STATUS_UNKNOWN_ERROR (2) on empty test db

Interestingly, seemingly if the 1146 (which did not seem relevant) does not happen, a STATUS_PERL_ERROR is resulting instead.

Additional testcase for STATUS_PERL_ERROR:

perl runall.pl --queries=100000000 --seed=random --threads=1 --duration=300 --querytimeout=10 --engine=MEMORY --mysqld=--max_heap_table_size=134217728 --grammar=/randgen/conf/optimizer/optimizer_access_exp.yy --gendata=/randgen/conf/replication/WL5092_data.zz --basedir1=/mysql/mysql-linux-x86_6 --vardir1=/tmp/test2/

Outcome:

=================
# 2011-09-01T11:56:23 `col_varchar_25_default_null` varchar(25) default null,
# 2011-09-01T11:56:23 `col_tinytext_binary_not_null` tinytext CHARACTER SET binary not null,
# 2011-09-01T11:56:23 /*Indices*/
# 2011-09-01T11:56:23 primary key (pk)) ENGINE=memory failed: 1163 The used table type doesn't support BLOB/TEXT columns. Further errors of this kind will be suppressed.
# 2011-09-01T11:56:23 Query: ALTER TABLE `table300_memory_int_autoinc` DISABLE KEYS failed: 1146 Table '%s' doesn't exist. Further errors of this kind will be suppressed.
# 2011-09-01T11:56:26 Loading Generator GenTest::Generator::FromGrammar.
# 2011-09-01T11:56:26 Caching schema metadata for dbi:mysql:host=127.0.0.1:port=19300:user=root:database=test
# 2011-09-01T11:56:29 Reporters: ErrorLog, Backtrace
# 2011-09-01T11:56:29 Validators: ErrorMessageCorruption
# 2011-09-01T11:56:29 Starting 1 processes, 100000000 queries each, duration 300 seconds.
# 2011-09-01T11:56:29 GenTest::ErrorFilter(26247) started
# 2011-09-01T11:56:29 Loading Validator GenTest::Validator::ErrorMessageCorruption.
Schema 'test' has no tables at lib/GenTest/Generator/FromGrammar.pm line 259
# 2011-09-01T11:56:29 Started periodic reporting process...
# 2011-09-01T11:56:30 Killing periodic reporting process with pid 26248...
# 2011-09-01T11:56:30 Kill GenTest::ErrorFilter(26247)
# 2011-09-01T11:56:30 Test completed with failure status STATUS_PERL_FAILURE (255)
# 2011-09-01T11:56:30 gentest.pl exited with exit status STATUS_PERL_FAILURE (255)
2011-09-01T11:56:30 [25716] runall.pl will exit with exit status STATUS_PERL_FAILURE (255)
=================

Hi Roel,

Am 01.09.2011 02:32, schrieb Roel Van de Paar:
> Public bug reported:
>
> Running this cmd line (there are more grammar combinations for which
> this bug is seen, this is just one example):
>
> perl runall.pl --queries=100000000 --seed=random --threads=1
> --duration=300 --mysqld=--log-output=none --engine=MEMORY
> --mysqld=--max_heap_table_size=134217728
> --grammar=/randgen/conf/optimizer/optimizer_no_subquery_portable.yy
> --gendata=/randgen/conf/replication/replication-6.0.zz --basedir1=/mysql
> /mysql-linux-x86_64 --vardir1=/tmp/test1
>
> Produces this output (reduced to relevant):
>
> =================
> [...]
> # 2011-09-01T11:28:45 key (`col_float_key` ),
> # 2011-09-01T11:28:45 key (`col_char_latin1_key` ),
> # 2011-09-01T11:28:45 key (`col_char_utf8_key` )) ENGINE=memory failed: 1163 The used table type doesn't support BLOB/TEXT columns. Further errors of this kind will be suppressed
> .
> # 2011-09-01T11:28:45 Query: ALTER TABLE `table0_memory` DISABLE KEYS failed: 1146 Table '%s' doesn't exist. Further errors of this kind will be suppressed.
> # 2011-09-01T11:28:45 # Creating MySQL table: test.table0_memory_int_autoinc; engine: memory; rows: 0 .
> # 2011-09-01T11:28:45 # Creating MySQL table: test.table0_memory_key_pk_parts_2; engine: memory; rows: 0 .
> # 2011-09-01T11:28:45 # Creating MySQL table: test.table0_memory_key_pk_parts_2_int_autoinc; engine: memory; rows: 0 .
> # 2011-09-01T11:28:45 # Creating MySQL table: test.table1_memory; engine: memory; rows: 1 .
> # 2011-09-01T11:28:45 # Creating MySQL table: test.table1_memory_int_autoinc; engine: memory; rows: 1 .
> # 2011-09-01T11:28:45 # Creating MySQL table: test.table1_memory_key_pk_parts_2; engine: memory; rows: 1 .
> # 2011-09-01T11:28:45 # Creating MySQL table: test.table1_memory_key_pk_parts_2_int_autoinc; engine: memory; rows: 1 .
> # 2011-09-01T11:28:45 # Creating MySQL table: test.table10_memory; engine: memory; rows: 10 .
> # 2011-09-01T11:28:46 # Creating MySQL table: test.table10_memory_int_autoinc; engine: memory; rows: 10 .
> # 2011-09-01T11:28:46 # Creating MySQL table: test.table10_memory_key_pk_parts_2; engine: memory; rows: 10 .
> # 2011-09-01T11:28:46 # Creating MySQL table: test.table10_memory_key_pk_parts_2_int_autoinc; engine: memory; rows: 10 .
> # 2011-09-01T11:28:46 # Creating MySQL table: test.table100_memory; engine: memory; rows: 100 .
> # 2011-09-01T11:28:49 # Creating MySQL table: test.table100_memory_int_autoinc; engine: memory; rows: 100 .
> # 2011-09-01T11:28:51 # Creating MySQL table: test.table100_memory_key_pk_parts_2; engine: memory; rows: 100 .
> # 2011-09-01T11:28:53 # Creating MySQL table: test.table100_memory_key_pk_parts_2_int_autoinc; engine: memory; rows: 100 .
> # 2011-09-01T11:28:56 Loading Generator GenTest::Generator::FromGrammar.
> # 2011-09-01T11:28:56 Caching schema metadata for dbi:mysql:host=127.0.0.1:port=19300:user=root:database=test
> # 2011-09-01T11:28:59 Reporters: ErrorLog, Backtrace
> # 2011-09-01T11:28:59 Validators: ErrorMessageCorruption
> # 2011-09-01T11:28:59 Starting 1 processes, 100000000 queries each, duration 300 seconds.
> # 2011-09-01T11:28:59 GenTest::ErrorFilter(18515) started
> # 2011-09-01T11...

Hi Roel,

Am 01.09.2011 02:56, schrieb Roel Van de Paar:
> Interestingly, seemingly if the 1146 (which did not seem relevant) does
> not happen, a STATUS_PERL_ERROR is resulting instead.
>
> Additional testcase for STATUS_PERL_ERROR:
>
> perl runall.pl --queries=100000000 --seed=random --threads=1
> --duration=300 --querytimeout=10 --engine=MEMORY
> --mysqld=--max_heap_table_size=134217728
> --grammar=/randgen/conf/optimizer/optimizer_access_exp.yy
> --gendata=/randgen/conf/replication/WL5092_data.zz --basedir1=/mysql
> /mysql-linux-x86_6 --vardir1=/tmp/test2/
>
> Outcome:
>
> =================
> # 2011-09-01T11:56:23 `col_varchar_25_default_null` varchar(25) default null,
> # 2011-09-01T11:56:23 `col_tinytext_binary_not_null` tinytext CHARACTER SET binary not null,
> # 2011-09-01T11:56:23 /*Indices*/
> # 2011-09-01T11:56:23 primary key (pk)) ENGINE=memory failed: 1163 The used table type doesn't support BLOB/TEXT columns. Further errors of this kind will be suppressed.
> # 2011-09-01T11:56:23 Query: ALTER TABLE `table300_memory_int_autoinc` DISABLE KEYS failed: 1146 Table '%s' doesn't exist. Further errors of this kind will be suppressed.
> # 2011-09-01T11:56:26 Loading Generator GenTest::Generator::FromGrammar.
> # 2011-09-01T11:56:26 Caching schema metadata for dbi:mysql:host=127.0.0.1:port=19300:user=root:database=test
> # 2011-09-01T11:56:29 Reporters: ErrorLog, Backtrace
> # 2011-09-01T11:56:29 Validators: ErrorMessageCorruption
> # 2011-09-01T11:56:29 Starting 1 processes, 100000000 queries each, duration 300 seconds.
> # 2011-09-01T11:56:29 GenTest::ErrorFilter(26247) started
> # 2011-09-01T11:56:29 Loading Validator GenTest::Validator::ErrorMessageCorruption.
> Schema 'test' has no tables at lib/GenTest/Generator/FromGrammar.pm line 259
> # 2011-09-01T11:56:29 Started periodic reporting process...
> # 2011-09-01T11:56:30 Killing periodic reporting process with pid 26248...
> # 2011-09-01T11:56:30 Kill GenTest::ErrorFilter(26247)
> # 2011-09-01T11:56:30 Test completed with failure status STATUS_PERL_FAILURE (255)
> # 2011-09-01T11:56:30 gentest.pl exited with exit status STATUS_PERL_FAILURE (255)
> 2011-09-01T11:56:30 [25716] runall.pl will exit with exit status STATUS_PERL_FAILURE (255)
> =================
>

just some guess.
What will happen in case of
- database does not contain any tables
   So the cache of table names for the database is empty.
- the grammar contains something like
   SELECT * FROM test . _table
   Because of the empty cache no table name can be
   selected by gentest.
?

Regards

Matthias

--
Matthias Leich
Senior QA Developer
Office: +49 30 4764614

ORACLE Deutschland B.V. & Co. KG
Hauptverwaltung: Riesstr. 25, D-80992 München
Registergericht: Amtsgericht München, HRA 95603

Komplementärin: ORACLE Deutschland Verwaltung B.V.
Rijnzathe 6, 3454PV De Meern, Niederlande
Handelsregister der Handelskammer Midden-Niederlande, Nr. 30143697
Geschäftsführer: Jürgen Kunz, Marcel van de Molen, Alexander van der Ven