On Thu, 10 Feb 2005 10:19:32 +0900, Batara Kesuma
<[EMAIL PROTECTED]> wrote:
> Hi Tobias,
>
> On Wed, 9 Feb 2005 14:48:16 +0100 (CET)
> Tobias Asplund <[EMAIL PROTECTED]> wrote:
>
> > > I try to install MySQL 4.1.9 (official RPM from mysql.com). My machine
> > > is running linux 2.6.9, and it has
Hi Tobias,
On Wed, 9 Feb 2005 14:48:16 +0100 (CET)
Tobias Asplund <[EMAIL PROTECTED]> wrote:
> > I try to install MySQL 4.1.9 (official RPM from mysql.com). My machine
> > is running linux 2.6.9, and it has 4GB of RAM. The problem is MySQL
> > won't start if I set innodb_buffer_pool_size to >= 2G
On Wed, 9 Feb 2005, Batara Kesuma wrote:
> Hi,
>
> I try to install MySQL 4.1.9 (official RPM from mysql.com). My machine
> is running linux 2.6.9, and it has 4GB of RAM. The problem is MySQL
> won't start if I set innodb_buffer_pool_size to >= 2GB. Here is my
> ulimit.
Are you trying this on a 3
> -Original Message-
> From: Dicky Wahyu Purnomo [mailto:[EMAIL PROTECTED]]
> Subject: Memory Limit
> And what is the calculation for the memory also
The formula you want is (this does not account for InnoDB buffers either):
key_buffer_size + (record_buffer + sort_buffer)*max_conne
On Wed, Feb 06, 2002 at 02:57:51PM +0200, Sinisa Milivojevic wrote:
> Franklin, Kevin writes:
> > We are running an extremely large instance of mysql version 3.23.46 on
> > Solaris 2.8. We are attempting to use a software version compiled 64 bit
> > and have been experiencing memory related serve
hi!
> "Sinisa" == Sinisa Milivojevic <[EMAIL PROTECTED]> writes:
Sinisa> Albert Chin writes:
>>
>> $ grep SIZEOF_LONG config.h
>> #define SIZEOF_LONG 8
>> #define SIZEOF_LONG_LONG 8
>>
>> $ grep SIZEOF_LONG incldue/my_config.h
>> #define SIZEOF_LONG 8
>> #define SIZEOF_LONG_LONG 8
>>
>>
On Wed, Feb 06, 2002 at 07:17:13PM +0200, Sinisa Milivojevic wrote:
> Albert Chin writes:
> >
> > $ grep SIZEOF_LONG config.h
> > #define SIZEOF_LONG 8
> > #define SIZEOF_LONG_LONG 8
> >
> > $ grep SIZEOF_LONG incldue/my_config.h
> > #define SIZEOF_LONG 8
> > #define SIZEOF_LONG_LONG 8
> >
> >
Albert Chin writes:
>
> Ok. Guess we'll wait for a fix. Any idea when a fix for MySQL will be
> available?
>
> --
> albert chin ([EMAIL PROTECTED])
>
Just look into Changelog's of the versions that come out.
--
Regards,
__ ___ ___ __
/ |/ /_ __/ __/ __ \/ /Mr. Sinisa M
On Wed, Feb 06, 2002 at 06:12:22PM +0200, Sinisa Milivojevic wrote:
> Albert Chin writes:
> >
> > MySQL 3.23.46 was built with the Sun C++ compiler:
> > $ CC -V
> > CC: Sun WorkShop 6 update 2 C++ 5.3 Patch 111685-03 2001/10/19
> >
> > It was built as follows:
> > CC=cc CFLAGS="-mr -Qn -xs
Albert Chin writes:
> On Wed, Feb 06, 2002 at 07:17:13PM +0200, Sinisa Milivojevic wrote:
> Yes, SIZEOF_INT is 4:
> $ grep SIZEOF_INT config.h
> #define SIZEOF_INT 4
> $ grep SIZEOF_INT include/my_config.h
> #define SIZEOF_INT 4
>
> However, according to
>
>http://docs.sun.com/ab2/coll.4
Albert Chin writes:
> On Wed, Feb 06, 2002 at 07:17:13PM +0200, Sinisa Milivojevic wrote:
>
> Yes, SIZEOF_INT is 4:
> $ grep SIZEOF_INT config.h
> #define SIZEOF_INT 4
> $ grep SIZEOF_INT include/my_config.h
> #define SIZEOF_INT 4
>
> --
> albert chin ([EMAIL PROTECTED])
>
We shall ha
Albert Chin writes:
>
> $ grep SIZEOF_LONG config.h
> #define SIZEOF_LONG 8
> #define SIZEOF_LONG_LONG 8
>
> $ grep SIZEOF_LONG incldue/my_config.h
> #define SIZEOF_LONG 8
> #define SIZEOF_LONG_LONG 8
>
> --
> albert chin ([EMAIL PROTECTED])
>
If SIZEOF_INT is 4, then it could be our bug.
P
On Wed, Feb 06, 2002 at 08:04:23PM +0200, Sinisa Milivojevic wrote:
> Albert Chin writes:
> > On Wed, Feb 06, 2002 at 07:17:13PM +0200, Sinisa Milivojevic wrote:
> > Yes, SIZEOF_INT is 4:
> > $ grep SIZEOF_INT config.h
> > #define SIZEOF_INT 4
> > $ grep SIZEOF_INT include/my_config.h
> >
hi!
> "Sinisa" == Sinisa Milivojevic <[EMAIL PROTECTED]> writes:
Sinisa> Albert Chin writes:
>>
>> $ grep SIZEOF_LONG config.h
>> #define SIZEOF_LONG 8
>> #define SIZEOF_LONG_LONG 8
>>
>> $ grep SIZEOF_LONG incldue/my_config.h
>> #define SIZEOF_LONG 8
>> #define SIZEOF_LONG_LONG 8
>>
>>
Albert Chin writes:
>
> Ok. Guess we'll wait for a fix. Any idea when a fix for MySQL will be
> available?
>
> --
> albert chin ([EMAIL PROTECTED])
>
Just look into Changelog's of the versions that come out.
--
Regards,
__ ___ ___ __
/ |/ /_ __/ __/ __ \/ /Mr. Sinisa M
On Wed, Feb 06, 2002 at 08:04:23PM +0200, Sinisa Milivojevic wrote:
> Albert Chin writes:
> > On Wed, Feb 06, 2002 at 07:17:13PM +0200, Sinisa Milivojevic wrote:
> > Yes, SIZEOF_INT is 4:
> > $ grep SIZEOF_INT config.h
> > #define SIZEOF_INT 4
> > $ grep SIZEOF_INT include/my_config.h
> >
On Wed, Feb 06, 2002 at 07:17:13PM +0200, Sinisa Milivojevic wrote:
> Albert Chin writes:
> >
> > $ grep SIZEOF_LONG config.h
> > #define SIZEOF_LONG 8
> > #define SIZEOF_LONG_LONG 8
> >
> > $ grep SIZEOF_LONG incldue/my_config.h
> > #define SIZEOF_LONG 8
> > #define SIZEOF_LONG_LONG 8
> >
> >
Albert Chin writes:
> On Wed, Feb 06, 2002 at 07:17:13PM +0200, Sinisa Milivojevic wrote:
> Yes, SIZEOF_INT is 4:
> $ grep SIZEOF_INT config.h
> #define SIZEOF_INT 4
> $ grep SIZEOF_INT include/my_config.h
> #define SIZEOF_INT 4
>
> However, according to
>
>http://docs.sun.com/ab2/coll.4
Albert Chin writes:
> On Wed, Feb 06, 2002 at 07:17:13PM +0200, Sinisa Milivojevic wrote:
>
> Yes, SIZEOF_INT is 4:
> $ grep SIZEOF_INT config.h
> #define SIZEOF_INT 4
> $ grep SIZEOF_INT include/my_config.h
> #define SIZEOF_INT 4
>
> --
> albert chin ([EMAIL PROTECTED])
>
We shall ha
Albert Chin writes:
>
> $ grep SIZEOF_LONG config.h
> #define SIZEOF_LONG 8
> #define SIZEOF_LONG_LONG 8
>
> $ grep SIZEOF_LONG incldue/my_config.h
> #define SIZEOF_LONG 8
> #define SIZEOF_LONG_LONG 8
>
> --
> albert chin ([EMAIL PROTECTED])
>
If SIZEOF_INT is 4, then it could be our bug.
P
On Wed, Feb 06, 2002 at 06:12:22PM +0200, Sinisa Milivojevic wrote:
> Albert Chin writes:
> >
> > MySQL 3.23.46 was built with the Sun C++ compiler:
> > $ CC -V
> > CC: Sun WorkShop 6 update 2 C++ 5.3 Patch 111685-03 2001/10/19
> >
> > It was built as follows:
> > CC=cc CFLAGS="-mr -Qn -xs
Albert Chin writes:
>
> MySQL 3.23.46 was built with the Sun C++ compiler:
> $ CC -V
> CC: Sun WorkShop 6 update 2 C++ 5.3 Patch 111685-03 2001/10/19
>
> It was built as follows:
> CC=cc CFLAGS="-mr -Qn -xstrconst -xO2 -xtarget=generic -xarch=v9" \
> CXX=CC CXXFLAGS="-xO2 -xtarget=generi
On Wed, Feb 06, 2002 at 02:57:51PM +0200, Sinisa Milivojevic wrote:
> Franklin, Kevin writes:
> > We are running an extremely large instance of mysql version 3.23.46 on
> > Solaris 2.8. We are attempting to use a software version compiled 64 bit
> > and have been experiencing memory related serve
Franklin, Kevin writes:
> We are running an extremely large instance of mysql version 3.23.46 on
> Solaris 2.8. We are attempting to use a software version compiled 64 bit
> and have been experiencing memory related server crashes.
>
> The behavior suggests that we are still unable to use more
Kevin,
The Mysql documentation suggests you use no more than 75% or 80% of
physical memory to allocate for key_buffer. As Heikki and Jeremy
confirmed for me this week, sort_buffer and record_buffer are the ones
that grow per thread - Heikki suggested 1Mb for each of those. Using
swap as RAM (
If you only have 2GB of RAM and you are allocating 5120M (=5GB) for the
key_buffer, you have a problem. That should make your system swap like
mad and everything slow to a crawl. Also, note that some of these config
options may effect per-thread allocation. I don't know which ones off the
26 matches
Mail list logo