-08-01T14:42:57 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 409,604,249 | 2,776 | alesapin | This behaviour is expected. As [documentation](https://clickhouse.yandex/docs/en/operations/table_engines/summingmergetree/) says:
> If the values were null in all of these columns, the row is deleted. (The exception is cases when the data part would not have any rows left in it.) | 2018-08-01T14:55:56 | 2018-08-01T14:55:56 | {} | MEMBER |
ClickHouse | ClickHouse | 409,659,775 | 2,691 | Amper | @xmariachi Can you give an example of what using a key might look like? | 2018-08-01T17:40:55 | 2018-08-01T17:40:55 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 408,337,393 | 2,521 | alexey-milovidov | Histogram will span between actual minimum and maximum of data value. | 2018-07-27T07:33:35 | 2018-07-27T07:33:35 | {} | MEMBER |
ClickHouse | ClickHouse | 408,344,978 | 2,728 | blinkov | Yep, it would be nice, though such tutorial will go well beyond just ClickHouse and will hardly be applicable "as is" to most companies. Tutorial on setting up small ClickHouse cluster lives here https://clickhouse.yandex/tutorial.html , but how to integrate it properly into existing company ecosystem is extremely indi... | 2018-07-27T08:07:09 | 2018-07-27T08:07:09 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 408,487,074 | 2,733 | blinkov | Why unavoidable? For example, someone could make a GUI that sets it up via step-by-step wizard or something, and then displays running tasks with progress bars.
I think ZooKeeper by itself will hardly be a problem if you don't have to interact with it directly. | 2018-07-27T17:31:36 | 2018-07-27T17:31:36 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 408,640,004 | 2,447 | alexey-milovidov | Now we have system tables for all required entities except language keywords. | 2018-07-28T22:57:02 | 2018-07-28T22:57:02 | {} | MEMBER |
ClickHouse | ClickHouse | 408,650,204 | 2,733 | zhang2014 | > Why unavoidable? For example, someone could make a GUI that sets it up via step-by-step wizard or something, and then displays running tasks with progress bars.
I think ZooKeeper by itself will hardly be a problem if you don't have to interact with it directly.
This is great, but we have some case is no zookeeper... | 2018-07-29T03:58:17 | 2018-07-29T03:58:17 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 408,806,286 | 1,661 | Slind14 | @blinkov the CPU power isn't even the bottleneck. A query runs for multiple minutes and the CPU usage stays below 30%. I assume this is due to the join since queries without a join get the CPU to 100%.
The SSDs are sleeping during this process as well and the memory is used 90% for cache. (512GB) | 2018-07-30T09:38:38 | 2018-07-30T09:38:38 | {} | NONE |
ClickHouse | ClickHouse | 408,351,683 | 2,521 | fizerkhan | @alexey-milovidov Great. Thanks | 2018-07-27T08:35:20 | 2018-07-27T08:35:20 | {} | NONE |
ClickHouse | ClickHouse | 408,410,032 | 2,728 | blinkov | It's not that straightforward even if we talk just about Kafka. There's Kafka table engine https://clickhouse.yandex/docs/en/operations/table_engines/kafka/ which allows to SELECT from topic and INSERT INTO persistent table in ClickHouse. It sounds like a good way to go, but in practice companies tend to have some appl... | 2018-07-27T12:54:04 | 2018-07-27T12:54:04 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 408,447,897 | 2,673 | dmitrydrynov | hm, after the next update, everything worked | 2018-07-27T15:09:39 | 2018-07-27T15:09:39 | {} | NONE |
ClickHouse | ClickHouse | 408,473,797 | 2,741 | alexey-milovidov | At least we need a test that something works. | 2018-07-27T16:40:05 | 2018-07-27T16:40:05 | {} | MEMBER |
ClickHouse | ClickHouse | 408,832,607 | 1,616 | alesapin | done. | 2018-07-30T11:30:22 | 2018-07-30T11:30:22 | {} | MEMBER |
ClickHouse | ClickHouse | 408,580,769 | 2,746 | vnnw | Sorry. The CHANGELOG.md already explains my question. | 2018-07-28T04:03:10 | 2018-07-28T04:03:10 | {} | NONE |
ClickHouse | ClickHouse | 408,679,890 | 1,661 | Slind14 | I have two datasets, one consistent of two tables with the latest data and one single table with all the data.
Once a day I write the data from the day before yesterday from the "live" table**s** to the "history" table.
The issue I'm still facing is with datasets which span around hundred of millions of records per... | 2018-07-29T14:03:11 | 2018-07-29T14:03:11 | {} | NONE |
ClickHouse | ClickHouse | 408,773,584 | 2,749 | blinkov | It definitely might be contributed by some company heavily relying on Windows in production, but at the moment we are not aware of any such effort.
As of just Visual Studio support, it usually appears with the first active developer who is used to it.
| 2018-07-30T07:34:13 | 2018-07-30T07:34:13 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 408,911,090 | 2,747 | alesapin | Look at [similar question in Google Groups](https://groups.google.com/forum/#!msg/clickhouse/9BrNXnCJGdE/Mnyqu-EsBgAJ;context-place=searchin/clickhouse/schema%7Csort:date). | 2018-07-30T15:46:45 | 2018-07-30T15:46:45 | {} | MEMBER |
ClickHouse | ClickHouse | 409,346,313 | 2,768 | alesapin | Sorry, but I cannot reproduce the problem. In my attempt I got single error message and same with yours stack trace. Can you provide full query, please? | 2018-07-31T19:51:41 | 2018-07-31T19:51:41 | {} | MEMBER |
ClickHouse | ClickHouse | 409,511,296 | 2,382 | millin | #2169 | 2018-08-01T09:22:49 | 2018-08-01T09:22:49 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 408,346,925 | 2,689 | blinkov | @sundy-li doesn't your company have some infrastructure to execute arbitrary command in parallel on a tagged subset of servers via ssh? If it does, than such stuff is easily done just by querying localhost with clickhouse-client or curl. | 2018-07-27T08:15:40 | 2018-07-27T08:15:40 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 408,353,820 | 712 | blinkov | @gelin this process is documented here: https://clickhouse.yandex/docs/en/operations/table_engines/replication/#recovery-after-failures | 2018-07-27T08:44:06 | 2018-07-27T08:44:06 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 408,375,296 | 1,125 | blinkov | @inkrement yep, step-by-step instruction on how to reproduce it would be nice. | 2018-07-27T10:11:02 | 2018-07-27T10:11:02 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 408,394,769 | 1,178 | blinkov | @lttPo do you have any further questions? | 2018-07-27T11:43:53 | 2018-07-27T11:43:53 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 408,411,289 | 1,661 | Slind14 | I'm still lacking a proper solution for this use case. I did a lot of work around's which add a lot of points of failure. If there is another better way which we haven't thought of, it would be great. | 2018-07-27T12:59:43 | 2018-07-27T12:59:43 | {} | NONE |
ClickHouse | ClickHouse | 408,438,029 | 2,733 | alexey-milovidov | > SEND PART part_name TO TABLE_FUNCTION
Table function `remote` is just a temporary `Distributed` table (a `StorageDistributed` object). How are you going to send data into it? The only possibility is to do the same as `INSERT INTO TABLE FUNCTION remote(...)`. | 2018-07-27T14:37:21 | 2018-07-27T14:38:32 | {} | MEMBER |
ClickHouse | ClickHouse | 408,440,467 | 2,712 | kevinli1 | Thanks so much. | 2018-07-27T14:45:15 | 2018-07-27T14:45:15 | {} | NONE |
ClickHouse | ClickHouse | 409,156,927 | 75 | kolsys | +1 to fix | 2018-07-31T09:28:11 | 2018-07-31T09:28:11 | {
"+1": 2
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 409,242,385 | 2,769 | ztlpn | This bug should be fixed in v18.4.0 and higher. Please try upgrading. | 2018-07-31T14:32:04 | 2018-07-31T14:32:04 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 409,261,022 | 2,748 | set-daemon | @alesapin
ii libengine-pkcs11-openssl 0.1.8-3 amd64 OpenSSL engine for PKCS#11 modules
ii libevent-openssl-2.0-5:amd64 2.0.21-stable-1ubuntu1.14.04.2 amd64 Asynchronous event notific... | 2018-07-31T15:23:58 | 2018-07-31T15:23:58 | {} | NONE |
ClickHouse | ClickHouse | 409,450,699 | 974 | alniks | The problem was due to lags in ClickHouse server under high load and different timeout settings in ClickHouse server and JDBC driver. When we have many select queries, performance of ClickHouse server degrades significantly and simple queries will run for more than 30 secs causing timeouts in JDBC driver | 2018-08-01T05:06:36 | 2018-08-01T05:06:36 | {} | NONE |
ClickHouse | ClickHouse | 408,380,450 | 1,336 | blinkov | @slw остались ли тут ещё вопросы? | 2018-07-27T10:33:49 | 2018-07-27T10:33:49 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 408,419,695 | 2,731 | den-crane | <max_query_size>41943040</max_query_size>
Code: 168. DB::Exception: Received from localhost:9000, ::1. DB::Exception: Received from localhost:9000, ::1. DB::Exception: AST is too big. Maximum: 50000. | 2018-07-27T13:33:40 | 2018-07-27T13:33:40 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 408,635,349 | 2,679 | vpanfilov | Your guery is not optimal. Try to group your data by `field1`, join tables and do some magic with `ARRAY JOIN` and higher order functions. `GROUP BY` clause will help you go multicore:
Here is a sample query for futher experiments:
```sql
SELECT
*
FROM
((SELECT
field1,
groupArray(timestamp) AS timestam... | 2018-07-28T21:15:33 | 2018-07-28T21:15:33 | {} | NONE |
ClickHouse | ClickHouse | 409,527,347 | 867 | BayoNet | The documentation is on the site in the russian version. It will be translated soon. | 2018-08-01T10:21:21 | 2018-08-01T10:21:21 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 409,533,748 | 2,748 | alesapin | Sorry, but I cannot reproduce this problem with the same ubuntu, compiler and openssl version. Can you provide `cmake` output, please? | 2018-08-01T10:47:54 | 2018-08-01T10:48:08 | {} | MEMBER |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.