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 | 633,860,604 | 11,194 | amy108 | i have solved it. | 2020-05-26T07:34:06 | 2020-05-26T07:34:06 | {} | NONE |
ClickHouse | ClickHouse | 633,925,571 | 8,979 | ropiteaux | @alesapin Hello, any news on this? Would it work with `20.3.8.53`? Thks | 2020-05-26T09:49:33 | 2020-05-26T09:49:33 | {} | NONE |
ClickHouse | ClickHouse | 634,473,418 | 11,209 | filimonov | You don't need Buffer tables after Kafka. Kafka flushes the data in blocks of `kafka_max_block_size`, or by timeout controlled by `stream_flush_interval_ms` setting.
Adjust those settings if you need to form bigger blocks. | 2020-05-27T07:11:10 | 2020-05-27T07:12:25 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 633,719,490 | 10,884 | filimonov | You did quite a lot of changes which seems unrelated
Did you read my message above? The only possible reason for the problem was mentioned, together with options to solve. Why do you change all those parameters which have nothing to it?
read that very carefully, please:
> your broker rejected commit because d... | 2020-05-25T21:20:39 | 2020-05-26T08:08:31 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 633,505,429 | 11,135 | zhuyoucui0909 | Any findings or update for this? | 2020-05-25T10:34:10 | 2020-05-25T10:34:10 | {} | NONE |
ClickHouse | ClickHouse | 633,521,598 | 11,135 | zhuyoucui0909 | Do you know which version introduce this issue? We are encountering a server hung issue on version 19.14.7.15. We can try to upgrade to other version that without this back_compatibility issue.
| 2020-05-25T11:16:44 | 2020-05-25T11:16:44 | {} | NONE |
ClickHouse | ClickHouse | 633,530,845 | 11,178 | 4ertus2 | I'd prefere remove this test at all. It's not an algorithmic part and already tested by functional tests. | 2020-05-25T11:41:51 | 2020-05-25T11:41:51 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 633,566,576 | 11,135 | winter7 | @4ertus2 I reproduced this and got it from `system.query_log`:
```
exception: Code: 352, e.displayText() = DB::Exception: Received from chi-ebstest-clickhouse-s0-r0-0.ebstest-clickhouse-shard0-0.clickhouse.svc.cluster.local:9000. DB::Exception: Ambiguous column 'id1'. Stack trace:
0. Poco::Exception::Ex... | 2020-05-25T13:14:12 | 2020-05-25T13:14:39 | {} | NONE |
ClickHouse | ClickHouse | 633,698,904 | 11,180 | alexey-milovidov | Answer: tests under sanitizers slowed down too much. | 2020-05-25T19:52:39 | 2020-05-25T19:52:39 | {} | MEMBER |
ClickHouse | ClickHouse | 633,723,824 | 11,006 | nikitamikhaylov | Integration tests are broken in master. | 2020-05-25T21:41:56 | 2020-05-25T21:41:56 | {} | MEMBER |
ClickHouse | ClickHouse | 634,074,649 | 11,041 | alexey-milovidov | First hand experience: one user faced the same issue; after upgrade of all replicas, the issue is gone. | 2020-05-26T14:51:17 | 2020-05-26T14:51:17 | {} | MEMBER |
ClickHouse | ClickHouse | 634,417,509 | 11,193 | BernieJiangDL | > Background merges are executed by thread from so called 'background thread pool'.
> Check the `background_pool_size` setting.
>
> And usually, if you face 'Too many parts ...' it's NOT caused by background pool. Most common reason is bad schema (too granular partitioning, too aggressive compression, etc) creating... | 2020-05-27T04:15:14 | 2020-05-27T04:15:14 | {} | NONE |
ClickHouse | ClickHouse | 633,520,537 | 8,889 | vumdao | The error appears in version 20.3.9 revision 54433
```
2020.05.25 11:01:08.746745 [ 148 ] {} <Fatal> BaseDaemon: ########################################
2020.05.25 11:01:08.746752 [ 110 ] {c5496c36-3509-4d58-a941-dba6077edcbb} <Trace> CreatingSetsBlockInputStream: Creating join.
2020.05.25 11:01:08.746757 [ 138 ... | 2020-05-25T11:13:57 | 2020-05-25T11:13:57 | {} | NONE |
ClickHouse | ClickHouse | 633,536,370 | 11,135 | 4ertus2 | ```it reports "DB::Exception: Ambiguous column***"```
Could you print stack trace after this line? You have omitted the most significant part of error. | 2020-05-25T11:57:02 | 2020-05-25T11:57:18 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 634,198,138 | 11,190 | azat | >Also it's worth to increase connect timeout for all tests globally...
Agree (maybe worth revert #11170 and increase timeouts globally for tests under sanitizers only) | 2020-05-26T18:27:28 | 2020-05-26T18:27:28 | {} | MEMBER |
ClickHouse | ClickHouse | 634,201,889 | 11,190 | alexey-milovidov | > increase timeouts globally
Also it can make some tests to run too slow (tests for failover). | 2020-05-26T18:34:21 | 2020-05-26T18:34:21 | {} | MEMBER |
ClickHouse | ClickHouse | 634,204,633 | 11,163 | azat | Previous HEAD's:
- 461fea813705038d930ad779b019497795fc4915 - stock 5.2.1
- b8b58dd62c7ebf53465907ebb2975ea6edf9754e and 4e3a43db93d71398f4087a645a0aa515a29deea1 - few patches reverted, slower
- 3adda87acfeb9202b41dc56fae07ec8568fc0fcc - mostly the same as latest
- dee04567b8f91fca08a483565e3255cc80a9e6b7 -- latest... | 2020-05-26T18:39:54 | 2020-05-26T18:39:54 | {} | MEMBER |
ClickHouse | ClickHouse | 634,222,249 | 11,163 | azat | >Yes, it is very useful because it provides good diagnostics for user. How it is possible to call it before malloc? Maybe it's worth it.
I though about changing [entry point](http://www.scoberlin.de/content/media/http/informatik/gcc_docs/ld_3.html#SEC10), but it will work only on linux (at least what I wanted to do)... | 2020-05-26T19:13:15 | 2020-05-26T19:13:15 | {} | MEMBER |
ClickHouse | ClickHouse | 619,440,787 | 10,373 | azat | So to summarize this should solve some issues that had been introduced by the #10341 and adds support for LIMIT/LIMIT BY/ORDER BY | 2020-04-25T21:06:50 | 2020-05-26T22:16:38 | {} | MEMBER |
ClickHouse | ClickHouse | 633,497,361 | 11,166 | alesapin | You can this query with `KILL MUTATION id_from_system.mutations`. But what is the expected behavior? | 2020-05-25T10:14:46 | 2020-05-25T10:14:46 | {} | MEMBER |
ClickHouse | ClickHouse | 633,861,505 | 1,576 | Siao-pin | Still facing same issue in 20.4.3.16 | 2020-05-26T07:36:18 | 2020-05-26T07:36:18 | {} | NONE |
ClickHouse | ClickHouse | 633,725,210 | 11,149 | filimonov | <s>Can't reproduce locally. Most probably the PR just pushed a strange balance of sleeps inside those Kafka tests, and some sleeps need to be reviewed/adjusted. Locally i have another test flapping now. </s>
UPD: https://github.com/ClickHouse/ClickHouse/pull/11149#issuecomment-633898915 | 2020-05-25T21:48:52 | 2020-05-26T11:32:32 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 634,065,628 | 11,041 | alexey-milovidov | Rolling update from 19.17 will lead to these error messages in log and temporary pause in replication. This is expected. Replication will continue after update of all replicas. | 2020-05-26T14:36:29 | 2020-05-26T14:36:29 | {
"+1": 2
} | MEMBER |
ClickHouse | ClickHouse | 634,180,014 | 11,163 | azat | >Does it perform a test at runtime that this feature is available in kernel?
Actually it does
https://github.com/jemalloc/jemalloc/blob/6cdac3c573de86c8d59d69fca8f1778bdbec25e0/src/pages.c#L661-L662
>Performance — 2 faster, 205 slower, 137 unstable, 3 errors
This, is how performance tests looks after upgrad... | 2020-05-26T17:55:09 | 2020-05-26T17:55:09 | {} | MEMBER |
ClickHouse | ClickHouse | 634,222,662 | 8,979 | filimonov | > copied into detached folder and then attach accordingly
How did you do that? Do you use FREEZE or stop the server ? When you back up a active working copy you can get inconsistant copy.
> B::Exception: No <column>.mrk file checksum for column <column> in part .../20190201_1_1_0/
Does the folder have that fil... | 2020-05-26T19:14:02 | 2020-05-26T19:14:02 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 634,277,214 | 11,214 | Potya | TODO: tests and mb some new data types | 2020-05-26T21:02:03 | 2020-05-26T21:02:03 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 633,503,304 | 11,180 | alesapin | Wait only for functional tests. | 2020-05-25T10:28:58 | 2020-05-25T10:28:58 | {} | MEMBER |
ClickHouse | ClickHouse | 633,723,722 | 11,187 | filimonov | Version? | 2020-05-25T21:41:27 | 2020-05-25T21:41:27 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 634,304,397 | 11,207 | alexey-milovidov | No. | 2020-05-26T22:02:46 | 2020-05-26T22:02:46 | {} | MEMBER |
ClickHouse | ClickHouse | 634,378,299 | 10,537 | CurtizJ | Arcadia still doesn't support some c++20 features | 2020-05-27T01:52:45 | 2020-05-27T01:52:45 | {} | MEMBER |
ClickHouse | ClickHouse | 633,545,543 | 11,135 | 4ertus2 | The 'Ambiguous column' error added cause you may have wrong result without it in some cases. Unfortunatelly we didn't find that it breaks joins with remote() tables. I cannot find quickly which version is first affected but the code in development branch committed at least 29.07.2019. So newer versions are probably aff... | 2020-05-25T12:21:28 | 2020-05-25T12:22:19 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 633,562,138 | 11,094 | den-crane | @shubham721 it's just an example. The documentation does not state that only these functions are supported. Almost any function can be used including combinators i.e. `groupUniqArrayArrayState`
https://gist.github.com/den-crane/048ce66bc59f5577d56e42d76934682e
https://gist.github.com/den-crane/6eff375752a236a456e1... | 2020-05-25T13:03:29 | 2020-05-25T13:03:29 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 633,626,947 | 11,169 | filimonov | Confirmed on 20.4.3. Thank you for the report! | 2020-05-25T15:39:26 | 2020-05-25T15:39:49 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 633,725,130 | 10,965 | nikitamikhaylov | Main PR is merged. Close. | 2020-05-25T21:48:27 | 2020-05-25T21:48:27 | {} | MEMBER |
ClickHouse | ClickHouse | 633,797,228 | 8,889 | jackpgao | some error info:
```
↖ Progress: 0.00 rows, 0.00 B (0.00 rows/s., 0.00 B/s.) Received exception from server (version 20.3.8):
Code: 1000. DB::Exception: Received from 127.0.0.1:9000. DB::Exception: mysqlxx::Exception: e.com ((nullptr):1851880817).
↖ Progress: 0.00 rows, 0.00 B (0.00 rows/s., 0.00 B/s.) Rece... | 2020-05-26T04:10:35 | 2020-05-26T04:11:16 | {} | NONE |
ClickHouse | ClickHouse | 633,957,199 | 10,923 | vitlibar | Please resolve the conflict and I will merge this PR. | 2020-05-26T11:03:10 | 2020-05-26T11:03:10 | {} | MEMBER |
ClickHouse | ClickHouse | 634,067,251 | 1,102 | alexey-milovidov | Now this feature can be easily implemented on top of DatabaseAtomic.
@tavplubix said that he is ready to take. | 2020-05-26T14:39:07 | 2020-05-26T14:39:07 | {
"hooray": 3
} | MEMBER |
ClickHouse | ClickHouse | 633,542,863 | 11,166 | donghy-coredumped | @alesapin Thank you, I think you should report an error because this conversion is not safe | 2020-05-25T12:14:37 | 2020-05-25T12:14:37 | {} | NONE |
ClickHouse | ClickHouse | 633,633,287 | 9,545 | qoega | we get error in ya make build:
```
ld: error: undefined symbol: __cxa_thread_atexit
>>> referenced by ThreadProfileEvents.cpp:166 (/-S/clickhouse/src/Common/ThreadProfileEvents.cpp:166)
>>> ThreadProfileEvents.cpp.o:(DB::PerfEventsCounters::processThreadLocalChanges(std::__y1::basic_string<char, std::... | 2020-05-25T15:57:23 | 2020-05-25T15:57:23 | {} | MEMBER |
ClickHouse | ClickHouse | 633,828,677 | 8,889 | jackpgao | It only occurs with Distributed engine, and the version `20.4.3.16` seems works well. | 2020-05-26T06:17:29 | 2020-05-26T06:17:29 | {} | NONE |
ClickHouse | ClickHouse | 633,858,516 | 11,184 | qoega | Test fail is probably related `test_system_queries/test.py::test_SYSTEM_FLUSH_LOGS`
| 2020-05-26T07:29:23 | 2020-05-26T07:29:23 | {} | MEMBER |
ClickHouse | ClickHouse | 634,131,390 | 11,201 | jackpgao | Same to #10994 | 2020-05-26T16:24:45 | 2020-05-26T16:24:45 | {} | NONE |
ClickHouse | ClickHouse | 634,231,323 | 11,028 | vitlibar | MySQL also supports `BIT[(M)]`, `BOOL` (as a synonyms for `TINYINT`), and treats `FLOAT(p)` as either `FLOAT` or `DOUBLE` depending on `p`. Should we support those types too? | 2020-05-26T19:31:10 | 2020-05-26T19:39:50 | {} | MEMBER |
ClickHouse | ClickHouse | 634,402,025 | 9,354 | filimonov | Probably we need a test? | 2020-05-27T03:14:30 | 2020-05-27T03:14:30 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 633,534,112 | 10,200 | KochetovNicolai | It's better to enable `joined_subquery_requires_alias` if you can, because this setting fixes some issues connected with column names. However, joins with non-aliased subquery will throw exception and should will be fixed.
Setting itself is most likely have no connection with segfaults. | 2020-05-25T11:50:39 | 2020-05-25T11:50:39 | {} | MEMBER |
ClickHouse | ClickHouse | 633,629,102 | 10,994 | Tri0L | Have the same issue after upgrade 20.1.2.4 => 20.4.2.9, `dictionaries_lazy_load=false` didn't help. | 2020-05-25T15:45:25 | 2020-05-25T15:45:25 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 633,725,560 | 11,149 | filimonov | > Also seems that now the upstream cppkafka can be used (your PR is merged, and other build fixes already there too)
Not in that PR. Cppkafka has some other changes too. That PR is a bugfix, want to keep it compact and limit side effects. | 2020-05-25T21:50:46 | 2020-05-25T21:50:46 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 633,746,428 | 11,191 | alexey-milovidov | @CurtizJ RFC. | 2020-05-25T23:59:30 | 2020-05-25T23:59:30 | {} | MEMBER |
ClickHouse | ClickHouse | 634,016,845 | 11,082 | Felixoid | It's impossible to implement it with the current code. The context of a task belongs to `default` user. https://github.com/clickhouse/ClickHouse/blob/7e18138/src/Interpreters/DDLWorker.cpp#L562-L565 | 2020-05-26T13:13:58 | 2020-05-26T13:14:56 | {} | MEMBER |
ClickHouse | ClickHouse | 634,228,833 | 11,171 | KochetovNicolai | Tests are failed mainly because of this: https://github.com/ClickHouse/ClickHouse/blob/master/src/Processors/QueryPipeline.cpp#L654
It will be fixed automatically if we update `PartialSortingTransform` instead of adding a new one. | 2020-05-26T19:25:50 | 2020-05-26T19:25:50 | {} | MEMBER |
ClickHouse | ClickHouse | 633,537,792 | 10,200 | KochetovNicolai | Also, you may try to install asan build to detect possible incorrect memory accesses.
It will be slower, but may reveal more information connected to your problem.
You can download asan packages from our ci checks (press `Details` on `ClickHouse build check` for commit and download package where `Sanitizer` column is... | 2020-05-25T12:00:46 | 2020-05-25T12:00:46 | {} | MEMBER |
ClickHouse | ClickHouse | 633,723,364 | 11,185 | filimonov | https://github.com/ClickHouse/ClickHouse/issues/7815
https://github.com/ClickHouse/ClickHouse/issues/9766 | 2020-05-25T21:39:38 | 2020-05-25T21:40:23 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 633,724,985 | 10,831 | nikitamikhaylov | Auto-merge did not work :sweat: | 2020-05-25T21:47:39 | 2020-05-25T21:47:39 | {} | MEMBER |
ClickHouse | ClickHouse | 634,140,451 | 10,241 | SaltTan | This is not fixed.
I've checked 20.4.4.18, all inserts into MVs are done by the same thread consecutively. | 2020-05-26T16:41:42 | 2020-05-26T16:41:42 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 634,208,081 | 11,163 | alexey-milovidov | > So this change makes that instruction check code pretty useless, is it useful?
Yes, it is very useful because it provides good diagnostics for user. How it is possible to call it before malloc? Maybe it's worth it. | 2020-05-26T18:46:33 | 2020-05-26T18:46:33 | {} | MEMBER |
ClickHouse | ClickHouse | 634,379,576 | 11,153 | jackpgao | same problem with `20.4.3.16` and `20.4.2.9` | 2020-05-27T01:57:27 | 2020-05-27T03:04:29 | {} | NONE |
ClickHouse | ClickHouse | 633,687,351 | 8,441 | den-crane | seems it fixed in 20.4.3.16 and clickhouse-client became faster than it was ever before
```
20.3.9.70
for i in `seq 1 50`; do curl -s -w 'Total: %{time_total}s\n' 'http://localhost:8123/?input_format_parallel_parsing=0&query=INSERT%20INTO%20log%20format%20TSV' --data-binary @test.json ; done;
Total: 2.244249s
... | 2020-05-25T19:06:09 | 2020-05-25T19:07:06 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 633,744,972 | 10,775 | alexey-milovidov | Continued in https://github.com/ClickHouse/ClickHouse/issues/11190 | 2020-05-25T23:49:57 | 2020-05-25T23:49:57 | {} | MEMBER |
ClickHouse | ClickHouse | 633,849,350 | 10,076 | Siao-pin | We use new 20.4.3.16 version of ClickHouse and faced the same problem - immortal query that cannot be killed. | 2020-05-26T07:08:29 | 2020-05-26T07:08:29 | {} | NONE |
ClickHouse | ClickHouse | 633,950,636 | 10,387 | vitlibar | This fixes https://github.com/ClickHouse/ClickHouse/issues/10019 | 2020-05-26T10:47:06 | 2020-05-26T10:47:06 | {} | MEMBER |
ClickHouse | ClickHouse | 634,047,701 | 10,561 | nikitamikhaylov | Stress tests (memory) fixed in master, usban assert in lz4 fixed in master | 2020-05-26T14:06:58 | 2020-05-26T14:06:58 | {} | MEMBER |
ClickHouse | ClickHouse | 634,164,379 | 11,183 | alesapin | test failures are expected. | 2020-05-26T17:25:16 | 2020-05-26T17:25:16 | {} | MEMBER |
ClickHouse | ClickHouse | 634,120,278 | 10,679 | amudong | > Maybe it should be implemented as `SYSTEM` query instead of `ALTER`?
> Something like `SYSTEM DROP REPLICA 'replica_name' FROM table_name` to remove dead replica of existing table and `SYSTEM DROP REPLICA 'replica_name' FROM '/path/to/table/in/zk/'` to remove dead replica of some table, which does not exists on the ... | 2020-05-26T16:04:57 | 2020-05-26T17:56:19 | {
"+1": 4
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 634,267,897 | 11,184 | blinkov | > @blinkov Are you going to fix by yourself or need some help?
I'll try | 2020-05-26T20:42:14 | 2020-05-26T20:42:14 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 634,480,237 | 11,119 | azat | @akuzm @alexey-milovidov can you take a look? | 2020-05-27T07:26:20 | 2020-05-27T07:26:20 | {} | MEMBER |
ClickHouse | ClickHouse | 633,528,263 | 10,683 | naorlov | @alexey-milovidov I want to do this task | 2020-05-25T11:35:01 | 2020-05-25T11:35:01 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 633,623,103 | 11,169 | filimonov | Hm. I can't reproduce that on master | 2020-05-25T15:29:01 | 2020-05-25T15:34:16 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 634,008,506 | 11,202 | akuzm | Thanks, I thought we had an issue for that already.
A workaround: `clickhouse-local -q 'select 1' --logger.log=log1.txt -- --application.baseName=local` -- but the logger settings won't work anyway. | 2020-05-26T12:58:40 | 2020-05-26T12:58:40 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 634,182,177 | 11,190 | azat | I saw some flackiness for the following test too:
- `test_insert_distributed_load_balancing` https://clickhouse-test-reports.s3.yandex.net/11114/aa656f95dd331d00d152e76ff8bca70d1bbd3441/integration_tests_(release).html#fail1
May be worth add it to the checklist?
> 01099_parallel_distributed_insert_select https:/... | 2020-05-26T17:59:15 | 2020-05-26T17:59:15 | {} | MEMBER |
ClickHouse | ClickHouse | 634,200,477 | 11,200 | KochetovNicolai | All broken checks are broken on master too. | 2020-05-26T18:31:35 | 2020-05-26T18:31:35 | {} | MEMBER |
ClickHouse | ClickHouse | 634,340,177 | 8,639 | nikitamikhaylov | stateless: tests for new functionality and for removed `findClusterIndex` function
stateless(ubsan): errors with lz4 fixed in master
stress(memory): fixed in master
integration: `test_ttl_move`, `test_multiple_disks` fixed in master, `test_quorum_insert` contains new feature | 2020-05-26T23:51:14 | 2020-05-26T23:51:14 | {} | MEMBER |
ClickHouse | ClickHouse | 633,592,591 | 11,177 | excitoon | Тут случилось смешное, после рибейза исчезло поле, которое использовалось в коде :laughing: | 2020-05-25T14:14:47 | 2020-05-25T14:14:47 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 633,638,515 | 11,180 | alexey-milovidov | What is the reason? | 2020-05-25T16:12:24 | 2020-05-25T16:12:24 | {} | MEMBER |
ClickHouse | ClickHouse | 633,716,986 | 11,149 | alexey-milovidov | @filimonov Do you have any ideas about test_storage_kafka/test.py::test_kafka_flush_by_block_size test fail? | 2020-05-25T21:09:28 | 2020-05-25T21:09:28 | {} | MEMBER |
ClickHouse | ClickHouse | 633,720,603 | 11,181 | filimonov | @KochetovNicolai | 2020-05-25T21:25:54 | 2020-05-25T21:25:54 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 633,898,915 | 11,149 | filimonov | Hm. It looks like a real issue (minor), <s>strange that it didn't hit me locally</s>.
It can exit the poll loop prematurely by eof, w/o retries. One line fix.
Will add one more test for that and fix it here too.
UPD. I've messed the builds. Reproduces well. | 2020-05-26T08:56:04 | 2020-05-26T11:31:46 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 634,196,072 | 11,190 | alexey-milovidov | > I saw some flackiness for the following test too
Yes, we add *every* test failure to the list, no exceptions.
> Fixed in #10940 (#10829 does not include it)
Ok, I'll mark it.
> Looks like can be fixed the same way as in #11170 (was this a temporary fix/quirk or permanent?)
Thank you. No it isn't a temp... | 2020-05-26T18:23:31 | 2020-05-26T18:23:31 | {} | MEMBER |
ClickHouse | ClickHouse | 634,253,581 | 8,979 | den-crane | I could not reproduce this issue.
Maybe it needs more than 1 replica
```
18.14.19
create table TX(A Int64, B String) Engine = MergeTree order by tuple();
insert into TX values(1, '1');
alter table TX add column Y String;
alter table TX freeze partition tuple();
ls shadow/.../all_1_1_0
A.bin A.mrk B.bin... | 2020-05-26T20:12:18 | 2020-05-26T20:23:00 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 633,492,252 | 11,173 | filimonov | Fixed in 20.4. | 2020-05-25T10:03:03 | 2020-05-25T10:03:03 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 634,104,594 | 1,576 | filimonov | > Still facing same issue in 20.4.3.16
Please post your query | 2020-05-26T15:39:37 | 2020-05-26T15:39:45 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 634,199,167 | 11,190 | alexey-milovidov | > Is due to unbundled build does not have patch for Poco
Thank you, I will add it to blacklist for "unbundled".
It has been already added by me :) I will mark it. | 2020-05-26T18:29:09 | 2020-05-26T18:30:10 | {} | MEMBER |
ClickHouse | ClickHouse | 634,275,492 | 8,639 | nikitamikhaylov | So, i win | 2020-05-26T20:58:13 | 2020-05-26T20:58:13 | {} | MEMBER |
ClickHouse | ClickHouse | 634,433,111 | 11,135 | zhuyoucui0909 | Can you tell us what's the earliest time to release the fix? It is a stoping issue for us. Please tell us the schedule so we can arrange our job.
Best Regards. | 2020-05-27T05:13:49 | 2020-05-27T05:13:49 | {} | NONE |
ClickHouse | ClickHouse | 633,652,515 | 11,169 | nikitamikhaylov | Will be fixed in next patch release. | 2020-05-25T16:58:00 | 2020-05-25T16:58:00 | {
"heart": 1
} | MEMBER |
ClickHouse | ClickHouse | 633,708,228 | 9,545 | alexey-milovidov | @qoega This method exists in C++ ABI since C++11 to support thread_local variables, but Arcadia build has specifically broken version of libc++ that lack this method. | 2020-05-25T20:31:30 | 2020-05-25T20:31:30 | {} | MEMBER |
ClickHouse | ClickHouse | 634,069,188 | 11,184 | alexey-milovidov | @blinkov Are you going to fix by yourself or need some help? | 2020-05-26T14:42:21 | 2020-05-26T14:42:21 | {} | MEMBER |
ClickHouse | ClickHouse | 634,100,391 | 11,193 | filimonov | Background merges are executed by thread from so called 'background thread pool'.
Check the `background_pool_size` setting.
And usually, if you face 'Too many parts ...' it's NOT caused by background pool. Most common reason is bad schema (too granular partitioning, too aggressive compression, etc) creating too m... | 2020-05-26T15:32:28 | 2020-05-26T15:32:28 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 634,301,581 | 11,207 | zlobober | Is test_version_update_after_mutation related to my changes?
| 2020-05-26T21:56:00 | 2020-05-26T21:56:00 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 634,465,423 | 11,190 | azat | - test_version_update_after_mutation - https://clickhouse-test-reports.s3.yandex.net/11163/ec894e7cb3e0e3f9f7c880b00d44ee3b83e028ad/integration_tests_(asan).html#fail1
- `No route to host` -- test_mysql_protocol - https://clickhouse-test-reports.s3.yandex.net/11163/ec894e7cb3e0e3f9f7c880b00d44ee3b83e028ad/integration_... | 2020-05-27T06:51:34 | 2020-05-27T06:51:34 | {} | MEMBER |
ClickHouse | ClickHouse | 633,651,012 | 9,545 | And42 | @akuzm @qoega as far as I can tell from the log, the issue is likely to be related to the "libc++abi" version used in the build process. This is mentioned in the boost sources, for example: https://www.boost.org/doc/libs/1_73_0/boost/config/stdlib/libcpp.hpp. And according to [this](https://github.com/intel/linux-sgx/i... | 2020-05-25T16:52:44 | 2020-05-25T16:52:44 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 633,681,214 | 10,621 | azat | @alexey-milovidov can you take a look? | 2020-05-25T18:41:30 | 2020-05-25T18:41:30 | {} | MEMBER |
ClickHouse | ClickHouse | 633,701,870 | 11,180 | azat | Plus @akuzm was noted that this may/will break parallel tests execution | 2020-05-25T20:04:44 | 2020-05-25T20:04:44 | {} | MEMBER |
ClickHouse | ClickHouse | 633,986,160 | 11,189 | alexey-milovidov | Yes, it becomes slow: https://clickhouse-test-reports.s3.yandex.net/11189/babf968027a30a9192adc1a9286ac7c041f45b72/performance_comparison/report.html#fail1 | 2020-05-26T12:12:09 | 2020-05-26T12:12:09 | {} | MEMBER |
ClickHouse | ClickHouse | 634,180,705 | 11,121 | azat | >arenas.muzzy_decay_ms
Note that current bundled version of jemalloc should not create muzzy pages, since `madviace(MADV_FREE)` is disabled:
https://github.com/ClickHouse/ClickHouse/blob/3ff28d0a7a4d53c2be2604bceae76149e3c5bf39/contrib/jemalloc-cmake/include_linux_x86_64/jemalloc/internal/jemalloc_internal_defs.h... | 2020-05-26T17:56:24 | 2020-05-26T17:56:24 | {} | MEMBER |
ClickHouse | ClickHouse | 634,196,542 | 11,190 | alexey-milovidov | Also it's worth to increase connect timeout for all tests globally... | 2020-05-26T18:24:24 | 2020-05-26T18:24:24 | {} | MEMBER |
ClickHouse | ClickHouse | 634,211,591 | 10,241 | den-crane | yes, no parallelism during MV inserts (one thread [ 202101 ] for all MV)
20.5.1.3496
insert into testX select number from numbers(100000) settings max_block_size=1000000;
parallel_view_processing=0 **Elapsed: 0.010 sec.**
parallel_view_processing=1 **Elapsed: 0.010 sec.** / EXPECTED 0.005 sec.
... | 2020-05-26T18:53:11 | 2020-05-26T19:03:09 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 634,418,787 | 11,135 | winter7 | I wonder if this really is a bug, and when do you expect to fix it?
| 2020-05-27T04:20:28 | 2020-05-27T04:20:28 | {} | NONE |
ClickHouse | ClickHouse | 634,463,916 | 8,889 | jackpgao | This MySQL engine bug occurs at a very late version `19.17.9.60` | 2020-05-27T06:47:40 | 2020-05-27T06:47:40 | {} | NONE |
ClickHouse | ClickHouse | 650,773,146 | 5,157 | ucasfl | Do we support mm.dd.yy or mm-dd-yy etc in US style, or just support mm/dd/yyyy? @alexey-milovidov | 2020-06-28T14:48:54 | 2020-06-28T15:19:35 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 650,814,881 | 11,913 | alexey-milovidov | Performance tests:
> merge_tree_many_partitions_2
Sped up two times.
> codecs...
Various difference. | 2020-06-28T20:05:44 | 2020-06-28T20:05:44 | {} | MEMBER |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.