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
490,120,961
5,155
filimonov
It looks like it's connected with prewhere. With ```set optimize_move_to_prewhere = 0;``` it works properly. Probably s2 column get 'optimized away' after prewhere, and after that s3 can't be calculated. /cc @alexey-milovidov
2019-05-07T15:08:47
2019-05-07T15:08:47
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,483,829
5,222
den-crane
1. Kafka? https://github.com/yandex/ClickHouse/issues/3198 MACOs? https://clickhouse.yandex/docs/en/operations/server_settings/settings/#max-open-files Wrong limits `cat /proc/{CH-Server-PID}/limits` 2. ,3. CPU/IO exhaustion ``` <merge_tree> <parts_to_delay_insert>150</parts_to_delay_insert> <part...
2019-05-08T13:23:34
2019-05-08T13:43:45
{ "+1": 2 }
CONTRIBUTOR
ClickHouse
ClickHouse
490,811,337
5,223
alexey-milovidov
I have provided description for you.
2019-05-09T08:35:26
2019-05-09T08:35:26
{}
MEMBER
ClickHouse
ClickHouse
491,255,331
5,237
filimonov
OK. Misunderstood the question. It should be executed on each node in parallel. If your initiator node is also a shard - certainly the query will be executed on it too, simultaneously with other nodes. Because the request is local it will work faster, but it should not block others. Do you claim that other no...
2019-05-10T11:25:01
2019-05-10T11:25:01
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,086,622
3,984
filimonov
@submergerock your mutations has status is_done = 0, so they are either failed, either still working. Select all columns from system.mutations to see what's exactly is going on.
2019-05-07T13:46:43
2019-05-07T13:46:43
{}
CONTRIBUTOR
ClickHouse
ClickHouse
483,600,846
3,174
filimonov
> @filimonov is there a way to for example list the parts that are waited for to be merged? If there are more than one part per partition and their sizes are smaller than maximum part size - that means they will be chosen for merge someday. The smaller part is the bigger chance that it will be merged soon. Check...
2019-04-16T10:19:40
2019-05-07T14:36:00
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,421,514
5,217
4ertus2
> I guess some perf test might be needed to make sure that ordinary any joins won't suffer from that additional `join.anyTakeLastRow()` check? Compiler should be clever enough to place bool value instead this call. And CPU will predict that it's always the same. Of course, we need regression tests for joins performa...
2019-05-08T09:43:36
2019-05-08T09:43:36
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,538,784
5,216
alexey-milovidov
1. Build has failed. 2. > Short description (up to few sentences): Fix error codes in DataPartsExchange This is not the proper description for changelog.
2019-05-08T15:41:56
2019-05-08T15:41:56
{}
MEMBER
ClickHouse
ClickHouse
490,626,529
5,222
den-crane
> PARTITION BY category CH was designed to use monthly partitioning (partition by toYYYYMM(somedate/time)) to manage data retention (drop old data). So the design suggests small number of partitions and a single digit number of partitions over batch. If you insert 5mil rows per insert and this batch has hundreds o...
2019-05-08T19:52:22
2019-05-08T19:52:22
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,824,688
4,953
filimonov
It looks like you have some problem with compiler installed on your system. Please ensure that `brew install` commands from https://clickhouse.yandex/docs/en/development/build_osx/ were successful. I was checking the build yesterday (on a clean install of older macOS 10.12.6 ), and it was succeeded after a few ...
2019-05-09T09:13:28
2019-05-09T09:32:11
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,863,038
3,546
filimonov
Relates #235
2019-05-09T11:20:50
2019-05-09T11:20:50
{}
CONTRIBUTOR
ClickHouse
ClickHouse
491,049,468
5,209
alexey-milovidov
Mostly Ok. A few concerns remain, you can address them in a "follow up" pull request.
2019-05-09T20:14:56
2019-05-09T20:14:56
{}
MEMBER
ClickHouse
ClickHouse
489,906,888
5,206
enqueue
> It is not so obvious to say what behaviour is correct. I agree. Particularly when time zones or leap seconds are involved :smile: I am not an expert on this issue, and if the current behavior fits most use cases or is consistent with SQL standard or with the behavior of many popular databases, it's OK. I would...
2019-05-07T04:49:09
2019-05-07T04:49:09
{ "eyes": 2 }
CONTRIBUTOR
ClickHouse
ClickHouse
489,985,247
5,112
blinkov
@BayoNet will there be an update?
2019-05-07T08:26:26
2019-05-07T08:26:26
{}
CONTRIBUTOR
ClickHouse
ClickHouse
489,996,054
2,707
MrBuddyCasino
I hit just this issue, too: `MATERIALIZED VIEW ... TO <target> POPULATE` throws an error. The SQL script is part of a migration, and the underlying table already contains a lot of data. Is there a way to trigger a `POPULATE` in a separate query? I didn't quite understand your "Легко обойти - можно просто INSERT ... SEL...
2019-05-07T08:58:09
2019-05-07T09:27:43
{}
NONE
ClickHouse
ClickHouse
490,431,203
5,213
alexm93
> Can you show Kafka table and MV creation queries? > > I think kafka has that column, and because of that default expression from MergeTree is not used. > > Maybe it will work if you will move DEFAULT expression there. Creation statements * Kafka Engine `CREATE TABLE default.test1 (`sign` FixedString(27),...
2019-05-08T10:15:55
2019-05-08T10:31:33
{}
NONE
ClickHouse
ClickHouse
490,476,330
5,141
den-crane
@0xTimT your problem is https://github.com/yandex/ClickHouse/issues/5066 Just use `clickhouse-client -A ` or downgrade to 19.4.3.11.
2019-05-08T13:00:53
2019-05-08T13:00:53
{ "+1": 1 }
CONTRIBUTOR
ClickHouse
ClickHouse
490,863,257
3,657
filimonov
Relates #235
2019-05-09T11:21:35
2019-05-09T11:21:35
{}
CONTRIBUTOR
ClickHouse
ClickHouse
491,069,566
4,558
code-of-kpp
One probable issue introduced is client program timeouts - when long operation is running sometimes one can get Timeout exception when it doesn't look like the right thing to happen.
2019-05-09T21:16:40
2019-05-09T21:16:40
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,540,956
5,225
filimonov
So probably ClickHouse server is already running? Check the contents /data/clickhouse/data/status file
2019-05-08T15:47:23
2019-05-08T15:47:23
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,960,852
5,230
den-crane
>DataTypeDecimal Please provide DDL for table_a , table_b show create table table_a; show create table table_b;
2019-05-09T15:51:17
2019-05-09T15:51:17
{}
CONTRIBUTOR
ClickHouse
ClickHouse
491,064,233
4,936
alexey-milovidov
Ok.
2019-05-09T20:59:25
2019-05-09T20:59:25
{}
MEMBER
ClickHouse
ClickHouse
491,207,431
79
filimonov
Let's continue the discussion in #469 to have a clearer backlog.
2019-05-10T08:33:29
2019-05-10T08:33:29
{}
CONTRIBUTOR
ClickHouse
ClickHouse
489,920,430
5,152
janplus
@alexey-milovidov Sorry for my late response. I have add an extra function to meet the new Nullable In's requirement. You may check it now.
2019-05-07T06:02:39
2019-05-07T06:02:39
{ "+1": 1 }
CONTRIBUTOR
ClickHouse
ClickHouse
489,920,008
5,182
filimonov
It is interesting subject. Currentry you can do manual analysis by connecting events by initial_query_id in query_log. And you can use application-generated query_id or get back query_id generated by ClickHouse to connect CH events with in-app events. But probably tracing could bring better control on that. @TH-...
2019-05-07T06:00:41
2019-05-07T07:52:02
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,831,222
5,220
andyyzh
@svladykin @filimonov fix it and add some test cases.
2019-05-09T09:32:29
2019-05-09T09:32:29
{ "+1": 1 }
CONTRIBUTOR
ClickHouse
ClickHouse
491,065,011
4,558
alexey-milovidov
?
2019-05-09T21:01:55
2019-05-09T21:01:55
{}
MEMBER
ClickHouse
ClickHouse
491,282,132
5,233
filimonov
It would be really cool if you will be able to create a minimal verifiable example. I.e. create table statements + some data (if needed) & needed config.
2019-05-10T13:06:59
2019-05-10T13:06:59
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,500,200
5,219
den-crane
0. The CH (OLAP) way is one wide table `Contact` (with Company attributes as columns). Join is very CPU/Memory heavy operation try to avoid JOINs as much as possible. 1. this is OK much much worse than №0 but OK. 2. a. You can shard Company and use `distributed_product_mode='allow'` but it worse than №1. b. You c...
2019-05-08T14:07:54
2019-05-08T14:09:44
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,544,349
4,686
vitlibar
Merged in https://github.com/yandex/ClickHouse/pull/5124
2019-05-08T15:56:15
2019-05-08T15:56:15
{}
MEMBER
ClickHouse
ClickHouse
490,861,335
360
filimonov
Relates #235
2019-05-09T11:14:35
2019-05-09T11:14:35
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,861,444
1,132
filimonov
Relates #235
2019-05-09T11:14:59
2019-05-09T11:14:59
{}
CONTRIBUTOR
ClickHouse
ClickHouse
491,062,771
4,935
alexey-milovidov
Do you really need this command? In my opinion, it's much better to use synchronous distributed inserts (the setting `insert_distributed_sync` = 1)
2019-05-09T20:54:57
2019-05-09T20:54:57
{}
MEMBER
ClickHouse
ClickHouse
491,252,625
5,237
Onlinehead
@filimonov Thanks for your answer. In the documentation, but it still a bit not clear. From a documentation: "ClickHouse applies this setting when the query contains the product of distributed tables, i.e. when the query for a distributed table contains a non-GLOBAL subquery for the distributed table.". В русской...
2019-05-10T11:13:46
2019-05-10T11:13:46
{}
NONE
ClickHouse
ClickHouse
491,263,396
5,237
Onlinehead
@filimonov yes, of course. Let me try to give your more info. If I missed something - please what exactly you need. - Clickhouse version - `19.5.2.6` - Cluster configuration (same on each node): ```xml <remote_servers> <logs> <shard> <internal_replication>false</internal_replication> ...
2019-05-10T11:58:02
2019-05-10T12:28:46
{}
NONE
ClickHouse
ClickHouse
490,081,362
1,252
filimonov
@victor-perov it's quite hard to follow/react comments apperared in closed issue :) Could you please open new issue about problematic TRUNCATE behavior? Does the problem reproduces? How big the inserts are? Can you share some sample of problematic sequence? I.e. create-tuncate-insert-data partially loss?
2019-05-07T13:33:29
2019-05-07T13:33:29
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,320,759
5,217
amosbird
Yeah, it seems reasonable to use a normal variable instead of a template one. I was following the design pattern that "We don't pay what we don't use" as `join_overwrite` is a rarely used flag and somehow only makes sense when using it as a dictionary. But it does introduce code bloat. I guess some perf test might be n...
2019-05-08T02:15:46
2019-05-08T02:15:46
{}
COLLABORATOR
ClickHouse
ClickHouse
490,507,195
3,268
c3mb0
Is this ever going to be fixed?
2019-05-08T14:25:48
2019-05-08T14:25:48
{}
NONE
ClickHouse
ClickHouse
490,882,989
5,229
NearTan
@den-crane Thanks for your information.
2019-05-09T12:30:00
2019-05-09T12:30:00
{}
NONE
ClickHouse
ClickHouse
490,972,566
5,152
alexey-milovidov
Ok. Now it corresponds to http://sqlfiddle.com/#!9/1b8410/1 http://sqlfiddle.com/#!17/2d173/1 http://sqlfiddle.com/#!5/2d173/1
2019-05-09T16:23:09
2019-05-09T16:25:39
{ "hooray": 1 }
MEMBER
ClickHouse
ClickHouse
491,286,200
5,233
blind-oracle
@den-crane I've seen this happen only once for now. I've built it myself using Altinity RPM builder scripts to include fixes for stuff like this https://github.com/yandex/ClickHouse/pull/5127 It was built on a generic CentOS 7 using devtoolkit-7 (gcc 7) @filimonov It's hard to verify because the table has around...
2019-05-10T13:20:34
2019-05-10T13:22:30
{}
NONE
ClickHouse
ClickHouse
490,863,429
4,202
filimonov
Relates #235
2019-05-09T11:22:13
2019-05-09T11:22:13
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,866,716
5,229
den-crane
duplicate of https://github.com/yandex/ClickHouse/issues/5066 Just use `clickhouse-client -A ` or downgrade to 19.4.3.11.
2019-05-09T11:34:08
2019-05-09T11:34:08
{}
CONTRIBUTOR
ClickHouse
ClickHouse
491,054,551
5,116
alexey-milovidov
> X-ClickHouse-Summary: {"read_rows":"0","read_bytes":"0","write_rows":"848484","write_bytes":"3393936","total_rows":"0"} In this example, the `total_rows` is absolutely misleading! We cannot proceed with that. What alternative do you suggest?
2019-05-09T20:30:01
2019-05-09T20:30:01
{}
MEMBER
ClickHouse
ClickHouse
491,269,517
5,237
filimonov
Please repeat the test in clickhouse-client with the `set send_logs_level = 'trace'` setting (it will collect logs from all 3 servers in the proper order) and attach results.
2019-05-10T12:21:21
2019-05-10T12:21:37
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,317,943
5,212
andyyzh
> @andyyzh what's the usecase for this? is it somehow related to Lunar calendar? A week number computing rule: the week contained January 1st is the first week, contained a day to make configurable. It has nothing to do with the lunar calendar. > USA, Canada, most of Latin America, Japan, Israel, South Korea, am...
2019-05-08T01:59:36
2019-05-08T01:59:36
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,488,571
5,224
den-crane
Duplicate of https://github.com/yandex/ClickHouse/issues/5031
2019-05-08T13:37:18
2019-05-08T13:37:18
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,846,743
5,152
janplus
Ping @alexey-milovidov
2019-05-09T10:21:11
2019-05-09T10:21:11
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,889,009
5,212
alexey-milovidov
> I plan to delete the 'CustomDate' parameter to make the purpose of the function clearer. Do you want to change the name of the function? called NorthWeek or USWeek? Ok. Yes, better to change the name. USWeek is not appropriate, because weeks start at Sunday in US. NorthWeek is too confusing. Better to implem...
2019-05-09T12:47:45
2019-05-09T12:47:45
{}
MEMBER
ClickHouse
ClickHouse
490,968,666
5,124
alexey-milovidov
It's not production ready (discussion): ``` Alexey Milovidov, [09.05.19 11:43] https://github.com/yandex/ClickHouse/pull/5124/files#diff-f0b05095130174676e7ba48481190657R348 @vitlibar @Danlark Какой способ определения возможностей CPU лучше? Alexey Milovidov, [09.05.19 11:45] https://github.com/yandex/Click...
2019-05-09T16:12:10
2019-05-09T16:12:10
{}
MEMBER
ClickHouse
ClickHouse
491,008,321
5,230
den-crane
>insert into table_a select col_a, col_b, col_c , sum(sale), sum(qty), .. > from table_b group col_a, col_b, col_c I don't see col_a col_c ?
2019-05-09T18:09:05
2019-05-09T18:09:14
{}
CONTRIBUTOR
ClickHouse
ClickHouse
491,328,261
5,237
filimonov
@alexey-milovidov sounds like a bug? Or it just a 'lack of feature'? :)
2019-05-10T15:23:53
2019-05-10T15:23:53
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,282,295
5,212
blinkov
@andyyzh what's the usecase for this? is it somehow related to Lunar calendar?
2019-05-07T22:48:34
2019-05-07T22:48:34
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,554,663
4,954
filimonov
Reopening that, just to have it in backlog, and don't forget about that problem
2019-05-08T16:23:08
2019-05-08T16:23:41
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,862,536
2,674
filimonov
Relates #235
2019-05-09T11:19:08
2019-05-09T11:19:08
{}
CONTRIBUTOR
ClickHouse
ClickHouse
491,270,227
5,233
blind-oracle
If some more info on table structure & environment needed - please let me know.
2019-05-10T12:23:53
2019-05-10T12:23:53
{}
NONE
ClickHouse
ClickHouse
491,276,322
5,231
filimonov
There was a serious bug https://github.com/yandex/ClickHouse/pull/4623 I'm not sure if [fix](https://github.com/yandex/ClickHouse/pull/4623/commits/73c571412cf5c711822f282d1c27f632aa43ec6f) was merged into 19.3 branch. We recommend to pick latest 19.4 for update.
2019-05-10T12:46:32
2019-05-10T12:46:32
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,082,637
3,388
filimonov
Exact duplicate #3526 and #2712 (there are more comments there).
2019-05-07T13:36:38
2019-05-07T13:54:10
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,098,098
4,165
filimonov
Same #3746
2019-05-07T14:14:57
2019-05-07T14:14:57
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,542,593
5,222
filimonov
1 & 2 can be related (sockets are also files). - Check @den-crane answer. "Too many parts" with high rate of inserts see https://github.com/yandex/ClickHouse/issues/3174#issuecomment-423435071
2019-05-08T15:51:40
2019-05-08T15:52:50
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,961,986
4,629
alexey-milovidov
I've reviewed this PR. Sorry for the delay. The code is mostly Ok, I have some minor complaints: `void setCustomization(DataTypeCustomDescPtr custom_desc_) const;` Better to return customized data type instead of modification of mutable member. `static const std::vector<String> supported_functions{"any", "any...
2019-05-09T15:54:18
2019-05-09T15:56:17
{}
MEMBER
ClickHouse
ClickHouse
491,068,506
4,558
code-of-kpp
I've implemented here all the suggestions from developers telegram channel but it doesn't seem to solve the problem. Failing tests are not related to new code. So I can rebease tomorrow morning and see if it is better at least in terms of tests.
2019-05-09T21:13:12
2019-05-09T21:13:12
{}
CONTRIBUTOR
ClickHouse
ClickHouse
491,186,884
5,232
eruiz
Thanks @javisantana, \N works. You're right, I won't parse it by hand, but I will use AWK instead of clickhouse-local, I work with huge datasets (>50GB) and I think AWK will give me higher speed.
2019-05-10T07:24:01
2019-05-10T07:24:01
{}
NONE
ClickHouse
ClickHouse
491,238,525
5,234
filimonov
There is ICU linking issue on MacOS. ICU is an optional component, please add `-DENABLE_ICU=OFF` to cmake call as workaround: ```cmake .. -DCMAKE_CXX_COMPILER=g++-8 -DCMAKE_C_COMPILER=gcc-8 -DENABLE_ICU=OFF``` See also: https://github.com/yandex/ClickHouse/pull/5223#issuecomment-490821056
2019-05-10T10:16:45
2019-05-10T10:16:45
{}
CONTRIBUTOR
ClickHouse
ClickHouse
491,174,146
5,210
Enmk
* Simplified documentation to meet style of other docs - removed superfluous sections. * Added new .md files to toc_{ru,en,zh,fa}.yml * Added docs in other languages as symbolic links to English docs. * Added a script that helps in adding links.
2019-05-10T06:35:30
2019-05-10T11:33:25
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,041,348
5,152
janplus
@alexey-milovidov It seems have passed all tests. You may review this PR now.
2019-05-07T11:31:25
2019-05-07T11:31:25
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,092,602
2,588
filimonov
На новых версиях можно использовать команду KILL MUTATION чтобы удалить ненужные мутации.
2019-05-07T14:01:21
2019-05-07T14:01:21
{ "+1": 1 }
CONTRIBUTOR
ClickHouse
ClickHouse
490,202,885
5,217
4ertus2
@amosbird, could you check all the changes are correct, please. There's too much specialisations for map. And I need some more logic there, cause ANY RIGHT JOIN is not correct. #5126
2019-05-07T18:28:34
2019-05-07T18:31:16
{}
CONTRIBUTOR
ClickHouse
ClickHouse
489,984,765
4,026
wawanawna
No, we didn't find a reason @filimonov We stopped using MV, found workaround.
2019-05-07T08:24:57
2019-05-07T08:24:57
{}
NONE
ClickHouse
ClickHouse
490,113,853
5,213
filimonov
Can you show Kafka table and MV creation queries? I think kafka has that column, and because of that default expression from MergeTree is not used. Maybe it will work if you will move DEFAULT expression there.
2019-05-07T14:52:13
2019-05-07T14:52:13
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,417,616
5,220
svladykin
@andyyzh Can you look at this one please?
2019-05-08T09:31:24
2019-05-08T09:31:24
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,422,919
5,141
0xTimT
I'm getting a similar seg fault on our server. This one occurred immediately after connecting using the official ClickHouse client, although that could have been a coincidence. I am also happy to provide additional information if needed. ```2019.05.08 10:59:05.665089 [ 71 ] {} <Error> BaseDaemon: ###################...
2019-05-08T09:47:57
2019-05-08T09:47:57
{}
NONE
ClickHouse
ClickHouse
490,844,850
5,212
andyyzh
> But you probably have noticed that the ISO week numbering starts from the first week containing January 4th. And you wanted to generalize this function, so you've added the possibility to start the numbering from any selected day within a year. Is this really needed? toCustomWeek(DateTime, [, CustomDate][, Timezone]...
2019-05-09T10:14:45
2019-05-09T10:15:10
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,861,187
141
filimonov
#235
2019-05-09T11:14:05
2019-05-09T11:14:05
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,861,518
1,488
filimonov
Relates #235
2019-05-09T11:15:18
2019-05-09T11:15:18
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,958,777
5,225
lzy305
I Solve it. THANK you!
2019-05-09T15:46:12
2019-05-09T15:46:12
{ "-1": 2 }
NONE
ClickHouse
ClickHouse
491,163,701
5,220
svladykin
@andyyzh Thanks a lot for your help!
2019-05-10T05:42:38
2019-05-10T05:42:38
{}
CONTRIBUTOR
ClickHouse
ClickHouse
489,981,338
5,210
blinkov
Also there are new errors in links integrity check: > INFO:root:Link to nowhere: #../../data_types/domains/IPv4/ > INFO:root:Link to nowhere: #../../data_types/domains/IPv6/ (probably because files are named in lowercase)
2019-05-07T08:14:12
2019-05-07T08:14:12
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,102,578
3,746
filimonov
Closing in advance to #4165
2019-05-07T14:25:43
2019-05-07T14:25:43
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,818,661
5,223
filimonov
For googlers. Fixes following: 1. Can't find ICU ```-- The following ICU libraries were not found: -- i18n (required) -- uc (required) -- data (required) -- Failed to find all ICU components (missing: ICU_INCLUDE_DIR ICU_LIBRARY _ICU_REQUIRED_LIBS_FOUND) ``` 2. After adding ICU_ROOT cmake policy need to...
2019-05-09T08:56:22
2019-05-09T09:05:04
{ "+1": 1 }
CONTRIBUTOR
ClickHouse
ClickHouse
490,836,362
5,212
alexey-milovidov
I'm trying to figure out the motivation for this implementation. (Sorry, it doesn't look clear for me.) It looks like that the initial motivation is to provide week numbering to start first week from January 1st. But you probably have noticed that the ISO week numbering starts from the first week containing Janua...
2019-05-09T09:48:07
2019-05-09T09:48:07
{}
MEMBER
ClickHouse
ClickHouse
490,840,143
5,212
alexey-milovidov
For week numbering system, at least the following customizations make sense: - start a week from Monday (as in Europe) or Sunday (as in USA) or Saturday; - allow numbered week to cross the year boundary (as in ISO-8601); or reset the numbering at January 1st; or use number zero instead of one if a week belongs to pre...
2019-05-09T09:59:49
2019-05-09T09:59:49
{}
MEMBER
ClickHouse
ClickHouse
490,861,586
1,630
filimonov
Relates #235
2019-05-09T11:15:35
2019-05-09T11:15:35
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,862,789
3,276
filimonov
Relates #235
2019-05-09T11:20:00
2019-05-09T11:20:00
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,863,094
3,655
filimonov
Relates #235
2019-05-09T11:21:01
2019-05-09T11:21:01
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,898,358
5,226
den-crane
Please show the output line with statistics from clickhouse-client (proccessed / mb/s) And `iostat -x 1 20` during the single query.
2019-05-09T13:12:27
2019-05-09T13:12:27
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,935,520
5,208
ghuname
Nobody knows?
2019-05-09T14:47:04
2019-05-09T14:47:04
{}
NONE
ClickHouse
ClickHouse
491,050,573
5,116
alexey-milovidov
write_* -> written_* ?
2019-05-09T20:18:08
2019-05-09T20:18:08
{}
MEMBER
ClickHouse
ClickHouse
491,204,928
5,232
filimonov
Duplicate #79 #469. Lets continue discussion there (if needed)
2019-05-10T08:25:43
2019-05-10T08:28:29
{}
CONTRIBUTOR
ClickHouse
ClickHouse
491,281,457
5,233
den-crane
Is it repeatable? Why did you build this by yourself? Check what thread [ 39 ] wrote in the log before crash. Try this https://clickhouse-builds.s3.yandex.net/4884/ac8ceade2096781a30586fcba24ccb38553318bc/report.html
2019-05-10T13:04:42
2019-05-10T13:04:42
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,613,532
5,219
Crazylearner30
Thanks for the reply let me try and come back
2019-05-08T19:11:49
2019-05-08T19:11:49
{}
NONE
ClickHouse
ClickHouse
490,810,377
5,223
alexey-milovidov
> No description provided. What build fixes?
2019-05-09T08:32:43
2019-05-09T08:32:43
{}
MEMBER
ClickHouse
ClickHouse
490,863,347
3,908
filimonov
Relates #235
2019-05-09T11:21:57
2019-05-09T11:21:57
{}
CONTRIBUTOR
ClickHouse
ClickHouse
491,250,090
5,237
filimonov
There is a setting controlling that behavior: https://clickhouse.yandex/docs/en/operations/settings/settings/#distributed-product-mode
2019-05-10T11:03:30
2019-05-10T11:03:30
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,111,000
5,214
filimonov
Reason: https://github.com/yandex/ClickHouse/issues/4165#issuecomment-458155881 + #1562
2019-05-07T14:45:40
2019-05-07T14:45:40
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,821,056
5,223
filimonov
Currenly Mac build with default settings fails on linking of ICU: ```[2/80] Linking CXX executable dbms/unit_tests_dbms FAILED: dbms/unit_tests_dbms : && /usr/local/bin/g++-8 -fdiagnostics-color=always -msse4.1 -msse4.2 -mpopcnt -fno-omit-frame-pointer -Wall -Wnon-virtual-dtor -Wextra -Werror -O2 -g -DNDEBUG ...
2019-05-09T09:02:57
2019-05-09T09:02:57
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,830,380
4,763
filimonov
there is no static libreadline for some reason? Please ensure that brew commands from official instructions succeeded. https://clickhouse.yandex/docs/en/development/build_osx/ I was checking the build yesterday (on older macOS 10.12.6 ), and it was succeeded after few small fixes (see https://github.com/yandex/Cl...
2019-05-09T09:30:05
2019-05-09T09:30:05
{}
CONTRIBUTOR
ClickHouse
ClickHouse
490,861,626
2,232
filimonov
Relates #235
2019-05-09T11:15:47
2019-05-09T11:15:47
{}
CONTRIBUTOR
ClickHouse
ClickHouse
491,180,907
5,232
javisantana
As per [CH doc](https://clickhouse.yandex/docs/en/interfaces/formats/#data-formatting) NULL is represented by a '\N' ```bash j@tinybird.co:~$ clickhouse client -q "CREATE TABLE dummy(id Int8, aDate Nullable(Date), other Int8) ENGINE = Log;" j@tinybird.co:~$ echo '"1",\N,"2"' | clickhouse client --input_format_allo...
2019-05-10T07:05:09
2019-05-10T07:05:09
{}
CONTRIBUTOR
ClickHouse
ClickHouse
491,324,739
5,237
den-crane
Seems initiator waits for query results from shards and then starts to solve the query locally CREATE TABLE events_test1 on cluster test (A Int64) ENGINE = MergeTree() ORDER BY tuple() CREATE TABLE events_test1_d on cluster test as events_test1 ENGINE = Distributed(test, currentDatabase(), events_test1, rand()); ...
2019-05-10T15:13:30
2019-05-10T15:13:30
{}
CONTRIBUTOR