Bug #53778 i found bug of memory
Submitted: 19 May 2010 7:13 Modified: 8 Nov 2010 12:42
Reporter: patcharee sirichavalit Email Updates:
Status: No Feedback Impact on me:
None 
Category:MySQL Server: Errors Severity:S1 (Critical)
Version:5.0.22, 5.1.46 OS:Linux
Assigned to: CPU Architecture:Any

[19 May 2010 7:13] patcharee sirichavalit
Description:
It is possible that mysqld could use up to
key_buffer_size + (read_buffer_size + sort_buffer_size)*max_connections = 77687 K
bytes of memory
Hope that's ok; if not, decrease some variables in the equation.

thd=(nil)
Attempting backtrace. You can use the following information to find out
where mysqld died. If you see no messages after this, something went
terribly wrong...
Cannot determine thread, fp=0x69f79e38, backtrace may not be correct.
Stack range sanity check OK, backtrace follows:
0x8174693
0x36b420
(nil)
0x8407561
0x83d092d
0x8318868
0xddd43b
0x2c1fde
New value of fp=(nil) failed sanity check, terminating stack trace!
Please read http://dev.mysql.com/doc/mysql/en/Using_stack_trace.html and follow instructions on how to resolve the stack trace. Resolved
stack trace is much more helpful in diagnosing the problem, so please do
resolve it
The manual page at http://www.mysql.com/doc/en/Crashing.html contains
information that should help you find out what is causing the crash.

Number of processes running now: 0
100519 14:02:26  mysqld restarted

How to repeat:
-
[19 May 2010 7:13] patcharee sirichavalit
version 5.0.22
[19 May 2010 8:05] Valeriy Kravchuk
Please, upgrade to a newer version, 5.0.91, and check if the crash will even happen with it.
[27 May 2010 6:48] patcharee sirichavalit
is it a bug ?

why you need me to upgrade ?
[27 May 2010 7:21] Valeriy Kravchuk
This crash can be a result of hardware failure, misconfiguration or a bug. I can not tell that based on the limited information provided. In case of a bug it can not be fixed in years old version 5.0.22, but only in some new one. 

Also it can be a bug that is already fixed long time ago (I can not identify it based on that unresolved stack trace). That's why I ask to check if the same crash is repeatable with 5.0.91. If it does not crash there we may have a fixed bug. If it crashes we get additional information, more complete stack trace likely etc.
[27 May 2010 7:23] patcharee sirichavalit
my manager would like to know how to prove it will not occur again , if we upgrade mysql.

Do you know how to prove it ?
[27 May 2010 7:54] Valeriy Kravchuk
This report looks more and more like free support request. I'd like to remind you and your manager that support on using our products is available both free in our forums at http://forums.mysql.com/ and for a reasonable fee direct from our skilled support engineers at http://www.mysql.com/support/. This site is for reporting bugs with repeatable test cases.

I can not prove anything based on the information already provided. To fix the bug we need to know how to repeat it on the recent version of MySQL server. You do not want to check if it is repeatable on 5.0.91 yourself, and you do not provide any test case for me to force the same crash. So, there is still no way to identify the bug (if any) and, thus, to fix it.
[27 May 2010 8:00] patcharee sirichavalit
If i know why it occur like this and how to make it occur again .
I will not open this bug report because i will try to avoid to make it occur instead .

so that why i ask you is it a bug or not .
If it not a bug , we will try to find another way to fix it .
we don't want to upgrade mysql ,if it can't solve this problem.
hope you will understand my point .
[27 May 2010 9:00] Valeriy Kravchuk
I understand that any upgrade in production is a problem. But, to really solve this problem somehow (or provide a valid workaround), you have to be able to reproduce it.

Can you tell me at least what statement server was executing at the moment of crash? Send your my.cnf also and the results of:

ulimit -a
free

Linux command from this server.
[27 May 2010 10:21] patcharee sirichavalit
how to send file to you ?
[27 May 2010 11:26] Valeriy Kravchuk
Use Files tab to upload it to the bug report.
[2 Jun 2010 10:29] patcharee sirichavalit
i alreadyupgrade mysql to 5.1.6  and move to 64 bit .

but  problem change to abort connection .

it can run only 1000 session .

why ? 
anyway i send file to you  for check .
[3 Jun 2010 5:34] Sveta Smirnova
Thank you for the feedback.

But there is no sense in upgrading to beta version. Please upgrade either to 5.0.91 or 5.1.47. We don't backport bugfixes to minor versions anyway.
[3 Jun 2010 6:13] patcharee sirichavalit
I'm so sorry . i upgraded to 5.1.46 .

for  version 5.1.46 , you can solve it for me or not ?
[3 Jun 2010 8:04] patcharee sirichavalit
Hi All,

for  version 5.1.46 , you can solve it for me or not ?

any update?
[10 Jun 2010 2:45] patcharee sirichavalit
Hi All,

Any update?
[10 Jun 2010 6:48] Sveta Smirnova
Thank you for the feedback.

To fix this we should know why this error happens. Unfortunately information which you provided does not contain this information. Please enabled core files at OS level and in mysqld configuration file(option --core), then send us core file and mysqld after MySQL server crashes again.
[10 Jun 2010 6:49] patcharee sirichavalit
if it not crash ,core file will growth or not ?
[10 Jun 2010 7:07] patcharee sirichavalit
Please enabled core files at OS level  <== how to do ?
my system engineer didn't know how to do it .

My OS is red hat
[8 Oct 2010 12:42] Valeriy Kravchuk
What exact Red Hat version do you use?

Please, send the results of

cat /etc/issue
[9 Nov 2010 0:00] Bugs System
No feedback was provided for this bug for over a month, so it is
being suspended automatically. If you are able to provide the
information that was originally requested, please do so and change
the status of the bug back to "Open".