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 | 614,159,299 | 10,288 | alesapin | Fixed in https://github.com/ClickHouse/ClickHouse/pull/10025. Patch will be available in release 20.3.6. | 2020-04-15T17:00:06 | 2020-04-15T17:18:19 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 614,210,092 | 10,292 | alesapin | > Ok but don't understand how it was working before and why it became broken...
It became broken, because now we don't override UInt64 type for id https://github.com/clickhouse/ClickHouse/blob/master/src/Storages/StorageDictionary.cpp#L77 with type from attributes https://github.com/clickhouse/ClickHouse/blob/master... | 2020-04-15T18:39:39 | 2020-04-15T18:39:39 | {} | MEMBER |
ClickHouse | ClickHouse | 614,682,267 | 8,682 | alexey-milovidov | We don't have this error message in master:
`grep -r -i -F enought src`
At least the typo was fixed. | 2020-04-16T14:18:24 | 2020-04-16T14:18:24 | {} | MEMBER |
ClickHouse | ClickHouse | 614,685,512 | 10,300 | alexey-milovidov | You should update the version either to 20.1.9+ or 20.3.6+. | 2020-04-16T14:23:43 | 2020-04-16T14:23:43 | {} | MEMBER |
ClickHouse | ClickHouse | 614,735,341 | 10,307 | alexey-milovidov | PVS check Ok. | 2020-04-16T15:48:35 | 2020-04-16T15:48:35 | {} | MEMBER |
ClickHouse | ClickHouse | 614,760,648 | 5,636 | alexey-milovidov | `clickhouse-performance-test` no longer exists. | 2020-04-16T16:33:54 | 2020-04-16T16:33:54 | {} | MEMBER |
ClickHouse | ClickHouse | 614,885,594 | 2,655 | alexey-milovidov | The issue is still relevant. Reference: #9596 | 2020-04-16T20:45:25 | 2020-04-16T20:45:25 | {} | MEMBER |
ClickHouse | ClickHouse | 613,842,655 | 10,273 | filimonov | BTW it is not clear how to use OrNull with stored aggregation function states. | 2020-04-15T06:27:43 | 2020-04-15T06:27:43 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 613,917,326 | 8,639 | alesapin | It seems ready for release, a summary about failed tests:
1. `Polymorphic Parts Enabled` stateless tests -- not implemented in 20.1
2. `test_settings_constraints_distributed` integration test -- `CREATE USER` not implemented in 20.1
3. `test_server_initialization` integration test -- test fixed in master, here it is... | 2020-04-15T09:08:35 | 2020-04-15T09:08:35 | {} | MEMBER |
ClickHouse | ClickHouse | 614,107,370 | 10,241 | nvartolomei | Broken in #3796 if I'm not mistaken. | 2020-04-15T15:26:37 | 2020-04-15T15:26:37 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 614,662,166 | 4,756 | 4ertus2 | It's not backported to 20.3 cause of code conflicts and backport complexity. It should be fixed in 20.4. Could you check it on muster build?
If you have version that passes this test https://github.com/ClickHouse/ClickHouse/blob/master/tests/queries/0_stateless/01104_distributed_one_test.sql but does not pass your t... | 2020-04-16T13:44:53 | 2020-04-16T13:44:53 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 614,880,808 | 2,469 | alexey-milovidov | https://github.com/ClickHouse/ClickHouse/pull/3900 | 2020-04-16T20:35:20 | 2020-04-16T20:35:20 | {} | MEMBER |
ClickHouse | ClickHouse | 614,883,991 | 948 | alexey-milovidov | We have two approaches to make it possible:
1. There is support for multiple storage volumes in ClickHouse with customizeable storage policies to control data placement. A volume may be mounted to tmpfs. https://presentations.clickhouse.tech/meetup36/new_features/#12
2. There is a feature in development to suppor... | 2020-04-16T20:41:57 | 2020-04-16T20:43:46 | {} | MEMBER |
ClickHouse | ClickHouse | 613,950,195 | 10,252 | vvchistyakov | #6462 | 2020-04-15T10:15:33 | 2020-04-15T10:15:33 | {} | NONE |
ClickHouse | ClickHouse | 614,174,696 | 10,290 | alexey-milovidov | Look at `convertFieldToType` for the pecularities of implementation. | 2020-04-15T17:29:10 | 2020-04-15T17:29:10 | {} | MEMBER |
ClickHouse | ClickHouse | 614,186,367 | 9,853 | 4ertus2 | PR with fix attached. There's also a workaround in new versions - `set multiple_joins_rewriter_version = 2`. | 2020-04-15T17:51:52 | 2020-04-15T17:51:52 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 614,295,039 | 10,216 | alexey-milovidov | @akuzm It didn't. | 2020-04-15T21:42:44 | 2020-04-15T21:42:44 | {} | MEMBER |
ClickHouse | ClickHouse | 614,298,668 | 10,252 | vvchistyakov | Maybe, I'm wrong.
I was continue investigate this trouble.
I found interest thing. I have the partition _20191230_961638_1264729_10079_1279734_. And, there is ttl.txt file with content
`{"table":{"min":1585753056,"max":1586044799}}`
1585753056 = 2020-04-01T14:57:36
1586044799 = 2020-04-04T23:59:59
Does it mean CH... | 2020-04-15T21:52:14 | 2020-04-15T21:52:14 | {} | NONE |
ClickHouse | ClickHouse | 614,682,900 | 8,682 | alexey-milovidov | `DB::Exception: Cannot reserve 1.35 GiB, not enought space.`
It means that all remaining space was reserved for other merges and this is normal. | 2020-04-16T14:19:25 | 2020-04-16T14:19:25 | {} | MEMBER |
ClickHouse | ClickHouse | 614,852,479 | 5,183 | alexey-milovidov | I've checked on master and I see that the issue is already solved. Now it looks like this:
2020.04.16 22:31:17.903683 [ 4920 ] {} <Error> ServerErrorHandler: Code: 210, e.displayText() = DB::NetException: Connection reset by peer, **while reading from socket ([::ffff:127.0.0.1]:35372)**, Stack trace (when copying th... | 2020-04-16T19:32:02 | 2020-04-16T19:33:02 | {
"+1": 1,
"-1": 1
} | MEMBER |
ClickHouse | ClickHouse | 614,853,434 | 5,183 | alexey-milovidov | Two changes:
- socket address is printed;
- stack trace is provided. | 2020-04-16T19:33:58 | 2020-04-16T19:34:04 | {} | MEMBER |
ClickHouse | ClickHouse | 614,817,905 | 10,274 | filimonov | Does it also solve https://github.com/ClickHouse/ClickHouse/issues/5183 ? | 2020-04-16T18:20:53 | 2020-04-16T18:20:53 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 613,844,693 | 10,224 | filimonov | See also https://github.com/ClickHouse/ClickHouse/issues/4101#issuecomment-613791667 | 2020-04-15T06:33:50 | 2020-04-15T06:33:50 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 614,143,926 | 10,283 | alexey-milovidov | We have a function `checkStackSize` for this purpose.
It's strange that it's not used in parser. | 2020-04-15T16:30:56 | 2020-04-15T16:30:56 | {} | MEMBER |
ClickHouse | ClickHouse | 614,194,290 | 10,031 | th0th | Thanks for the comments! | 2020-04-15T18:07:57 | 2020-04-15T18:07:57 | {} | NONE |
ClickHouse | ClickHouse | 614,209,351 | 10,278 | maison2710 | The table is partitioned with `toDate(timestamp)` (timestamp column has Int32 type)
I found out that this issue happens for records having huge timestamp in order to make toDate(timestamp) = '2106-02-07'.
I tried to run `select toDate(very_huge_number)` and the result is always `2106-02-07`. Seems like `2106-02-07` i... | 2020-04-15T18:38:15 | 2020-04-15T18:45:50 | {} | NONE |
ClickHouse | ClickHouse | 614,311,353 | 7,701 | vvchistyakov | It looks like the bug for me too.
I have few tables with
```sql
PARTITION BY toMonday(timestamp)
TTL timestamp + toIntervalDay(90)
```
And, the old data is not deleted also =(
My version is `19.14.7.15` | 2020-04-15T22:28:03 | 2020-04-15T22:28:03 | {} | NONE |
ClickHouse | ClickHouse | 614,684,753 | 10,300 | alexey-milovidov | Fixed in https://github.com/ClickHouse/ClickHouse/pull/9142 | 2020-04-16T14:22:33 | 2020-04-16T14:22:33 | {} | MEMBER |
ClickHouse | ClickHouse | 614,817,735 | 8,986 | alexey-milovidov | @achimbab We are doing well, thank you!
This PR was not in high priority. But it's almost ready to merge. | 2020-04-16T18:20:33 | 2020-04-16T18:20:33 | {} | MEMBER |
ClickHouse | ClickHouse | 614,875,976 | 10,317 | alexey-milovidov | Side note: pipe character is not escaped in output. And the escaping schema may be not markdown compatible. This is considered normal. | 2020-04-16T20:24:44 | 2020-04-16T20:24:44 | {} | MEMBER |
ClickHouse | ClickHouse | 314,851,516 | 979 | inv2004 | I suppose that Time type with ms (or nano?) instead of UInt32 would help to track events more precise and would have some sugar to make work with time more accurate.
For better understanding, could you please show how to work with time in current implementation?
SELECT event WHERE date = '2017.02.02' AND time < '... | 2017-07-12T18:09:38 | 2020-04-16T20:36:53 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 614,889,316 | 332 | alexey-milovidov | @azat If we do GROUP BY injective function of sharding key, and the setting `skip_unavailable_shards` is active, we can set `distributed_group_by_no_merge` setting automatically.
If we will do that, it will solve the issue (almost). | 2020-04-16T20:52:55 | 2020-04-16T20:55:05 | {} | MEMBER |
ClickHouse | ClickHouse | 613,799,282 | 1,851 | alexey-milovidov | Stack trace:
```
2020.04.15 06:46:02.620312 [ 16691 ] {} <Error> ServerErrorHandler: Poco::Exception. Code: 1000, e.code() = 107, e.displayText() = Net Exception: Socket is not connected, Stack trace (when copying this message, always include the lines below):
0. /home/milovidov/work/ClickHouse/build/../contrib/... | 2020-04-15T03:46:37 | 2020-04-15T03:46:37 | {} | MEMBER |
ClickHouse | ClickHouse | 614,044,239 | 3,621 | alexey-milovidov | The fact that MySQL client silently using unix socket when the user specified `localhost` - it is an infamous usability issue that we have to be aware. | 2020-04-15T13:34:53 | 2020-04-15T13:34:53 | {} | MEMBER |
ClickHouse | ClickHouse | 614,142,151 | 10,283 | alexey-milovidov | Reproduced successfully! | 2020-04-15T16:27:38 | 2020-04-15T16:27:38 | {} | MEMBER |
ClickHouse | ClickHouse | 614,653,353 | 4,756 | hohaftf | As I understand this must be fixed in v 20.3.6.40. But case A still does not work.
query
```
select `default`.bug_distr1.id
from `default`.bug_distr1
inner join `default`.bug_distr2 on `default`.bug_distr1.id = `default`.bug_distr2.id
where `default`.bug_distr1.id > 0
order by `default`.bug_... | 2020-04-16T13:29:24 | 2020-04-16T13:36:11 | {} | NONE |
ClickHouse | ClickHouse | 614,845,524 | 10,274 | alexey-milovidov | No, it's a different issue. | 2020-04-16T19:16:47 | 2020-04-16T19:16:47 | {} | MEMBER |
ClickHouse | ClickHouse | 614,881,949 | 979 | alexey-milovidov | Now we have DateTime64 data type with support up to nanosecond precision.
We still don't have dedicated Time (without Date) type but it is "won't fix". | 2020-04-16T20:37:46 | 2020-04-16T20:37:58 | {} | MEMBER |
ClickHouse | ClickHouse | 613,802,358 | 10,274 | alexey-milovidov | Tested manually with https://github.com/ClickHouse/ClickHouse/blob/master/src/Client/tests/test_connect.cpp | 2020-04-15T03:59:06 | 2020-04-15T03:59:06 | {} | MEMBER |
ClickHouse | ClickHouse | 613,845,098 | 10,250 | maison2710 | @alexey-milovidov
I pulled your latest update.
However, I still got the problem that the build always tried to g++-8. It seems like `dpkg-architecture cannot determine CC system type then falling back to default (native compilation)`. Then it installed g++-8 and gcc-8 in some tmp location and started to use the... | 2020-04-15T06:34:57 | 2020-04-15T06:38:22 | {} | NONE |
ClickHouse | ClickHouse | 614,106,662 | 10,241 | nvartolomei | #7195 | 2020-04-15T15:25:31 | 2020-04-15T15:25:31 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 614,209,337 | 10,283 | alexey-milovidov | No, it was before. | 2020-04-15T18:38:13 | 2020-04-15T18:38:13 | {} | MEMBER |
ClickHouse | ClickHouse | 614,733,561 | 10,308 | alesapin | Maybe related https://github.com/ClickHouse/ClickHouse/issues/10309 | 2020-04-16T15:45:22 | 2020-04-16T15:45:22 | {} | MEMBER |
ClickHouse | ClickHouse | 614,787,309 | 10,309 | alesapin | Introduced here https://github.com/ClickHouse/ClickHouse/pull/9232 | 2020-04-16T17:24:51 | 2020-04-16T17:24:51 | {} | MEMBER |
ClickHouse | ClickHouse | 614,856,910 | 10,000 | akuzm | Made a simple benchmark locally, with `hits` table from Yandex.Metrica dataset:
```
echo "select sumMap([OS], [UserAgent]) from hits;" | build-gcc9-rel/programs/clickhouse-benchmark --cumulative --timelimit 60 -d0
```
The median query run time in seconds is as follows:
| Query | master | branch | with allocation... | 2020-04-16T19:41:37 | 2020-04-16T19:43:26 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 613,808,268 | 3,621 | alexey-milovidov | It's easy to make it configurable, see https://github.com/ClickHouse/ClickHouse/pull/983 | 2020-04-15T04:24:13 | 2020-04-15T04:24:38 | {} | MEMBER |
ClickHouse | ClickHouse | 614,563,448 | 10,301 | thyn | All builds are from
https://hub.docker.com/r/yandex/clickhouse-server/ | 2020-04-16T10:31:04 | 2020-04-16T10:31:12 | {} | NONE |
ClickHouse | ClickHouse | 614,760,573 | 5,635 | alexey-milovidov | `clickhouse-performance-test` no longer exists. | 2020-04-16T16:33:46 | 2020-04-16T16:33:46 | {} | MEMBER |
ClickHouse | ClickHouse | 614,812,565 | 10,302 | 4ertus2 | Backported to 20.3. 20.1 is not affected. | 2020-04-16T18:11:09 | 2020-04-16T18:11:09 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 614,877,633 | 10,002 | akuzm | Transitive is no good, really. Maybe `additive function` would do? https://en.wikipedia.org/wiki/Additive_map
I understand that we don't support all of them now, but in principle it's possible to support any such function. I don't see this mentioned anywhere -- the docs don't explain why this column type it is support... | 2020-04-16T20:28:19 | 2020-04-16T20:28:19 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 614,878,014 | 10,243 | drook | Whel building from a source tree with was initialized from git with **--recursive** flag and v20.3.5.21-stable tag checked out I get:
Cmake build args:
-DUNBUNDLED=1 \
-DUSE_STATIC_LIBRARIES=0 \
-DENABLE_EMBEDDED_COMPILER=1 \
-DNO_WERROR=1 \
-DCMAKE_SYSTEM_INCLUDE_PATH=/usr/include:/usr/local/include \
-DCITYH... | 2020-04-16T20:29:09 | 2020-04-16T20:29:09 | {} | NONE |
ClickHouse | ClickHouse | 613,797,937 | 2,509 | alexey-milovidov | Closed in favor of #10273. | 2020-04-15T03:41:02 | 2020-04-15T03:41:02 | {} | MEMBER |
ClickHouse | ClickHouse | 614,214,959 | 10,216 | Avogar | @alexey-milovidov
After a small research I've found that similar errors have already occured:
https://github.com/mymarilyn/clickhouse-driver/issues/121
https://stackoverflow.com/questions/57786801/testing-containers-and-clickhouse-driver-errorunexpected-eof-while-reading-byte
For this reason I can suggest that p... | 2020-04-15T18:47:35 | 2020-04-15T18:47:35 | {} | MEMBER |
ClickHouse | ClickHouse | 614,687,098 | 6,613 | lwolters | Hi Alexey,
What's the status of this feature? We noted that when running limited queries with joins, the 'rows_before_limit_at_least' outputs a correct value. However, whenever running 'simple' queries on a single table the 'rows_before_limit_at_least' outputs a wrong number (smaller than the 'exact' value).
It l... | 2020-04-16T14:26:18 | 2020-04-16T14:26:18 | {} | NONE |
ClickHouse | ClickHouse | 613,799,022 | 1,851 | alexey-milovidov | Reproducing in master (20.4).
And we have a test for that: https://github.com/ClickHouse/ClickHouse/blob/master/src/Client/tests/test_connect.cpp | 2020-04-15T03:45:40 | 2020-04-15T03:45:40 | {} | MEMBER |
ClickHouse | ClickHouse | 613,806,070 | 1,776 | alexey-milovidov | @skaurus The issue of unintentional summing of enum values is fixed long time ago.
The script is no longer available and, nevertheless, it's a different issue. | 2020-04-15T04:14:59 | 2020-04-15T04:14:59 | {} | MEMBER |
ClickHouse | ClickHouse | 613,904,299 | 1,839 | korservick | 2 years ago... :) Just remember that I catch this `+inf` from Node.js code. | 2020-04-15T08:43:20 | 2020-04-15T08:43:20 | {} | NONE |
ClickHouse | ClickHouse | 614,021,159 | 10,259 | qoega | Probably bug introduced in #10218 | 2020-04-15T12:53:30 | 2020-04-15T12:53:30 | {} | MEMBER |
ClickHouse | ClickHouse | 614,143,190 | 10,283 | alexey-milovidov | Stack overflow. | 2020-04-15T16:29:36 | 2020-04-15T16:29:36 | {} | MEMBER |
ClickHouse | ClickHouse | 614,328,715 | 10,252 | den-crane | it means that TTL should start to remove rows after 2020-04-01T14:57:36 | 2020-04-15T23:25:28 | 2020-04-15T23:25:28 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 614,714,469 | 10,002 | akuzm | Probably we should write it out explicitly that SimpleAggregationFunction works for transitive (not sure?) functions, where `fun([x1, ..., xn]) = fun([fun([x1,...,xm]), fun([xm+1, ..., xn])])`. Square brackets denote row sets and not arrays here. For such functions, we don't have to store the state, and it's enough to ... | 2020-04-16T15:11:52 | 2020-04-16T15:11:52 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 614,764,170 | 8,676 | alexey-milovidov | @zhang2014 I think that we need to make this feature work explicit.
There is also another planned feature named "replicated database engine" that will do something similar within database. | 2020-04-16T16:40:18 | 2020-04-16T16:40:18 | {} | MEMBER |
ClickHouse | ClickHouse | 613,804,662 | 10,184 | alexey-milovidov | Test has passed. | 2020-04-15T04:09:03 | 2020-04-15T04:09:03 | {} | MEMBER |
ClickHouse | ClickHouse | 613,859,691 | 10,263 | azat | >test_settings_constraints_distributed can be fixed with SYSTEM FLUSH DISTRIBUTED
Added | 2020-04-15T07:11:48 | 2020-04-15T07:11:48 | {} | MEMBER |
ClickHouse | ClickHouse | 613,930,400 | 9,637 | Akazz | Hey, @toby080508050805!
Please, have a look at #10073, it should have resolved your issue with false positives. | 2020-04-15T09:34:00 | 2020-04-15T09:34:00 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 614,042,086 | 10,275 | alexey-milovidov | We don't use Lunr anymore. | 2020-04-15T13:30:53 | 2020-04-15T13:30:53 | {} | MEMBER |
ClickHouse | ClickHouse | 614,297,355 | 10,294 | Schaefer-Thomas | not yet, send him an email (and a t-shirt?)! | 2020-04-15T21:48:37 | 2020-04-15T21:48:53 | {} | NONE |
ClickHouse | ClickHouse | 613,842,153 | 3,621 | filimonov | Not clear why closed? It is not implemented. | 2020-04-15T06:26:13 | 2020-04-15T06:26:13 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 613,841,277 | 10,263 | azat | >test_settings_constraints_distributed/test.py::test_insert_clamps_settings
>test_dictionaries_mysql/test.py::test_load_mysql_dictionaries
Fails in upstream/master too
`test_settings_constraints_distributed` can be fixed with `SYSTEM FLUSH DISTRIBUTED` | 2020-04-15T06:23:40 | 2020-04-15T06:26:35 | {} | MEMBER |
ClickHouse | ClickHouse | 614,172,964 | 10,289 | den-crane | https://en.wikipedia.org/wiki/Tz_database#Area
Area
Area is the name of a continent, an ocean, or "Etc". The continents and oceans currently used are Africa, America, Antarctica, Arctic, Asia, Atlantic, Australia, Europe, Indian, and Pacific.
The oceans are included since some islands are hard to connect to a ... | 2020-04-15T17:25:55 | 2020-04-15T17:25:55 | {
"heart": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 614,294,012 | 10,295 | alexey-milovidov | FYI clang-tidy has an option to prohibit default parameters at all. But it's too hard to enforce in our code base. Currently we have turned on an option to prohibit default parameters in virtual methods. | 2020-04-15T21:40:03 | 2020-04-15T21:40:03 | {} | MEMBER |
ClickHouse | ClickHouse | 614,506,550 | 8,968 | gubinjie | @filimonov @den-crane Hello
We use hdfs engine: table1, data volume 400 million data volume, more than 500 columns
Insert table1 data into ClickHouse's MergeTree table, there are too many parts abnormal problems, are there any parameters or other solutions that need to be changed?
Our version number: 20.1.4.14
Tha... | 2020-04-16T08:45:49 | 2020-04-16T08:45:49 | {} | NONE |
ClickHouse | ClickHouse | 614,513,451 | 7,701 | alex-zaitsev | @vvchistiakov TTLs has been significantly improved since 19.14. Please try with 19.16.16 or 20.1.x release.
Also look at merge tree settings related to TTLs:
select * from system.merge_tree_settings where name like '%ttl%'; | 2020-04-16T08:58:19 | 2020-04-16T08:58:19 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 614,877,548 | 10,311 | 4ertus2 | Useless without #9830 that is not backported to 20.1 | 2020-04-16T20:28:07 | 2020-04-16T20:28:07 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 613,908,317 | 10,264 | alesapin | Let's look at `SELECT * FROM system.replcas WHERE table = 'foo'`. Also information from zk nodes `/path_to_table/leader_election` and `/path_to_table/replicas/replica_name/is_active` will be useful. | 2020-04-15T08:51:25 | 2020-04-15T08:51:25 | {} | MEMBER |
ClickHouse | ClickHouse | 614,173,940 | 9,624 | alesapin | `test_settings_constraints_distributed/test.py::test_select_clamps_settings` -- `ALTER USER` not implemented in this branch.
`test_server_initialization/test.py::test_partially_dropped_tables` -- test fixed in master.
`test_dictionaries_mysql/test.py::test_load_mysql_dictionaries` -- fails in master, but looks suspic... | 2020-04-15T17:27:53 | 2020-04-15T17:27:53 | {} | MEMBER |
ClickHouse | ClickHouse | 614,224,900 | 10,216 | akuzm | > @alexey-milovidov
> After a small research I've found that similar errors have already occured:
> [mymarilyn/clickhouse-driver#121](https://github.com/mymarilyn/clickhouse-driver/issues/121)
> https://stackoverflow.com/questions/57786801/testing-containers-and-clickhouse-driver-errorunexpected-eof-while-reading-by... | 2020-04-15T19:06:58 | 2020-04-15T19:06:58 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 614,295,283 | 10,216 | alexey-milovidov | I'll merge this PR without addition to perf test... | 2020-04-15T21:43:21 | 2020-04-15T21:43:21 | {} | MEMBER |
ClickHouse | ClickHouse | 614,296,724 | 10,234 | alexey-milovidov | Some functional tests have to be updated. | 2020-04-15T21:47:03 | 2020-04-15T21:47:03 | {} | MEMBER |
ClickHouse | ClickHouse | 614,298,610 | 10,221 | alexey-milovidov | The test of cluster copier was failed due to timeout... looks completely unrelated. | 2020-04-15T21:52:02 | 2020-04-15T21:52:02 | {} | MEMBER |
ClickHouse | ClickHouse | 614,729,905 | 9,758 | abyss7 | Closed in favour of #10307 | 2020-04-16T15:38:59 | 2020-04-16T15:38:59 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 614,044,913 | 10,259 | alexey-milovidov | @qoega Please elaborate. | 2020-04-15T13:36:08 | 2020-04-15T13:36:08 | {} | MEMBER |
ClickHouse | ClickHouse | 614,047,731 | 10,272 | alexey-milovidov | Performance tests Ok. | 2020-04-15T13:41:34 | 2020-04-15T13:41:34 | {} | MEMBER |
ClickHouse | ClickHouse | 614,048,865 | 10,024 | alexey-milovidov | To add support for these options we have to understand their names. | 2020-04-15T13:43:41 | 2020-04-15T13:43:41 | {} | MEMBER |
ClickHouse | ClickHouse | 614,175,407 | 10,283 | alexey-milovidov | Looks like the bug was introduced in https://github.com/ClickHouse/ClickHouse/pull/8647 | 2020-04-15T17:30:27 | 2020-04-15T17:30:27 | {} | MEMBER |
ClickHouse | ClickHouse | 614,298,214 | 10,221 | alexey-milovidov | One integration test has failed, need to investigate... | 2020-04-15T21:50:58 | 2020-04-15T21:50:58 | {} | MEMBER |
ClickHouse | ClickHouse | 614,620,550 | 4,101 | abraithwaite | FWIW, for those coming to this issue, we also increased our insertion batch size to 150k events, and moved from inserting into a distributed table, to inserting directly into the shards. This relieves a lot of the load since inserting into a distributed table re-shards the data again across your number of shards given... | 2020-04-16T12:26:28 | 2020-04-16T12:26:28 | {
"+1": 1
} | NONE |
ClickHouse | ClickHouse | 614,139,887 | 10,284 | alexey-milovidov | What is the engine of the table that you are renaming?
And what is the version of clickhouse-server you run?
Also you can investigate the issue by turning on logs on client:
```
SET send_logs_level = 'trace'
```
before executing RENAME query.
Please attach these logs here. | 2020-04-15T16:23:27 | 2020-04-15T16:24:01 | {} | MEMBER |
ClickHouse | ClickHouse | 614,683,271 | 8,682 | alexey-milovidov | This is not an error. | 2020-04-16T14:20:05 | 2020-04-16T14:20:05 | {} | MEMBER |
ClickHouse | ClickHouse | 614,744,590 | 10,305 | den-crane | But AVG needs state?
How are you going calculate AVG from two other AVG values? | 2020-04-16T16:04:51 | 2020-04-16T16:04:51 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 614,761,024 | 4,197 | alexey-milovidov | Fixed by @amosbird long time ago. | 2020-04-16T16:34:39 | 2020-04-16T16:34:39 | {} | MEMBER |
ClickHouse | ClickHouse | 614,865,022 | 10,002 | azat | > Probably we should write it out explicitly that SimpleAggregationFunction works for transitive (not sure?) functions, where fun([x1, ..., xn]) = fun([fun([x1,...,xm]), fun([xm+1, ..., xn])]).
Well, `SimpleAggregateFunction` does not support all the functions that can be supported actually (see #10000), and beside... | 2020-04-16T19:59:54 | 2020-04-16T19:59:54 | {} | MEMBER |
ClickHouse | ClickHouse | 613,880,635 | 10,252 | vvchistyakov | Sorry @den-crane. It still is not clear for me. Is it ok when partitions have these same statuses on all replicas?
Could you suggest a direction where I should search problem with TTL? I've checked error log for few days and haven't found any interest errors. | 2020-04-15T07:57:21 | 2020-04-15T07:57:21 | {} | NONE |
ClickHouse | ClickHouse | 613,934,075 | 10,003 | alesapin | Continuation https://github.com/ClickHouse/ClickHouse/issues/10277 | 2020-04-15T09:41:06 | 2020-04-15T09:41:06 | {} | MEMBER |
ClickHouse | ClickHouse | 613,891,955 | 10,216 | Avogar | It's really strange... I've successfully ran this test locally (corresponding to https://github.com/ClickHouse/ClickHouse/blob/master/docker/test/performance-comparison/performance_comparison.md#run-a-single-test-on-the-already-configured-servers).
There is a parse_engine_file-err.log:
```
Traceback (most recent c... | 2020-04-15T08:18:53 | 2020-04-15T13:57:54 | {} | MEMBER |
ClickHouse | ClickHouse | 614,274,606 | 10,294 | alexey-milovidov | Ok, but the tool does not look available: https://github.com/sqlancer/sqlancer | 2020-04-15T20:53:46 | 2020-04-15T20:53:46 | {} | MEMBER |
ClickHouse | ClickHouse | 614,472,128 | 8,682 | blinkov | @jonhsonjs, is this question still relevant? | 2020-04-16T07:40:41 | 2020-04-16T07:40:41 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 614,496,210 | 10,252 | vvchistyakov | Oh...I see
Look's like trouble with TTL for me is #7701
So, as I could understand using `is_frozen` flag on all replicas at the same time for partitions without creating snapshots is normal behaviour for CH.
I guess I've finished my investigation here.
Thanks @den-crane | 2020-04-16T08:27:06 | 2020-04-16T08:27:06 | {} | NONE |
ClickHouse | ClickHouse | 626,916,302 | 10,486 | azat | Looks like the `tmp` was a file (this is the only reason for such error in `Poco::File::createDirectory`) | 2020-05-11T19:41:03 | 2020-05-11T19:41:03 | {} | MEMBER |
ClickHouse | ClickHouse | 626,940,512 | 7,984 | alexey-milovidov | SAP ASE.
Looks like we will merge without a test. | 2020-05-11T20:24:42 | 2020-05-11T20:24:42 | {} | MEMBER |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.