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 | 700,376,417 | 15,265 | liangli1213 | @BohuTANG , got below error message, can you reproduce it?
```
backup@bdcmysqldev2:(none)>show grants;
+------------------------------------------------+
| Grants for backup@% |
+------------------------------------------------+
| GRANT REPLICATION SLAVE ON *.* TO 'backup'@'%' |
+---... | 2020-09-29T01:50:12 | 2020-09-29T01:50:12 | {} | NONE |
ClickHouse | ClickHouse | 700,488,254 | 4,821 | bharatnc | > @nvartolomei I think this issue can be closed. This was implemented via #14252 and the `system.query_log` and `system.query_thread_log` tables should have `query_start_time_microseconds` fields containing the `microseconds` resolution.
Just to note that relevant changes for this feature is addressed by both #14252... | 2020-09-29T06:48:53 | 2020-09-29T06:48:53 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 700,490,904 | 15,405 | alesapin | Build failure == OOM | 2020-09-29T06:55:10 | 2020-09-29T06:55:10 | {} | MEMBER |
ClickHouse | ClickHouse | 700,491,985 | 15,428 | amosbird | > but it may require to change the IStorage interface
Yes. It's reasonable to have the current query context passed into this method. Then you have the related settings available. | 2020-09-29T06:57:44 | 2020-09-29T06:57:44 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 700,623,927 | 15,433 | alesapin | The only stress test with tsan is important | 2020-09-29T10:49:35 | 2020-09-29T10:49:35 | {} | MEMBER |
ClickHouse | ClickHouse | 700,643,807 | 15,397 | nikitamikhaylov | This feature was only in master. | 2020-09-29T11:33:55 | 2020-09-29T11:33:55 | {} | MEMBER |
ClickHouse | ClickHouse | 700,173,057 | 14,265 | ilejn | > Reverted because it breaks integration tests: `Integration tests (asan) — fail: 0, passed: 0, error: 0 `
@tavplubix , any suggestions or ideas what went wrong? | 2020-09-28T17:23:48 | 2020-09-28T17:23:48 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 700,400,849 | 15,355 | zhang2014 | Implemented in #14535
Maybe close this issue. | 2020-09-29T03:16:56 | 2020-09-29T03:16:56 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 700,428,910 | 15,428 | haoch | @amosbird I agree, but it may require to change the IStorage interface, do you think it's ok? | 2020-09-29T05:06:16 | 2020-09-29T05:06:16 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 700,643,527 | 15,399 | nikitamikhaylov | This feature was only in master. | 2020-09-29T11:33:16 | 2020-09-29T11:33:16 | {} | MEMBER |
ClickHouse | ClickHouse | 700,690,145 | 15,427 | amosbird | In general it requires processing over encoded columns, and in near future there might be a proposal named `Column Type` which covers related optimizations. | 2020-09-29T13:07:45 | 2020-09-29T13:07:45 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 700,710,185 | 15,431 | 744570676 | > `ColumnUInt64 *col = ColumnUInt64::create().detach();` if you realy want.
ok, thank you~ :) | 2020-09-29T13:39:33 | 2020-09-29T13:39:33 | {} | NONE |
ClickHouse | ClickHouse | 700,850,254 | 11,755 | 4ertus2 | Sample to reproduce
```
CREATE TABLE cat_hist (categoryId UUID, categoryName String) ENGINE Memory;
CREATE TABLE prod_hist (categoryId UUID, productId UUID) ENGINE = MergeTree ORDER BY productId;
CREATE TABLE products_l AS prod_hist ENGINE = Distributed(test_shard_localhost, currentDatabase(), prod_hist);
CREATE... | 2020-09-29T17:03:37 | 2020-09-29T17:03:37 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 700,155,981 | 15,376 | akuzm | Can we have a test that covers this variant of grammar? | 2020-09-28T16:52:33 | 2020-09-28T16:52:33 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 700,199,529 | 15,421 | keliss | Well, I jumped the gun here, just got the same exception with the same out-of-nowhere limit of 56.15GB :( | 2020-09-28T18:15:44 | 2020-09-28T18:15:44 | {} | NONE |
ClickHouse | ClickHouse | 700,215,181 | 15,421 | keliss | Is there a way to see if this got applied? I don't see such a setting in the system.settings table and in preprocessed/ everything looks fine. Should I restart the server for this setting to take effect? | 2020-09-28T18:47:29 | 2020-09-28T19:00:42 | {} | NONE |
ClickHouse | ClickHouse | 700,390,335 | 15,369 | BernieJiangDL | > > Need any more info?
>
> Yes. Full row from `system.mutations` please.
>
> Also - you need to update at least to the latest 20.3 release (20.3.19 currently).
Yes， our ch version is 20.3.8, and the full row of this table I pasted on my last comment. | 2020-09-29T02:36:28 | 2020-09-29T02:36:28 | {} | NONE |
ClickHouse | ClickHouse | 700,483,664 | 4,821 | nvartolomei | @bharatnc thank you! | 2020-09-29T06:37:53 | 2020-09-29T06:38:08 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 700,545,701 | 14,265 | tavplubix | See the logs: https://clickhouse-test-reports.s3.yandex.net/14265/e2ee97202bf7b81cae3ab05458b6c4d8609c2617/integration_tests_(asan)/integration_run.1.txt.out.log
```
docker: invalid reference format: repository name must be lowercase.
...
subprocess.CalledProcessError: Command 'docker run --net=host --rm --name ... | 2020-09-29T08:43:30 | 2020-09-29T08:43:30 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 700,857,654 | 15,177 | azat | >You are saying that ThreadFromGlobalPool::join() may exit before captured lambda was destroyed, so any resource which was captured by labmbda may do something in it's destructor. So, actually, DB::ThreadGroupStatus::~ThreadGroupStatus() could be called later then ThreadFromGlobalPool::join().
Exactly
>I suppose ... | 2020-09-29T17:16:10 | 2020-09-29T17:16:10 | {} | MEMBER |
ClickHouse | ClickHouse | 700,176,095 | 14,936 | KochetovNicolai | ```
EXPLAIN
SELECT sum(number)
FROM remote('127.0.0.{1,2}', numbers(10))
FORMAT TSV
Expression (Projection)
Expression (Before ORDER BY and SELECT)
MergingAggregated
SettingQuotaAndLimits (Set limits and quota after reading from storage)
Union
Converting (Convert block structur... | 2020-09-28T17:29:35 | 2020-09-28T17:29:35 | {} | MEMBER |
ClickHouse | ClickHouse | 700,388,400 | 15,369 | BernieJiangDL | > @BernieJiangDL What about other system.mutations columns? **latest_fail_reason, latest_fail_time, parts_to_do** for example?
The full row of this table is shown below which seems desc
│ test│ scene_cnt_table │ 0000038280 │DELETE WHERE (minute = 202009271705) AND (hour = 2020092717) │ 2020-09-27 17:12:43 │ ['20... | 2020-09-29T02:29:10 | 2020-09-29T02:29:10 | {} | NONE |
ClickHouse | ClickHouse | 700,756,831 | 15,431 | alexey-milovidov | @Sasasu @744570676 I'm afraid that's not the best answer.
Because it will release the ownership of the object.
Just write `col.get()` to obtain non-owning pointer whille keeping the ownership by MutablePtr object. | 2020-09-29T14:47:32 | 2020-09-29T14:48:13 | {} | MEMBER |
ClickHouse | ClickHouse | 700,780,637 | 15,127 | Akazz | Hey there, @ucasFL!
Currently `StorageMemory` uses the following mechanisms for resolving possible conflicts between concurrent operations:
- SELECTs vs DROP/TRUNCATEs are resolved via table-level locks (`lockForShare()` vs `lockExclusively()`)
- SELECTs vs INSERTs and INSERTs vs DROP/TRUNCATEs are resolved via ... | 2020-09-29T15:23:13 | 2020-09-29T15:23:13 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 699,920,041 | 15,356 | filimonov | ~~I think it's caused by #14549~~
@alesapin build scripts / docker images versioning don't work yet for 20.3? | 2020-09-28T10:22:40 | 2020-09-28T16:31:45 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 700,147,720 | 15,356 | filimonov | Binary size changed:
```
# 20.3.17
root@0bc4d3888cc4:/# ls -la /usr/bin/clickhouse
-rwxr-xr-x 1 root root 284583976 Aug 14 17:58 /usr/bin/clickhouse
# 20.3.18
root@6315cd265f86:/# ls -la /usr/bin/clickhouse
-rwxr-xr-x 1 root root 246814792 Sep 8 03:20 /usr/bin/clickhouse
```
Diff in build options
 again and still don't understand why `ThreadPool` in `PipelineExecutor` fixes something.
What I see from report:
1. Logger was deleted when destructor for global `Logger` object was called.
```
2020-09-22 00:48:... | 2020-09-29T10:01:16 | 2020-09-29T10:01:16 | {} | MEMBER |
ClickHouse | ClickHouse | 700,247,358 | 15,414 | filimonov | (kinda) related https://github.com/ClickHouse/ClickHouse/pull/9900 (similar nature of the issue) | 2020-09-28T19:53:49 | 2020-09-28T19:54:19 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 700,294,329 | 15,315 | den-crane | > Apparently MATERIALIZED VIEW don't supporting SELECT with aggregate functions for the multiple rows.
MV supports SELECT with aggregate functions, but aggregations from different inserts should be done by a table engine.
In your case you have used `ENGINE = MergeTree() PARTITION BY (date) ORDER BY (date)` and thi... | 2020-09-28T21:36:31 | 2020-09-28T21:36:31 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 700,368,580 | 15,118 | chang0391 | @zhang2014
How is the bug handling going on ? | 2020-09-29T01:23:02 | 2020-09-29T01:23:02 | {} | NONE |
ClickHouse | ClickHouse | 700,391,691 | 15,430 | den-crane | `OPTIMIZE TABLE prod_stats.PriceChanges2` initiates an unplanned merge. It's not designed to remove duplicates.
For example your table has 100 partitions and each partition has 100 parts. OPTIMIZE TABLE can just merge 2 of the parts inside one partitions and leave 10000 unmerged parts. Furthermore it can do nothing b... | 2020-09-29T02:41:52 | 2020-09-29T02:41:52 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 700,436,371 | 15,369 | filimonov | Please update to 20.3.19, let us know if issue will remain | 2020-09-29T05:31:20 | 2020-09-29T05:31:20 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 700,540,888 | 15,369 | BernieJiangDL | > Please update to 20.3.19, let us know if issue will remain
Got it! | 2020-09-29T08:35:09 | 2020-09-29T08:35:09 | {} | NONE |
ClickHouse | ClickHouse | 700,748,157 | 14,939 | alexey-milovidov | @nikitamikhaylov But the original intention of the PR https://github.com/ClickHouse/ClickHouse/pull/14890/files was to add backquotes. | 2020-09-29T14:34:56 | 2020-09-29T14:34:56 | {} | MEMBER |
ClickHouse | ClickHouse | 700,816,075 | 11,755 | 4ertus2 | It could be related to #15242 | 2020-09-29T16:19:31 | 2020-09-29T16:28:48 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 700,931,132 | 15,007 | azat | If upstream/master will not go ahead then this maybe enough | 2020-09-29T19:24:37 | 2020-09-29T19:24:37 | {} | MEMBER |
ClickHouse | ClickHouse | 700,962,912 | 14,731 | azat | >No news so far.
Then how about split out 56187e694fa47eac710a1f616c782e5a053ced0c ? This will reduce number of conflicts later | 2020-09-29T20:18:09 | 2020-09-29T20:18:09 | {} | MEMBER |
ClickHouse | ClickHouse | 700,207,172 | 15,421 | keliss | Thank you. | 2020-09-28T18:30:42 | 2020-09-28T18:30:42 | {} | NONE |
ClickHouse | ClickHouse | 700,635,267 | 15,420 | OmarBazaraa | I cannot consistently reproduce the issue, but it only happens to me after a pure installation of ClickHouse (on a completely new VM).
When I drop the table and try to create it again, the creation completes instantaneously.
Here is the output when I set `send_logs_level = 'trace'`:
```
CREATE TABLE IF NOT EXISTS... | 2020-09-29T11:14:45 | 2020-09-29T11:14:45 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 701,031,439 | 15,312 | alexey-milovidov | Stress test (thread) - known issue. | 2020-09-29T22:46:06 | 2020-09-29T22:46:06 | {} | MEMBER |
ClickHouse | ClickHouse | 701,095,098 | 15,074 | alexey-milovidov | > test_rename_column
This test was broken and is already fixed in master | 2020-09-30T00:49:12 | 2020-09-30T00:49:12 | {} | MEMBER |
ClickHouse | ClickHouse | 700,203,955 | 15,421 | keliss | I see. So the real default value for max_server_memory_usage_to_ram_ratio is 0.9, right? | 2020-09-28T18:24:18 | 2020-09-28T18:28:31 | {} | NONE |
ClickHouse | ClickHouse | 700,207,021 | 15,421 | den-crane | >So the real default value for max_server_memory_usage_to_ram_ratio is 0.9, right?
Right. You can set max_server_memory_usage_to_ram_ratio = 100 for example (if you need unlimited).
I've made a PR to fix the doc. | 2020-09-28T18:30:24 | 2020-09-28T18:30:47 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 700,235,178 | 15,292 | den-crane | I see you got some help in Telegram.
@nikk0 Do you need a further assistance? Or it's solved? | 2020-09-28T19:28:24 | 2020-09-28T19:28:24 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 700,368,012 | 15,179 | chang0391 | @zhang2014
How is the bug handling going on ? | 2020-09-29T01:21:06 | 2020-09-29T01:21:06 | {} | NONE |
ClickHouse | ClickHouse | 700,633,241 | 13,617 | akuzm | https://github.com/ClickHouse/ClickHouse/pull/15239 | 2020-09-29T11:10:17 | 2020-09-29T11:10:17 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 701,110,662 | 14,443 | alexey-milovidov | @filimonov I'm Ok with this change if you will check it manually. | 2020-09-30T01:49:34 | 2020-09-30T01:49:34 | {} | MEMBER |
ClickHouse | ClickHouse | 700,116,653 | 15,348 | tavplubix | Functional stateless tests (thread) - 00652_replicated_mutations_zookeeper - flacky | 2020-09-28T15:46:09 | 2020-09-28T15:46:09 | {} | MEMBER |
ClickHouse | ClickHouse | 700,162,308 | 15,363 | akuzm | A line in the docs about this would be helpful.
Failures in stress tests are known and not related. | 2020-09-28T17:03:23 | 2020-09-28T17:03:23 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 700,186,931 | 15,420 | den-crane | This query should instant. Please provide CH server logs around this `CREATE TABLE default.daemon_heartbeat` or full CH logs /var/log/clickhouse-server/clickhouse-server.log
If it reproducible please show an output from CH client for
```
set send_logs_level = 'trace';
CREATE TABLE IF NOT EXISTS default.daemon_h... | 2020-09-28T17:51:00 | 2020-09-28T17:51:00 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 700,249,745 | 15,369 | filimonov | > Need any more info?
Yes. Full row from `system.mutations` please.
Also - you need to update at least to the latest 20.3 release (20.3.19 currently). | 2020-09-28T19:59:00 | 2020-09-28T19:59:00 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 701,020,754 | 15,426 | alexey-milovidov | AST Fuzzer: #13281 | 2020-09-29T22:14:02 | 2020-09-29T22:14:02 | {} | MEMBER |
ClickHouse | ClickHouse | 700,200,777 | 15,421 | den-crane | >The documentation says that if this setting is not set, then there's no memory limit.
documentation is wrong
https://clickhouse.tech/docs/en/operations/server-configuration-parameters/settings/#max_server_memory_usage
**The default max_server_memory_usage value is calculated as memory_amount * max_server_memory... | 2020-09-28T18:18:05 | 2020-09-28T18:22:59 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 700,212,526 | 15,421 | keliss | Looks like I'm still doing something wrong. I've set <max_server_memory_usage_to_ram_ratio>0.95</max_server_memory_usage_to_ram_ratio> in my config.xml, and it doesn't work, I'm still getting the same exception with the same limit. | 2020-09-28T18:41:56 | 2020-09-28T18:42:59 | {} | NONE |
ClickHouse | ClickHouse | 700,233,858 | 15,173 | den-crane | @keliss Can you reproduce a CH hanging or the error "Possible deadlock avoided" with 20.7+ ?
| 2020-09-28T19:25:40 | 2020-09-28T19:25:40 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 700,358,606 | 4,821 | bharatnc | @nvartolomei I think this issue can be closed. This was implemented via #14252 and the `system.query_log` and `system.query_thread_log` tables should have `query_start_time_microseconds` fields containing the `microseconds` resolution. | 2020-09-29T00:47:14 | 2020-09-29T00:47:14 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 700,430,403 | 15,428 | haoch | In fact, we got another similar exceptions while executing `select total_rows, total_bytes from system.tables` caused by the same problem. If we could change `totalRows` interface, it will be better and we could resolve the bug in [`StorageSystemTables`](https://github.com/ClickHouse/ClickHouse/blob/master/src/Storage... | 2020-09-29T05:11:50 | 2020-09-29T05:23:43 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 700,583,067 | 15,392 | tavplubix | Integration tests (release) - test_dictionaries_dependency_xml/test.py::test_get_data - flacky | 2020-09-29T09:26:58 | 2020-09-29T09:26:58 | {} | MEMBER |
ClickHouse | ClickHouse | 700,634,346 | 15,407 | akuzm | `test_replication_after_partition` failure not related, reported https://github.com/ClickHouse/ClickHouse/issues/14414 | 2020-09-29T11:12:44 | 2020-09-29T11:12:44 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 700,634,964 | 14,939 | nikitamikhaylov | I've removed backquotes from primary key column names. | 2020-09-29T11:14:06 | 2020-09-29T11:14:06 | {} | MEMBER |
ClickHouse | ClickHouse | 700,662,171 | 15,433 | alesapin | Looks like it help. | 2020-09-29T12:14:10 | 2020-09-29T12:14:10 | {} | MEMBER |
ClickHouse | ClickHouse | 700,820,068 | 15,420 | den-crane | ```
2020.09.29 10:53:46.578344 [ 2661 ] {4b420014-ece6-4f4e-9cce-1452cb4d1404} <Debug> executeQuery: (from [::ffff:127.0.0.1]:39234) CREATE TABLE IF NOT EXISTS default.daemon_heartbeat ( date Date DEFAULT toDate(time), time DateTime, daemon String, server String, port Int32, build_version String, build_commit String, ... | 2020-09-29T16:26:26 | 2020-09-29T16:26:38 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 700,836,709 | 14,499 | nvartolomei | I need some clarifications on the comments I left previously in order to continue/finish this. | 2020-09-29T16:40:26 | 2020-09-29T16:40:26 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 701,103,749 | 9,071 | sszc | I also encountered this problem, the version is 20.6.6.7 | 2020-09-30T01:22:13 | 2020-09-30T01:22:13 | {} | NONE |
ClickHouse | ClickHouse | 700,392,254 | 15,378 | amosbird | Weird JIT crash https://clickhouse-test-reports.s3.yandex.net/15378/8eb81752583e50e27bfdf66b2ab258f17b388d67/fuzzer/fuzzer.log | 2020-09-29T02:44:01 | 2020-09-29T02:44:01 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 700,643,650 | 15,400 | nikitamikhaylov | This feature was only in master. | 2020-09-29T11:33:33 | 2020-09-29T11:33:33 | {} | MEMBER |
ClickHouse | ClickHouse | 700,643,927 | 15,398 | nikitamikhaylov | This feature was only in master. | 2020-09-29T11:34:10 | 2020-09-29T11:34:10 | {} | MEMBER |
ClickHouse | ClickHouse | 701,014,012 | 15,455 | alexey-milovidov | It's about `avg_value_size_hint` in `DataTypeString::deserializeBinaryBulk`. | 2020-09-29T21:58:28 | 2020-09-29T21:58:28 | {} | MEMBER |
ClickHouse | ClickHouse | 701,106,872 | 14,971 | alexey-milovidov | Pipeline stuck - @KochetovNicolai | 2020-09-30T01:33:59 | 2020-09-30T01:33:59 | {} | MEMBER |
ClickHouse | ClickHouse | 701,110,299 | 14,693 | alexey-milovidov | https://zookeeper.apache.org/doc/r3.6.1/zookeeperProgrammers.html#ch_zkWatches
> With regard to watches, ZooKeeper maintains these guarantees:
> - Watches are ordered with respect to other events, other watches, and asynchronous replies.
Ok. | 2020-09-30T01:47:57 | 2020-09-30T01:48:09 | {} | MEMBER |
ClickHouse | ClickHouse | 700,210,110 | 15,371 | den-crane | related https://github.com/ClickHouse/ClickHouse/issues/14201 | 2020-09-28T18:36:52 | 2020-09-28T18:36:52 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 700,236,707 | 15,369 | den-crane | @BernieJiangDL What about other system.mutations columns? **latest_fail_reason, latest_fail_time, parts_to_do** for example? | 2020-09-28T19:31:36 | 2020-09-28T19:31:36 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 700,670,684 | 14,831 | bharatnc | Also relevant to: #14252 | 2020-09-29T12:31:27 | 2020-09-29T12:31:27 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 700,767,547 | 15,066 | den-crane | Did you miss data for this particular time 20200920214000, or it can be reproduced with any time period ? | 2020-09-29T15:03:14 | 2020-09-29T15:03:14 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 700,878,780 | 15,418 | 4ertus2 | Nothing but perf test is changed in the last commit. Other tests are OK. | 2020-09-29T17:52:25 | 2020-09-29T17:52:25 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 700,145,941 | 11,512 | gopala000 | Thanks @alexey-milovidov. Do you know approximate date when the docker image for 20.10 will be available? | 2020-09-28T16:33:35 | 2020-09-28T16:33:35 | {} | NONE |
ClickHouse | ClickHouse | 700,183,288 | 15,421 | den-crane | **max_memory_usage_for_all_queries** is deprecated.
https://github.com/ClickHouse/ClickHouse/blob/master/CHANGELOG.md
ClickHouse release v20.4.2.9, 2020-05-12
_Added a setting max_server_memory_usage to limit total memory usage of the server. The metric MemoryTracking is now calculated without a drift. The setti... | 2020-09-28T17:43:46 | 2020-09-28T17:43:46 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 700,212,298 | 15,383 | den-crane | > @filimonov
> I/O is falling during insertion
> 
Try new CH and --max_insert_threads =4 The maximum number of threads to execute the INSERT SELECT query. Values 0 or 1 means that INSERT SELECT is not run... | 2020-09-28T18:41:30 | 2020-09-28T18:41:30 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 700,381,992 | 15,428 | amosbird | It's better to provide a variant `totalRows()` without sequential consistency, and imitate the behavior of the `read()` method https://github.com/ClickHouse/ClickHouse/blob/master/src/Storages/StorageReplicatedMergeTree.cpp#L3529 | 2020-09-29T02:04:39 | 2020-09-29T02:04:39 | {} | COLLABORATOR |
ClickHouse | ClickHouse | 700,631,368 | 15,324 | Slach | really nice feature added to docs ;) thanks | 2020-09-29T11:06:18 | 2020-09-29T11:06:18 | {
"hooray": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 700,195,237 | 15,421 | keliss | Thanks for the quick answer, looks like it works with max_server_memory_usage. But why? The documentation says that if this setting is not set, then there's no memory limit. | 2020-09-28T18:07:10 | 2020-09-28T18:07:10 | {} | NONE |
ClickHouse | ClickHouse | 700,196,835 | 15,422 | KochetovNicolai | ```
EXPLAIN header = 1
SELECT 1 AS x
FORMAT TSV
Expression (Projection)
Header: x UInt8
Expression (Before ORDER BY and SELECT)
Header: _dummy UInt8
1 UInt8
ReadFromStorage (Read from SystemOne)
Header: dummy UInt8
``` | 2020-09-28T18:10:23 | 2020-09-28T18:10:23 | {} | MEMBER |
ClickHouse | ClickHouse | 700,394,904 | 15,369 | den-crane | It seems it need to process 3208 more parts.
Check several times `select parts_to_do from system.mutations where not is_done`. Is parts_to_do changing ? | 2020-09-29T02:54:08 | 2020-09-29T02:54:08 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 700,404,328 | 15,369 | BernieJiangDL | > It seems it need to process 3208 more parts.
>
> Check several times `select parts_to_do from system.mutations where not is_done`. Is parts_to_do changing ?
Nope, parts_to_do remains the same number 3208 for 2 days.
If clickhouse processes this mutation correctly, I guess the action will be logged. But there... | 2020-09-29T03:30:00 | 2020-09-29T03:38:06 | {} | NONE |
ClickHouse | ClickHouse | 700,493,438 | 11,568 | alexey-milovidov | @bookeezhou Why close? Is it solved? | 2020-09-29T07:01:08 | 2020-09-29T07:01:08 | {} | MEMBER |
ClickHouse | ClickHouse | 700,752,367 | 14,939 | nikitamikhaylov | > @nikitamikhaylov But the original intention of the PR https://github.com/ClickHouse/ClickHouse/pull/14890/files was to add backquotes.
Lol, I didn't noticed that. We investigate the problem more. | 2020-09-29T14:41:04 | 2020-09-29T14:41:04 | {} | MEMBER |
ClickHouse | ClickHouse | 701,067,105 | 15,285 | alexey-milovidov | > Stress test (thread) — Sanitizer assert (in stderr.log)
Known bug in OpenSSL: https://github.com/openssl/openssl/issues/11974 | 2020-09-30T00:28:23 | 2020-09-30T00:28:23 | {} | MEMBER |
ClickHouse | ClickHouse | 700,197,464 | 15,395 | den-crane | it seems this bug existed for very long time.
```
18.14.19
┌─cnt─┬─name──┬───────────percent─┐
│ 79 │ name1 │ 33.33333333333333 │
│ 62 │ name2 │ 33.33333333333333 │
│ 44 │ name3 │ 33.33333333333333 │
└─────┴───────┴───────────────────┘
```
WA: different lambda var. name. ` t ->` replaced with ` x -> `
... | 2020-09-28T18:11:44 | 2020-09-28T18:12:01 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 700,230,413 | 15,421 | den-crane | Restart is needed.
You can see applied settings in /var/log/clickhouse-server/clickhouse-server.log right after start
```
2020.09.28 19:15:47.006720 [ 2595 ] {} <Information> : Starting ClickHouse 20.10.1.4751 with revision 54441, build id: 08039EA9DD6F26B0, PID 2595
....
2020.09.28 19:15:47.021840 [ 2595 ] {} <... | 2020-09-28T19:18:13 | 2020-09-28T19:18:13 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 700,606,461 | 15,177 | KochetovNicolai | @azat
Also I, probably, get your point.
You are saying that `ThreadFromGlobalPool::join()` may exit **before** captured lambda was destroyed, so any resource which was captured by labmbda may do something in it's destructor. So, actually, `DB::ThreadGroupStatus::~ThreadGroupStatus()` could be called later then `Thre... | 2020-09-29T10:11:57 | 2020-09-29T10:11:57 | {} | MEMBER |
ClickHouse | ClickHouse | 701,068,114 | 15,285 | alexey-milovidov | > AST fuzzer — Task failed: $?=8
Temporary error in CI:
> 2020-09-26 11:35:56 + wget -nv -nd -c https://clickhouse-builds.s3.yandex.net/15285/6d5cd95d3f073442acb93afaab9d12514e75414b/clickhouse_build_check/clang-10_debug_none_bundled_unsplitted_disable_False_binary/clickhouse
2020-09-26 11:35:56 https://click... | 2020-09-30T00:29:10 | 2020-09-30T00:29:10 | {} | MEMBER |
ClickHouse | ClickHouse | 701,107,300 | 14,776 | alexey-milovidov | > test_rename_column
Already fixed. | 2020-09-30T01:35:39 | 2020-09-30T01:35:39 | {} | MEMBER |
ClickHouse | ClickHouse | 700,113,718 | 15,239 | alesapin | Wow, functional tests failed in alter query. | 2020-09-28T15:44:30 | 2020-09-28T15:44:30 | {} | MEMBER |
ClickHouse | ClickHouse | 700,214,773 | 15,360 | den-crane | It's not related to a sparse index.
>Any good ideas?
Two tables with duplicated data:
CREATE TABLE test1 (aString,b String) ENGINE = MergeTree() ORDER BY (a, b)
CREATE TABLE test2 (aString,b String) ENGINE = MergeTree() ORDER BY (b, a)
There is no solution for this without inverse indexes and without Z inde... | 2020-09-28T18:46:37 | 2020-09-28T18:46:56 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 700,506,685 | 15,431 | Sasasu | `ColumnUInt64 *col = ColumnUInt64::create().detach();` if you realy want. | 2020-09-29T07:29:16 | 2020-09-29T07:29:16 | {} | NONE |
ClickHouse | ClickHouse | 701,135,290 | 15,458 | amosbird | > Usually they just copy-paste cluster definition under different name with all servers as replicas.
Perhaps we can have cluster transformers, like oneShard(my_cluster) and oneReplica(my_cluster).
| 2020-09-30T03:28:22 | 2020-09-30T03:28:22 | {
"+1": 3
} | COLLABORATOR |
ClickHouse | ClickHouse | 701,139,967 | 15,442 | alexey-milovidov | Not a bugfix. Maybe "improvement"? | 2020-09-30T03:47:44 | 2020-09-30T03:47:44 | {} | MEMBER |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.