| Bug #47128 | Test "binlog_unsafe" exceds default stack allocation | ||
|---|---|---|---|
| Submitted: | 4 Sep 2009 10:20 | Modified: | 15 Oct 2012 14:48 |
| Reporter: | Joerg Bruehe | Email Updates: | |
| Status: | Unsupported | Impact on me: | |
| Category: | Tests: Server | Severity: | S3 (Non-critical) |
| Version: | 5.4.2-beta, cluster-6.3.27a, 5.5.0-beta,5.1.43 | OS: | Any (see text) |
| Assigned to: | Kent Boortz | CPU Architecture: | Any |
[29 Oct 2009 16:40]
Joerg Bruehe
In the build of clster-6.3.27a, that very same problem also arises in test "sp": ===== CURRENT_TEST: main.sp mysqltest: At line NNN: query 'call bug10100t(5)' failed: 1436: Thread stack overrun: 68936 bytes used of a 196608 byte stack, and 128000 bytes needed. Use 'mysqld -O thread_stack=#' to specify a bigger stack. The result from queries just before the failure was: < snip > if level < lim then select level; fetch c into lv; select lv; update t3 set a=level+lv; FLUSH TABLES; call bug10100pc(level+1, lim); else select * from t3; end if; close c; end| set @@max_sp_recursion_depth=4| select @@max_sp_recursion_depth| @@max_sp_recursion_depth 4 select bug10100f(3)| ERROR HY000: Recursive stored functions and triggers are not allowed. select bug10100f(6)| ERROR HY000: Recursive stored functions and triggers are not allowed. ===== Here, it affects Solaris 9 and 10, Sparc, 32 bit, debug build.
[18 Nov 2009 17:55]
Joerg Bruehe
Sme problem in the build of 5.5.0-beta.
[18 Nov 2009 18:03]
Joerg Bruehe
In 5.5.0-beta, test "signal_demo3" has the same problem.
[26 Jan 2010 23:03]
Joerg Bruehe
Same problem in the 5.1.43 release build.
[16 Mar 2010 9:34]
Bugs System
A patch for this bug has been committed. After review, it may be pushed to the relevant source trees for release in the next version. You can access the patch from: http://lists.mysql.com/commits/103375 3436 Bernd Ocklin 2010-03-09 make failing solaris test cases experimental Bug#47128, Bug#46895
[4 Jan 2011 10:44]
Joerg Bruehe
Situation in the release build of 5.6.1-m5: The failure occurs in tests "signal_demo3" and "max_sp_recursion_depth_func" with RPMs for RHEL or SLES on IA64 (compiler icc) in debug mode. (A similar error on Solaris in test "subquery_sj_none" is filed as bug#47791.)
Description: Found during the 5.4.2-beta release build: ===== binlog.binlog_unsafe 'stmt' [ retry-fail ] Test ended at YYYY-MM-DD HH:MM:SS CURRENT_TEST: binlog.binlog_unsafe mysqltest: At line NNN: query 'EXECUTE prep6' failed: 1436: Thread stack overrun: 69456 bytes used of a 196608 byte stack, and 128000 bytes needed. Use 'mysqld -O thread_stack=#' to specify a bigger stack. The result from queries just before the failure was: < snip > BEGIN CALL proc1(); RETURN 0; END| CREATE TRIGGER trig3 BEFORE INSERT ON trigger_table2 FOR EACH ROW BEGIN DECLARE tmp INT; SELECT func2() INTO tmp; END| CREATE PROCEDURE proc4() INSERT INTO trigger_table2 VALUES (1)| CREATE FUNCTION func5() RETURNS INT BEGIN CALL proc4; RETURN 0; END| PREPARE prep6 FROM 'SELECT func5()'| More results from queries before failure can be found in /PATH/mysql-test/var/log/binlog_unsafe.log - saving '/PATH/mysql-test/var/log/binlog.binlog_unsafe-stmt/' to '/PATH/mysql-test/var/log/binlog.binlog_uns afe-stmt/' ===== This happened (with slightly differing stack usage values) on Linux x86 ICC and in the debug builds on Solaris 8 x86 Solaris 10 Sparc 32bit Linux IA64 (RPMs for RH 3, 4, and 5, SuSE 9 and 10) How to repeat: Run the test suite on any of these. Suggested fix: Simplify the test, so that "stack used" is reduced or Raise the default stack size or Check whether we really need to assume 128,000 byte for a new Sp stack frame.