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 | 770,416,857 | 19,384 | RO-29 | @kanadaj can you please give me an example of your tables that worked for you and what should I fix here in my task.xml (https://github.com/ClickHouse/ClickHouse/issues/16867#issuecomment-770384176) | 2021-01-31T17:22:07 | 2021-01-31T17:22:07 | {} | NONE |
ClickHouse | ClickHouse | 770,553,156 | 19,897 | alexey-milovidov | ```
The project creates multiple executables. Please select one from the list below
to detect races.
In the future, you can specify the executable using the option "-e" if you know
which ones you want to analyze.
coderrect -e your_executable_name1,your_executable_name2 your_build_command_line
1) uti... | 2021-02-01T04:20:50 | 2021-02-01T04:20:50 | {} | MEMBER |
ClickHouse | ClickHouse | 770,653,983 | 19,857 | akuzm | This is expected for now -- the window functions are processed at the latest stage of the query execution, so we can't calculate any expressions with them, and have to use a subselect.
Proper support for this will be complicated: we'll have to account for window functions in `ActionsDAG`, then split it into alternatin... | 2021-02-01T08:00:26 | 2021-02-01T08:00:26 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 770,694,974 | 12,481 | alexey-milovidov | @hanguyen6 We need more details on this issue, at lease the stack trace. | 2021-02-01T09:02:56 | 2021-02-01T09:02:56 | {} | MEMBER |
ClickHouse | ClickHouse | 770,700,378 | 11,423 | alexey-milovidov | Sorry, I don't understand the question. Could you please provide more details? | 2021-02-01T09:10:12 | 2021-02-01T09:10:12 | {} | MEMBER |
ClickHouse | ClickHouse | 770,743,430 | 19,790 | tavplubix | AST fuzzer (UBSan) - #19913
AST fuzzer (debug) - #19914
Integration tests - flacky | 2021-02-01T10:16:55 | 2021-02-01T10:16:55 | {} | MEMBER |
ClickHouse | ClickHouse | 770,384,176 | 16,867 | RO-29 | Here's my full task.xml
```
<yandex>
<!-- Configuration of clusters as in an ordinary server config -->
<remote_servers>
<source_cluster>
<!--
source cluster & destination clusters accept exactly the same
parameters as parameters for the usual Distri... | 2021-01-31T13:40:18 | 2021-01-31T13:40:18 | {} | NONE |
ClickHouse | ClickHouse | 770,483,854 | 19,872 | alexey-milovidov | > AST fuzzer (ASan) — Received signal -3
OOM. Not an error: when ASan is enabled there is no memory tracking for new/delete and OOM is possible despite the memory limits.
`SELECT groupArrayForEach(arr) FROM (SELECT intDiv(number, 1048575) AS k, intDiv(number, 1048575) AND 100 AND NULL, range(number) AS arr FROM ... | 2021-02-01T00:33:09 | 2021-02-01T00:33:09 | {} | MEMBER |
ClickHouse | ClickHouse | 770,487,755 | 19,677 | alexey-milovidov | Arcadia does not support dictionaries, need to add this test to the arcadia_skip_list. | 2021-02-01T00:46:39 | 2021-02-01T00:46:39 | {} | MEMBER |
ClickHouse | ClickHouse | 770,532,982 | 19,883 | fzyzcjy | @Alex-Burmak Thanks for the info! So IMHO I should (1) look at those incompatible changes (2) wait a bit for 21.1.3? | 2021-02-01T03:07:17 | 2021-02-01T03:07:17 | {} | NONE |
ClickHouse | ClickHouse | 770,534,761 | 19,747 | Sallery-X | > Part name includes the block number that is synchronized with ZooKeeper and guaranteed to be distinct for distinct parts.
Concurrently write to different replicas of the same shard，The tmp directory is likely to be duplicated？ | 2021-02-01T03:13:41 | 2021-02-01T03:13:41 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 770,698,262 | 11,763 | alexey-milovidov | Fixed long time ago. | 2021-02-01T09:07:24 | 2021-02-01T09:07:24 | {} | MEMBER |
ClickHouse | ClickHouse | 770,444,843 | 19,518 | bharatnc | > IMHO It would be a bit safer to use new name for that function to avoid compatibility issues. But can go like that too.
>
> But now we need a "mirror" function - which will take IPv6 type and convert to IPv6 or IPv4 string format depending on what is stored
Probably. There were generally two proposals (found in... | 2021-01-31T20:25:21 | 2021-01-31T20:25:21 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 770,522,429 | 19,384 | kanadaj | @RO-29 I didn't fix the task. You use `ALTER TABLE` to remove TTL (`ALTER TABLE table_name REMOVE TTL
` and `ALTER TABLE table_with_ttl MODIFY COLUMN column_ttl REMOVE TTL` ) and indices (`ALTER TABLE [db].name DROP INDEX name`), copy the tables with clickhouse-copier, then modify the tables again at the destination.... | 2021-02-01T02:31:50 | 2021-02-01T02:31:50 | {} | NONE |
ClickHouse | ClickHouse | 770,539,810 | 19,603 | alexey-milovidov | The code LGTM. | 2021-02-01T03:30:41 | 2021-02-01T03:30:41 | {} | MEMBER |
ClickHouse | ClickHouse | 770,386,295 | 19,384 | RO-29 | Hi, @kanadaj how did you fix this error? seems like the distributed engine is created with an engine definition of source table which is failing `clickhouse-copier` | 2021-01-31T13:56:00 | 2021-01-31T13:56:00 | {} | NONE |
ClickHouse | ClickHouse | 770,406,664 | 19,518 | filimonov | IMHO It would be a bit safer to use new name for that function to avoid compatibility issues. But can go like that too.
But now we need a "mirror" function - which will take IPv6 type and convert to IPv6 or IPv4 string format depending on what is stored | 2021-01-31T16:16:18 | 2021-01-31T16:16:41 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 770,637,585 | 19,884 | alexey-milovidov | > Stress test (thread) — Sanitizer assert (in stderr.log)
https://github.com/ClickHouse/ClickHouse/pull/19473 | 2021-02-01T07:31:22 | 2021-02-01T07:31:22 | {} | MEMBER |
ClickHouse | ClickHouse | 770,658,013 | 19,896 | depressed-pho | Like `Codec(Default, Encrypted('AES-128-GCM-SIV'))`? Okay, I will do it. | 2021-02-01T08:07:35 | 2021-02-01T08:07:35 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 770,377,803 | 17,847 | alexey-milovidov | There are multiple variants: LZSSE2, LZSSE8 with different tradeoffs.
It can be slow on compression but good on decompression.
"x64" is the same as "x86_64" or "amd64" - it is 100% of our servers.
| 2021-01-31T12:51:38 | 2021-01-31T12:51:47 | {} | MEMBER |
ClickHouse | ClickHouse | 770,386,350 | 19,384 | RO-29 | https://github.com/ClickHouse/ClickHouse/issues/16867#issuecomment-770384885 | 2021-01-31T13:56:30 | 2021-01-31T13:56:30 | {} | NONE |
ClickHouse | ClickHouse | 770,528,599 | 19,883 | fzyzcjy | Aha, sounds great!
P.S. Do you suggest using 21.1 in **production** environment - is it stable enough? | 2021-02-01T02:52:03 | 2021-02-01T02:52:11 | {} | NONE |
ClickHouse | ClickHouse | 770,533,623 | 19,883 | alexey-milovidov | > (1) look at those incompatible changes
Yes, this is necessarily.
> (2) wait a bit for 21.1.3
Most likely no. There are no regressions in 21.1 comparing to 20.12 (in the list above). | 2021-02-01T03:09:41 | 2021-02-01T03:09:41 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 770,539,361 | 19,603 | alexey-milovidov | No need to modify other languages, english is enough. | 2021-02-01T03:28:57 | 2021-02-01T03:28:57 | {} | MEMBER |
ClickHouse | ClickHouse | 770,539,525 | 19,603 | alexey-milovidov | Need to adjust integration tests. | 2021-02-01T03:29:37 | 2021-02-01T03:29:37 | {} | MEMBER |
ClickHouse | ClickHouse | 770,666,666 | 19,896 | alexey-milovidov | It should be conditionally compiled under `#if USE_SSL`, see examples in code. | 2021-02-01T08:21:23 | 2021-02-01T08:21:23 | {} | MEMBER |
ClickHouse | ClickHouse | 770,359,910 | 19,860 | alexey-milovidov | > AST fuzzer (UBSan) — ../src/Functions/DivisionUtils.h:90:40: runtime error: nan is outside the range of representable values of type 'long'
#19876 | 2021-01-31T10:26:56 | 2021-01-31T10:26:56 | {} | MEMBER |
ClickHouse | ClickHouse | 770,652,457 | 19,896 | alex-zaitsev | @depressed-pho , this is nice, but can we have it more generic? I.e. instead of AES codec define it as Encrypted with parameters. That will allow to make it more generic, as current 'encrypt' function: https://clickhouse.tech/docs/en/sql-reference/functions/encryption-functions/
For now it will only support AES_128_... | 2021-02-01T07:57:58 | 2021-02-01T07:57:58 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 770,773,406 | 11,862 | Dlougach | > It performs ARRAY JOIN before LIMIT BY and it actually requires that amount of memory.
> Reason: lack of possible optimization.
Sorry, that doesn’t make sense, if it was the lack of optimisation, there wouldn’t be any difference with the workaround. | 2021-02-01T11:05:31 | 2021-02-01T11:05:31 | {} | NONE |
ClickHouse | ClickHouse | 770,378,019 | 19,852 | vzakaznikov | @alexey-milovidov, this PR is ready to be merged. | 2021-01-31T12:53:13 | 2021-01-31T12:53:13 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 770,459,010 | 11,068 | sah4ez | I got the same error, but when I add the database name, a view was created. | 2021-01-31T22:05:18 | 2021-01-31T22:05:18 | {
"laugh": 1
} | NONE |
ClickHouse | ClickHouse | 770,534,084 | 19,883 | fzyzcjy | @alexey-milovidov Thanks! | 2021-02-01T03:11:23 | 2021-02-01T03:11:23 | {} | NONE |
ClickHouse | ClickHouse | 770,633,031 | 19,691 | chenk818 | > Hello, @chenk818, I checked that we did not have support for `reinterpretAsUInt256`, `reinterpretAsInt128`, `reinterpretAsInt256`. So I added it in scope of #19858.
> When it will be merged, this functions can be used for this task.
>
> ```
> SELECT reinterpretAsUInt256(hash), sipHash128('test') as hash
> ```
... | 2021-02-01T07:22:54 | 2021-02-01T07:22:54 | {} | NONE |
ClickHouse | ClickHouse | 770,639,974 | 19,603 | JackyWoo | Understand. | 2021-02-01T07:36:13 | 2021-02-01T07:36:13 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 770,695,951 | 11,999 | alexey-milovidov | We need to completely remove the notion of "Data Type domains". | 2021-02-01T09:04:18 | 2021-02-01T09:04:18 | {} | MEMBER |
ClickHouse | ClickHouse | 770,745,350 | 19,258 | kitaisreal | Answer is provided in https://github.com/ClickHouse/ClickHouse/issues/19517. | 2021-02-01T10:20:07 | 2021-02-01T10:20:07 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 770,368,682 | 19,881 | astudnev | Symlink of the complete /var/lib/clickhouse did not help, either
| 2021-01-31T11:39:49 | 2021-01-31T11:39:49 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 770,384,885 | 16,867 | RO-29 | Stack trace by clickhouse-copier command
`clickhouse-copier --config zookeeper.xml --task-upload-force=true task-upload-force --task-path /clickhouse --task-file ch-task.xml --base-dir ch-logs`
```
2021.01.31 14:29:59.533633 [ 6106 ] {} <Error> ClusterCopier: Will retry: Code: 36, e.displayText() = DB::Exception... | 2021-01-31T13:45:33 | 2021-01-31T13:45:33 | {} | NONE |
ClickHouse | ClickHouse | 770,439,366 | 19,229 | azat | @chirag-applift Also it is interesting to know does your environment has some virtualization layer? | 2021-01-31T19:47:05 | 2021-01-31T19:47:05 | {} | MEMBER |
ClickHouse | ClickHouse | 770,484,323 | 19,872 | alexey-milovidov | > Stress test (thread) — Sanitizer assert (in stderr.log)
#19887 | 2021-02-01T00:34:57 | 2021-02-01T00:34:57 | {} | MEMBER |
ClickHouse | ClickHouse | 770,544,721 | 19,384 | RO-29 | I confirm it works after removing the index from my source table, thx for this "hack" but it would be so nice if it works without it | 2021-02-01T03:49:11 | 2021-02-01T03:49:11 | {} | NONE |
ClickHouse | ClickHouse | 770,546,363 | 19,603 | JackyWoo | Thanks. I will commit the integration test in a few days.
> @vitlibar I think we can generalize Quotas further (after merging this PR) by allowing to use all the existing ProfileEvents. What do you think?
As I understanding you mean add all ProfileEvents to quotas. In my opintion it will expand user contorls, but... | 2021-02-01T03:55:19 | 2021-02-01T03:55:19 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 770,558,407 | 19,895 | alexey-milovidov | One number is used for tmp directory and another - when directory is renamed.
Only final block number is synchronized between replicas.
Temporary data parts are local and does not necessarily have distinct names on distinct replicas. | 2021-02-01T04:40:10 | 2021-02-01T04:40:10 | {} | MEMBER |
ClickHouse | ClickHouse | 770,618,896 | 19,603 | alexey-milovidov | Code complexity / maintainability. We want some uniform approach to events, metrics, quotas, limits, resource sharing... This task is currently unsolved. | 2021-02-01T06:54:37 | 2021-02-01T06:54:37 | {} | MEMBER |
ClickHouse | ClickHouse | 770,620,993 | 19,863 | alexey-milovidov | I have fixed Arcadia, waiting for checks... | 2021-02-01T06:58:45 | 2021-02-01T06:58:45 | {} | MEMBER |
ClickHouse | ClickHouse | 770,701,428 | 11,333 | alexey-milovidov | @alesapin Looks like easy to do. Could you please take a look? | 2021-02-01T09:11:47 | 2021-02-01T09:11:47 | {} | MEMBER |
ClickHouse | ClickHouse | 770,360,003 | 19,863 | alexey-milovidov | > AST fuzzer (UBSan) — ../src/Functions/FunctionDateOrDateTimeAddInterval.h:141:18: runtime error: signed integer overflow: 47764 + 9223372036854775807 cannot be r
#19859 | 2021-01-31T10:27:38 | 2021-01-31T10:27:38 | {} | MEMBER |
ClickHouse | ClickHouse | 770,360,923 | 19,584 | alexey-milovidov | > AST fuzzer (UBSan) — ../src/Functions/DivisionUtils.h:90:40: runtime error: nan is outside the range of re
#19876 | 2021-01-31T10:35:25 | 2021-01-31T10:35:25 | {} | MEMBER |
ClickHouse | ClickHouse | 770,488,983 | 19,584 | alexey-milovidov | @azat Could you please take a look at the `test_materialize_mysql_database`? | 2021-02-01T00:48:37 | 2021-02-01T00:49:40 | {} | MEMBER |
ClickHouse | ClickHouse | 770,574,509 | 19,603 | alexey-milovidov | > As I understanding you mean add all ProfileEvents to quotas.
Yes. For some profile events it does not make sense, but it's very tempting to add all of them "for free".
For example, we will be able to set up quota on disk IO, CPU in userspace, network bytes transferred, etc...
| 2021-02-01T05:23:11 | 2021-02-01T05:24:05 | {} | MEMBER |
ClickHouse | ClickHouse | 770,446,247 | 9,075 | miggi | @RO-29 nope, new version of the clickhouse-copier seemed crappy to me :) I've finally used copying select * from into with process under screen. It took some time) to move many Tb of data. But it's predictable and reliable.
Let's hope for fixes. | 2021-01-31T20:35:02 | 2021-01-31T20:36:07 | {} | NONE |
ClickHouse | ClickHouse | 770,448,300 | 9,075 | RO-29 | > select * from into
For rebalancing the shards, let's say I have to move 10Tb, approx how many hours will it take?
Assuming All my shards and cluster are on the same network (5GiB/s) | 2021-01-31T20:48:19 | 2021-01-31T20:48:35 | {} | NONE |
ClickHouse | ClickHouse | 770,490,658 | 19,584 | alexey-milovidov | > AST fuzzer (debug) — Logical error: 'Bad cast from type DB::ColumnVector<unsigned long> to DB::ColumnTuple'.
@kitaisreal This looks like a recent issue. | 2021-02-01T00:49:31 | 2021-02-01T00:49:31 | {} | MEMBER |
ClickHouse | ClickHouse | 770,554,597 | 19,897 | alexey-milovidov | Bitcode files are generated correctly:
```
milovidov@milovidov-desktop:~/work/ClickHouse/build$ du -hs .coderrect
11G .coderrect
milovidov@milovidov-desktop:~/work/ClickHouse/build$ find . -name *.bc | wc -l
8639
``` | 2021-02-01T04:25:57 | 2021-02-01T04:25:57 | {} | MEMBER |
ClickHouse | ClickHouse | 770,698,890 | 11,727 | alexey-milovidov | @azat Looks like it is fixed in master, can you confirm? | 2021-02-01T09:08:15 | 2021-02-01T09:08:15 | {} | MEMBER |
ClickHouse | ClickHouse | 770,779,964 | 19,857 | akuzm | Not actively working on this ATM. | 2021-02-01T11:16:17 | 2021-02-01T11:16:17 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 770,453,564 | 9,075 | miggi | sorry, I meant insert into table A select * from table B . I used it for moving data of course. It took 5-8 days to move ~12 Tb (~150 billons of records). But its's very dependant on your bandwidth, cpu and other factors. I had different architecture on destination cluster, so I used insertion into distributed table (f... | 2021-01-31T21:25:27 | 2021-01-31T21:27:15 | {
"+1": 1
} | NONE |
ClickHouse | ClickHouse | 770,484,926 | 19,888 | alexey-milovidov | @azat Could you please take a look? (this is insertion into Distributed table). | 2021-02-01T00:37:16 | 2021-02-01T00:37:23 | {} | MEMBER |
ClickHouse | ClickHouse | 770,486,424 | 19,868 | alexey-milovidov | > AST fuzzer (debug) — Logical error: 'Block structure mismatch in RemoteBlockOutputStream stream: different types:
#19888 | 2021-02-01T00:42:43 | 2021-02-01T00:42:43 | {} | MEMBER |
ClickHouse | ClickHouse | 770,486,621 | 19,863 | alexey-milovidov | This PR is ready but blocked by Arcadia :( | 2021-02-01T00:43:24 | 2021-02-01T00:43:24 | {} | MEMBER |
ClickHouse | ClickHouse | 770,528,156 | 19,883 | alexey-milovidov | This issue is relevant in 20.12 but it is fixed in 21.1 (by this PR: #17206).
> Does it reproduce on recent release?
The latest release is 21.1. | 2021-02-01T02:50:22 | 2021-02-01T03:05:17 | {} | MEMBER |
ClickHouse | ClickHouse | 770,535,310 | 19,332 | alexey-milovidov | Minor issue. Not merged correctly - won't backport. | 2021-02-01T03:15:25 | 2021-02-01T03:15:25 | {} | MEMBER |
ClickHouse | ClickHouse | 770,537,882 | 19,603 | JackyWoo | Can anybody help. I just modify the English docs in “docs/en” folder. How can I modify other languages conviniently? | 2021-02-01T03:23:30 | 2021-02-01T03:23:30 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 770,545,578 | 19,897 | alexey-milovidov | ```
$ cd build
$ ninja clean
$ coderrect ninja
```
It takes only slightly longer than usual build. | 2021-02-01T03:52:05 | 2021-02-01T04:27:14 | {} | MEMBER |
ClickHouse | ClickHouse | 770,563,970 | 19,884 | alexey-milovidov | clickhouse-test is creating unique temporary databases for every test automatically.
(It is needed for parallel test run) | 2021-02-01T04:56:36 | 2021-02-01T04:56:36 | {} | MEMBER |
ClickHouse | ClickHouse | 770,687,947 | 19,896 | alexey-milovidov | Unit tests are optional but functional tests are mandatory. | 2021-02-01T08:53:15 | 2021-02-01T08:53:15 | {} | MEMBER |
ClickHouse | ClickHouse | 770,816,703 | 16,132 | qoega | @zhang2014 will you have time to return to this PR? | 2021-02-01T12:21:09 | 2021-02-01T12:21:09 | {} | MEMBER |
ClickHouse | ClickHouse | 770,347,638 | 19,557 | sv3t0sl4v | Does not show rows with values of 0 when grouping. | 2021-01-31T08:41:33 | 2021-01-31T08:41:33 | {} | NONE |
ClickHouse | ClickHouse | 770,354,723 | 17,623 | baibaichen | @alexey-milovidov see https://github.com/ClickHouse/ClickHouse/issues/19873
| 2021-01-31T09:43:21 | 2021-01-31T09:43:21 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 770,390,883 | 19,881 | astudnev | ```
studnev@ch1-1:~$ sudo df -h /var/lib/clickhouse
Filesystem Size Used Avail Use% Mounted on
/dev/md3 1.7T 1.3T 329G 80% /spool1
studnev@ch1-1:~$ sudo df -h /var/lib/clickhouse/data
Filesystem Size Used Avail Use% Mounted on
/dev/md3 1.7T 1.3T 329G 80% /spool1
studnev@ch1-1:~$ s... | 2021-01-31T14:29:09 | 2021-01-31T14:29:20 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 770,417,806 | 19,518 | filimonov | Also - did you benchmark the speed difference for the pure IPv6 use case ? | 2021-01-31T17:28:22 | 2021-01-31T17:28:45 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 770,635,447 | 19,863 | alexey-milovidov | Arcadia did not publish the results. Here they are:
| 2021-02-01T07:27:30 | 2021-02-01T07:27:30 | {} | MEMBER |
ClickHouse | ClickHouse | 770,668,688 | 19,896 | Enmk | @depressed-pho thanks, a very nice feature! I noticed that there are no tests, could you please add some? To simplify this, you may want to reuse the unit-test suite from other codes, please see [`CodecTestCompatibility`](https://github.com/ClickHouse/ClickHouse/blob/ac989f3a8fe8e2135e25ac7d6e340b048092e906/src/Compres... | 2021-02-01T08:24:36 | 2021-02-01T08:24:36 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 770,697,261 | 11,862 | alexey-milovidov | It performs ARRAY JOIN before LIMIT BY and it actually requires that amount of memory.
Reason: lack of possible optimization. | 2021-02-01T09:06:04 | 2021-02-01T09:06:04 | {} | MEMBER |
ClickHouse | ClickHouse | 770,701,649 | 19,896 | depressed-pho | But I'm not sure how to write a functional test for this. Would it be enough to just add something like `<encryption><key_command>echo "Some static key in base64"</key_command></encryption>` to `tests/server-test.xml` and a stateless SQL test using the codec? | 2021-02-01T09:12:06 | 2021-02-01T09:12:06 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 770,817,551 | 19,889 | KochetovNicolai | ```
2021.01.31 17:30:37.097617 [ 2390 ] {de25703a-a0f8-4b61-a39e-ae59c7673156} <Debug> executeQuery: (from [::1]:54656, using production parser) (comment: '/usr/share/clickhouse-test/queries/0_stateless/01661_test_toDayOfWeek_mysql_compatibility.sql') SELECT * FROM system.tables
2021.01.31 17:30:37.099770 [ 2390 ] {d... | 2021-02-01T12:22:41 | 2021-02-01T12:25:14 | {} | MEMBER |
ClickHouse | ClickHouse | 770,349,540 | 19,862 | alexey-milovidov | ```
docker: Error response from daemon: toomanyrequests: You have reached your pull rate limit. You may increase the limit by authenticating and upgrading: https://www.docker.com/increase-rate-limit.
See 'docker run --help'.
``` | 2021-01-31T08:59:50 | 2021-01-31T08:59:50 | {} | MEMBER |
ClickHouse | ClickHouse | 770,380,960 | 16,867 | RO-29 | Hi, how did you fix this error? seems like some distributed engine is created with an engine definition of source table which is failing `clickhouse-copier`
Any simple workaround?
One workaround I found was to define source table as `table_distributed` whose engine type is distributed engine but I don't think its... | 2021-01-31T13:15:14 | 2021-01-31T13:16:43 | {} | NONE |
ClickHouse | ClickHouse | 770,411,414 | 19,384 | kanadaj | @RO-29 The bad way - I've temporarily removed my indices and TTLs from the table to copy it, then re-added it at the destination. | 2021-01-31T16:46:55 | 2021-01-31T16:46:55 | {} | NONE |
ClickHouse | ClickHouse | 770,644,103 | 18,247 | ghuname | @SaltTan This is situation right now:
Now all 4 number are the same = 113. They used to be (left to right, top to bottom) 124, 124, 1, 1 (not the same), when I reported the istuation last time.
The f... | 2021-02-01T07:43:28 | 2021-02-01T07:43:28 | {} | NONE |
ClickHouse | ClickHouse | 770,705,224 | 19,896 | alexey-milovidov | @depressed-pho Yes, you can add a new config with a script for static key, see `tests/config/README.md` | 2021-02-01T09:17:24 | 2021-02-01T09:17:38 | {} | MEMBER |
ClickHouse | ClickHouse | 770,557,539 | 19,897 | alexey-milovidov | It's not possible to analyze 11G bitcode files in 0 seconds, something is definitely incorrect. | 2021-02-01T04:37:06 | 2021-02-01T04:37:06 | {} | MEMBER |
ClickHouse | ClickHouse | 770,637,986 | 19,747 | alexey-milovidov | tmp directory is local on each replica. | 2021-02-01T07:32:12 | 2021-02-01T07:32:12 | {} | MEMBER |
ClickHouse | ClickHouse | 770,364,574 | 19,868 | alexey-milovidov | > AST fuzzer (MSan) — SUMMARY: MemorySanitizer: use-of-uninitialized-value obj-x86_64-linux-gnu/../src/Columns/ColumnUnique.h:524:18 in COW<DB::IColumn>::mutable_
https://github.com/ClickHouse/ClickHouse/issues/19877 | 2021-01-31T11:07:53 | 2021-01-31T11:07:53 | {} | MEMBER |
ClickHouse | ClickHouse | 770,380,600 | 10,571 | fzyzcjy | Hi I have the same problem. Is there any updates? Thanks! @filimonov
This makes my Elastic Stack's MetricBeat cannot gather Clickhouse's Prometheus data :( | 2021-01-31T13:12:27 | 2021-01-31T13:13:06 | {} | NONE |
ClickHouse | ClickHouse | 770,391,744 | 19,881 | astudnev | It replicated, despite. of these messages.
Most probably the issue was in /tmp space | 2021-01-31T14:34:18 | 2021-01-31T14:34:18 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 770,443,584 | 9,075 | RO-29 | >Also I've cleaned up all the tables manually and had to stop this.
>That's seems to me like downgrade :) you can consider this like a feedback guys.
@miggi did you test with latest versions? I am using click house latest version v21.1.2 and still facing a slow copy issue
| 2021-01-31T20:16:04 | 2021-01-31T20:16:04 | {} | NONE |
ClickHouse | ClickHouse | 770,531,271 | 19,883 | alexey-milovidov | Version 21.1 was released to stable on Jan 18. It was in prestable (staging) since Dec 10 and we have fixed all the issues that were found during that month: #17951
Please read the changelog in "backward incompatible changes" - there are a few: https://github.com/ClickHouse/ClickHouse/blob/master/CHANGELOG.md#backwa... | 2021-02-01T03:01:31 | 2021-02-01T03:03:19 | {} | MEMBER |
ClickHouse | ClickHouse | 770,540,081 | 19,603 | alexey-milovidov | @vitlibar I think we can generalize Quotas further (after merging this PR) by allowing to use all the existing ProfileEvents. What do you think? | 2021-02-01T03:31:51 | 2021-02-01T03:31:51 | {} | MEMBER |
ClickHouse | ClickHouse | 770,544,369 | 16,867 | RO-29 | For anyone looking for solutions looking to make it work, you have to strip source tables of indexes and ttl columns and ...
It works after that but I would really love for clickhouse-copy to work without this hack
https://github.com/ClickHouse/ClickHouse/issues/19384#issuecomment-770522429 | 2021-02-01T03:47:54 | 2021-02-01T03:50:12 | {} | NONE |
ClickHouse | ClickHouse | 770,614,815 | 19,603 | JackyWoo | Adding disk IO, CPU in userspace, network bytes transferred etc... to quotas is excellent. What I am wondering is that why add all but not what we need. | 2021-02-01T06:46:42 | 2021-02-01T06:46:42 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 770,662,911 | 15,162 | vlauzeckas | yes, new version of clickhouse corrupts number of queries:
`SELECT * FROM (SELECT NULL AS a, 1 AS b) AS foo
RIGHT JOIN (SELECT 1024 AS b) AS bar
ON 1 = foo.b;`
throws error: Not equi-join ON expression: 1 = b. No columns in one of equality side.: While processing 1 = b
whats wrong in joining on value=colu... | 2021-02-01T08:15:28 | 2021-02-01T08:15:28 | {} | NONE |
ClickHouse | ClickHouse | 770,727,466 | 19,252 | alexey-milovidov | 21.2 is prestable #19912 | 2021-02-01T09:52:14 | 2021-02-01T09:52:35 | {} | MEMBER |
ClickHouse | ClickHouse | 770,813,018 | 19,129 | qoega | https://clickhouse-test-reports.s3.yandex.net/19884/2459777f57a265378e949d7ab9926c29fc7b4de6/stress_test_(thread).html#fail1 | 2021-02-01T12:14:16 | 2021-02-01T12:14:16 | {} | MEMBER |
ClickHouse | ClickHouse | 770,352,532 | 17,759 | hombit | Not sure someone will test it, but the whole table can be downloaded from google drive (100 GB tar)
https://drive.google.com/file/d/1lRqzJx650MrTuY8SGfMTO9E_5U8xldL-/view?usp=sharing
Original post is updated | 2021-01-31T09:25:15 | 2021-01-31T09:45:03 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 770,358,014 | 19,862 | alexey-milovidov | > AST fuzzer (ASan) — 0x603000e4ab91 is located 1 bytes to the right of 32-byte region [0x603000e4ab70,0x603000e4ab90)
#19233 | 2021-01-31T10:10:49 | 2021-01-31T10:10:49 | {} | MEMBER |
ClickHouse | ClickHouse | 770,491,904 | 19,852 | alexey-milovidov | > AST fuzzer (UBSan) — ../src/Functions/DivisionUtils.h:90:40: runtime error: nan is outside the range of representable values of type 'long'
#19876 | 2021-02-01T00:52:02 | 2021-02-01T00:52:02 | {} | MEMBER |
ClickHouse | ClickHouse | 756,604,508 | 18,818 | templarzq | create table test_bm(
dim UInt64,
id UInt64
)
ENGINE = MergeTree()
ORDER BY ( dim,id )
SETTINGS index_granularity = 8192;
create table test_bm_join(
dim UInt64,
id UInt64
)
ENGINE = MergeTree()
ORDER BY ( dim,id )
SETTINGS index_granularity = 8192;
insert into test_bm VALUES (1,1),(2,2),(3,3),... | 2021-01-08T07:44:55 | 2021-01-08T07:44:55 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 756,657,981 | 18,865 | IVitamin-C | 
| 2021-01-08T09:45:03 | 2021-01-08T09:45:03 | {} | NONE |
ClickHouse | ClickHouse | 756,714,340 | 18,721 | UnamedRus | > It seems a certificate issue, and the host of S3 in China regions is different from global regions, for example, it's 'https://s3.cn-north-1.amazonaws.com.cn'.
Yes, it looks like.
> I would like to know is this a known issue or need a fix on this?
I don't think that is a known issue, one of the possible rea... | 2021-01-08T11:47:38 | 2021-01-08T11:47:38 | {} | CONTRIBUTOR |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.