User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/29.0.1547.57 Safari/537.36
Build Identifier:
I have data and scripts ready to reproduce this, but the data is quite big (14g uncompressed), so I cannot attach it here. I also tried on smaller data samples, but then I cannot reproduce it.
Please write me an email to get a link to data and scripts, I rather not make it public.
The bug seems to pop up only in Feb2013_SP4 (it works fine on another installation that is up to changeset 49170:c0bf1946fc19), and this is its summary:
Two queries, Q1 and Q2, are executed within the same transaction. They both (although slightly differently) join a small table against a large table, and take the "top" 2 tuples. They both have no side-effects, they are simple SELECT queries.
What happens:
Case 1:
START TRANSACTION;
Q1; -- 2 tuples out, as expected
Case 2:
START TRANSACTION;
Q2; -- 2 tuples out, as expected
Case 3:
START TRANSACTION;
Q1; -- 2 tuples out, as expected
Q2; -- BUG: 8 tuples out !!!
Reproducible: Always
Compiled from Mercurial repo, changeset 49170:c0bf1946fc19
MonetDB 5 server v11.15.16 (64-bit, 64-bit oids)
This is an unreleased version
Copyright (c) 1993-July 2008 CWI
Copyright (c) August 2008-2013 MonetDB B.V., all rights reserved
Visit http://www.monetdb.org/ for further information
Found 15.6GiB available memory, 8 available cpu cores
Libraries:
libpcre: 8.32 2012-11-30 (compiled with 8.32)
openssl: OpenSSL 1.0.1e 11 Feb 2013 (compiled with OpenSSL 1.0.1e-fips 11 Feb 2013)
libxml2: 2.9.1 (compiled with 2.9.1)
Compiled by: roberto@photon.spinque.com (x86_64-unknown-linux-gnu)
Compilation: gcc -g -Werror -Wall -Wextra -W -Werror-implicit-function-declaration -Wpointer-arith -Wdeclaration-after-statement -Wundef -Wformat=2 -Wno-format-nonliteral -Winit-self -Winvalid-pch -Wmissing-declarations -Wmissing-format-attribute -Wmissing-prototypes -Wold-style-definition -Wpacked -Wunknown-pragmas -Wvariadic-macros -fstack-protector-all -Wstack-protector -Wpacked-bitfield-compat -Wsync-nand -Wjump-misses-init -Wmissing-include-dirs -Wlogical-op -Wunreachable-code
Linking : /usr/bin/ld -m elf_x86_64
Date: 2013-09-30 11:44:30 +0200
From: @swingbit
To: SQL devs <>
Version: 11.15.3 (Feb2013-SP1)
CC: @njnes
Last updated: 2013-10-22 14:42:10 +0200
Comment 19217
Date: 2013-09-30 11:44:30 +0200
From: @swingbit
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/29.0.1547.57 Safari/537.36
Build Identifier:
I have data and scripts ready to reproduce this, but the data is quite big (14g uncompressed), so I cannot attach it here. I also tried on smaller data samples, but then I cannot reproduce it.
Please write me an email to get a link to data and scripts, I rather not make it public.
The bug seems to pop up only in Feb2013_SP4 (it works fine on another installation that is up to changeset 49170:c0bf1946fc19), and this is its summary:
Two queries, Q1 and Q2, are executed within the same transaction. They both (although slightly differently) join a small table against a large table, and take the "top" 2 tuples. They both have no side-effects, they are simple SELECT queries.
What happens:
Case 1:
START TRANSACTION;
Q1; -- 2 tuples out, as expected
Case 2:
START TRANSACTION;
Q2; -- 2 tuples out, as expected
Case 3:
START TRANSACTION;
Q1; -- 2 tuples out, as expected
Q2; -- BUG: 8 tuples out !!!
Reproducible: Always
Compiled from Mercurial repo, changeset 49170:c0bf1946fc19
MonetDB 5 server v11.15.16 (64-bit, 64-bit oids)
This is an unreleased version
Copyright (c) 1993-July 2008 CWI
Copyright (c) August 2008-2013 MonetDB B.V., all rights reserved
Visit http://www.monetdb.org/ for further information
Found 15.6GiB available memory, 8 available cpu cores
Libraries:
libpcre: 8.32 2012-11-30 (compiled with 8.32)
openssl: OpenSSL 1.0.1e 11 Feb 2013 (compiled with OpenSSL 1.0.1e-fips 11 Feb 2013)
libxml2: 2.9.1 (compiled with 2.9.1)
Compiled by: roberto@photon.spinque.com (x86_64-unknown-linux-gnu)
Compilation: gcc -g -Werror -Wall -Wextra -W -Werror-implicit-function-declaration -Wpointer-arith -Wdeclaration-after-statement -Wundef -Wformat=2 -Wno-format-nonliteral -Winit-self -Winvalid-pch -Wmissing-declarations -Wmissing-format-attribute -Wmissing-prototypes -Wold-style-definition -Wpacked -Wunknown-pragmas -Wvariadic-macros -fstack-protector-all -Wstack-protector -Wpacked-bitfield-compat -Wsync-nand -Wjump-misses-init -Wmissing-include-dirs -Wlogical-op -Wunreachable-code
Linking : /usr/bin/ld -m elf_x86_64
Comment 19232
Date: 2013-10-02 16:56:38 +0200
From: @njnes
fixed by making sure a view only shares a hash with its parent if this
full column is used (ie not a slice)
The text was updated successfully, but these errors were encountered: