List:General Discussion« Previous MessageNext Message »
From:Sasha Pachev Date:May 7 2001 3:44pm
Subject:Re: MySQL-Server getting signal 11 ... !!
View as plain text  
On Sunday 06 May 2001 20:06, Stephan Skusa wrote:
> Hello,
> 
> I've got a problem:
> 
> My MySQL-Server gets signal 11 almost every day arround 0:00 am ... I have a
> nice
> mrtg - graph showing this ... ;o)
> 
> In fact every time the server crashes the statement shown in the server.log
> (or a
> similar one with other memID) is executed.
> 
> topsitesTEMP is a temporary table. Every time I execute this statement
> manualy
> using tool mysql everything works fine.
> 
> TIA
> 	Stephan Skusa
> 
> 
> mysqld got signal 11;
> The manual section 'Debugging a MySQL server' tells you how to use a
> stack trace and/or the core file to produce a readable backtrace that may
> help in finding out why mysqld died
> Attemping backtrace. You can use the following information to find out
> where mysqld died.  If you see no messages after this, something went
> terribly wrong
> stack range sanity check, ok, backtrace follows
> 0x4009365d
> 0x80d7664
> 0x80b7d2d
> 0x80dab94
> 0x80dab19
> 0x80c3629
> 0x80c67e2
> 0x80c27f3
> 0x80c1b57
> stack trace successful, now will try to get some
> variables. Some pointers may be invalid and cause dump abort
> thd->query at 0x84af148 = SELECT myROWNUM, histRANKING FROM topsitesTEMP
> WHERE memID=12800
> thd->thread_id = 17276
> successfully dumped variables, if you ran with --log
>  take a look at the details of what thread 17276 did to cause the crash.
> In some cases of really bad corruption, this value can be invalid
> Please use the information above to create a repeatable
> test case for the crash, and send it to bugs@stripped
> 
> Number of processes running now: 0
> 010504 23:56:38  mysqld restarted
> /usr/local/mysql-3.23.33/libexec/mysqld: Warten auf Verbindungen.
> 
>   
>

Some things to do:

* Can you resolve those stack traces with resolve_stack_dump ? Follow 
instructions in http://www.mysql.com/doc/U/s/Using_stack_trace.html

* What is the structure and type of topsitesTEMP?

* Can you write a script to take a backup of topsitesTEMP  every time before 
you expect a crash? This will be tricky if topsitesTEMP is a heap or 
temporary table.

* You may also want to check if changing the table to a different type helps 
avoid the coredump - this will give us an idea where the bug is.

* Does the bug go away if you use our latest ( 3.23.37) binary?

* How big is your key buffer, record_buffer, and sort_buffer, and how many 
concurrent connections at the time of the crash? I am asking this because it 
looks like you have built the binary yourself, which means that most likely 
you have linked it against default LinuxThreads, and it has problems when you 
get up to 600 connections, or even less if mysqld is using up a lot of memory 
already.



-- 
MySQL Development Team
   __  ___     ___ ____  __ 
  /  |/  /_ __/ __/ __ \/ /   Sasha Pachev <sasha@stripped>
 / /|_/ / // /\ \/ /_/ / /__  MySQL AB, http://www.mysql.com/
/_/  /_/\_, /___/\___\_\___/  Provo, Utah, USA
       <___/                  
Thread
MySQL-Server getting signal 11 ... !!Stephan Skusa7 May
  • Re: MySQL-Server getting signal 11 ... !!Gerald Clark7 May
    • AW: MySQL-Server getting signal 11 ... !!Stephan Skusa7 May
      • Re: AW: MySQL-Server getting signal 11 ... !!Gerald Clark9 May
Re: MySQL-Server getting signal 11 ... !!Sasha Pachev7 May