[MDEV-13843] rocksdb.rocksdb_parts failed in buildbot (sporadic) Created: 2017-09-19  Updated: 2023-04-27

Status: Open
Project: MariaDB Server
Component/s: Storage Engine - RocksDB, Tests
Affects Version/s: 10.2, 10.3
Fix Version/s: 10.4

Type: Bug Priority: Major
Reporter: Alice Sherepa Assignee: Sergei Petrunia
Resolution: Unresolved Votes: 0
Labels: None


 Description   

http://buildbot.askmonty.org/buildbot/builders/kvm-bintar-trusty-amd64/builds/2722/steps/test/logs/stdio

rocksdb.rocksdb_parts                    w1 [ fail ]
 
        Test ended at 2017-09-18 14:31:33
 
 
 
CURRENT_TEST: rocksdb.rocksdb_parts
 
--- /usr/local/mariadb-10.3.2-linux-x86_64/mysql-test/plugin/rocksdb/rocksdb/r/rocksdb_parts.result	2017-09-18 12:09:47.000000000 +0000
 
+++ /usr/local/mariadb-10.3.2-linux-x86_64/mysql-test/plugin/rocksdb/rocksdb/r/rocksdb_parts.reject	2017-09-18 14:31:33.018940452 +0000
 
@@ -76,7 +76,7 @@
 
 # The following must use "Using index"
 
 explain select pk from t1 force index(col1) where col1=10;
 
 id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
 
-1	SIMPLE	t1	ref	col1	col1	5	const	2000	Using index
 
+1	SIMPLE	t1	ref	col1	col1	5	const	1001	Using index
 
 drop table t1;
 
 #
 
 # Issue #214: subqueries cause crash
 
 
 
mysqltest: Result content mismatch



 Comments   
Comment by Alice Sherepa [ 2017-09-25 ]

also on 10.2 http://buildbot.askmonty.org/buildbot/builders/kvm-bintar-trusty-amd64/builds/2824/steps/test/logs/stdio

Comment by Sergei Petrunia [ 2017-10-29 ]

Can produce different estimate numbers:

http://buildbot.askmonty.org/buildbot/builders/kvm-fulltest/builds/10590/steps/test_2/logs/stdio

-1	SIMPLE	t1	ref	col1	col1	5	const	2000	Using index
 
+1	SIMPLE	t1	ref	col1	col1	5	const	1001	Using index

http://buildbot.askmonty.org/buildbot/builders/kvm-deb-jessie-ppc64le/builds/1583/steps/mtr/logs/stdio

-1	SIMPLE	t1	ref	col1	col1	5	const	2000	Using index
 
+1	SIMPLE	t1	ref	col1	col1	5	const	3000	Using index

This is while rocksdb_parts-master.opt has 

--rocksdb_debug_optimizer_n_rows=1000

And the test case has:

create table t1 (
 
pk int primary key,
 
col1 int,
 
col2 int,
 
key (col1)
 
) engine=rocksdb partition by hash(pk) partitions 2;
 
insert into t1 values (1,10,10);
 
insert into t1 values (2,10,10);
 
insert into t1 values (11,20,20);
 
insert into t1 values (12,20,20);
 
explain select pk from t1 force index(col1) where col1=10;

It should be using two partitions and produce rows=2000.
It's not clear how an estimate of 3000 could be produced...

Comment by Elena Stepanova [ 2018-09-15 ]

Still happens in buildbot on current 10.2:
http://buildbot.askmonty.org/buildbot/builders/kvm-deb-bionic-amd64/builds/724/steps/mtr/logs/stdio

Comment by Sergei Golubchik [ 2018-12-21 ]

two failures in just one push (tarbuildnum 23476 on rpm-centos73-amd64 and on rpm-centos74-amd64). Looks like it fails fairly often.

Generated at Thu Feb 08 08:08:49 UTC 2024 using Jira 8.20.16#820016-sha1:9d11dbea5f4be3d4cc21f03a88dd11d8c8687422.