Commit Graph

36 Commits

Author SHA1 Message Date
alexey-milovidov
4e11e7cfa9
Merge pull request #18772 from azat/optimize-memory-tracking-fix
[RFC] Fix memory tracking for OPTIMIZE TABLE/merges
2021-01-22 03:48:06 +03:00
ygrek
8f2a830d83
add zstd long range option (#17184)
* add zstd long compression option

* tests: add zstd long read-write test

Co-authored-by: Joris Giovannangeli <joris.giovannangeli@ahrefs.com>
Co-authored-by: ip <igor@ahrefs.com>
2021-01-13 16:22:59 +03:00
Azat Khuzhin
05608687d6 Get back memory tracking blocker in calculateAndSerializePrimaryIndex()
But reduce scope, to avoid leaking too much memory, since there are old
values in last_block_index_columns.

The scope of the MemoryTracker::BlockerInThread has been increased in #8290
2021-01-12 01:00:37 +03:00
Azat Khuzhin
bd05d9db2f Fix memory tracking for OPTIMIZE TABLE queries
Because of BlockerInThread in
MergeTreeDataPartWriterOnDisk::calculateAndSerializePrimaryIndex memory
was accounted incorrectly and grows constantly.

And IIUC there is no need in that blocker, since INSERT SELECT shares
the same thread group.
2021-01-12 00:49:40 +03:00
Alexey Milovidov
6eb5a5f4d9 Remove useless code 2021-01-10 03:28:59 +03:00
alesapin
cd598034cd Fix add tests 2020-12-18 17:44:31 +03:00
alesapin
c1e7d5d3c1 Working code 2020-12-18 16:49:45 +03:00
alesapin
8670836573 Add some comments 2020-12-15 13:34:28 +03:00
alesapin
a50187c995 Slightly better 2020-12-15 11:04:37 +03:00
alesapin
47e3f619e5 Remove redundant code 2020-12-14 16:01:01 +03:00
alesapin
303c12d2e4 Slightly better 2020-12-14 15:51:14 +03:00
alesapin
6fb23da851 Refactoring 2020-12-14 15:03:49 +03:00
alesapin
4b83707c63 Compileable code for some reason 2020-12-11 18:00:58 +03:00
alesapin
57c30c4164 Garbage 2020-12-11 16:20:19 +03:00
alesapin
7ffd76a605 Fix bug 2020-12-11 11:41:02 +03:00
alesapin
f3d1dc360c Fix style 2020-12-10 12:24:52 +03:00
alesapin
0704b478ee Less constuctors 2020-12-10 12:22:43 +03:00
alesapin
3b88c077ed Mechanical refactoring 2020-12-09 21:19:49 +03:00
alesapin
af73db93e6 Mechanical refactoring 2020-12-09 21:10:09 +03:00
alesapin
7783ddb3af
Revert "Fix index granularity calculation on block borders" 2020-12-09 14:05:48 +03:00
alesapin
7080f424e2 Fix bug for skip indices and make code more complex 2020-11-18 15:04:13 +03:00
alesapin
fbda80ac89 Simplier storage of last index column 2020-11-02 15:19:37 +03:00
Alexey Milovidov
2619efadc8 Fix multiple issues with memory tracking 2020-10-21 03:31:34 +03:00
Pavel Kovalenko
69b4bc6f33 Proper error handling during insert into MergeTree with S3. 2020-10-06 12:38:00 +03:00
Anton Popov
3cadc9033a fsyncs for metadata files of part 2020-09-01 18:53:22 +03:00
Anton Popov
4df5c72878 Merge remote-tracking branch 'upstream/master' into HEAD 2020-08-24 12:57:27 +03:00
alexey-milovidov
fa1c4d1eef
Merge pull request #13538 from CurtizJ/fix-skip-indices
Fix secondary indices corruption in compact parts
2020-08-09 20:13:13 +03:00
Anton Popov
2280619ad0 fix secondary indices corruption in compact parts 2020-08-09 04:45:10 +03:00
Alexey Milovidov
12f66fa82c Fix 99% of typos 2020-08-08 04:01:47 +03:00
alesapin
6f1824f0ea Correct merge with master 2020-06-26 14:30:23 +03:00
Anton Popov
1ce09e1faa Merge remote-tracking branch 'upstream/master' into polymorphic-parts 2020-06-03 16:27:54 +03:00
Anton Popov
6d8749b100 in-memory parts: faster computation of checksums 2020-05-21 01:16:08 +03:00
Anton Popov
4c03f48703 in-memory parts: better restoring 2020-05-15 03:39:15 +03:00
Anton Popov
aa723d245f Merge remote-tracking branch 'upstream/master' into polymorphic-parts 2020-04-30 18:27:39 +03:00
Anton Popov
42997bce86 im-memory parts: replication 2020-04-29 20:14:49 +03:00
Anton Popov
9a2855fae7 in-memory parts: preparation 2020-04-17 15:15:20 +03:00